Model-free learning on robot kinematic chains using a nested multi-agent topology

This paper proposes a model-free learning scheme for the developmental acquisition of robot kinematic control and dexterous manipulation skills. The approach is based on a nested-hierarchical multi-agent architecture that intuitively encapsulates the topology of robot kinematic chains, where the activity of each independent degree-of-freedom (DOF) is finally mapped onto a distinct agent. Each one of those agents progressively evolves a local kinematic control strategy in a game-theoretic sense, that is, based on a partial (local) view of the whole system topology, which is incrementally updated through a recursive communication process according to the nested-hierarchical topology. Learning is thus approached not through demonstration and training but through an autonomous self-exploration process. A fuzzy reinforcement learning scheme is employed within each agent to enable efficient exploration in a continuous state–action domain. This paper constitutes in fact a proof of concept, demonstrating that global dexterous manipulation skills can indeed evolve through such a distributed iterative learning of local agent sensorimotor mappings. The main motivation behind the development of such an incremental multi-agent topology is to enhance system modularity, to facilitate extensibility to more complex problem domains and to improve robustness with respect to structural variations including unpredictable internal failures. These attributes of the proposed system are assessed in this paper through numerical experiments in different robot manipulation task scenarios, involving both single and multi-robot kinematic chains. The generalisation capacity of the learning scheme is experimentally assessed and robustness properties of the multi-agent system are also evaluated with respect to unpredictable variations in the kinematic topology. Furthermore, these numerical experiments demonstrate the scalability properties of the proposed nested-hierarchical architecture, where new agents can be recursively added in the hierarchy to encapsulate individual active DOFs. The results presented in this paper demonstrate the feasibility of such a distributed multi-agent control framework, showing that the solutions which emerge are plausible and near-optimal. Numerical efficiency and computational cost issues are also discussed.     

End‐Point Regulation and Vibration Suppression of a Flexible Robotic Manipulator

In this paper, end‐point regulation and vibration suppression are investigated for a flexible robotic manipulator subject to the external disturbances. The dynamics of the flexible robotic manipulator is represented by one partial differential equation (PDE) and five ordinary differential equations (ODEs). Based on the Lyapunov's direct method, boundary control is developed to drive the manipulator to the desired set‐point and simultaneously suppress the vibrations of the flexible manipulator. Considering the unknown spatiotemporally varying disturbance, uniform boundedness of the closed‐loop system is achieved. The control performance of the closed‐loop system is guaranteed by suitably choosing the design parameters. Simulations are provided to illustrate the effectiveness of the proposed control.     

Towards Focused Plan Monitoring: A Technique and an Application to Mobile Robots

Until recently, techniques for AI plan generation relied on highly restrictive assumptions that were almost always violated in real-world environments; consequently, robot designers adopted reactive architectures and avoided AI planning techniques. Some recent research efforts have focused on obviating such assumptions by developing techniques that enable the generation and execution of plans in dynamic, uncertain environments. In this paper, we discuss one such technique, rationale-based monitoring, originally introduced by Veloso, Pollack, and Cox (Proceedings for the Fourth International Conference on AI Planning Systems, Pittsburgh, PA, 1998, pp. 171–179) and we describe our use of it in a simple mobile robot environment. We review the original approach, describe how it can be adapted for a causal-link planner, and provide experimental results demonstrating that it can lead to improved plans without consuming excessive overhead. We also describe our use of rationale-based monitoring in a mobile robot office-assistant project currently in progress.     

Industrial Robot Navigation and Obstacle Avoidance Employing Fuzzy Logic

This paper proposes a novel conceptual approach based on fuzzy logic to solve the local navigation and obstacle avoidance problem for industrial 3-dof robotic manipulators. The proposed system is divided into separate fuzzy units, which control individually each manipulator link. The fuzzy rule-base of each unit combines a repelling influence, which is related to the distance between the manipulator and the nearby obstacles, with the attracting influence produced by the angular difference, between the actual and the final manipulator configuration, to generate a new actuating command for each link. It can be considered as an on-line local navigation method for the generation of instantaneous collision-free trajectories. The strategy has been successfully applied to manipulators in different simulated workspace environments providing collision-free paths. Some of the simulation results obtained are included.     

Optimal design and motion control of biomimetic robotic fish

This paper is concerned with the design, optimization, and motion control of a radiocontrolled, multi-link, free-swimming biomimetic robotic fish based on an optimized kinematic and dynamic model of fish swimming. The performance of the robotic fish is determined by both the fish＇s morphological characteristics and kinematic parameters. By applying ichthyologic theories of propulsion, a design framework that takes into consideration both mechatronic constraints in physical realization and feasibility of control methods is presented, under which a multiple linked robotic fish that integrates both the carangiform and anguilliform swimming modes can be easily developed. Taking account of both theoretic hydrodynamic issues and practical problems in engineering realization, the optimal link-lengthratios are numerically calculated by an improved constrained cyclic variable method, which are successfully applied to a series of real robotic fishes. The rhythmic movements of swimming are driven by a central pattern generator （CPG） based on nonlinear oscillations, and up-and-down motion by regulating the rotating angle of pectoral fins. The experimental results verify that the presented scheme and method are effective in design and implementation.     

