Keywords:
Abstract: Novel solutions for the robotic manipulation of deformable objects can enable new real-world tasks in sectors of strategic interest. These include the manufacturing industry, agriculture, or medicine. However, deformable objects are complex to model and require the manipulation system to have high adaptability and the ability to operate under uncertainty. These are important challenges and, although exciting progress has been made over the past years, there remain many open technical questions in this area regarding perception, planning and control. This workshop brings together researchers to showcase recent work, explore connections between the targeted topics of interest and spark discussion about current and future research in the area.

This video is an introduction to the workshop with information on its program, which consists of a series of talks from invited speakers and contributed paper authors.

The workshop's website is: http://commandia.unizar.es/irosworkshop2020/

The workshop's contact email is: romadoworkshop@gmail.com

Keywords:
Abstract: Berk Calli is an Assistant Professor at Worcester Polytechnic Institute (WPI). His research primarily focuses on problems related to robotic manipulation, which is a key functionality largely missing from the current state of the art in robotics for unstructured environments, including homes, modern warehouses, and collaborative manufacturing stations. He develops multi-modal robotic manipulation strategies mainly focusing on the role of vision feedback for coping with uncertainties of unstructured environments. He integrates advanced control methods, active vision framework, machine learning and intelligent mechanical design to provide robust dexterous manipulation capabilities. Berk obtained a PhD degree from Delft University of Technology. Prior to WPI, he worked in the Grab Lab at Yale University on robust within-hand manipulation techniques. He is also one of the founders and the main administrator of the Yale-CMU-Berkeley (YCB) object set project, which facilitates benchmarking efforts worldwide for robotic manipulation. His current focus is to utilize robots in sustainability projects (e.g. sorting for recycling) by solving complicated manipulation problems therein.

Keywords:
Abstract: Danica Kragic is a Professor at the School of Computer Science and Communication at the Royal Institute of Technology, KTH. She received MSc in Mechanical Engineering from the Technical University of Rijeka, Croatia in 1995 and PhD in Computer Science from KTH in 2001. She has been a visiting researcher at Columbia University, Johns Hopkins University and INRIA Rennes. She is the Director of the Centre for Autonomous Systems. Danica received the 2007 IEEE Robotics and Automation Society Early Academic Career Award. She is a member of the Royal Swedish Academy of Sciences, Royal Swedish Academy of Engineering Sciences and Young Academy of Sweden. She holds a Honorary Doctorate from the Lappeenranta University of Technology. She chaired IEEE RAS Technical Committee on Computer and Robot Vision and served as an IEEE RAS AdCom member. Her research is in the area of robotics, computer vision and machine learning. In 2012, she received an ERC Starting Grant. Her research is supported by the EU, Knut and Alice Wallenberg Foundation, Swedish Foundation for Strategic Research and Swedish Research Council. She is an IEEE Fellow.

Ioanna Mitsioni is a PhD student in the division of Robotics, Perception and Learning at KTH, under the supervision of Danica Kragic and the co-supervision of Yiannis Karayiannidis. She received her Diploma of Electrical and Computer Engineering from the Aristotle University of Thessaloniki. Currently, she is investigating how Deep Learning can enhance more traditional Interaction Control approaches in contact-rich manipulation tasks.

Keywords:
Abstract: David Navarro-Alarcon is an Assistant Professor of Robotics at the Department of Mechanical Engineering of The Hong Kong Polytechnic University, and the Principal Investigator of the Robotics and Machine Intelligence Laboratory. Before joining PolyU, he worked at CUHK from February 2014 to June 2017, first as a Postdoctoral Fellow in soft object manipulation, and then as a Research Assistant Professor at the T Stone Robotics Institute. He received his PhD degree in mechanical and automation engineering from CUHK in January 2014. His research interests are mostly in the area of robotics, machine intelligence, adaptive systems, and control engineering. His work has been published in the top academic journals on robotics such as the International Journal of Robotics Research (IJRR) and the IEEE Transactions on Robotics (T-RO). He has participated in several ITF, RGC, and industry-sponsored robotics projects. He is an Associate Editor of the journal Frontiers in Robotics and AI, Specialty Section on Soft Robotics, Associate Editor of the IEEE Int. Conf. Robotics and Automation 2019, Guest Editor of the Journal of Robotics and Autonomous Systems, Organiser of the IROS 2018 Special Session on Methods and Algorithms for Automatic Manipulation of Deformable Objects, and Co-Organiser of the IROS Workshop Series on “Multimodal Sensor-Based Robot Control”. He is a Member of IEEE, the Robotics and Automation Society, and the Computational Intelligence Society.

Keywords:
Abstract: Raul Suarez received the Electronic Engineer (Hons.) degree from the National University of San Juan, San Juan, Argentina, in 1984, and the Ph.D. (cum laude) degree from the Universitat Politecnica de Catalunya (UPC), Barcelona, Spain, in 1993. He is a Research Supervisor with the Institute of Industrial and Control Engineering (IOC), UPC. He was responsible for the research line on process control at the IOC from 1998 to 2003, the Deputy Director of the IOC from 2003 to 2009, and its Director from 2009 to 2016 (re-elected in 2012). Dr. Suarez is the coordinator of the PhD Program "Automatic Control, Robotics and Computer Vision" at UPC. He is the president of the IEEE RAS Spanish Chapter. His research activities have been mainly focused on the area of robotized assembly, especially in the field of fine motion planning in the presence of uncertainty, and more recently in the field of robotized grasping and manipulation. His current research interests include grasping and manipulation, mechanical hands, fixturing, assembly, task planning, telemanipulation, and manufacturing automation.

Keywords:
Abstract: Professor Kaspar Althoefer is an electronics engineer, leading research on Robotics at Queen Mary University of London. After graduating with a degree in Electronic Engineering from the University of Technology Aachen, Germany, and obtaining a PhD in Robot Motion Planning from Kings College London, he joined the Kings Robotics Group in 1996 as a Lecturer. Made a Senior Lecturer in 2006, he was promoted to Reader and Professor in 2009 and 2011, respectively. In April 2016, he joined Queen Mary as full Professor of Robotics Engineering. He leads competitively-funded research on soft robotics, intelligent micro-sensing systems and interaction dynamics modelling with applications in minimally invasive surgery, assistive technologies and human-robot interaction, having acquired more than £5.5M as PI from national/international funding bodies and successfully completed 21 PhD projects. Prof Althoefer has authored/co-authored more than 200 peer-reviewed papers. The majority of his journal papers (over 60%) are in the top journals of the field. He is named inventor on five patent applications. He is currently heading a team of five postdoctoral Research Assistants / Fellows and six PhD students. His current research interests are in the areas of robot autonomy, soft robotics, modelling of tool-environment interaction dynamics, sensing and neuro-fuzzy-based sensor signal classification with applications in robot-assisted minimally invasive surgery, rehabilitation, assistive technologies and human-robot interactions in the manufacturing environment.

Keywords:
Abstract: Dr. Javier Alonso-Mora is an Associate Professor at the Cognitive Robotics department of the Delft University of Technology, where he leads the Autonomous Multi-robots Laboratory. He is a Principal Investigator at the Amsterdam Institute for Advanced Metropolitan Solutions (AMS) and co-founder of Routable AI. He is actively involved in the Delft robotics ecosystem, including the Robotics Institute, the Transportation Institute and Robovalley. Before joining TU Delft, Dr. Alonso-Mora was a Postdoctoral Associate at the Computer Science and Artificial Intelligence Lab (CSAIL) of the Massachusetts Institute of Technology (MIT). He received his Ph.D. degree in robotics from ETH Zurich, where he worked in the Autonomous Systems Lab, and in partnership with Disney Research Zurich. Dr. Alonso-Mora holds a Diploma in Engineering and a Diploma in Mathematics from the Technical University of Barcelona (UPC), where he was part of the Interdisciplinary Higher Education Centre (CFIS) and the Robotics Institute (IRI). His main research interest is in navigation, motion planning and control of autonomous mobile robots, with a special emphasis on multi-robot systems, on-demand transportation and robots that interact with other robots and humans in dynamic and uncertain environments. He has published over 60 papers in premier venues and his work has appeared in major media outlets. He is the recipient of multiple prizes and grants, including the ICRA Best Paper Award on Multi-robot Systems (2019), an Amazon Research Award (2019) and a talent scheme VENI award from the Netherlands Organisation for Scientific Research (2017).

Keywords:
Abstract: Veronique Perdereau is a Professor of Robotics at the Faculty of Science and Engineering at Sorbonne University. She has been responsible for several master specialties from 2004 to 2017 and on the academic commission of the Sorbonne University Institute of Health Engineering from 2014 to 2017. She has been heavily invested in the EIT KIC Health since 2015, where she is director of education for France and head of masters and doctorates for Europe. She conducts research in robotics at the Institute of Intelligent Systems and Robotics (ISIR, a shared Sorbonne University/CNRS institute). She coordinated the European project HANDLE from 2009 to 2013 for which she received the Stars of Europe Award in 2014. Her research interests include hybrid force/position control of robot manipulators, multi-robot cooperation, modelling and control of a multi-fingered robotic hand, planning and control of in-hand dexterous manipulation, grasping force optimization and grasping strategies for unknown objects.

Keywords:
Abstract: The present paper details our manipulation planning-based approach to tethered tool manipulation for dual-armed robots. Our approach is aimed at preventing entanglements and cable collisions during tethered tool manipulation tasks that require tool re-grasping and handover. Our framework implements constraints and bi-manual simultaneous tool-cable manipulation to avoid excess bending and cable collisions. Simulations and real-world experiments help validate our approaches.

Keywords:
Abstract: Through the use of tactile perception, a manipulator can estimate the stability of its grip, among others. However, tactile sensors are only activated upon contact. In contrast, humans can estimate the feeling of touching an object from its visual appearance. Providing robots with this ability to generate tactile perception from vision is desirable to achieve autonomy. To accomplish this, we propose using a Generative Adversarial Network. Our system learns to generate tactile responses using as stimulus a visual representation of the object and target grasping data. Since collecting labeled samples of robotic tactile responses consumes hardware resources and time, we apply semi-supervised techniques. For this work, we collected 4000 samples with 4 deformable items and experiment with 4 tactile modalities.

Keywords:
Abstract: Deformable object manipulation tasks have long been regarded as challenging robotic problems. However, until recently, very little work had been done on the subject, with most robotic manipulation methods being developed for rigid objects. As machine learning methods are becoming more powerful, there are new model-free strategies to explore for these objects, which are notoriously hard to model. This paper focuses on shape control problems for Deformable Linear Objects (DLOs). Despite being one of the most researched classes of DLOs in terms of geometry, no other paper has focused on materials with elastoplastic properties. Therefore, a novel shape control task, requiring permanent plastic deformation is implemented in a simulation environment. Reinforcement Learning methods are used to learn a continuous control policy. To that end, a discrete curvature measure is used as a low-dimensional state representation and as part of an intuitive reward function. Finally, three state-of-the-art actor-critic algorithms are compared on the proposed environment and successfully achieve the goal shape.

