【作者】 马力波； 黄席樾； 陈东义；

【Author】 MA Li-bo HUANG Xi-yue CHEN Dong-yi (College of Automation, Chongqing University, Chongqing 400044, China; College of Science and Technology, Guizhou T.V. University, Guiyang 550004, China)

【机构】 重庆大学自动化学院；

【摘要】 本文根据决策控制变换策略思想,提出一个人-机智能系统自主性变换的新方法,以解决人-机交互问题。控制变换策略由两类条件序列行动构成,即智能体转移控制决策权给另一智能体的控制行动和改变多智能体之间预定义的协调约束控制机制。为此,建立一个控制变换策略的数学模型,该模型使用Markov 决策过程来构造和引导决策操作,即根据未知环境以及个体或智能群体的时间开销来选择一个优化策略。通过远程机器人的人-机交互仿真实验,证实了自主性变换控制模型的可行性和有效性。更多还原

【Abstract】 In this paper, a new approach of autonomy transform is introduced in a human-machine intelligence system based on the concept of transform strategies of a decision-making control in order to resolve an existing problem of human-machine interaction. The control transform strategy consists of two types of sequence actions: namely.a control action that a agent transfers a decision-making to another and a control mechanism of changing pre-specified coordination constraints between multi-agent or team. Therefore,a mathematical model of the control transform strategy is established by using a Markov decision process to construct and guide the operationalization of the strategies, which select an optimal strategy according to an uncertain environment and costs to the individuals or teams. it is validated by a human-machine interaction simulation on a tele-robotic control system that the control model of autonomy transform is feasible and effective.更多还原