Last updated on March 1, 2015. This conference program is tentative and subject to change
The system also incorporates software that recognizes and localizes objects based on an acquired 3D point cloud and a CAD model of the object to search for. The core matching algorithm is based on oriented point pairs and a Hough-like voting scheme. The method has been improved with a robust clustering algorithm as well as methods for pose verification and pose refinement that significantly increase the accuracy and robustness of the system.
As an application of the system, an industrial prototype workcell is presented. The task is to recognize, grasp and transfer parts of office chairs, such as seats, seat backs and armrests, from a pallet to a cardboard container. It demonstrates how the vision system is easily set up to recognize three different components by using CAD models. The prototype workcell further demonstrates how the pose estimation results can be fed directly to a separate handling robot in order to grasp a chair seat. A series of ten experiments was performed where the chair seat was placed in arbitrary poses in the pallet. Pose estimations were performed in just over one second per experiment, and the obtained accuracy was well within the tolerances for the grasp operation in all ten cases.
In this paper, we propose a purely data-driven method to tackle the pose estimation problem. Motivated by photometric stereo, we build an imaging system with multiple lights where each image channel is obtained under different lightning conditions. In an offline stage, we capture images of an object in several poses. Then, we train random ferns to map the appearance of small image patches into votes on the pose space.
At runtime, each patch of the input image votes on possible pose hypotheses. We further show how to increase the accuracy of the object poses from our discretized pose hypotheses. Our experiments show that the proposed method can detect and estimate poses of textureless and shiny objects accurately and robustly within half a second.
We propose the Incremental Posterior Joint Compatibility (IPJC) test. While equivalent to JCBB on linear problems, it is significantly more accurate on non-linear problems. When used for feature-cloud matching (an important special case), IPJC is also dramatically faster than JCBB. We demonstrate the advantages of IPJC over JCBB and other commonly-used methods on both synthetic and real-world datasets.
Traditional occupancy grid (OG) mapping makes two assumptions for computational efficiency. This paper formulates the full Bayesian solution, which makes no assumptions. This solution is computationally intractable for reasonably sized maps. Therefore, we introduce a novel patch map that approximates the full Bayesian solution. The patch map computes the full Bayesian solution for a small patch of the map surrounding the cell being updated. The patch map computes the occupancy of each cell in turn and requires prior knowledge of a relatively accurate map outside of the patch (ground truth or maximum a posteriori (MAP) estimate). The patch map can be computed for large maps where the full solution would be computationally intractable, and thus can be used as a benchmark for the evaluation of other occupancy mapping algorithms.
This paper shows that the patch map approximates the full solution for a simple one-dimensional test case, whereas traditional occupancy grid mapping does not. The patch map is then shown to work on two-dimensional cases, where the full solution cannot be computed. The paper concludes that the patch map is a better benchmark for occupancy grid mapping than existing solutions and that traditional OG mapping can be overconfident. The proposed benchmark could be used to quantify/optimize future online OG mapping methods.
In this work, we propose the use of Gaussian Process regression to model lens distortion. With the use of a squared exponential covariance function, a Gaussian Process (GP) can describe the space of smooth distortion functions; kernel hyper-parameter selection in this space then analogous to performing explicit model selection between possible parametric models.
Our evaluation shows that this Gaussian Process formulation of lens distortion performs on par with parametric distortion models.
In this paper we evaluate the performance of various reordering techniques on benchmark SLAM data sets and provide definitive recommendations based on our results. We also compare these state of the art algorithms against our simple and easy to implement algorithm which achieves comparable performance. Finally, we provide empirical evidence that few gains remain with respect to variants of minimum degree ordering.
The proposed SPG consists of two parts: i) estimating the single direction of possible motion based on filtering of the measured end effector velocities; and ii) choosing an appropriate set-point for the underlying IFC, resulting in an effective operation of the mechanism. A major aspect of our approach is to explore the kinematic constraints with the manipulators desired set-point, avoiding direct force control, as the required interaction force is unknown and hence there is no definite reference force.
The presented approach is a generalization of our previous work on constrained manipulation of unknown mechanisms and extends the applicability to a wider class of manipulators by considering joint-level IFC and taking into account the applied forces in yielding a robust and effective controller.
The approach is evaluated in various experiments on a manipulator, providing joint space compliance.
However, properly controlling the motion along the scanning trajectory is a major problem. Indeed, the tissue exhibits deformations under friction forces exerted by the probe leading to deformed mosaics. In this paper we propose a visual servoing approach for controlling the probe movements relative to the tissue while rejecting the tissue deformation disturbance. The probe displacement with respect to the tissue is firstly estimated using the confocal images and an image registration real-time algorithm. Secondly, from this real-time image-based position measurement, the probe motion is controlled thanks to a simple proportional-integral compensator and a feedforward term. Ex vivo experiments using a Staubli TX40 robot and a Mauna Kea Technologies Cellvizio imaging device demonstrate the effectiveness of the approach on liver and muscle tissue.
In this video, we introduce a collective decision-making method for swarms of robots that is based on positive feedback. The method enables a swarm of robots to choose the fastest action from a set of possible actions. The method is based solely on the local observation of the opinions of other robots. Therefore, the method can be applied in swarms of very simple robots that lack sophisticated communication capabilities.