Keywords:
Abstract: Tactile feedback during cloth manipulation could be crucial in addressing the huge challenges involved in closing the loop during execution, complementing vision. However, up to our knowledge, tactile sensing has only been successfully used in cloth manipulation to classify type of fabrics, detect how many layers were grasped, and estimate the grasping force. In this work, we want to explore its potential to also provide information about whether the task is executed as expected. Two types of experiments are performed, in which a robot carries out 4 simple tasks, all involving a single finger manipulating a flat cloth on a table. Firstly, we analyze the sensor's signals once the cloth manipulation has finished using Dynamic Time Warping (DTW) to see if they are informative enough to classify the tasks. Our results show that tactile feedback depends highly on the type of manipulated fabric. Secondly, we analyze the tactile feedback during the manipulation of the cloth using a recurrent neural network (RNN). For each sensor measurement, the RNN recognizes if the finger slides over the cloth, pulls it, flattens a fold in it, or if it's about to lose contact, with 95.7% accuracy. These are promising results that show how tactile sensing has the potential of providing crucial information that would be very difficult to obtain with vision only.

Keywords:
Abstract: The intelligent handling of deformable objects by robotic manipulators has enabled the automation of various delicate or labour-intensive tasks, but still faces many challenges when it comes to more complex situations. In particular, the lack of a general framework to manage deformation makes it difficult to manipulate unknown objects, especially in the 3D case. This work aims to provide a path forward to solve some of the challenges pertaining to the robotic manipulation of deformable objects and integrate these solutions in a flexible system able to manipulate various deformable objects in a task-independent way.

Keywords:
Abstract: Currently, perception and manipulation of dynamic objects represent an open research problem. In this paper, we show a proof of concept of a multi-camera robotic setup which is intended to perform coverage of dynamic objects. The system includes a set of RGB-D cameras, which are positioned and oriented to cover the object's contour as required in terms of visibility. An algorithm of a previous study allows us to minimize and configure the cameras so that collisions and occlusions are avoided. We test the validity of the platform with the Robot Operating System (ROS) in simulations with the software Gazebo and in real experiments with Intel RealSense modules.

Keywords:
Abstract: Estimating the state of deformable objects is vital for manipulation, while it is also challenging due to high degrees of freedom and the nonlinearity of the dynamics model. In order to achieve robust state estimation, we propose a novel framework shown in Fig. 1, which includes point cloud recovery, node registration, feedback linearization controller, and dynamical simulation modules. Compared with previous works, the point cloud recovery step is able to robustly provide a complete point cloud of the object even under massive occlusions. In addition, the feedback linearization controller is able to stabilize the rope tracking procedure via applying a control law that first cancels higher-order terms in the dynamic equation and then uses an additional PD control law to control the remaining linear dynamics. Experimental results validate the effectiveness and robustness of the proposed framework. The experimental videos can be found at [1].

Keywords:
Abstract: The video presents recent work on deformable object manipulation carried out at SIGMA Clermont / Institut Pascal. Several works are included on the topics of monocular shape tracking and shape servoing in 3D for isometrically deforming objects, dual arm coordination for shape control of linear objects, multirobot coordination based on deformation, and manipulation of deformable objects based on force sensing and finite element modeling.

Participants:

- Miguel Aranda, Postdoctoral researcher at SIGMA Clermont / Institut Pascal

- Mohammad Reza Shetab, PhD student at UCA / Institut Pascal

- Rohit Chandra, Postdoctoral researcher at SIGMA Clermont / Institut Pascal

Keywords:
Abstract: Zhanat Kappassov is an Assistant Professor in the Department of Robotics and Mechatronics at Nazarbayev University. He graduated as Engineer in Radioelectronics from Tomsk State University of Control Systems and Radioelectronics (TUSUR). He obtained a PhD degree in Robotics from Paris 6 University (UPMC) in 2016. He received a Best PhD Thesis award from GDR Robotique France (2nd place). He was a visiting researcher at SIGMA Clermont / Institut Pascal and at the Key State Institute of Robotics. He founded and is the head of the Tactile Robotics Laboratory at Nazarbayev University. His research work is focused on investigation and development of new control algorithms for dexterous autonomous manipulation with robot hands equipped with tactile sensing modalities. Topics of current interest in his research include soft object recognition, tactile servoing of flexible objects, and slip detection.

Keywords:
Abstract: This video presents the project COMMANDIA, which supports the organization of this workshop, and shows some of the project's recent results.

COMMANDIA (SOE2/P1/F0638) is a project of the Interreg Sudoe Programme cofinanced by the European Regional Development Fund (ERDF). The acronym COMMANDIA stands for Collaborative Robotic Mobile Manipulation of Deformable Objects in Industrial Applications.

The main goal of this project is to improve the competitiveness and work conditions of industries where deformable objects have to be manipulated directly by human operators in order to control their shapes during production. This project proposes to develop and disseminate a set of new techniques and technologies in robotic perception, control and planning for dynamically handling deformable objects with collaborative mobile manipulators.

The project consortium consists of:

- SIGMA Clermont, France

- INESCOP, Spain

- Universidad de Zaragoza, Spain

- Universidad de Alicante, Spain

- Universidade de Coimbra, Portugal

- ADIV (Associate Partner), France

Project website: http://commandia.unizar.es/

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT Soft robotics and its subfields are increasingly becoming relevant in the academic discipline, but lagging behind in its transferability to the industry. This is partly because of the nascency of the field and the lack of knowledge transfer amongst the two. By approaching the modelling-and-control of soft robots from an application-oriented viewpoint, novel problems and technologies can be discovered. This workshop attempts to bring together experts from various research and application domains working towards real-world control problems to identify the industrial requirements in soft-bodied modelling, state estimation, model identification, path planning, control and design optimization; striving towards immediate applications in medicine, entertainment, industrial inspection, space robotics, etc. This workshop is an attempt to help steer the broad field of modelling-and-control of soft robots towards immediate commercial applications.

Keywords:
Abstract: ABSTRACT Model-based optimal control of large-scale soft robots is difficult for many reasons. Two of the main limitations are the difficulty of modeling such systems, and the tractability of the optimal control problem given the large number of degrees of freedom. First principle-based models can be difficult to derive and validate for soft robots. In addition, if the form of the soft robot platform changes, the method to model the platform may also need to change drastically. While any attempt to include additional states to improve the model accuracy also increases the complexity and decreases the tractability of the optimal control problem. In this talk, with an eye towards enabling multiple soft robots to coordinate motion for co-manipulation tasks, we present two approaches to solve these problems. First, we present learned models and DNN architectures that enable closed-loop optimal control of large-scale soft robots. Second, we present a sampling-based method to find high-quality solutions in a model predictive control formulation. Finally, we also present the limitations of this work and needed changes to reach the full potential of model-based control for soft robots.

Keywords:
Abstract: ABSTRACT

This talk will review modeling for some new compliant robotic structures. First, a new actuation concept - concentric precurved bellows - will be described and demonstrated, including simple constant curvature models, and some recent constant-curvature modeling discussions in the literature will be highlighted. Next, general kinetostatic problem formulations will be categorized and discussed. Recent work shows that model-based estimation of continuous load distributions along soft robots is feasible and useful. Finally, the talk will highlight real-time dynamic simulation work for all types of continuum and soft robots, including concentric-tube robot snap-through behavior.

Keywords:
Abstract: ABSTRACT

In medical interventions, there are advantages to using soft materials for both implanted devices and therapeutic robots. This talk explores their use in treating blockages of the airways for which two fundamental procedures are stenting and balloon dilation. An in vivo molding technique will be presented which enables patients to receive a stent tailored to the size and shape of their airway. In this approach, a soft stent with a liquid UV-curable core is balloon expanded so that it conforms to the airway walls and is then quickly cured to this shape. A soft robot will also be presented that can navigate through the airways serving as both an endoscope and balloon dilator. Combining these functions which are normally performed using two instruments avoids the visual occlusion problems encountered during balloon dilations. Fabricated using inexpensive materials and processes and powered by pressurized air and vacuum sources that are found in standard medical operating rooms, the design lends itself to use as a disposable.

Keywords:
Abstract: ABSTRACT

Unlike their rigid cousins, soft robots conform naturally to the environment. This makes them inherently safe and robust to uncertainty, and therefore promising for applications requiring close interaction with people, such as elderly care, medicine, and rehabilitation. However, these benefits are not for free. The same softness that makes soft robots so exciting also renders them unintuitive to design, and difficult to control. For example, to answer the seemingly simple question of calculating a soft robot's inverse kinematics, we must account for highly nonlinear soft body physics and contact dynamics. A fundamental challenge in soft robotics therefore is to make it easy to predict and control soft robot motion. We propose a differentiable soft robot simulator based on the Finite Element Method. I will show how to leverage this simulator to answer not only the classic questions of forward kinematics and inverse kinematics for soft robots, but also to start answering the harder questions of soft locomotion, soft manipulation, and more.

Keywords:
Abstract: ABSTRACT

Soft, growing robots achieve locomotion by material extending from their tip. They are inherently compliant and can safely navigate through tight turns and highly constrained environments that prove challenging for traditional robots. Despite the potential of these robots, there remain a number of challenges to practical implementation, particularly related to shape estimation and control, that must be addressed. In this talk, we present a low-cost, wireless, permanent magnet-based method for localizing the tip of these robots. A permanent magnet is placed at the robot tip, and an array of magneto-inductive sensors is used to measure the change in magnetic field as the robot moves through its workspace. We develop an approach to localization that combines analytical and machine learning techniques and show that it outperforms existing methods. In addition, we explore the potential to use this localization method for closed-loop control of a steerable growing robot.

Keywords:
Abstract: ABSTRACT

The talk will discuss modeling and control of soft robots with emphasis on applications as steerable implantable devices. Simplified approaches for modeling and control of steerable slender continuum devices made of elastomers and their possible applications for cochlear implant surgery will be used to motivate this talk. The talk will also discuss how some of these approaches can inform indirect sensing and steering of electrode arrays for cochlear implant surgery and other applications. Recent extensions of these approaches to contact detection using pneumatic braided muscle actuators will also be discussed along with challenges and suggestions for future research directions.

