论文标题:壁面机器人路径规划与控制系统研究
STUDY ON PATH PLANNING AND CONTROL SYSTEM OF WALL-CLIMBING ROBOT
论文作者 史庆国
论文导师 王爱民,论文学位 硕士,论文专业 机械制造及其自动化
论文单位 河北工业大学,点击次数 73,论文页数 82页File Size2701k
2000-01-01论文网 http://www.lw23.com/lunwen_138890302/ 分布式控制系统;单片机;冗余控制;神经网络;路径规划; 自适应寻优
Distributed Control System;Single-chip Computer; Redundancy Control;Neural Network Path Planning; Self-adaptive Optimizing
本文主要对真空吸附式壁面机器人的路径规划和控制系统进行了研究。 在对步行式机器人的特点和控制方法充分研究的基础上，研制出了分布式单片微型计算机控制系统。该系统工作可靠，能够很好地完成对机器人各种运动动作的控制，实现所要求的功能。本文所提出的双真空泵冗余控制系统切实可行，效果显著。控制软件可以预测故障的到来，极大地提高了壁面机器人的安全性。 本文开发了基于人工神经网络的移动机器人路径规划算法，应用自适应加权寻优的方法解决了在具有多个障碍物的复杂环境中获得最优路径的问题。此外，还提出了局部路径修正和路径“张紧”两种实现路径最优化的方法。实验表明该算法效果明显，能够快速得到最优路径，并为智能壁面机器人的行为规划奠定了基础。
This paper mainly study on the path planning and control system of wall-climbing robot that based on absorption. On the base of adequate study on the properties and control method of walking- robot, we designed a control system of sir~le-chip computer based on Distributed Control Theory. It works reliably and can firish controlling task to all moving actions very well, fulfill the function remanded. The Redundancy Control System of Double Vacuum Pump (RCSDVP) developed in this paper is feasible and has remarkable effect. The controlling software has the ability of predicting fault, so it improves the safety of wall-climbing robot greatly. This paper develops a moveable robot path-planning algorithm based on artificial neural network and it solves the problem of finding optimal path in complex environment with many obstacles. In addition, two optimizing methods are proposed, that is: revising path partially and path 搕ension?, which do great help in optimizing path. Experiments shows the algorithm in this paper is very effective, it can get an optimal path very quickly and lay foundation for the action planning of intelligent climbing robot.
|
