基于Internet的远程控制机器人系统结构设计探讨

主要由框架设计中的硬件与软件两方面着手,分析设计远程控制机器人系统的一些方法与内容,以望能起到抛砖引玉的作用。

METHODS OF IMPROVING COMPUTATIONAL ACCURACY OF ROBOT DYNAMICS

In this paper the geometric meaning of robot systems is expounded based on the theory of multibody system. The error accumulation for the known algorithm is analyzed and the cause of ‘Energy consumption’ is revealed, the relationship between the coefficients of dynamic equation is derived so as to establish the canonical equations. The error accumulation of dynamics can be eliminated by using canonical equations and the symplectic integral method so that the computational accuracy can be ensured effectively. As an example, a planar robotics system is considered.

Self-organized architecture for outdoor mobile robot navigation

This paper proposed a multi-agent based architecture for outdoor mobile robot navigation where event-driven control is used to handle the dynamically changing of the environment. With the support of a distributed communication infrastructure and an event-driven situation evaluation agent, the robot can initiate action adaptive to the dynamical changes in the environment through reorganize its internal architecture. Adaptiveness and feasibility of the proposed architecture is validated through navi- gation experiments on the robot in a variety of natural outdoor environments.

Safety architecture of internet based multi-robot teleoperation system

Safety subsystem is one of the important parts in robot teleoperation system. In this paper, a safety architecture of safety subsystem in Internet based multi-operator-multi-robot (MOMR) teleoperation system is presented. The subsystem is divided into three layers in its logic architecture: interactive monitor layer, collaborative control layer and real-time control layer. The safety problems and the related strategy are clarified by detailed analysis of each layer and relationship among the layers. So we can obtain a high performance MOMR teleoperation system with multi-layer safety architecture.

Metamorphic Algorithm of Self-reconfigurable Modular Robotic System

A self-reconfigurable robot is a non-linear complex system composed of a large number of modules. The complexity caused by non-linearity makes it difficult to solve the problem of module motion planning and shape-changing control with the traditional algorithm. In this paper, a full-discrete metamorphic algorithm is proposed. The modules concurrently process the local sensing information, update their eigenvector, and act by the same predetermined logical rules. Then a reasonable motion sequence for modules and the global metamorphosis can be obtained. Therefore, the complexity of metamorphic algorithm is reduced, the metamorphic procedure is simplified, and the self-organizing metamorphosis can be obtained. The algorithm cases of several typical systems are studied and evaluated through simulation program of 2-D planar homogeneous modular systems.

Test Bench for Robotics and Autonomy： Overview and Test Results

In this paper, the TAS-I （Thales Alenia Space-Italy） Test Bench for Robotics and Autonomy （TBRA） is presented. It is based on a flexible and modular software architecture （Framework Engine）, in which each functional module （representing the GNC subsystems） implements a key functionality of the GNC （Guidance Navigation and Control）. Modules communicate by means of standardised interfaces designed for exchange of necessary information among the modules composing the entire system. This approach permits the interchange-ability of each subsystem without affecting the overall functionalities of the GNC system. In this paper, the TBRA system, together with the implemented functional modules will be described. Tests results will be reported and future development will be discussed.

Three-layer intelligence of planetary exploration wheeled mobile robots:Robint,virtint,and humint

The great success of the Sojourner rover in the Mars Pathfinder mission set off a global upsurge of planetary exploration with autonomous wheeled mobile robots(WMRs),or rovers.Planetary WMRs are among the most intelligent space systems that combine robotic intelligence(robint),virtual intelligence(virtint),and human intelligence(humint) synergetically.This article extends the architecture of the three-layer intelligence stemming from successful Mars rovers and related technologies in order to support the R&D of future tele-operated robotic systems.Double-layer human-machine interfaces are suggested to support the integration of humint from scientists and engineers through supervisory(Mars rovers) or three-dimensional(3D) predictive direct tele-operation(lunar rovers).The concept of multilevel autonomy to realize robint,in particular,the Coupled-Layer Architecture for Robotic Autonomy developed for Mars rovers,is introduced.The challenging issues of intelligent perception(proprioception and exteroception),navigation,and motion control of rovers are discussed,where the terrains' mechanical properties and wheel-terrain interaction mechanics are considered to be key.Double-level virtual simulation architecture to realize virtint is proposed.Key technologies of virtint are summarized:virtual planetary terrain modeling,virtual intelligent rover,and wheel-terrain interaction mechanics.This generalized three-layer intelligence framework is also applicable to other systems that require human intervention,such as space robotic arms,robonauts,unmanned deep-sea vehicles,and rescue robots,particularly when there is considerable time delay.