Keywords:
Abstract: ABSTRACT As the field of Soft Robotics matures, the complexity of the tackled problems will inevitably increase, and computational tools for simulation and optimization will become key aspects of soft robot design and control. In this talk, I will highlight how computation paves the way toward industrial-grade robots that are lightweight and inexpensive, yet functional. Specifically, I will discuss how we can leverage ​differentiable simulation​ to (1) accurately ​characterize​ soft robotic materials, (3) computationally ​control​ very soft robotic systems while suppressing visible mechanical oscillations, and (3) automate the ​design​ of proprioceptive soft robots with desired functionality.

Keywords:
Abstract: ABSTRACT

The talk discusses issues in the control of continuum robots. Beginning with an introduction to core continuum robot modules, or sections, examples of alternative types of controllers proposed for them are presented and discussed. Questions regarding which type and complexity of controllers are most appropriate for compliant continuum robotic structures are introduced and addressed.

Keywords:
Abstract: ABSTRACT

Dummy video

Keywords:
Abstract: Ambulation with conventional prostheses is slower, less stable, and less efficient than able-bodied ambulation, causing reduced mobility and quality of life. Robotic powered prostheses have the potential to close the gap between the performance of existing lower limb prostheses and human legs. In contrast to conventional passive prostheses, powered prostheses can provide biomechanically accurate kinetics and kinematics including during activities that require energy injection. Powered prostheses have evolved from devices tethered to a power supply or computer to devices that have onboard electronics and batteries. Powered prosthetic devices are now able to assist during walking, ambulation on stairs and ramps, and sit-stand transitions. However, there are significant challenges that the field needs to address for powered prostheses to fully realize their potential. Powered prostheses are often heavy, bulky, noisy, and fragile compared to conventional prostheses. Even the most advanced controllers cannot match the agility of the human body. Powered prostheses have made many advances but there are still challenges ahead of us.

The goal of this workshop is to discuss the state of the art and open challenges in powered prosthetics. There will be talks from top researchers in the field to discuss their current and future approaches to the design and control of powered prostheses. Additionally, there will be demonstrations of powered prosthetics from leaders in the field to provide hands-on experience with leading technology. This will be the first time that the top lower limb robotic prostheses will be in the same room. Lastly, there will be a poster session for young and new researchers in the field to present their current work. This workshop will allow established researchers, as well as those new to the field of robotic prostheses, to experience firsthand where the field is today and what challenges lie ahead.

Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics.

Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics.

Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: wearable robotics; rehabilitation robotics; powered prosthesis; robotic leg; design; control; actuation; gait; biomechanics. Workshop website: https://belab.mech.utah.edu/iros2020/

Keywords:
Abstract: Failures are an essential part of every innovation and development process. The road that leads to most successful human undertakings is usually paved by a large amount of mistakes, failed experiments and incorrect assumptions. This reality is even more remarkable in scientific research, since making new discoveries can sometimes be seen as a leap in the dark, where the methods to be used and even the outcomes to achieve typically cannot be known a priori, which makes trial-and-error approaches a constituent part of the scientific method. Despite their importance, failures are often overlooked in the research process. This might be due to the research community’s unwillingness to admit and discuss failure, as well as problems that arise from the refusal of publishing negative results. This produces an environment where little time is devoted to reasoning and reporting unsuccessful trials. The complexity of robotic systems presents a number of challenges in identifying failure cases and addressing their causes. First because the number of different components present in a robotic system translates into a multitude of possible failures (e.g. the vision might fail because of lighting, the IMU because of magnetic interference, etc.). Troubleshooting these complex systems becomes a difficult and time-consuming task. Besides, there are usually a multitude of single points of failure, which are the components of a system where a failure will stop the entire system from working.

This workshop aims to serve as a discussion forum where the participants will have the opportunity to:

Learn from the experience of other participants, finding explanations for problems that they encounter in their work, as well as possible solutions. They may also avoid repeating the same mistakes, when an approach is proven to be ineffective. Share their own unsuccessful ideas, hypotheses, and experiments, gaining some insights and possible solutions from participants working on similar problems. Network with other participants, joining efforts to address common problems or employing one’s know-how to solve another’s persistent failure case. Create new research problems, based on the discussions during the workshop, finding the underlying causes and solutions for common failure cases.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: In this talk, Prof. Jeannette Bohg reviews her past work and presents how past failures and limitations of her methods informed her views on what are the crucial principles of robot manipulation that can avoid and recover from failures.

Keywords:
Abstract: Why Robots Fail to Grasp? Juxi Leitner

Keywords:
Abstract: Why Robots Fail to Grasp? Ken Goldberg - Addressing Persistent Failures in Warehouse Grasping

Keywords:
Abstract: In this talk, Lorenzo Natale shares his experience of grasping objects with the iCub humanoid robot.

Keywords:
Abstract: Why Robots Fail to Grasp? Rob Platt

Keywords:
Abstract: In this talk Andrew Sharp demonstrates some manipulation abilities of the NASA robonaut humanoid robot, and some failures and limitations that have been identified.

Keywords:
Abstract: In this talk, Toni Oliver describes how the team at Shadow Robot Company has been addressing the challenges and finding solutions some of the failures they have encountered when executing grasps, including while teleoperating.

Keywords:
Abstract: Why Robots Fail to Grasp? Dieter Fox

Keywords:
Abstract: Why Robots Fail to Grasp? Discussion Part 1

Keywords:
Abstract: Why Robots Fail to Grasp? Discussion Part 2

Keywords:
Abstract: Why Robots Fail to Grasp? Discussion Part 3

Keywords:
Abstract: Why Robots Fail to Grasp? Discussion Part 4

Keywords:
Abstract: At the moment, robots are mainly employed for industrial applications where accurate positioning and precise tracking are needed. This led to the development of robotic systems that were stiff and heavy. As such, they can only operate in highly structured environments void of any physical interaction with humans. This limitation has motivated the robotic research community to develop novel theoretical and technological solutions to allow robots to operate amongst and with humans and to safely move in unknown and unstructured environments. To guarantee human safety, the stability and integrity of the robot must be preserved during physical interaction. A common approach is to introduce a certain degree of compliance to the robot, which allows it to account for external disturbances. Compliance can be embedded in robots either passively or actively. For instance, passive visco-elastic elements can be integrated into the robot design. Alternatively, a controller can shape the mechanical impedance of the robot (e.g., stiffness, damping, inertia). This means robot behavior can be planned not only in the kinematic domain (i.e. motion planning) but also in terms of its dynamic response. These approaches have been proven effective in managing physical interaction with its surrounding environment and humans. Still, knowing what the desired robot compliance should be for a given scenario is an open problem. The primary goal of this full-day workshop is to critically discuss the current and new approaches used to identify the proper robot compliance for a given task, interaction, level of uncertainty, etc. We invited speakers to discuss how the selection of impedance parameters can be formulated as an optimization problem, as well as speakers who use learning strategies to understand and generalize task-specific impedance regulation. We have also invited speakers from the human motor control community to discuss how humans are able to robustly manage physical interaction by modulating their mechanical impedance.

Keywords:
Abstract: Humans are able to modulate their mechanical impedance depending on the task they are performing. This capability has been transferred also to robotic systems, either via software or using hardware solutions. Impedance modulation in robotics found fertile application to solve multiple problems ranging from manipulation to rehabilitation, from locomotion to prosthetics. In this talk, I will first review how humans exploit their capabilities in changing and adapting their limb impedance. Then, I will discuss different solutions to mirror this behavior in machines. Finally, I will present and results in planning, controlling and learning robot impedance to solve a variety of practical issues.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: Human muscles are tension actuators that lie outside the bones. Stabilizing that endo-skeletal anatomy requires muscle stiffness to increase at least in proportion to muscle tension. That, in turn, enables mechanical impedance modulation by antagonist muscle co-activation, especially important for the most rapid events, before feedback control has time to respond. But some tasks may require low impedance—e.g. complying with a constraint. If force production is also needed, low impedance appears to be unavailable. A ‘work-around’ may be found in the kinematic complexity of the skeleton. Pose modulates all aspects of mechanical impedance—stiffness, damping, inertia, and more. Measurements of the stiffness of the wrist and ankle—which support our major points of interaction with the world—show that both are highly anisotropic. Wrist stiffness is extremely low in the direction of the ‘dart-throwers’ motion, and it appears that muscle action does little to increase it. Thus a key part of motor skill acquisition may be learning the most favorable pose. But controlling pose is complicated by the skeleton’s very many degrees-of-freedom. Fortunately, redundant degrees of freedom may be managed without inverting kinematic equations by exploiting the compositionality of mechanical impedance.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: In this presentation, I will argue that the main role of impedance modulation is to trade-off the need to reject external disturbances and the unavoidable uncertainty in contact locations. Stiffness helps increase accuracy but can lead to catastrophic results if contact with the environment is made unexpectedly. On the other hand, compliance and damping helps create safe contacts but can result in reduced tracking performance. To support this claim, I will present an algorithm to efficiently compute optimal impedance schedules that explicitly trades-off contact uncertainty and external disturbances and show that this can significantly increase robustness during locomotion on unknown terrains. Then I will discuss a reinforcement learning approach that explicitly incorporates structure in the learned control policy to enforce explicit impedance learning. In particular, I will show that the method can create behaviors that are robust to contact uncertainty (location, stiffness, friction) and transfer on real robots. All results will be demonstrated on our novel open-source quadruped robot capable of a large range of impedance modulation.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: Physical interaction of human and robot (pHRI) in shared environments and joint tasks poses many challenges. Collaborative carrying tasks, which rely on complementary advantages of human and robot, cannot be accomplished individually by a single human or robot. In this talk, we will introduce a hybrid framework using visual and force sensing for human-robot co-carrying tasks. Further, an adaptive neural network impedance-based control strategy will be presented. Motion synchronization can be achieved and this approach yields a stable and efficient interaction behavior between human and robot, decreases human control effort and avoids interference to human during the interaction. Finally, co-carrying tasks in simulations and experiments will be illustrated to verify our proposed method.

Keywords:
Abstract: At the moment, robots are mainly employed for industrial applications where accurate positioning and precise tracking are needed. This led to the development of robotic systems that were stiff and heavy. As such, they can only operate in highly structured environments void of any physical interaction with humans. This limitation has motivated the robotic research community to develop novel theoretical and technological solutions to allow robots to operate amongst and with humans and to safely move in unknown and unstructured environments. To guarantee human safety, the stability and integrity of the robot must be preserved during physical interaction. A common approach is to introduce a certain degree of compliance to the robot, which allows it to account for external disturbances. Compliance can be embedded in robots either passively or actively. For instance, passive visco-elastic elements can be integrated into the robot design. Alternatively, a controller can shape the mechanical impedance of the robot (e.g., stiffness, damping, inertia). This means robot behavior can be planned not only in the kinematic domain (i.e. motion planning) but also in terms of its dynamic response. These approaches have been proven effective in managing physical interaction with its surrounding environment and humans. Still, knowing what the desired robot compliance should be for a given scenario is an open problem. The primary goal of this full-day workshop is to critically discuss the current and new approaches used to identify the proper robot compliance for a given task, interaction, level of uncertainty, etc. We invited speakers to discuss how the selection of impedance parameters can be formulated as an optimization problem, as well as speakers who use learning strategies to understand and generalize task-specific impedance regulation. We have also invited speakers from the human motor control community to discuss how humans are able to robustly manage physical interaction by modulating their mechanical impedance.