In the last years monocular SLAM has been widely used to obtain highly accurate maps and trajectory estimations of a moving camera. However, one of the issues of this approach is that, due to the impossibility of the depth being measured in a single image, global scale is not observable and scene and camera motion can only be recovered up to scale. This problem gets aggravated as we deal with larger scenes since it is more likely that scale drift arises between different map portions and their corresponding motion estimates. To compute the absolute scale we need to know some kind of dimension of the scene (e.g., actual size of an element of the scene, velocity of the camera or baseline between two frames) and somehow integrate it in the SLAM estimation. In this paper, we present a method to recover the scale of the scene using an omnidirectional camera mounted on a helmet. The high precision of visual SLAM allows the head vertical oscillation during walking to be perceived in the trajectory estimation. By performing a spectral analysis on the camera vertical displacement, we can measure the step frequency. We relate the step frequency to the speed of the camera by an empirical formula based on biomedical experiments on human walking. This speed measurement is integrated in a particle filter to estimate the current scale factor and the 3D motion estimation with its true scale. We evaluated our approach using image sequences acquired while a person walks. Our experiments show that the proposed approach is able to cope with scale drift.
We tackle this problem by extending previous work that stores the map as a hash table containing occupied voxels at multiple resolutions. We apply Bloom filters to the problem of spatial querying and voxel maps for the example application of SLAM. Their efficacy is demonstrated building 3D maps with both simulated and real 3D point cloud data. Looking up whether a voxel is occupied is three times faster than the hash table and within 10% of the speed of querying a dense 3D array, potentially the upper limit to query speed. Map generation was done with scan to map alignment on simulated depth images, for which the true pose is available. The calculated poses exhibited sub-voxel error of 0.02m and 0.3 degrees for a typical indoor scene with a map resolution of 0.04m.
In this paper, a probabilistic version of the Generalized Voronoi Diagram (GVD), called the PGVD, is used to determine the relative transformation between maps and fuse them. The new method is effective for finding relative transformations quickly and reliably. In addition, the novel approach accounts for all map uncertainties in the fusion process.
the k' closest nodes to a specified node and then selects k
of those nodes at random. Intuitively, LocalRand attempts to benefit from random sampling while maintaining the higher levels of local planner success inherent to selecting more local neighbors. We provide a methodology for selecting the parameters k and k', and in an experimental comparison for both rigid and articulated robots show that LocalRand results in roadmaps with improved connectivity, with comparable computation cost, over the traditional k-closest and a purely random neighbor selection policy.
Recent micro electro-mechanical microphones in conjunction with reconfigurable logic tackle the weight, size, power consumption and cost constraints of robotic systems. A novel, automatic array shape calibration algorithm is developed for 2D and 3D arrays to face common experimental problems such as reverberation and poor signal-to-noise ratio when calibrating the array. The special case of a 2D array calibrated using sources moving in 3D is addressed. No prior information on array geometry is required, the process is fully automated and does not require any specific calibration equipment.
The example application of an acoustic camera is presented as a proof of concept. High-quality acoustic images are computed in real-time by generalized inverse beamforming. This demonstrates the effectiveness of the proposed design and illustrates the usefulness of such sensing capabilities for various robotic applications.
Results are shown from tests performed on benchmark data sets and real-world experiments with multiple robotic platforms.
Within this work a state-space representation that forms a locally singularity-free Atlas of the admissible configuration-space is presented. Based on this state-space description a switching based controller is developed that incorporates the former singular regions into the used configuration space and thus allows to exploit the full flexibility of non-holonomic, omnidirectional undercarriages. The implemented controller is quantitatively and qualitatively evaluated and compared to one approach that avoids the singular regions and one that completely neglects the non-holonomic bindings.
The leader is a Parrot AR.Drone which is controlled by an iPad App utilizing the visual odometry provided by the quadrocopter and pilots it autonomously. The follower is an Asctec Hummingbird which is controlled by an onboard 8-bit microcontroller. Neither communication nor external sensors are required. A custom-built pan/tilt unit and the camera of a Nintendo Wii remote tracks a pattern of infrared lights and allows for online pose estimation. A base station allows for monitoring the behavior but is not required for autonomous flights.
Our efficient solution of the perspective-3-point problem allows for estimating the pose of the camera relative to the pattern in six degrees of freedom at a high frequency on the microcontroller. The presented experiments include a scenario in which the follower follows the leader with a constant distance of two meters flying different shapes in narrow, GPS-denied indoor environment.
Here a proof-of-concept prototype is introduced that shows the feasibility of learning single strokes. For the proof-of-concept prototype only the thickness of the stroke is learned and it is assumed that the thickness of the line is only depending on the distance between the brush and the paper, i.e. on the z-coordinate of the robot's end effector.
On a Nao humanoid, we apply kinesthetic teaching to learn single stepping motions for the ramp. As we show in the experiments, by using the learned motions and integrating monocular vision and inertial data, the Nao is able to autonomously walk down a 2.10 m long ramp at an inclination of 20 degrees. The accompanying video shows the complete process of locating the beginning of the ramp using visual observations, walking down with regular corrections based on the inertial data, and finally determining the end of the ramp by detecting the ending edge before exiting the ramp.