Keywords: Neural networks
Abstract: Piezoelectric phenomenon is commonly used in microsystems. Many sensors as well as actuators are based on this phenomenon. Because of the nonlinear character of the piezo phenomenon, exact measuring of fast dynamic systems is difficult with piezoelectric sensors. Piezo-based actuators on the other hand need feedback for the exact motion. This has increased the size of the system as well as the power consumption, which are undesirable characteristics in microworld. In this paper a solution for the problem is determined by modeling. First, a third order transfer function is generated to model the piezoactuator at the operating point. The parameters of a grey box-model are implemented as dynamic, because of the nonlinearity of the piezo actuator. This is the way to capture the characters of the transfer function to fit the real actuator at each operating point. A multilayer perception neural network is used to model the behavior of the system. The training data for the network is measured at different operating points. The model is validated by test data at different operating points. The agreement with the model and the measurements is excellent.

Keywords: Neural networks
Abstract: Electronic mail is inarguably the most widely used Internet technology today. With the massive amount of information and speed the Internet is able to handle, communication has been revolutionized with email and other online communication systems. However, some computer users have abused the technology used to drive these communications, by sending out thousands and thousands of spam emails with little or no purpose other than to increase traffic or decrease bandwidth. This paper evaluates the effectiveness of email classifiers based on the feedforward backpropagation neural network and Baysian classifiers. Results are evaluated using accuracy and sensitivity metrics. The results show that the feed-forward backpropagation network algorithm classifier provides relatively high accuracy and sensitivity that makes it competitive to the best known classifiers. On the other hand, though Baysian classifiers are not as accurate they are very easy to construct and can easily adapt to changes in spam patterns.

Keywords: Neural networks, Adaptive and learning control
Abstract: In this paper, we introduced new adaptive learning algorithms and related networks to extract optimal features from multidimensional data in order to reduce the data dimensionality while preserving class separability. For this purpose, new adaptive algorithms for the computation of the square root of the inverse covariance matrix are introduced. We introduce a new cost function related to the given adaptive learning algorithms in order to prove their convergence. Self organized networks are constructed based on these algorithms. By cascading network and an adaptive principal component analysis (APCA) network, we present new adaptive self organized LDA feature extraction network. Adaptive nature of the new optimal feature extraction method makes it appropriate for on-line incremental pattern classification and machine learning applications. Both networks in the proposed structure are trained simultaneously, using a stream of input data. Existence of cost function, make it available to compute learning rate efficiently in every iteration in order to increase the convergence rate. Experimental results using synthetic multi-class multi-dimensional sequence of data, demonstrated the effectiveness of the new adaptive self organized feature extraction networks.

Keywords: Neural networks, Adaptive and learning control, Soft computing
Abstract: In supervised training, we often try to find out a set of weights and biases for a pattern recognition neural network in order to classify all patterns in a training data set. However, it would be difficult if the neural network was not big enough for learning a large training data set. In this paper, we propose a training method and a design of pattern recognition neural network that is not big but still able to classify all the training patterns exactly. The neural network is designed with a reject output to separate the training data set into some parts for classifying more easily. The training method helps the neural network to find out not only one but many sets of weights and biases for classifying all the training patterns, controlling the recognizing rejection and reducing the error rate. On the other hand, with this design we can reduce the size of the neural network implemented on a FPGA chip in order to make fast smart sensors for the robots.

Keywords: Neural networks, Computer vision, Adaptive and learning control
Abstract: In this paper, we present a new incremental face recognition (IFR) system based on new adaptive learning algorithms and networks. We introduce new adaptive linear discriminant analysis (LDA) algorithm and related network for optimal facial feature extraction and use them to construct a new IFR system. Convergence proof of all algorithms is given using an appropriate cost function and discussing about its initial conditions. Application of the new IFR on feature extraction from facial image sequences is given in two steps: i) image preprocessing, which includes normalization, histogram equalization, mean centering and background removal, ii) adaptive LDA feature estimation. In the preprocessing stage, all input images are cropped and prepared for the next step. Outputs of the preprocessing stage are used as a sequence of inputs for IFR system. The proposed system was tested on YALE face database. Experimental results on this database demonstrated the effectiveness of the proposed system for adaptive estimation of the feature space for online face recognition.