Keywords:
Abstract: ABSTRACTAt the moment, robots are mainly employed for industrial applications where accurate positioning and precise tracking are needed. This led to the development of robotic systems that were stiff and heavy. As such, they can only operate in highly structured environments void of any physical interaction with humans. This limitation has motivated the robotic research community to develop novel theoretical and technological solutions to allow robots to operate amongst and with humans and to safely move in unknown and unstructured environments. To guarantee human safety, the stability and integrity of the robot must be preserved during physical interaction. A common approach is to introduce a certain degree of compliance to the robot, which allows it to account for external disturbances. Compliance can be embedded in robots either passively or actively. For instance, passive visco-elastic elements can be integrated into the robot design. Alternatively, a controller can shape the mechanical impedance of the robot (e.g., stiffness, damping, inertia). This means robot behavior can be planned not only in the kinematic domain (i.e. motion planning) but also in terms of its dynamic response. These approaches have been proven effective in managing physical interaction with its surrounding environment and humans. Still, knowing what the desired robot compliance should be for a given scenario is an open problem. The primary goal of this full-day workshop is to critically discuss the current and new approaches used to identify the proper robot compliance for a given task, interaction, level of uncertainty, etc. We invited speakers to discuss how the selection of impedance parameters can be formulated as an optimization problem, as well as speakers who use learning strategies to understand and generalize task-specific impedance regulation. We have also invited speakers from the human motor control community to discuss how humans are able to robustly manage physical interaction by modulating their mechanical impedance.

Keywords:
Abstract: At the moment, robots are mainly employed for industrial applications where accurate positioning and precise tracking are needed. This led to the development of robotic systems that were stiff and heavy. As such, they can only operate in highly structured environments void of any physical interaction with humans. This limitation has motivated the robotic research community to develop novel theoretical and technological solutions to allow robots to operate amongst and with humans and to safely move in unknown and unstructured environments. To guarantee human safety, the stability and integrity of the robot must be preserved during physical interaction. A common approach is to introduce a certain degree of compliance to the robot, which allows it to account for external disturbances. Compliance can be embedded in robots either passively or actively. For instance, passive visco-elastic elements can be integrated into the robot design. Alternatively, a controller can shape the mechanical impedance of the robot (e.g., stiffness, damping, inertia). This means robot behavior can be planned not only in the kinematic domain (i.e. motion planning) but also in terms of its dynamic response. These approaches have been proven effective in managing physical interaction with its surrounding environment and humans. Still, knowing what the desired robot compliance should be for a given scenario is an open problem. The primary goal of this full-day workshop is to critically discuss the current and new approaches used to identify the proper robot compliance for a given task, interaction, level of uncertainty, etc. We invited speakers to discuss how the selection of impedance parameters can be formulated as an optimization problem, as well as speakers who use learning strategies to understand and generalize task-specific impedance regulation. We have also invited speakers from the human motor control community to discuss how humans are able to robustly manage physical interaction by modulating their mechanical impedance.

Keywords:
Abstract: At the moment, robots are mainly employed for industrial applications where accurate positioning and precise tracking are needed. This led to the development of robotic systems that were stiff and heavy. As such, they can only operate in highly structured environments void of any physical interaction with humans. This limitation has motivated the robotic research community to develop novel theoretical and technological solutions to allow robots to operate amongst and with humans and to safely move in unknown and unstructured environments. To guarantee human safety, the stability and integrity of the robot must be preserved during physical interaction. A common approach is to introduce a certain degree of compliance to the robot, which allows it to account for external disturbances. Compliance can be embedded in robots either passively or actively. For instance, passive visco-elastic elements can be integrated into the robot design. Alternatively, a controller can shape the mechanical impedance of the robot (e.g., stiffness, damping, inertia). This means robot behavior can be planned not only in the kinematic domain (i.e. motion planning) but also in terms of its dynamic response. These approaches have been proven effective in managing physical interaction with its surrounding environment and humans. Still, knowing what the desired robot compliance should be for a given scenario is an open problem. The primary goal of this full-day workshop is to critically discuss the current and new approaches used to identify the proper robot compliance for a given task, interaction, level of uncertainty, etc. We invited speakers to discuss how the selection of impedance parameters can be formulated as an optimization problem, as well as speakers who use learning strategies to understand and generalize task-specific impedance regulation. We have also invited speakers from the human motor control community to discuss how humans are able to robustly manage physical interaction by modulating their mechanical impedance.

Keywords:
Abstract: In order to provide help to elderly people with cognitive impairment, a robot assistant must understand two things, what the person wants to perform, and if the person requires cognitive help to perform it. Approaches from Plan Recognition can be options to address both of these problems. In this article, we propose to use the particle filter approach of plan recognition over plan libraries to detect anomalies in activities. These anomalies can be used to detect cognitive distress. We use the Expected Next Observation compute by this approach to determine how likely the observation of an action is. If this likelihood is beneath a threshold we consider that this is an anomaly. Empirical results on simulated problems show that this approach is effective in case of perfect observability or low level of noisy observations, but the approach is still too sensitive to noisy observation for real-life applications on mobile robots.

Keywords:
Abstract: In this paper, we propose a method for building sophisticated costmaps based on spatio-temporal world information. This enables mobile robots to gain adequate movement behaviors in unconstrained environments, such as hospitals and other healthcare facilities. Here robots needs to be extra cautious as they encounter some of the most fragile members of society, who are likely unfamiliar with robots. We show that our method improves standard navigation methods by keeping a larger distance to people, by predicting people’s future trajectories and acting appropriately to these. Also, we use f-formations to group people who are likely interacting socially, thereby enabling the robot to avoid interrupting them.

Keywords:
Abstract: This paper describes the system architecture for an autonomous interactive drink serving robot developed for use in public spaces and at social events. The proposed system design focuses on finding the balance between technological readiness and social readiness levels thus enabling a technology that can be deployed in real-world environments along with social acceptance. We describe different components required for designing such a system and discuss both their technical feasibility and social acceptance aspects. We also present a behavior tree based software architecture for integrating these components, which results in modular design and promotes their re-usability.

Keywords:
Abstract: In nursing, nurses are burdened with many physically demanding tasks. Especially in ambulatory care, nurses are alone with their patients and are under great time pressure. Also, a suitable nursing aid is not always available or at hand for the patient. A flexible robotic solution can help here. In this paper we consider the positioning of the patient on the side. In particular, the patient mobilized on the side should be held in this position. We analyze how a nurse would perform this activity and how a robot can be used for this. For this we use the Whole Arm Manipulation (WAM). WAM is a suitable means of ensuring that a manipulator can apply the necessary forces and reduce the risk of injury to the patient. We developed an algorithm for WAM based on the kinematics of the manipulator and the geometry of the scene. For simplification an elliptical cylinder is used as a human model. This is derived from the real dimensions of the patient. The results show that the algorithm delivers valid results in different constellations and is robust against small deviations from assumptions made for development.

Keywords:
Abstract: Robots have the potential to augment the care provided by humans in enabling older adults to lead a healthy and independent lifestyle. However, to be ubiquitous, robots must build trust with older adults and provide a safe interaction environment, while still maintaining functionality. Current robots built using rigid links, actuators, and sensors provide accurate movements and can handle large loads, but the benefits of these features are compromised by cost as well as retrofitting safety features, and a complex control policy. On the other hand, the new field of soft robots is inherently safe but suffers from reduced load bearing ability, speed, and accuracy. This work proposes a new concept called morphological switching that can overcome the limitations of external rigid and soft robot configurations. Morphological switching is the ability of the robot to morph its structure from a rigid configuration to a soft continuum configuration. We present the challenges to translate the concept of morphological switching for adult care and support tasks.

Keywords:
Abstract: Social robots have the potential to help the elderly through activities of daily living provided they can obtain enough situational awareness and build a rapport with their human partner. In the past, context-aware robots were able to gather information about the elderly’s activities and the objects in the environments but lacked a natural conversational interface and a long-term memory representation for visual perception to support extended interactions. This paper provides proof of concept demonstrations of two use-cases of an autonomous mobile robot helping the elderly in the home. In the first, the robot helps the elder, especially with dementia, to find household objects, that they might have a hard time remembering. In the second, the robot engages in a dialog to guide the person through cooking a recipe. We are looking forward to participants in this workshop giving feedback on these use cases regarding the suitability of using a robot for such tasks and the potential failures of the proposed approach with the target user.

Keywords:
Abstract: This work initially proposed the deployment of a socially-assistive robot (SAR) in low-income elder care facilities based on the findings of prior studies. Previous studies considered relevant tasks to meet the needs of all stakeholders (clinicians, caregivers and older adults) to be performed by the robot. This study was combined with the investigation of low-cost modular hardware and software. Although the need for further testing of a low-cost mobile platform capable of human-robot-interaction (HRI) in elder care settings remains, the current global COVID-19 pandemic and its extreme effects on older adults (especially in nursing homes and PACE centers) led to a pivot in our research question. We are combining this assistive technology and its use in aiding the wellness screening for COVID-19 symptoms in older adults in these settings. In lieu of the robot actively and physically interacting with users considering previous ranked tasks, a series of scenarios based on current procedures adopted by clinicians in screening older adults will be considered. A comparison study will investigate older adults’ preference in having robot or human screening and monitoring for COVID symptoms by actively checking vital signs or through engaging in dialog and verbal instructions for preventive purposes. Post-interaction surveys with participants will inform their preferences for the type of interaction and possible improvements on the current version of the robot

Keywords:
Abstract: This paper examines the framework conditions for safe work in physical human-robot interaction to reduce the workload of caregivers in the nursing domain. For this particular case, the transmission of forces play a major role. Because of that, currently existing standards to transmit forces to the human body are discussed. Further on, testing early robot prototypes where the robot comes into direct contact with humans is impossible due to safety concerns and thus has to be done on patient simulators. First, we analyzed the most important nursing activities to be supported by a robot manipulator. Then we conducted two experiments to find out whether a patient simulator behaves similar in comparison to a human while being mobilised by a caregiver during a nursing activity and whether conventional collaborative lightweight robots are up to the task of handling a patient without external help despite having a rather low payload capacity. The experimental results show that moving a patient simulator is more physically demanding compared to moving a human with similar weight and that conventional collaborative lightweight robots are able to push and move a patient simulator weighing 80 kg which is far higher than the robot’s actual payload suggests.

