Abstract: The random time-delay exists in the Internet can influence the real-time interaction between the operator and telerobot.And it also reduces the stability and the manipulability of the system.This paper proposes a novel approach based on the dynamic neural group model.With analysis of the control instruction sequence from operator，it is able to realize the inference about the operator intention.Under the random time-delay，the telerobot can realize the expected tasks through the local-autonomy control，which according to the operator intention and current environment information.Meanwhile，the control mode can switch with the event-based method to guarantee the stability of the system and improve the dynamic performance of the system.The experiment is demonstrated on the Internet-based telerobot soccer system.Simulation results prove the validity and feasibility of the proposed method.

Key words: Internet-based teleoperation, event based, dynamic neural group, operator intention

摘要： 互联网传输过程中存在的随机时延，影响了操作者与遥操作机器人之间的实时交互，降低了系统稳定性和操作性能。论文提出一种新的方法，利用动态神经元群模型对操作者发送的控制指令序列进行分析，实现对操作者意图的推断。在随机时延条件下，遥操作机器人能够根据操作者意图和当前环境信息，通过局部自主控制完成期望任务动作。同时可以与主端操作者基于事件的控制指令进行切换，来保证系统的稳定性，提高整个控制系统的操作性能和效率。最后，通过互联网足球机器人平台进行实验，仿真结果验证了所提模型与方法的有效性和可行性。

关键词: 网络遥操作, 基于事件, 动态神经元群, 操作者意图