Keywords: Manipulator control, Sensor fusion, Robot manipulators
Abstract: This paper analyzes the signals captured during impacts and vibrations of a mechanical manipulator. To test the impacts, a flexible beam is clamped to the end-effector of a manipulator that is programmed in a way such that the rod moves against a rigid surface. Several signals are captured and theirs Fourier Transforms are calculated and approximated by trendlines based on a power law formula. A sensor classification scheme based on the frequency spectrum behavior is presented.

Keywords: Robot manipulators, Manipulator control
Abstract: Parametric identification consists in estimating values of physical parameters of systems. The most popular methods consist in using the least squares regressions because of their simplicity. However, we know that these techniques are not intrinsically robust. An alternative consists in justifying the use of the LS technique. This paper focuses on this issue and introduces a derivation of the CESTAC method which will be applied to a 6 degrees of freedom (DOF) serial robot.

Keywords: Microhandling and -manipulation, Robot manipulators
Abstract: Flexure joints are frequently used in precision mechanisms, from motion stages to micro-robots. Due to their monolithic construction and superior wear and loss properties, flexure joints can be used to reduce the mechanism size and increase the positioning accuracy. The compliance of flexure joints, however, can affect the static and dynamic characteristics of the overall mechanism. The objective of this paper is to develop analysis and design tools for general platform type parallel mechanisms that contain flexure joints. In this paper, a micro planar parallel mechanism that has 3-DOF motion is examined by considering the configuration change effect. Usually, MEMS mechanism is designed for a specific purpose with a motion restriction. Here, using the conventional planar parallel robotics

Keywords: Robot manipulators
Abstract: Parametric identification requires a good know-how and an accurate analysis. The most popular methods consist in using simply the least squares techniques because of their simplicity. However, these techniques are not intrinsically robust. An alternative consists in helping them with an appropriate data treatment while another choice consists in applying a robust identification method. This paper focuses on an analysis of a robust method called “the simple refined instrumental variable method”. This method will be applied to a single degree of freedom master arm developed by the CEA Interactive Robotic Unit.

Keywords: Robot manipulators
Abstract: The Interactive Robotics Unit of CEA LIST has developed a very challenging robotic carrier (called P.A.C.) which is able to perform high range intervention tasks inside blind hot cells. This long reach multi-link carrier has 11 degrees of freedom (DOF) and weighs less than 30 kg. Due to its size and weight, this large robot manipulator holds lots of elastic and geometric deformations. Hence, it presents very low position accuracy. A mechanical model is developed to take into account the flexibilities of the structure. A calibration method of the robot’s parameters is used to further reduce the positioning error of the end effector. This method is first tested by simulation. After encouraging simulation results, an experimental field is made for the calibration of the PAC manipulator. Results show that the adopted flexible model, with the new calibrated parameters, is indeed able to correct and reduce the system errors.

Keywords: Modular robotics, Biologically inspired robotics, Mobile robotics
Abstract: The ability to self-replicate is one of the distinctive features of living organisms. Robots capable of self-replication would have a profound impact on the field of robotics by improving lifetime and robustness. In the past our lab has built several prototypes of self-replicating robotic systems including semi-autonomous and fully-autonomous robots with microprocessor-based control, and a self-replicating electromechanical circuit composed of basic electronic elements (transistors, resistors, etc.). These previous efforts demonstrated that man-made systems with simple behaviors are capable of self-replication. Extending our previous results, in this paper, we present an autonomous self-replicating robotic system with distributed electronic components in a structured environment. Using simple discrete electronic components allows for a more uniform decomposition of each robot into simpler parts than for microprocessor-controlled systems. Ultimately, we would like to demonstrate robots that replicate from the most basic parts, and this paper represents one more step toward acheiving this goal.