Keywords:
Abstract: ABSTRACT

This is an introduction video to the 2020 IROS Workshop on Wearable SuperLimbs. It describes the scope of the workshop, and introduces organizers and speakers as well as a sponsor. It addresses the new frontier of research on SuperLimbs, and discusses related disciplines and emerging technologies applied to the development of SuperLimbs. Finally this video briefly describes the program of the workshop.

Keywords:
Abstract: ABSTRACT This talk at the 2020 IROS workshop on Wearable SuperLimbs will address a methodology for developing SuperLimbs based on data analysis of human movements. It is critically important in SuperLimbs design that we understand how the wearable robot works in concert with the human. Human movements, however, are complex. Instead of designing a system based on engineering intuition, here we propose to measure human movements and extract robot structure and control algorithm from the data.

Keywords:
Abstract: ABSTRACT Soft robotics has been extensively explored to enable safe interactions with the users and environments, by utilizing materials that are inherently compliant, low-cost, and lightweight. Wearable soft robots can assist the users in different ways, either as exosuits to provide joint support or as supernumerary limbs to augment the users. In this talk, we will present several designs of soft robotic supernumerary arms (we then called Soft Poly Limbs, SPLs), inspired by elephant trunks. The SPLs are enabled by fluidic-driven soft actuators made of elastomeric materials and multi-layered fabrics. The soft actuators are mechanically programmed to achieve bending, elongation, twisting, and a combination of these motion patterns upon inflation. We will introduce the actuator design, fabrication, characterization, computational modeling, and system integration of the SPLs. We will also discuss our work on the extended rigid-arm dynamic models, system identification, and learning controller design for individual soft actuators and the entire SPL to generate versatile motion patterns, and precisely manipulate objects using end effectors and its soft body. Our experimental results highlight the SPLs’ abilities to safely interact with the user while demonstrating promising performance in assisting with a wide variety of tasks in the workplace and daily life. This talk will be concluded by proposing some new opportunities and challenges for building synergies between the human users and SuperLimbs outside the lab environment.

Keywords:
Abstract: ABSTRACT The goal of this paper is to describe our studies on the Sixth Finger and present the different solutions available to control and to get haptic feedback from the system. The path toward the final solutions adopted has been paved with different intuitions about the designers and, more importantly, with continuous feedback from patients that tested all the proposed solutions a led us to the final setup. The paper is organised as follows. We will first recall the main characteristic of the Robotic Sixth Finger. Than we will go through the proposed solutions recalling the main

Keywords:
Abstract: ABSTRACT This talk will provide historical background and current status of human augmentation technology and virtual cyborgs for breaking through the limitations of the human body. The Moonshot Project, a Japanese government research and development project on cybernetic avatars, will also be presented.

Keywords:
Abstract: ABSTRACT Supernumerary robotic limbs: research overview and actuator challenges

Keywords:
Abstract: ABSTRACT Safety Critical Design, Dynamics, and Control of Human-Wearable Extra Robotic Legs

Keywords:
Abstract: ABSTRACT This video introduces to Manas Robotics, a start-up company producing wearable supernumerary grippers and a new type of muscle activity sensor, M2Sensor. The new sensor can potentially replace traditional EMG. It is wearable. You can put it even over a shirt. It can reliably detect your muscle activity, so it will be an ideal sensor for Wearable SuperLimbs.

Keywords:
Abstract: ABSTRACT Panel discussion part 1.

Keywords:
Abstract: ABSTRACT Panel discussion 1 Part 2 (2-5)

Keywords:
Abstract: ABSTRACT Panel discussion 1 Part 3 (3-5)

Keywords:
Abstract: ABSTRACT Panel discussion 2 Part 4 (4-5)

Keywords:
Abstract: ABSTRACT Panel discussion 2 Part 5 (5-5)

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: The workshop targets researchers from rehabilitation, workplace ergonomics, sports medicine, orthopaedics, physical therapy, humanoid robotics, entertainment robotics, computer animation, and machine learning. Especially, the workshop aims to bring experts from different field together, exchange a fruitful discussion between these communities and inspire researchers to create new idea. Social network channels will be opened and used to advertise the workshop. Link to the website of this workshop will be opened in TC: Human Movement Understanding to advertise this workshop. To enhance discussion and interaction between experts and early-career researchers, we encourage experts to communicate with early-career researchers during their poster session. To increase the level of interaction between researchers, active discussion and further communication on the website will be maintained before and after the workshop.

Keywords:
Abstract: Modern developments of construction robotics generally utilize a robot-oriented design approach to develop viable systems for the building industry. This has led to highly sophisticated automation of conventional, but at best slightly altered construction processes. In this paper, we argue for a material-robot oriented design process for the creation of novel construction robotic systems, which can expand the repertoire of current building practice and architectural possibilities. The co-design of a modular material-robot kinematic chain construction system in which the material, robot, and process inform the overall system is introduced from the architectural design, robotic mechatronic development, and task and motion planning perspectives. We present initial research on how material-robot kinematic chains can work in parallel to assemble, disassemble and rearrange large structures.

Keywords:
Abstract: We present here the concept - and contextualization of a robotic fabrication system for bespoke, monolithic timber slabs for multi-story building structures. Wood architecture is seen as the most promising technology for the sustainable development of growing urban areas around the world, resulting in increasing legislative support. Still, wood building systems and construction techniques are currently hardly competitive outside of modular construction paradigms – restricting the material’s use in the majority of building typologies. In many projects, the use of concrete is still preferred, as slabs can be poured into monolithic slabs of various geometries and steel reinforcement can be freely arranged according to structural requirements. In order to allow engineered wood structures to be used across all building types, we propose the co-design of a wood construction systems leveraging a respective robotic fabrication platform in two onsite fabrication scenarios. A mobile robot is developed that can incrementally reinforce multi-directional, bespoke timber slabs of unlimited dimensions and continuous stiffness gradients.

Keywords:
Abstract: As collaborative robots (cobots) enter the field of construction, successful multimodal communication will be essential to safe and effective human-robot cooperation. Auditory communication holds particular promise for successful interaction, but this topic in human-robot interaction is not well understood, and many robots lack the audio systems needed for this type of communication. We present exploratory work on audio system design and integration for the Husky robot. In these efforts, we designed, built, and tested an speaker system appropriate for equipping the Husky robot with audio communication capabilities suitable for construction environments. The prototype system uses 59 W to produce broadband sound at up to 107.9 dB measured 1 m from the robot, though the frequency response and sound pressure level vary depending on the recording location. Robot designers and researchers may benefit by implementing similar systems with additional consideration for speaker placement, power supply interfacing, and broadband sound alarms.

Keywords:
Abstract: In this paper, we used interview data collected at building construction project to examine how different construction stakeholders perceive robot adoption. Our data analysis uncovered notable differences between different construction trades as well as between different positions in the project organization. In pursuit of investigating the underlying factors that engendered the diverse perceptions of and attitudes toward robot adoption, we speculated how various contextual factors in job settings affected their perceptions of robots. Based on the findings, we derived implications for construction robot deployment and human-robot collaboration. This study can enable organizations to understand principled foundation on human-robot interaction in the context of construction and to establish better guidelines for designing the construction workplaces.

Keywords:
Abstract: Service robots with social intelligence are desired to be integrated into our everyday lives. Service robots are intended to improve aspects of quality of life as well as efficiency in human-robot interactive applications. Especially, human-care service robots are desired to have more robust and stable interaction capability not only with human but also with daily objects. The interaction with human is more toward the traditional human-robot interaction and the interaction with daily objects is toward the robotic manipulation in human-care settings. We are organizing an exciting workshop at IROS 2020 that is oriented towards sharing the ideas and experiences amongst participants with diverse backgrounds ranging from Human-Robot Interaction design, social intelligence, decision making, task-motion planning, social psychology and aspects, robotics manipulation and social skills. The purpose of this workshop is to explore how social AI for Human-Robot Interaction can be applied to service robots into our daily lives. This workshop focuses on three social aspects of human-robot interaction of human-care service robots: (1) technical implementation of social AI, (2) form, function and behavior of human-care service robots and (3) human behavior and expectations as a means to understand the social aspects of interacting with these robots and products. The workshop is a single-track, half-day meeting that includes the topics of Human-Robot Interaction design, social intelligence, decision making, social psychology and robotic manipulation skills. We invite world-recognized speakers to present their work and participate in panel discussions that will present good opportunities for the workshop participants, including students. We will also have a poster session and give opportunities to the workshop participants to share and discuss their work. Demonstrations are also possible in this workshop. After the workshop, all the presentations and papers will be available on the workshop webpage.

Keywords:
Abstract: The Covid-19 pandemic found us unprepared in multiple levels of our everyday lives. After the first lockdown pe-riod we are facing the challenge to create a new normality by combining our habits and social norms with precautions against the virus. The educational field is one of the top priority fields with the double role, to educate and keep the students safe. In the current paper, we propose the use of service robots as teach-ing assistants for students aged 13-17. In a real school classroom environment, we conducted a set of experiments with a course conducted by a teacher and a service robot in the role of his/her assistant. Students’ responses in the given questionnaires before and after the course shown that after the interaction with the robot, they statistically significantly believe that the use of robots can make difficult courses more interesting, easier understand-able and motivate them to follow a relevant career in the future. Additionally, apart from the educational benefits we propose that it will eliminate the spread of the virus in the class since the teacher will minimize his/her physical contact with the students and the students will have the chance to collaborate through their robot agent.

Keywords:
Abstract: Nowadays, caregivers are focusing on the health information of newborn infants. It is very important to discover a neurological disease in infants. Especially, Early diagnosis must be made, in the case of premature infants as the possibility of neurological diseases is high. There are many systems for monitoring infants, but few systems classify the status of infants through artificial intelligence analysis. In this paper, the normal and abnormal movements were classified to help diagnose neurological diseases of infants by using Fully Convolutional Network (FCN) according to the skeleton information of infants. Besides, only the labels of normal and abnormal can produce the following weakly-supervised three characteristics. 1) whether the infant's condition is normal or abnormal) 2) which segment of the video shows the infant's movement is abnormal, 3) which body part of the infant is found to be abnormal. The proposed network is simple and suitable for application to robot devices that observe infants in the home environment. This study was conducted using the skeleton database extracted from the video of infants.

