Internet-based Teleoperation System Control

Technology of Internet-based telerobot is to carry observation and operation through robot to long distance environment with Internet as the communication channel, which can expand human's ability in carrying out tasks in dangerous environments such as in the sky, under the sea and in the radiating situations etc. Teleoperated robotic technology is catching much attention from researchers all over the world. Teleoperation control of an Internet-based robot is one aspect of the teleoperated robotic technology. The main contributions of this thesis are as follows:1. It summarizes the situation and the research hot points of teleoperated robotic technology with the model of master-slave teleoperation system and the simplified model, the thesis analyses the structure of the system, and mainly discuss the stability problem of Internet-based robot control system caused by Internet time-delay. Finally, the valuable control system schemes are given.2. Internet time-delay is the key inherent problem in implementation of an Internet robot control system. Random time delays may cause instability in the system. Referring to time data getting from WAN transmission experiments, some statistical characters of time-delay are obtained such as time-delay jumpiness, losing data, boundary of time-delay. The paper adopts two methods to model time-delay, one is fixed time-delay method, the other is neural network prediction. The emphasis is placed on the representations and applicable situations of linear neural network and radial basis function neural network in time-delay prediction .3. Based on the analysis of the related literatures in time-delay control system, we explain the general concepts of a time-delay system and the methods for controlling the system, and introduce the state-of -the-art in time-delay control system and its application, with the emphasis placed on the Smith predictor and several improved Smith predictors.4. Varying time delay in an Internet-based robot control system often leads to the instability of the system. We present an improved Smith predictor to tackle this problem. First, an RBF (Radial Basis Function) neural network is employed to estimate the time delay of the signal transmission over the Internet. Then, based onthe principle of the Smith predictor, two improved control systems with one Smith predictor and two Smith predictors respectively are presented to deal with the varying time delay. Theoretical analysis and simulation results show that the improved Smith predictor can compensate the varying time delay and stabilize the system.Theoretic study and simulation experiments of this paper tests the feasibility of the teleoperation system with telepresence based on Internet, and lay the foundation for the subsequent experiments and real teleoperation system.