多足球機器人之策略設計

Strategy Design for Multiple Soccer Robots

王侯禕(Hou-Yi Wang)

指導教授 ： 翁慶昌

淡江大學/工學院/電機工程學系博士班/博士(2008年)

https://doi.org/10.6846/TKU.2008.01060

摘要

本論文提出一多足球機器人之策略架構來讓多機器人之間可以有效的相互合作來進行機器人足球賽。此架構主要有六個模組：(1)環境資訊模組，(2)資訊分析模組，(3)團體策略模組，(4)自我策略模組，(5)路徑規劃模組及(6)移動控制模組。本論文分別將所提之多足球機器人策略架構應用在FIRA與RoboCup之中型足球機器人賽上，從實際競賽結果可以驗證所提之策略確實具有不錯之效果。此外，本論文提出一個具有控制階段、模糊鑑別階段與控制器學習階段等三個階段的批次學習架構來自動產生一個可以有效控制輪型機器人移動之模糊控制器。此架構主要有兩個模糊系統，其中一個為用來鑑別輪型機器人之模糊鑑別器，另一個為用來控制機器人移動之模糊移動控制器。在批次學習架構中，模糊系統的一些可調參數被視為一個參數集，本論文提出一個混合粒子群最佳化與遺傳演算法(HPSOGA)來分別找出具有最佳逼近性能之模糊鑑別器與具有最佳控制性能之模糊控制器，使得輪型機器人之控制具有最佳的移動性能。從FIRA足球機器人模擬器之模擬以及中型足球機器人系統之實作結果可以驗證所提批次學習架構與方法確實具有快速學習建立移動控制器之能力。

關鍵字

機器人足球競賽 ； 足球機器人 ； 自主機器人

並列摘要

A strategy structure for multiple soccer robots is proposed in this thesis. This structure includes six modules: (1) Environmental Information Module, (2) Information Analysis Module, (3) Group Strategy Module, (4) Self Strategy Module, (5) Path Planning Module, and (6) Motion Control Module. The proposed structure has been applied in the robot soccer games of the RoboSot league of FIRA and the middle size league of RoboCup. We can see the proposed strategy has a good performance in the robot soccer game. Furthermore, a batch learning structure is proposed to automatically determine a motion fuzzy controller so that the controlled wheeled robot has a good motion. This structure in each generation can be separated into three states: a control state, a system identification state, and a controller learning state. There are two fuzzy systems in this structure. One is a fuzzy identifier to identify the model of wheeled robot and the other is a fuzzy controller to control the motion of wheeled robot. The antecedent and consequent parameters of the fuzzy system are viewed as a parameter set and a Hybrid Particle Swarm Optimization and Genetic Algorithm (HPSOGA) is proposed to choose appropriate parameter sets so that the selected fuzzy identifier has a good approximation and the selected fuzzy controller has a good control performance. Some simulation results in FIRA 3D robot soccer simulator are used to illustrate the proposed learning structure is effective.

並列關鍵字

Robot Soccer Game ； Soccer Robot ； Autonomous Robot

參考文獻

[5] W. G. Han, S. M. Baek, and T. Y. Kuc, “GA based online path planning of mobile robots playing soccer games,” The 40th Midwest Symposium on Circuits and Systems, vol. 1, pp. 522-525, 1997.

[7] K. Kostiadis and H. Hu, “Reinforcement learning and co-operation in a simulated multi-agent system,” IEEE/RSJ International Conference on Intelligent Robots and Systems, vol. 2, pp. 990-995, 1999.

[9] H. K Lam, T. H. Lee, F. H. F. Leung, and P. K. S. Tam, “Decision maker for robot soccer,” The 27th Annual Conference of the Industrial Electronics Society, vol. 1, pp. 43-48, 2001.

[11] H.L. Sng G. Sen Gupta, and C.H. Messom, “Strategy for collaboration in robot soccer,” IEEE International Workshop on Electronic Design, pp. 347-351, 2002.

[12] M.Veloso and P. Stone, “Individual and collaborative behaviors in a team of homogeneous robotic soccer agents,” International Conference on Multi Agent Systems, pp. 309 -316, 1998