Keywords:
Abstract: The application of healthcare robots in home and medical settings is approaching reality. In order to ensure that interactions between healthcare robots and users are successful however, these robots need to be considered from a behavioral aspect. Ensuring robots act in a way that is both useful and acceptable is particularly important when considering healthcare robots, which will likely be interacting with individuals who are injured, unwell, or otherwise vulnerable. One way in which to inform research exploring appropriate behaviours for healthcare robots, is to study literature examining physician behaviours that are found to be effective in ensuring successful physician-patient interactions. One such behaviour, found to be associated with positive patient outcomes, is the use of physician humour. No research thus far however, has studied the use of humour by a healthcare robot in human-robot interactions. The current study examined the use of humour by a healthcare robot during a scripted human-robot interaction, within a simulated medical environment. Ninety-one participants took part in this study and were allocated to either a humorous or neutral condition. Participants who took part in the interaction with the humorous robot rated the robot significantly higher in terms of empathy, likability, safety, animacy, and sociability. Participants were also significantly more likely to laugh during interactions with the humorous robot, compared to the neutral robot. These findings support the use of healthcare robot humour as a potential way in which to increase user perceptions and acceptability.

Keywords:
Abstract: Alzheimer’s disease (AD) has become one of the most common neurodegenerative disorder, currently affecting nearly twenty-five million worldwide. There is no cure available at this moment, however early diagnosis, socially assistive robotics therapy showed promising results in slowing down the progression of the diseases. This paper discusses recent developments in early AD diagnosis and non-pharmacological treatment options.

Keywords:
Abstract: This paper introduces a large-scale Korean speech dataset, called VOTE400, that can be used for analyzing and recognizing voices of the elderly people. The dataset includes about 300 hours of continuous dialog speech and 100 hours of read speech, both recorded by the elderly people aged 65 years or over. A preliminary experiment showed that speech recognition system trained with VOTE400 can outperform conventional systems in speech recognition of elderly people’s voice. This work is a multi-organizational effort led by ETRI and MINDs Lab Inc. for the purpose of advancing the speech recognition performance of the elderly-care robots.

Keywords:
Abstract: Observing another’s affective state and adjusting one’s behavior to respond to it, is the basic functionality of empathy. To enable robots to do this, they need a mechanism to learn how to provide the most appropriate empathic behavior through continuous interaction with humans. To this end, we propose a reinforcement learning based framework for cognitive empathy, which uses reinforcement learning to learn the most appropriate empathic behavior for different emotional states during human-robot interactions. To verify the proposed framework, an experiment is conducted with the humanoid robot Pepper over 28 participants, where their facial emotion expression is tracked continuously and used to select appropriate empathic behaviors. The obtained results show the proposed reinforcement learning model converges to the optimal empathic behaviors for all emotions that were expressed a sufficient number of times, which helps the participants feel more positive emotions like happiness.

Keywords:
Abstract: Image-based user recognition has become a technology that is now used in many ways. This is because it is possible to use only a camera without the need for any other device. In addition, due to the advantage of contactless, there are relatively many people who prefer it. However, normal recognition is not possible if some of the face information is lost due to the user's posture or the wearing of masks caused by the recent prevalent disease. This is because the information needed for recognition is largely lost due to obscurity. In some platforms, although performance is improved through incremental updates, it is still inconvenient. It takes a lot of time to update information about the occluded faces, and the overall accuracy is lowered. In this paper, we propose a method to respond more actively to these situations. First, determine whether the obscurity occurs, and improve the stability by calculating the feature vector using only a significant area when the obscurity occurs. By recycling the existing recognition model, without incurring little additional costs, it was confirmed the results of reducing the recognition performance drop in certain situations. Since only one layer has been added to the existing recognizer, there is no burden in terms of computational speed or memory. Using this technique, we confirmed the performance improvement of about 1.5% in the situation where some information is lost. Although performance is not dramatically improved, it is the biggest advantage that it can improve recognition performance by utilizing existing systems.

Keywords:
Abstract: Deep learning, based on which many modern algorithms operate, is well known to be data-hungry. In particular, the datasets appropriate for the intended application are difficult to obtain. To cope with this situation, we introduce a new dataset called ETRI-Activity3D, focusing on the daily activities of the elderly in robot-view. The major characteristics of the new dataset are as follows: 1) practical action categories that are selected from the close observation of the daily lives of the elderly; 2) realistic data collection, which reflects the robot’s working environment and service situations; and 3) a large-scale dataset that overcomes the limitations of the current 3D activity analysis benchmark datasets. The proposed dataset contains 112,620 samples including RGB videos, depth maps, and skeleton sequences. During the data acquisition, 100 subjects were asked to perform 55 daily activities. Further, the domain difference between both groups of age was verified experimentally.

Keywords:
Abstract: Emotion change detection via facial expression information is an important clue to non-verbal communication that can uncover emotional context. In a human-robot interaction scenario, figuring out the timing of emotion changes using facial expression information from an user has three advantages on: 1) providing a start point to obtain timeconsistent multi-modal information, 2) reducing search space in time series, and 3) producing feedback information to improve the scenario. In this regard, we introduce an initial investigation to an automatic emotion change detection framework in the field of human-robot interaction. To tackle this issue, we propose a novel method of deep emotion change detection for inferencing emotion status and detecting multiple points of emotion changes. Incorporating these ideas, we provide evaluation methods to validate the framework and the baseline performance of our approach.

Keywords:
Abstract: Our research aim is, to investigate the problems of dynamic environment understanding for service robot navigation in social areas. This paper presents an efficiency analysis of multi-head attention models for real-time social dynamics prediction on limited computational hardware.

Keywords:
Abstract: Acquiring a motor ability is a complex process, whether for an athlete working toward peak performance or a post-stroke patient re-learning to control a limb. Curriculum selection is the process of choosing a sequence of sub-tasks, their training order, and their frequency in order to achieve a complex target task. Currently, in motor skill training, no systematic method exists for selecting curricula, and can result in long, costly and often unsuccessful training. At the same time, recent advances in artificial intelligence have introduced curriculum learning using Reinforcement Learning, which has enabled some impressive speed-ups in artificial agents’ abilities to learn complex tasks. This paper delineates how the computational approaches used in curriculum learning for reinforcement learning can be modified, to represent the learning process of people in motor tasks. This paper also presents some preliminary results on a dynamic motor game designed to evaluate the process of motor task learning and the efficacy of different curricula.

Keywords:
Abstract: Despite the recent rapid advancement of applying reinforcement learning in social robotics, the current state-ofthe- art algorithms are still not able to adapt quickly when facing a similar but new social scenario. In this study, we examine meta-reinforcement learning (Meta-RL) methods in a task where a robot needs to learn to approach a group of human agents. For the robot to learn socially aware behavior quickly and effectively in multiple similar but different robot approaching scenarios, we incorporate proximal policy optimization (PPO) with the model-agnostic meta-learning frameworks, i.e., MAML and Reptile, in the training process. Our results show that the Meta-RL methods are feasible to be adopted in learning robot approaching behavior toward groups and they outperform the baseline PPO algorithm (train from scratch), measured by accumulated reward and time.

Keywords:
Abstract: Development of Human-care Robot Technology for Aging Society

Keywords:
Abstract: Social robots with the ability to read the room and work the crowd hold potential for more effective and engaging interactions with people, but such human performerinspired abilities for robots are currently limited. We propose that audiovisual data and comedian-inspired decision-making can enhance the performance of social robots in settings from live performance to healthcare. The present paper describes a preliminary robotic system for comedy performance, which includes integration with OpenFace facial behavior analysis software for real-time detection of Facial Action Units and gaze information. In parallel to the robotic system creation and initial validation, we are planning to survey human comedy performers to learn more about the audience cues they use to assess joke success and decide on performance adaptations. Preliminary results show promise for OpenFace to aid with the proposed real-time recognition, and future steps will more thoroughly assess proposed behavior recognition models and robot performance adaptation approaches. The products of this work can inform social roboticists who wish to create more compelling and entertaining systems.

Keywords:
Abstract: In many industrial countries, work-related musculoskeletal disorders (WMSD) are associated with high costs to employers such as lost productivity, worker's compensation costs, and so on. WMSD risk factors include awkward posture, repetition, mechanical compression, vibration, etc. To tackle such risk factor, a workplace ergonomics assessment comprises of intervention descriptions for WMSD and the intervention strategy for reducing, eliminating, or controlling worker exposure to the WMDS risk factor. These measures aim to ensure workers' health and safety. Human-Robot Collaboration (HRC) a promising concept that can potentially help with maintaining and improving ergonomics and working conditions of human co-workers. However, the traditional HRC technologies mostly focus safety in terms of collision avoidance and impact safety, while ergonomics potential is largely under-exploited. Therefore some of the major challenges and opportunities to exploit such potential are: collaborative robots have to be aware of co-worker's state and predict his/her actions; the interactive motion has to be able to reconfigure the human working conditions; the methods must operate online.

The above-mentioned challenges and opportunities present a novel research topic for the community. We previously organized two successful workshops to establish the field and to discuss this topic. First, we held a workshop at ICRA 2018 to introduce and discuss the concept in terms of collaborative robotics. Then we held a continuation workshop at IROS 2019 to review the initial research progress based on the goals set at the first workshop and identify major ongoing research problems. Now it is time to discuss those problems and come up with solutions to make progress towards robust and applicable methods for ergonomic human-robot collaboration.

The proposed workshop will first review the progress of the research and development in the new field that was achieved since the last workshop at IROS 2019. Next, we will focus on how to solve the problems that emerged during the research and development based on the goals that were set at the previous workshop. In addition, we will discuss potential novel research directions, ergonomic metric and robot control methods. Finally, we will discuss and design new validation methods that can be used for objective benchmarking within the field. Such an approach requires experts from various research fields and interdisciplinary discussion. For this purpose, we assemble a diverse set of organizers and speakers, who are leading experts in their respective areas that are highly relevant to this workshop topic.

Keywords:
Abstract: In many industrial countries, work-related musculoskeletal disorders (WMSD) are associated with high costs to employers such as lost productivity, worker's compensation costs, and so on. WMSD risk factors include awkward posture, repetition, mechanical compression, vibration, etc. To tackle such risk factor, a workplace ergonomics assessment comprises of intervention descriptions for WMSD and the intervention strategy for reducing, eliminating, or controlling worker exposure to the WMDS risk factor. These measures aim to ensure workers' health and safety. Human-Robot Collaboration (HRC) a promising concept that can potentially help with maintaining and improving ergonomics and working conditions of human co-workers. However, the traditional HRC technologies mostly focus safety in terms of collision avoidance and impact safety, while ergonomics potential is largely under-exploited. Therefore some of the major challenges and opportunities to exploit such potential are: collaborative robots have to be aware of co-worker's state and predict his/her actions; the interactive motion has to be able to reconfigure the human working conditions; the methods must operate online.

