Fuzzy Controller Design by PSO for Two Mobile Robots Cooperation Control

• Main Authors: Wan-Ting Zeng, 曾婉婷
• Other Authors: 莊家峰
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/24964161686257302457

碩士 === 國立中興大學 === 電機工程學系所 === 100 === This thesis proposes navigation of two cooperative mobile robots in unknown environments using a particle swarm optimization (PSO) designed fuzzy controller (FC). In the navigation problem, a leading and a following robot cooperatively carry an object and simultaneously perform either a cooperative boundary following (BF) or a cooperative target seeking (TS) behavior to reach a target. The PSO-designed FC is first applied to the control of a single robot for the obstacle BF learning using only reinforcement signals. The learning approach is then applied to address the BF problem of two cooperative robots, where an auxiliary FC is deigned for the following robot. A simple learning environment is created for building the BF behavior without an exhaustive collection of input-output training pairs in advance. All of the free parameters in the FC are learned through PSO, which avoids the time-consuming manual design task. A rule for coordination of the two robots in the cooperative TS behavior is proposed. A behavior supervisor is proposed to combine the learned cooperative BF behavior and the cooperative TS behavior for navigation, and the problem of dead cycles is considered. Successful navigation of two cooperative mobile robots carrying an object in several complex environments verifies the PSO-designed FC navigation approach.