小型全方位足球機器人之行為式控制

Behavior-based Control for Small-size Omni-directional Soccer Robots

許吳輔(Wu-Fu Hsu)

指導教授 ： 王銀添

淡江大學/工學院/機械與機電工程學系碩士班/碩士(2006年)

https://doi.org/10.6846/TKU.2006.00134

摘要

本論文針對RoboCup小型機器人足球比賽系統規劃行為式控制比賽策略。策略方面，是依據足球比賽的複雜程度劃分為群隊策略、機器人角色、機器人行為和機器人動作等四個層級的階層式架構。首先，推導全方位機器人的動態模式，根據動態模式設計逆動力學控制器與PID控制器。並且以Matlab進行模擬，測試控制器對於各種命令的反應是否能夠達到穩定。其次，將控制器與影像迴授、無線傳輸、以及機器人驅動系統做整合，以便進行實際測試與驗證控制器的可行性。並針對系統的硬體限制所造成的問題，調整控制器與韌體的設定。最後以驗證可行之控制器為基礎，對於階層式策略架構中最下層之動作當中的移至球後動作進行設計並測試。

關鍵字

足球機器人系統 ； 全方位移動足球機器人 ； 逆動力學控制 ； PID控制 ； 行為式控制

並列摘要

A soccer game strategy is proposed in this thesis for RoboCup small-sized soccer robot. Based on the degree of complexity of the soccer game, the strategy is planned as a four-layer hierarchical structure, namely, the layers of team behavior, robot roles, robot behaviors, and robot actions. First stage of this research, the dynamic model of an omnidirectional robot is analyzed, and inverse dynamic controller and PID controller are designed based on this model. These controllers are simulated by using Matlab in order to check whether their responses to different inputs are stable. Second, these controllers are integrated with image processing, radio frequency transmission, and robot driving system for testing their workability. Adjustment of controller and firmware are made because of the limitation of the robot hardware during problem solving procedure. Finally, by the proven workable controller, robot actions are designed and tested in real soccer robot system.

並列關鍵字

Soccer Robot System ； omni-directional Soccer Robot ； Inverse Dynamic Control ； PID Control ； Behavior-based Control

參考文獻

[27] 李志偉，2005，全方位移動機器人之影像伺服運動控制，淡江大學機械與機電工程學系碩士論文。

[28] 黃勻良，2005，全方位移動機器人之機構與驅動設計，淡江大學機械與機電工程學系碩士論文。

[29] 謝如蕓，2006，全方位移動機器人之運動控制，淡江大學機械與機電工程學系碩士論文。

[1] A. Abreu and L. Correia, “Behavior based decision control in autonomous vehicles:A fuzzy approach using khepera,” Proc. of 9th IEEE Int. Conf. Fuzzy System, vol. 1, 2000, pp. 140-145.

[4] J. Y. jen Hsu, D. C. Lo, and S.C. Hsu, “Fuzzy control for behavior-based mobile robots,” presented at the NAFIPS/IFIS/NASA’94, San Antonio, TX, Dec. 1994.