The above-mentioned challenges and opportunities present a novel research topic for the community. We previously organized two successful workshops to establish the field and to discuss this topic. First, we held a workshop at ICRA 2018 to introduce and discuss the concept in terms of collaborative robotics. Then we held a continuation workshop at IROS 2019 to review the initial research progress based on the goals set at the first workshop and identify major ongoing research problems. Now it is time to discuss those problems and come up with solutions to make progress towards robust and applicable methods for ergonomic human-robot collaboration.

The proposed workshop will first review the progress of the research and development in the new field that was achieved since the last workshop at IROS 2019. Next, we will focus on how to solve the problems that emerged during the research and development based on the goals that were set at the previous workshop. In addition, we will discuss potential novel research directions, ergonomic metric and robot control methods. Finally, we will discuss and design new validation methods that can be used for objective benchmarking within the field. Such an approach requires experts from various research fields and interdisciplinary discussion. For this purpose, we assemble a diverse set of organizers and speakers, who are leading experts in their respective areas that are highly relevant to this workshop topic.

Keywords:
Abstract: Using collaborative robots (“cobots”) for supporting work tasks with high physical and psychological demands (“un-ergonomic tasks”) is a very promising approach for compensating the aging workforce in Europe and improving health and productivity in the industrial sector. However, it is not easy to design collaborative work systems according to safety, ergonomics and productivity standards from a technical and a social point of view. Many details of the applied technologies have to be taken into account and workers should be allowed to participate in the work system design proactively in order to consider their practical knowledge and increase their acceptance for the implemented solution. “ema Work Designer” is a software tool that enables to design manual and semi-automatic (= collaborative) work systems according to safety, ergonomics and productivity standards like DIN ISO/TS 15066, EAWS and MTM. The interactive 3D visualization supports technical experts in work station design and facilitates communication with workers. This presentation will show a use case that was developed in the EU-founded SOPHIA-Project with the company Hankamp Gears. Based on the current manual work station, a solution for integrating the UR16e cobot was designed as a collaborative work system together with technology specialists, social scientists and representatives from the target company. ema Work Designer software was used to validate the geometrical set-up of the work station in interaction with the cobot platform. Moreover, it was used to assess the performance of the current and the future work system, for example showing a reduction of active manual production time (measured with MTM) by 18 seconds/cycle (= -28%) and ergonomic work load (measured with EAWS) by 31,5 points (= -59%).

Keywords:
Abstract: How can robots physically interact with humans in a meaningful and healthy way? Existing approaches to specifying close-contact, physical interactions between humans and robots focus solely on successful task completion. However, these approaches completely neglect the biomechanical and ergonomic ramifications of robot actions on the human body. An action which may seem momentarily effective may result in stresses to the human musculoskeletal system and even serious injuries. In this talk, I will discuss how machine learning enables healthy, bi-directional, and biomechanically-safe interactions between humans and machines that can be sustained over long periods of time. Specifically, I will present Bayesian Interaction Primitives -- a probabilistic framework that enables learning and inference for HRI scenarios. Bayesian Interaction Primitives encode the mutual dependencies between interaction partners and can be used to 1.) predict human motion and sensor values, 2.) infer task-relevant biomechanical variables, and 3.) generate appropriate robot responses. Used within a model-predictive control loop, Bayesian Interaction Primitives generate actions that minimize long-term impact on the musculoskeletal system of the human partner. To demonstrate the approach, I will present a number of applications in prosthetics and social robotics.

Keywords:
Abstract: In this talk I first present our work on using machine learning to “compile” complex biomechanical simulations into compact models which can be quickly queried. These models are used to optimize robotic behavior with respect to relevant musculo-skeletal variables of humans that physically interact with the robot. In our current work, we are applying this approach to compensating for the so called “limb position effect” in prosthetic control. Finally, I will provide an outlook on using high-level planning to automatically distribute tasks between robots and humans, which aim at factory-wide optimization, rather than the behavior of individual robots.

Keywords:
Abstract: The talk will present several robot control methods for co-manipulation with humans, where the key element is an optimisation of human ergonomics. The optimization process incorporates biomechanical models and real-time measurements to track and improve various metrics, such as: muscle fatigue, joint torques and arm manipulability. In the first part, the focus is on the application to co-manipulation during physical human-robot collaboration in various practical tasks (e.g., collaborative sawing, polishing, valve turning, etc.). The second part will focus on the application in exoskeletons, where co-manipulation is done while human and robot limbs are physically coupled. Finally, the last part will examine the application in teleoperation, where co-manipulation pertains to the remote robot being commanded by a human operator. In particular, we will present an analysis of impedance-command interfaces in force-feedback tele-impedance.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: The talk will present a control approach to human-robot co-manipulation that accounts for human ergonomics and related fatigues. We will first present a human monitoring method that includes a overloading joint torque estimation technique, as well as the feedback methods to provide a guidance to human. The robot by means the mobile manipulator platform MObile Collaborative robotic Assistant (MOCA) then uses the lower level control framework in conjunction with the proposed higher level methods that can anticipate human ergonomic states. The robot then uses the proposed methods to control its own behaviour in a way that offloads the excessive effort of the human and ensures ergonomic working conditions, as well as the coexistence in the workplace.

Keywords:
Abstract: Cobots allow to combine the dexterity, flexibility and problem-solving ability of humans, with the strength, endurance, quality consistency and precision of robots and can be redeployed to other tasks much more easily. By also placing health, the working conditions and the sustainable development goals central, we can ensure ourselves to reap the full benefits not only economically but also for our society. This talk will discuss the multidisciplinary approach to develop a new generation of cobots and exoskeletons and control algorithms for improving the ergonomics of workers in the manufacturing industry.

Keywords:
Abstract: Demands are increasing for robots that can cooperate with humans more naturally and comfortably. We have been conducting research on human-centered analysis focusing on observation of humans' reaction when doing cooperative tasks with a robot. We first analyzed factors causing discomfort, and experiments in a coworker scenario implied that perceived uncertainty in a robot’s movement is a key determinant of how much humans are comfortable with them due to robot motion uncertainty. Through another task of handover, we have found that humans are combining feedforward and feedback control, whose parameter varies depending on personality and physical dimensions. Finally, we are conducting studies on active physical human-robot cooperation. In a physically cooperative task scenario, we obtained first results on factors alternating humans' behavior when the robot interacts actively with them. We will keep addressing design principle of behavior of future collaborative robots by reinforcing those experimental studies with increased involvement of robots.

Keywords:
Abstract: In this talk I will present our recent results in prediction of human movement for improving human-robot collaboration and optimization of human movement with respect to ergonomics criteria.

Keywords:
Abstract: ABSTRACT

Keywords:
Abstract: The Bio-inspired Robotics session will deliver an overview of this rapidly evolving field with an emphasis on recent developments. Its goal is to provide alternative solutions to existing real-world problems by looking at engineering and robotics from the perspective of biology and nature. With a panel of well renowned experts in the field, the session will cultivate points of inspirations and solutions for individuals of different backgrounds and discipline in both science and engineering. Get ready to be inspired, bioinspired!

Keywords:
Abstract: The tutorial presents an introduction to bio-inspired robotics from a biologist’s viewpoint. It frames the question an engineer should ask a biologist to make certain that they are receiving the most effective transfer of biological principles. It covers the issues of biological discovery, scaling, constraint, selection, and complexity. It does so with examples from the research of the presenter on legged in locomotion. The lessons from biology include energy management, embodied control, and learning. The presentation concludes with the introduction of a new multi-university research initiative focusing on the science of learning, innovation, and control that is embodied (SLICE), ultimately aspiring to robotic squirrel-like capabilities.

Keywords:
Abstract: This tutorial covers the bio-inspired principles behind building 10 cm scale dynamic legged robots. Using principles of dynamic similarity to a cockroach lead to the VelociRoACH robot, which can run at 2.7 meters per second. The tutorial next looks at using mechanical advantage and series elastic actuation to construct a robot with high vertical jumping agility, Salto. Finally, folding fabrication techniques for rapidly prototyping some of these millirobots is described, with pointers to a completely open source hexapod: OpenRoACH.

Keywords:
Abstract: The octopus, with its soft, muscular hydrostat body and arms, is proficient in locomotion and complex motor functions. Their versatility, "infinite" degrees of freedom, and dexterity have made them an inspiration for soft robotics and synthetic adhesion mechanisms. Octopus suckers have been observed to be utilized for body anchoring, swift maneuvering, object examination and manipulation, and chemo and tactile sensing. Most of these tasks would be impossible without the sucker’s main functionality, the attachment mechanism. This study was designed to determine: 1) how pull-off force is impacted if there is no communication with the brain and 2) how much of the attachment mechanism depends on suction versus adhesion. While these parameters may have been qualitatively described in previous studies, they have never been addressed quantitatively. Experiments were conducted on ten Octopus bimaculoides (five female and five male). Pull-off force was measured on intact arms, amputated arms, and amputated arms with a punctured sucker to gain insights into sucker functionality and control mechanisms. The results of these experiments can be used to design efficient synthetic underwater attachment mechanisms. Coupled with soft robotic arms, these synthetic suckers can be utilized to maneuver on and through aquatic environments for exploration and environmental monitoring.

Keywords:
Abstract: Mobile robots have the potential to address many of today’s pressing problems ranging from search-and-rescue to environmental monitoring to disaster relief. However, the traditional approaches indoors do not perform well in unstructured environments. The key to solving many of these challenges is to explore new, non-traditional designs. Fortunately, nature surrounds us with examples of novel ways to navigate and interact with the real world. Dr. Tolley’s Bioinspired Robotics and Design Lab seeks to borrow the key principles of operation from biological systems and apply them to robotic design. This talk will give an overview of recent projects demonstrating approaches to the design, fabrication, and control of soft mobile robots. These projects seek to develop bioinspired systems capable of navigating the world by walking and swimming.

Keywords:
Abstract: The biosonar system of bats is an example for a biological sensory system that enables dexterous autonomy in complex natural environments. The capabilities of the animals could inspire new solutions to critical problems in creating bioinspired robots that could navigate in natural environment with the same high levels of complexity as the habitats of bats. In particular, bats can provide insights into how to encode the relevant sensory information, extraction of patterns that are hidden in complex data, and the integration of sensing and actuation.

Keywords:
Abstract: In this talk, I describe our new soft burrowing robot. The talk has three sections. First, I introduce the challenges associated with burrowing and how we can look to biology to glean some principles for reducing and controlling the resistive forces that arise while burrowing. Second, I describe a set of experimental results from testing hypotheses that we derived from our look to biology. Third, I present the design and characterization of the robot.

