基于D-S证据理论的机器人多传感器信息融合方法研究

Method research of robot multi-sensor information fusion based on D-S evidence theory

基于D-S证据理论,采用加权平均和证据相似度思想对Dempster合成法则进行改进,然后对光电编码器、激光传感器和CCD摄像机探测到的信息进行融合.融合结果表明,改进后的方法可以消除噪声干扰,实现信号平滑处理;机器人可以自主导航,进行障碍检测、轨迹规划,从而实现安全避障.

Based on the D-S evidence theory,the dempster synthesis law was improved using the ideas of weighted average and evidence similarity,and then the information detected by photoelectric encoder,CCD camera and the laser sensors was fused.The experiment results showed that the improved D-S evidence theory can eliminate the noise interference,handle signals smoothly.Therefore robot can implement obstacle avoidance safely by doing autonomous navigation,obstacle detecting and track planning.