Keywords: Modular robotics
Abstract: Ubiquitous Robotic Companion (URC) is defined as a networked-based intelligent service robot and it is in the early stage on commercial. The service quality and safety of URC should be guaranteed because URC provides various services to a user at close hand compare to industrial robot. This paper focuses on a test specification and certification system of URC to guarantee the service quality and safety.

Keywords: Co-operative robotics, Robot manipulators, Manipulator control
Abstract: The paper focuses on design and control of a new anthropomorphic robot arm enabling torque measurement in each joint to ensure safety while performing tasks of physical interaction with human and environment. When the contact of arm with an object occurs, local impedance algorithm provides active compliance of corresponding robot arm joint. Thus, the whole structure of the manipulator can safely interact with unstructured environment. A novel variable control strategy was elaborated to increase the robot functionality as well as to archive humanlike dynamics of interaction. In the paper, we describe the detailed design procedure of 4-DOF robot arm and optical torque sensors. The experimental results of variable joint impedance control show that proposed approach provides not only safe interaction of entire structure of robot arm with human being, but also improves the effectiveness of contact task performance, and enables to contact with environment in delicate manner.

Keywords: Sensor fusion, Computer vision, Mobile robotics
Abstract: This paper describes a visual simultaneous localization and mapping (VSLAM) method for a pipe inspection robot that can serve as a carrier for nondestructive testing (NDT) sensors inside in-service water mains. The proposed method features a multi-sensor fusion system that simultaneously executes two important functions of the robot: construct a global image of the internal surface of the pipe and provide the location of the NDT sensors. The fusion involves a synergy between image mosaicing and dead reckoning. More precisely, the sequential images of a digital camera are stitched together with the help of an inertial navigation system (INS). In the proposed model, the processing load of image mosaicing is reduced significantly, and at the same time the accumulation of error of the INS is prevented. The functionality of the model is verified using a simulator that mimics the motion of a pipe inspection robot inside a water main.

Keywords: Soft computing, Evolutionary computing, Adaptive and learning control
Abstract: A heuristic search for stable high-order delta sigma modulators is presented. Searches guided by the Linear Model found stable modulators with higher SNR performance compared to classical filter design techniques. The presented design approach is both more automatic than classical design and can also be weighted such to direct the search toward designs for special purpose. A novel adaptive modulator concept is also presented.

Keywords: Manipulator control, Microhandling and -manipulation, Robot manipulators
Abstract: Actuators using advance materials like piezoelectric and shape memory alloy are gaining popularity in applications involving high frequency, high precision and also where there’s a need in compactness. As time is required for the switching of polarization, the phenomena hysteretic behavior of these materials changes with rate. Most present hysteresis models are based on rate-independent assumption and cannot be applied for non-periodic applications. To make matters worse, the hysteresis actually becomes ill-conditioned when the velocity is high at the turning point. This paper proposes a phenomena rate-dependent model using a modified Prandtl-Ishlinskii (PI) operator without singularity to model the behavior of piezoelectric actuators, even when subjected to varying frequency signals. Past work had shown that the weights of the Prandtl-Ishlinskii operators vary linearly with velocity when the velocity is less than 900µm/s. In this paper, experimental data are obtained to confirm the findings. As the first weight becomes negative when operating at higher frequencies, the threshold value has to be kept large to avoid the singularity problem when computing the inverse Prandtl-Ishlinskii model. Similar ill-conditioned problems also arise when the actuators are subjected to heavy loads. Thus, this paper proposes extensions to the PI operator by mapping the hysteresis data through a linear transformation onto a domain where the singularity problem is removed. The inverse weights are obtained and subsequently used to compute the inverse hysteresis model and implemented as an open-loop feed-forward control of a piezoelectric actuator.

Keywords: Manipulator control
Abstract: This paper provides a minimum time algorithm to intercept an object on the conveyor belt by robotic manipulator. The goal is that the robot is able to intercept objects on a conveyor line moving at a given speed in minimum time. In order to formulate the problem, we introduce the robot and object arrival time functions, and conclude that the optimal point occurs at the intersection of these two functions. The search algorithm for finding the intersection point between the robot and object arrival time functions are also presented to find the optimal point in real-time.

