论文部分内容阅读
基于机器人未知环境探索的智能行为,以先锋Pioneer3-DX移动机器人为对象,构建了基于多传感器信息融合的机器人智能行为系统以实现其路径规划行为。构建了一个双层信息融合路径规划智能行为系统,数据层融合采用基于联邦kalman滤波融合的信息融合方法,通过多传感器融合降低信息的不确定性,为机器人提供更可靠和更准确的环境信息;决策层融合采用模糊推理方法,根据数据层融合结果作为模糊控制器的输入,构建基于模糊推理的决策层融合模块,其设计思想是模仿人的智能行为进行决策,使其不仅决策出机器人的行走方向,同时决策出机器人的行走速度,实现了动态路径规划。基于Matlab的仿真实验,验证了方法的可行性。
Based on the intelligent behaviors explored in the unknown environment of the robot, a Pioneer 3-DX mobile robot is designed to construct a robot intelligent behavior system based on multi-sensor information fusion to realize its path planning. A bi-level information fusion routing intelligent behavior system is constructed. The data layer fusion uses the information fusion method based on federated kalman filtering fusion to reduce the uncertainty of information through multi-sensor fusion to provide robots with more reliable and accurate environment information. Based on the data layer fusion result as the input of fuzzy controller, the decision-making fusion is to build a decision-making fusion module based on fuzzy inference. The design idea of the decision-making layer is to imitate human’s intelligent behavior to make decisions not only to decide the robot’s walking Direction, at the same time make the robot’s walking speed, to achieve a dynamic path planning. Based on Matlab simulation, the feasibility of the method is verified.