Behavior selection and coordination of the Internet-based multi-robot teleoperation system by a neural group network approach

In recent years, the Internet-based teleoperation control has become a research focus. Compared with single robot, multi-robot system can be applied to many fields. It has the ability to complete the complicated tasks more rapidly and effectively. The combination form of the Internet-based multi-robot teleoperation system has attracted the widespread attention. However, due to the existence of the random time-delay, it is difficult for the robots to obtain the real-time operation commands. This may seriously decrease the stability and reliability of the system. Properly increasing the local-autonomy of the telerobots under the human operation via the Internet is an effective way. In this paper, a neural group network approach to autonomous behavior selection and coordination is proposed. It is capable of controlling a group of Internet-based mobile telerobots to achieve multiple tasks automatically when data transmission is failed. The architecture of the Internet-based multi-robot teleoperation system and a neural group network controller model are presented. Then, we describe the behavior coordination process based on the selective strategy. The validity of our approach is demonstrated on the Internet-based telerobot soccer system by simulation experiments. The results and future work are described in the end.