Robust tracking control design for uncertain robotic systems with persistent bounded disturbances

In this study, a robust nonlinear L_∞‐gain tracking control design for uncertain robotic systems is proposed under persistent bounded disturbances. The design objective is that the peak of the tracking error in time domain must be as small as possible under persistent bounded disturbances. Since the nonlinear L_∞ ‐gain optimal tracking control cannot be solved directly, the nonlinear L_∞ ‐gain optimal tracking problem is transformed into a nonlinear L_∞ ‐gain tracking problem by given a prescribed disturbance attenuation level for the L_∞ ‐gain tracking performance. To guarantee that the L_∞ ‐gain tracking performance can be achieved for the uncertain robotic systems, a sliding‐mode scheme is introduced to eliminate the effect of the parameter uncertainties. By virtue of the skew‐symmetric property of the robotic systems, sufficient conditions are developed for solving the robust L_∞ ‐gain tracking control problems in terms of an algebraic equation instead of a differential equation. The proposed method is simple and the algebraic equation can be solved analytically. Therefore, the proposed robust L_∞ ‐gain tracking control scheme is suitable for practical control design of uncertain robotic systems. Copyright © 2008 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

A distributed control scheme for multiple robotic vehicles to make group formations

This paper presents a novel distributed control scheme of multiple robotic vehicles. Each robotic vehicle in this scheme has its own coordinate system, and it senses its relative position and orientation to others, in order to make group formations. Although there exists no supervisor and each robotic vehicle has only relative position feedback from the others in the local area around itself, all the robotic vehicles are stabilized, which we have succeeded in proving mathematically only in the cases where the attractions between the robots are symmetrical. Each robotic vehicle especially has a two-dimensional control input referred to as a “formation vector” and the formation is controllable by the vectors. The validity of this scheme is supported by computer simulations.     

Shape Morphing-Based Control of Robotic Visual Servoing

We present an approach for controlling robotic interactions with objects, using synthetic images generated by morphing shapes. In particular, we attempt the problem of positioning an eye-in-hand robotic system with respect to objects in the workspace for grasping and manipulation. In our formulation, the grasp position (and consequently the approach trajectory of the manipulator), varies with each object. The proposed solution to the problem consists of two parts. First, based on a model-based object recognition framework, images of the objects taken at the desired grasp pose are stored in a database. The recognition and identification of the grasp position for an unknown input object (selected from the family of recognizable objects) occurs by morphing its contour to the templates in the database and using the virtual energy spent during the morph as a dissimilarity measure. In the second step, the images synthesized during the morph are used to guide the eye-in-hand system and execute the grasp. The proposed method requires minimal calibration of the system. Furthermore, it conjoins techniques from shape recognition, computer graphics, and vision-based robot control in a unified engineering amework. Potential applications range from recognition and positioning with respect to partially-occluded or deformable objects to planning robotic grasping based on human demonstration.     

Design and Testing of a Robotic System for mr Image-guided Interventions

The effective integration of robotics together with magnetic resonance (mr) technology is expected to facilitate the real-time guidance of various diagnostic and therapeutic interventions. Specially designed robotic manipulators are required for this purpose, the development of which is a challenging task given the strong magnetic fields and the space limitations that characterize the mr scanning environment. A prototype mr-compatible manipulator is presented, designed to operate inside cylindrical mr scanners. It was developed for the study of minimally invasive diagnostic and therapeutic interventions in the abdominal and thoracic area with real-time mr image guidance. Initial tests were performed inside a high-field clinical mr scanner and included mr-compatibility tests and phantom studies on image-guided targeting.     

微操作机器人的视觉伺服控制

视觉伺服控制是微操作机器人实现精确运动,完成自动操作的必要手段．本文介绍 了实现微操作机器人视觉伺服控制的方法．首先论述了微操作机器人的视觉伺服结构,并以 建立的面向生物工程的双手微操作机器人系统为例,介绍了基于二维显微视觉信息的三自由 度柔性铰链微操作机器人的运动学建模方法,针对压电驱动器控制器的特点提出了基本的PI D视觉伺服控制规律实现方法,并进行了点到点运动和圆轨迹跟踪实验．实验结果表明,视 觉伺服控制克服了由于标定以及环境等因素导致的运动模型不准确而引入的误差．