论文部分内容阅读
本文采用模糊认知图模拟机器人在复杂环境下的动态行为,来对机器人进行高层规划.基于模糊认知图的机器人高层规划系统不仅在一定程度上克服了基于符号机器人规划存在的不适用于含有许多独立子问题或子系统的大系统及机器人之间的合作与协调机制较难寻找等问题,而且还能有效克服基于进化机器人、基于神经网络机器人进行动态行为规划时算法和实现的高度复杂性、需要大量的训练样本及时间、易产生干扰等问题.实验表明把模糊认知图应用到机器人高层规划系统能较好的解决以上问题,且具有简单、鲁棒等特点.
In this paper, the fuzzy cognitive map is used to simulate the robot’s dynamic behavior in complex environment to make the robot’s high-level planning.High-level planning system based on fuzzy cognitive graph not only overcomes the existing problem of the robot based on the symbolic robot planning, Many independent sub-problems or sub-systems of large systems and robots cooperation and coordination mechanism more difficult to find such issues, but also can effectively overcome the evolutionary robot, neural network based on the dynamic behavior of the robot when the algorithm and the realization of the high complexity , Requiring a lot of training samples and time, easy to produce interference and so on.Experiments show that applying the fuzzy cognitive graph to the robot high-level planning system can better solve the above problems, and has the characteristics of simplicity and robustness.