Centralized and Distributed Human-Swarm Control System By Abstracted Task Function

• Main Authors: Chao-WeiLin, 林晁瑋
• Other Authors: Yen-Chen Liu
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/71098066307117191523

碩士 === 國立成功大學 === 機械工程學系 === 103 === Enduring human intelligence with traditional multi-robot systems can significantly improve flexibility and maneuverability. However, kinematic dissimilarity between master robots, manipulated by human operators, and mobile robots, implementing tasks in remote environment is the most important issue in the study of human-swarm systems. In this thesis, centralized and distributed control frameworks for a human operator to remotely manipulate a group of mobile robots over unreliable communication network are proposed. By controlling the task abstraction of the multi-robot system to overcome the kinematic dissimilarity, the human user is able to control the movements of a group of mobile robots. In the centralized system, stability and tracking performance are guaranteed when the communication network is subjected to time-varying delays. The distributed control framework can be accomplished by assigning a portion of agents as the receiver(s) such that the entire group of mobile robots can be commanded by the human operator. The position tracking between master and swarm robots is guaranteed under constant time delay by using the estimated value. Both of the human-swarm systems and control algorithms are validated in this thesis via numerical simulation and experiments to demonstrate stability and performance.