Keywords: Manipulator control, Adaptive and learning control, Intelligent systems
Abstract: This article presents an approach to Programming by Demonstration (PbD) to simplify programming of industrial manipulators. By using a set of task primitives for a known task type, the demonstration is interpreted and a manipulator program is automatically generated. A pick-and-place task is analyzed, based on the velocity profile, and decomposed in task primitives. Task primitives are basic actions of the robot/gripper, which can be executed in a sequence to form a complete a task. For modeling and generation of the demonstrated trajectory, fuzzy time clustering is used, resulting in smooth and accurate motions. To illustrate our approach, we carried out our experiments on a real industrial manipulator.

Keywords: Manipulator control
Abstract: This paper addresses the design method for robust PID like controllers which guarantee the quadratic stability, performance in terms of H_2 and H_infty specifications, pole locations and maximum output control. The approach is based on the transformation of the PID controller design problem to that of state feedback controller design thereby the convex optimization approaches can be adapted. Real time experimental results on a double inverted pendulum system demonstrates the validity and applicability of the proposed approach.

Keywords: Manipulator control, Robot manipulators
Abstract: Researchers have studied teleoperation systems essentially for implementation in critical tasks. Although teleoperation systems require stability apriori, time delays over the communications lines become a major source of instability. During the last two decades several researchers have focused on the time delay problem. The wave variable technique is one of the teleoperation controllers developed to ensure stability under time delays. In this paper, a modified algorithm of the wave variable technique is introduced. The purpose of the modification is to ensure that the slave manipulator tracks the position demand of the master even if there is a communication loss or the wave variable technique fails for a limited time period.

Keywords: Health-care robotics, Biologically inspired robotics, Evolutionary computing
Abstract: An approach to modeling and signal processing of the human heart is presented. First the anatomy of the heart is described and the anatomical functions performed are outlined. A preliminary model of the cardio-systemic circuit system is described. The aim is to further refine the model to correlate to measured responses and characterization of those responses. The APB signal analysis method is quantified and empirical examples are provided and analyzed.

Keywords: Health-care robotics, Intelligent systems, Mobile robotics
Abstract: Hospitals face with a heavy traffic of goods everyday, where transportation tasks are mainly carried by human. Analysis of the current situation of transportation in a typical hospital showed several transportation tasks are suitable for automation. This paper presents a system, consisting of a fleet of robot vehicles, automatic stations and smart containers for automation of transportation of goods in hospitals. Design of semi-autonomous robot vehicles, containers and stations are presented and the overall system architecture is described. Implementing such a system in an existing hospital showed the need of necessary modifications to the hospital infrastructure.

Keywords: Health-care robotics, Manipulator control, Adaptive and learning control
Abstract: In order to reduce fluoroscope usage in endovascular surgery there is a need to develop autonomous catheter insertion systems. We propose a system for tracking the position of the catheter using a magnetic motion capture sensor to provide feedback to a catheter driving mechanism to perform autonomous catheter insertion in major vasculature. Catheter insertion path reconstruction experiments were performed with the system inside silicone model of major vasculature to simulate surgery. As result the system reproduced a path inside the silicone blood vessel phantom with less than 7 mm of error. We found that error in path reconstruction depends on the model’s cross-section diameter, the properties of the catheter insertion mechanism, the magnetic sensor and the system guidance technique.

Keywords: Co-operative robotics, Health-care robotics, Agent technology
Abstract: An unmanned robotic operating room consists of a tele-operated surgical robot and various other subsystems that perform surgery support functions such as dispensing tools and supplies. Execution Engine for such an operating room has to coordinate the subsystems to perform surgery support functions. A support function consists of a series of tasks along with other constructs such as conditional statements. A task, in turn, is composed of subsystem level actions connected by complex synchronization constraints. Previous research has focused on developing a language for specifying tasks and modeling the constraints between them. We present an approach where a task is specified as a data-structure rather than a language. This has the advantage of implicitly encoding the synchronization constraints in the data-structure itself, making it easy to write, debug and maintain task programs. We model a task as a directed acyclic graph and propose an algorithm to execute the task graph. The algorithm is general and language independent, thus eliminating the need for a special purpose language to specify tasks. We discuss implementation results on an emulated system.

