This work considers the problem of force/position regulation for a robotic finger in compliant contact with an unknown curved surface resulting in uncertain force and position control subspaces. The proposed controller is an adaptive control scheme of a simple structure that achieves the desired target by the on-line tuning of the position and force control actions to their corresponding actual subspaces at the desired point using motion state feedback. The local asymptotic stability of the system equilibrium point is proved and an estimate of the region of attraction is given. The controller performance is illustrated by a simulation example.

In this study, we consider a model based robust control scheme for kinematically redundant robot manipulators that also enables the use of self motion of the manipulator to perform multiple sub-tasks (e.g., maintaining manipulability, avoidance of mechanical joint limits, and obstacle avoidance). The controller proposed ensures uniformly ultimately bounded end-effector and sub-task tracking despite the parametric uncertainty associated with the dynamic model. A Lyapunov based approach has been utilized in the controller design and extension to a non minimum set of parameters for orientation representation has been presented to illustrate the flexibility of the approach. Extensive simulation studies performed initially on a 3 link planar robot arm (for the planar case) and on a six degree of freedom (DOF) Puma type robot arm (for the 3D case with quaternion feedback) are presented to demonstrate the capabilities and the performance of the controller. The results were then experimentally tested on an actual Puma 560 robot to illustrate the feasibility of the proposed method.

Operational space control of industrial robots is addressed in this document. We analyze a two-loop hierarchical control with the resolved motion rate controller (RMRC) as outer loop and the joint velocity PI controller as inner loop; the latter is the typical velocity controller used in industrial robots. We prove, by the first time, that these simple controllers make the solutions of the closed-loop system uniformly ultimately bounded. Additionally, we give some simple guidelines for the selection of the control gains so as to ensure an explicit bound of the tracking error.

The paper is concerned with the relationship between the equation of elastic line motion, the “Euler-Bernoulli approach” (EBA), and equation of motion at the point of elastic line tip, the “Lumped-mass approach” (LMA). The Euler–Bernoulli equations (which have for a long time been used in the literature) should be expanded according to the requirements of the motion complexity of elastic robotic systems. The Euler–Bernoulli equation (based on the known laws of dynamics) should be supplemented with all the forces that are participating in the formation of the bending moment of the considered mode. This yields the difference in the structure of Euler–Bernoulli equations for each mode. The stiffness matrix is a full matrix. Mathematical model of the actuators also comprises coupling between elasticity forces. Particular integral of Daniel Bernoulli should be supplemented with the stationary character of elastic deformation of any point of the considered mode, caused by the present forces. General form of the elastic line is a direct outcome of the system motion dynamics, and can not be described by one scalar equation but by three equations for position and three equations for orientation of every point on that elastic line. Simulation results are shown for a selected robotic example involving the simultaneous presence of elasticity of the joint and of the link (two modes), as well the environment force dynamics.

A general new methodology using evolutionary algorithms viz., Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) and Multi-objective Differential Evolution (MODE), for obtaining optimal trajectory planning of an industrial robot manipulator (PUMA 560 robot) in the presence of fixed and moving obstacles with payload constraint is presented. The problem has a multi-criterion character in which six objective functions, 32 constraints and 288 variables are considered. A cubic NURBS curve is used to define the trajectory. The average fuzzy membership function method is used to select the best optimal solution from Pareto optimal fronts. Two multi-objective performance measures namely solution spread measure and ratio of non-dominated individuals are used to evaluate the strength of Pareto optimal fronts. Two more multi-objective performance measures namely optimiser overhead and algorithm effort are used to find computational effort of the NSGA-II and MODE algorithms. The Pareto optimal fronts and results obtained from various techniques are compared and analysed. Both NSGA-II and MODE are best for this problem.