Keywords:
Abstract: An elephant eats 200 kg per day, or the equivalent of 200 grams every minute. How does an elephant feed so quickly? We present experiments with African elephants at the Atlanta Zoo and dissections of elephant trunks at the Smithsonian Institution. We demonstrate three ways that elephants can feed quickly: sucking like a vacuum cleaner, squeezing food items together, and wrapping their trunk around objects to get a better grip. We use mathematical models to rationalize the forces and geometries required by the trunk to perform these feats.

Keywords:
Abstract: This video introduces and provides an overview of the IROS 2020 Introduction to Next Generation Haptics for XR tutorial session. The tutorial session will primarily take place synchronously on October 29th as described in the video. The video outlines the schedule for the session on the 29th, the individuals involved in the session, and how to get more information.

Keywords:
Abstract: Human touch represents a marvelous sensory system, with a large variety of receptors and sensing elements. The interplay of these elements enables to shape the exploration and the interaction with the external world, and it also significantly contributes to human proprioception and perceptual organization of motion and space. It is hence clear that reproducing human tactile capabilities in artificial systems or haptic interfaces is an extremely challenging task: indeed, it is extremely hard for a haptic interface to fulfill both modality matching MM (i.e. an artificially delivered cue should be mediated by a stimulus sharing the same sensory modality as the one that would be naturally felt) and somatotopic matching SM (i.e. the stimulus should be experienced at the same location where it would be naturally experienced). The fulfillment of the latter requirement is not easy in tactile augmented reality, where the user’s hands should be able to naturally touch the physical items, and at the same time, experience artificially superimposed tactile stimuli. In this talk, I will discuss an approach to achieve both MM and SM in wearable tactile systems for advanced human machine interaction, augmented reality, and tele-robotics. This approach moves from the neuroscientific investigation of human touch and its modelling using a language that can be easily implemented in an artificial body. At the same, it leverages upon the usage of fabrics for the interaction surface with the user’s skin since it does not impair human touch-related sensory capabilities while touching physical items and maximizes device wearability. This combined approach opens interesting perspectives for tactile augmented reality applications.

Keywords:
Abstract: How to Design HX: Applying Interaction Design to Haptic Experiences Oliver Schneider, Assistant Professor, University of Waterloo Abstract Interaction designers follow an established process to create effective systems. Haptic technology, now maturing, can now fit into this process, but presents unique challenges. To create haptic experiences, designers need to know what changes and how to work with this emerging modality. In this presentation, I will outline the major activities used in interaction design, and describe how each can to be adapted to haptics. Drawing from a several studies conducted with hapticians, or makers of haptics, I will describe the problems you are likely to encounter with working with haptics, and possible solutions when you try to create haptic experiences. After this presentation, you will be well-equipped to think about interaction design with haptic technology. Speaker Oliver Schneider is an Assistant Professor, Human-Computer Interaction (HCI) researcher, and haptician at the University of Waterloo (Faculty of Engineering, Department of Management Sciences). His work combines HCI and haptics, focusing on understanding the challenges facing hapticians (makers of haptics) and developing tools to support them in creating haptic experiences. Oliver completed his PhD in Computer Science under the supervision of Karon MacLean (2016, University of British Columbia) and holds an MSc (2012, University of British Columbia) and BSc Honours (2010, University of Saskatchewan). From 2017-2018, Oliver was a Postdoctoral Scholar with Patrick Baudisch at the Hasso Plattner Institute in Potsdam, Germany, partially supported by a Natural Sciences and Engineering Council of Canada Postdoctoral Fellowship (NSERC PDF). Oliver has also collaborated with Disney Research on novel haptic interactions as an intern and then consultant. Oliver’s work has been published in venues including ACM CHI and UIST; IEEE Haptics Symposium, EuroHaptics, and World Haptics; and the International Journal of Human-Computer Studies (IJHCS).

Keywords:
Abstract: Haptic devices allow touch-based information transfer between humans and intelligent systems, enabling communication in a salient but private manner that frees other sensory channels. For such devices to become ubiquitous, their physical and computational aspects must be intuitive and unobtrusive. The amount of information that can be transmitted through touch is limited in large part by the location, distribution, and sensitivity of human mechanoreceptors. Not surprisingly, many haptic devices are designed to be held or worn at the highly sensitive fingertips, yet stimulation using a device attached to the fingertips precludes natural use of the hands. Thus, in this tutorial we will examine the design of a wide array of haptic feedback mechanisms, ranging from devices that can be actively touched by the fingertips to multi-modal haptic actuation mounted on the arm. We demonstrate how these devices are effective in virtual reality, human-machine communication, and human-human communication.

Keywords:
Abstract: The human body is capable of dexterous manipulation in many different environments. However, some environments are challenging to access because of distance, scale, and limitations of the body itself. In many of these situations, telerobots can effectively restore access. Dexterous manipulation through these telerobots can only occur when the operator receives sensory feedback of the telerobot's interactions in the environment. In this talk, I will discuss recent work from our lab on the application of haptic feedback in the context of telerobotics. I will begin with an overview of two different telerobotic applications, telesurgical robots, and upper-extremity prosthetic devices. I will then discuss lessons learned from our attempts at haptic interaction design for these applications in a manner applicable to researchers hoping to incorporate haptics into robotic, XR, and HCI applications.

Keywords:
Abstract: The morning speaker panel will take place on October 29th during the synchronous part of the Introduction to Next Generation Haptics for XR tutorial session. The speaker panel will feature speakers from the morning part of the tutorial session. In this panel, speakers will discuss haptics for XR from various perspectives and answer questions.

Keywords:
Abstract: The ability to identify and perceive the properties of objects relies on our sense of touch or more accurately active touch or haptic sensing. However, it is not only tactile cues arising from mechanoreceptors in the skin that provide us with spatial and temporal information about objects, but also signals from thermoreceptors that sense changes in skin temperature associated with contacting objects made from different materials. Tactile and thermal displays designed to replicate these sensations in virtual environments or for teleoperated robotic systems have typically focused on a single modality rather than the multisensory experience associated with object contact. In our work we have sought to understand the multisensory interactions that occur in the cutaneous senses by examining how tactile and thermal information is processed. We are particularly interested in understanding how these independent sensory systems function synergistically, given their profound differences in temporal and spatial processing.

Keywords:
Abstract: Wearable haptic devices present many challenges in actuator selection and design. In addition to providing a sufficiently strong signal to be easily felt by the user, the actuators must also not encumber the user’s motion. These limitations motivate our research towards providing complex haptic signals using minimal hardware. We approach this problem from two directions by using haptic illusions and by creating novel actuators that can provide multiple haptic sensations simultaneously. Haptic illusions seek to fool our sense of touch into thinking we are feeling something different than what is being displayed by taking advantage of the inexactness of our perception. In our research, we employ haptic illusions to allow small, lightweight actuators to create sensations that would normally require actuators that are impractical for a wearable device. I will discuss two haptic illusions and their uses: asymmetric vibrations for generating ungrounded forces and sequential indentation for creating lateral motions. I will also present our work in creating multi-modal hardware, which allows for complex haptic sensations using minimal actuator size and weight.

Keywords:
Abstract: As haptics have become an ingrained part of our wearable experience, particularly through phones, smartwatches, and fitness trackers, significant research effort has been conducted to find new ways of using wearable haptics to convey information, especially while we are on-the-go. In this paper, instead of focusing on aspects of haptic information design, such as tacton encoding methods, actuators, and technical fabrication of devices, we address the more general recurring issues and “gotchas” that arise when moving from core haptic perceptual studies and in-lab wearable experiments to real world testing of wearable vibrotactile haptic systems. We summarize key issues for practitioners to take into account when designing and carrying out in-the-wild wearable haptic user studies, as well as for user studies in a lab environment that seek to simulate real world conditions. We include not only examples from published work and commercial sources, but also hard-won illustrative examples derived from issues and failures from our own haptic studies. By providing a broad-based, accessible overview of recurring issues, we expect that both novice and experienced haptic researchers will find suggestions that will improve their own mobile wearable haptic studies.

Keywords:
Abstract: In the last decade, there has been a rapid growth in the use of modern technologies in XR landscape — thereby reducing the friction between the user and digital mediums by incorporating user actions, their environments and their well-being in the digital context and thereof making the digital content more accessible, engaging and immersive to the users. Haptic feedback can add a “physical” dimension to user interactions, where the digital information is communicated to the users through their sense of touch. In numerous research studies, haptic feedback technologies have shown to increase the bandwidth of information communicated and to enhance the fidelity of interactions between the user and their digital mediums, however haptic feedback has seen limited success in wide scale consumer grade devices and applications. In this talk, I will highlight some of the challenges incurred during the maturity and dissemination of haptic technologies for consumer use and will discuss a few opportunities to overcome these challenges.

Keywords:
Abstract: For the past eight years, I have been part of a team at the University of Michigan that is exploring new techniques for building a full-page tactile display. Though no such devices have reached the market, there are a number of manufacturers who are getting close to releasing products. It is safe to say that in the next five years there will be at least one product released. However, we are only now beginning to think about what interacting with content on these displays might involve.

In this talk, I will discuss this journey, placing emphasis on how important it is to cultivate both deep roots and broad branches when you are designing technologies for a community of use to which you may not yourself belong.

Keywords:
Abstract: Vibrotactile feedback has seen several exciting new developments in recent years as researchers have begun to explore more complex cue synthesis, higher density tactile arrays, and virtual and augmented reality applications. Traditional vibrotactor control methods, whether through commercial controllers or custom integrated circuit designs, present tradeoffs between cost, flexibility, and speed of implementation. In this workshop, we present a method of vibrotactor control based on digital audio interfaces that accommodates emerging research trends while striking a balance between these tradeoffs. We describe an open-source software and hardware package, Syntacts, that lowers the technical barrier to rendering vibrations with audio. The workshop includes tutorials for the Syntacts amplifier kit, programming APIs, and graphical user interface. We conclude with a real-world application of Syntacts, where we use the framework to deliver haptic feedback through a bracelet interface for hand interactions in virtual reality.

Keywords:
Abstract: The afternoon speaker panel will take place on October 29th during the synchronous part of the Introduction to Next Generation Haptics for XR tutorial session. The speaker panel will feature speakers from the afternoon part of the tutorial session. In this panel, speakers will discuss haptics for XR from various perspectives and answer questions.

Keywords:
Abstract: The tutorial closing will take place on October 29th during the synchronous part of the Introduction to Next Generation Haptics for XR tutorial session. The recording of the tutorial closing will be uploaded after the 29th.