Keywords: Health-care robotics, Manipulator control, Robot manipulators
Abstract: Disturbance / vibration reduction is critical in many applications. One example is disk drive industry. With the advance in optical disk drive technology, the off-track or off-focusing error caused by disturbance is limiting the performance. Instead of going with the more familiar approach like vibration absorber, a real-time disturbance compensation without much phase lag is proposed. In this paper, to illustrate the idea of real-time compensation, a one dof cancellation of tremor is performed. Instantaneous motion due to the disturbance is sensed by an accelerometer inertial measurement unit (IMU). Modeling of the accelerometer is done to provide better readings. The compensation motion is provided by manipulating a piezoelectric-driven flexure-based mechanism. Due to the existence of backslash in revolute joints, flexure-based mechanism is proposed to achieve better result. Piezoelectric actuator is proposed because of its high frequency and precision. The hysteretic non-linearity of the piezoelectric actuator is modeled using Prandtl-Ishlinskii operator. Based on this model, an open-loop inverse feed-forward controller is implemented to accurately provide the required motion.

Keywords: Genetic algorithms
Abstract: Cross Docking is a distribution system in which the merchandise received at a warehouse or distribution center is not stocked but immediately prepared for onward shipment. In other words, cross docking is the transfer of inward deliveries from the point of reception directly to the point of delivery with limited or no interim storage. One way to reduce cost in a cross dock terminal is to park incoming and outgoing trucks so that the loads can be efficiently moved across the dock. This means that the distance from loading and unloading is the shortest possible distance. The problem with generating an efficient schedule for the door assignments is that for n incoming and m outgoing trucks there are n!*m! possible solutions.

Keywords: Genetic algorithms, Evolutionary computing
Abstract: A hybrid and flexible genetic algorithm (HybFlexGA) is presented for the job-shop scheduling problem (JSSP). The JSSP is one of the most difficult production scheduling problems in industry because it requires very large combinational search space and the precedence constraint between machines. The computational results on the well-known benchmark instances show the proposed HybFlexGA is very effective and competitive with other methods presented in the literature.

Keywords: Genetic algorithms, Mobile robotics, Soft computing
Abstract: Autonomous vehicles travel through a varying environment that is not limited to the physical terrain but also includes the “RF terrain”. The autonomous vehicle must be able to adapt to the varying RF conditions. “Cognitive radios” are being developed that address this issue. This paper discusses the use of Genetic Algorithms (GA) to implement the adaptive processes for a cognitive radio on an autonomous vehicle. Specifically GA’s are used to solve the optimization of RF parameters for a wireless network. In particular, a Fitness Measure is derived which provides a figure of merit for the performance of the GA in relation to overall RF performance. Additionally, a chromosome structure is derived which consists of “RF genes”. Each gene is a binary string representing some aspect or parameter of the RF environment. Finally the GA determines a set of RF parameters for optimal radio communications in the varying RF environment.

Keywords: Intelligent systems, Genetic algorithms, Agent technology
Abstract: Computational grid offers a great potential solution to parallel meta-heuristics toward combinatorial optimization. However, it is quite difficult for specialists in combinatorial optimization to develop parallel meta-heuristics in extremely heterogeneous computational environment, starting from scratch without any toolkit. This paper presents a Problem Solving Environment for Combinatorial Optimization Based on Parallel Meta-heuristics (PSEPMH) to help specialists to harness heterogene-ous computational resources and handle dynamic granularity control. PSEPMH requires specialist to decompose one problem into two sub-problems with divide-and-conquer framework just as generic sequential algorithm. Then compiler of PSEPMH generates mobile agent code that automatically forms adaptive multi-granularity parallel computing at run-time by cloning himself and distributing along dynamic, com-plex grid environment with the support of PSEPMH. Not only can PSEPMH relieve specialists’ burden, but also make use of the computational resources more efficiently.