Knowledge of the physical properties of terrain surrounding a planetary exploration rover can be used to allow a rover system to fully exploit its mobility capabilities. Terrain classification methods provide semantic descriptions of the physical nature of a given terrain region. These descriptions can be associated with nominal numerical physical parameters, and/or nominal traversability estimates, to improve mobility prediction accuracy. Here we study the performance of multisensor classification methods in the context of Mars surface exploration. The performance of two classification algorithms for color, texture, and range features are presented based on maximum likelihood estimation and support vector machines. In addition, a classification method based on vibration features derived from rover wheel–terrain interaction is briefly described. Two techniques for merging the results of these “low-level” classifiers are presented that rely on Bayesian fusion and meta-classifier fusion. The performance of these algorithms is studied using images from NASA's Mars Exploration Rover mission and through experiments on a four-wheeled test-bed rover operating in Mars-analog terrain. Also a novel approach to terrain sensing based on fused tactile and visual features is presented. It is shown that accurate terrain classification can be achieved via classifier fusion from visual and tactile features.

A two-channel bilateral controller is proposed for teleoperation systems. The controller takes into account both the free space motion and the constrained motion. Specifically, the force-position (F-P) architecture is applied during the constrained motion, while the position-position (P-P) architecture is applied during the free space motion. Theoretically, perfect transparency can be achieved. In addition, it does not need to switch the control modes of the master and the slave controllers during the transition between the free space motion and the constrained motion. Experiments are conducted to validate the analysis, and to demonstrate the effectiveness of the proposed bilateral controller. The limitations of the proposed controller are also discussed.

This paper is organized in two parts. In Part I, the wrench polytope concept is presented and wrench performance indices are introduced for planar parallel manipulators (PPMs). In Part II, the concept of wrench capabilities is extended to redundant manipulators and the wrench workspace of different PPMs is analyzed. The end-effector of a PPM is subject to the interaction of forces and moments. Wrench capabilities represent the maximum forces and moments that can be applied or sustained by the manipulator. The wrench capabilities of PPMs are determined by a linear mapping of the actuator output capabilities from the joint space to the task space. The analysis is based upon properly adjusting the actuator outputs to their extreme capabilities. The linear mapping results in a wrench polytope. It is shown that for non-redundant PPMs, one actuator output capability constrains the maximum wrench that can be applied (or sustained) with a plane in the wrench space yielding a facet of the polytope. Herein, the determination of wrench performance indices is presented without the expensive task of generating polytopes. Six study cases are presented and performance indices are derived for each study case.

This part of the paper investigates the wrench capabilities of redundantly actuated planar parallel manipulators (PPMs). The wrench capabilities of PPMs are determined by mapping a hypercube from the torque space into a polytope in the wrench space. For redundant PPMs, one actuator output capability constrains the wrench space with a smaller polytope that is contained inside the overall polytope. Performance indices are derived from six study cases. These indices are employed to analyze the wrench workspace for constant orientation of the mobile platform of the non-redundant 3-RRR PPM, and actuation redundant 4-RRR and 3-RRR PPMs, where the underline indicates the actuated joints. A comparison of the results shows that both of the redundantly-actuated PPMs give better wrench capabilities than the non-redundant PPM. However, it is shown that scaled for the operational cost (wrench capabilities divided by total actuation output) the non-redundant 3-RRR PPM provides the highest maximum reachable force, the 3-RRR PPM produces the highest isotropic force, and the 4-RRR yields the highest reachable moment.

This paper describes a hybrid approach to a fast and accurate localization method for legged robots based on Fuzzy-Markov (FM) and Extended Kalman Filters (EKF). Both FM and EKF techniques have been used in robot localization and exhibit different characteristics in terms of processing time, convergence, and accuracy. We propose a Fuzzy-Markov–Kalman (FM–EKF) localization method as a combined solution for a poor predictable platform such as Sony Aibo walking robots. The experimental results show the performance of EKF, FM, and FM-EKF in a localization task with simple movements, combined behaviors, and kidnapped situations. An overhead tracking system was adopted to provide a ground truth to verify the performance of the proposed method.