Keywords: Soft computing, Sensor fusion
Abstract: Kernel methods are a mathematical tool that provides a generally higher dimensional representation of given data set in feature space for feature recognition and image analysis problems. Typically, the kernel trick is thought of as a method for converting a linear classification learning algorithm into non-linear one, by mapping the original observations into a higher-dimensional non-linear space so that linear classification in the new space is equivalent to non-linear classification in the original space. Moreover, optimal kernels can be designed to capture the natural variation present in the data. In this paper we present the performance results of fifteen novel kernel functions and their respective performance for kernel principal component analysis on five select databases. Empirical results show that our kernels perform as well and better than existing kernels on these databases.

Keywords: Mobile robotics, Intelligent systems, Adaptive and learning control
Abstract: In the former researches [5][6][7], we suggested an algorithm which can be used for deriving the nonlinear inverse kinematic problem in unmanned bicycle system by using iterative method. In this short note, we reinforce the former method and propose a new iterative algorithm for a certain type of nonlinear dynamic equation such as inverse kinematics of bicycle system. The idea of proposed algorithm is similar to Piccard's Iterative method in basic concept. We also propose a robust control strategy for tracking problem in bicycle system based on nonlinear compensation. In this control algorithm, we invent and attach the load mass balance system for the self stabilization with more ease. From the computer simulation results, we can see that the proposed control algorithm can be applied to real system.

Keywords: Mobile robotics, Intelligent systems, Agent technology
Abstract: In order to realize the dual arm mobile robot control and simulation, a distributed robotic system is presented. The system provides the robot’s model building, path planning, robot task planning, simulation and actual robot control functions in an indoor environment. This paper is focused on two major issues: the distributed system’s modularity and the key technologies based on the robot’s key-values in robot construction and multi-robot simulation. Experiments were conducted in order to compare the performances of the proposed system with the performances of a former centralized system. The results show that the distributed system uses less system resources, has better real time performances and satisfies the requests of dual arm mobile robot.

Keywords: Mobile robotics, Intelligent systems, Soft computing
Abstract: Path planning for mobile robots has received a great deal of attention over the past two decades. The basic Rapidly-exploring Random Tree (RRT) algorithm is famous in this field because it excels at exploring free space in the large environments and it is parallelizable. In this paper, we applied modified RRT algorithm to local navigation of a mobile robot in unknown environment. The biased direction from RRT algorithm is used to reduce the nodes and therefore the amount of calculation is decreased for real-time computation. We conducted the simulation and simple real-time experiments to verify the proposed method in real unknown environment.

Keywords: Mobile robotics, Sensor fusion
Abstract: The ANSER project (Airport Night Surveillance Expert Robot) is described, exploiting a mobile robot for autonomous surveillance in civilian airports and similar wide outdoor areas. The paper focuses on the localization subsystem of the patrolling robot, composed of a non-differential GPS unit and a laser rangefinder for map-based localization (inertial sensors are absent). Moreover, it shows that an augmented state vector approach and an Extended Kalman filter can be successfully employed to estimate the colored components in GPS noise, thus getting closer to the conditions for the EKF to be applicable.

Keywords: Mobile robotics, Sensor fusion, Intelligent systems
Abstract: Wireless sensor networks have been widely used in many applications such as environment monitoring, surveillance systems and unmanned space explorations. However, poor deployment of sensor devices leads (1) bad network connectivity which makes data communication or data collection very hard; or (2) redundancy of coverage which wastes energy of sensors and causes redundant data in the network. Thus, in this paper, we propose using a mobile robot to assist the sensor deployment and data collection for unmanned explorations or monitoring. We assume that the robot can carry and deploy the sensor devices, and also have certain communication capacity to collect the data from the sensor devices. Given a set of interest points in an area, we study the following interesting problems: (1) how to decide minimum number of sensor devices to cover all the interest points; (2) how to schedule the robot to place these sensor devices in certain position so that the path of the robot is minimum; and (3) after the deployment of sensors, how to schedule the robot to visit and communicate with these sensor devices to collect data so that the path of the robot is minimum. We propose a complete set of heuristics for all these problems and verify the performances via simulation.