摘要
针对敏感皮肤结构及其实时为智能机器人提供周围环境信息的特点,提出了一种基于容错技术的多传感器在线自适应加权融合算法.该算法首先对敏感皮肤模块测量数据进行预处理,获得具有容错特性的虚拟传感器数据值,然后依据各虚拟传感器方差的变化,及时以自适应方式调整参与融合的各虚拟传感器的最优权系数,使融合系统的均方误差始终最小.实验结果表明:该算法能有效处理冲突信息和传感器故障产生的错误信息,同时保证融合结果的精确性和可靠性,从而使敏感皮肤系统具有良好的容错性和稳健性.
For the structure of sensitive skin and the characteristic of real-timely providing environmental information of intelligent robot, an on-line adaptive weighted fusion algorithm based on technology of fault tolerance is presented. The measuring data of sensitive skin board is preproscessed and virtual sensor value of fault-tolerance trait is gained. According to the changes in virtual sensor＇ s variance,the algorithm can adjust the fused virtual sensor＇s optimal weight by using the adaptive way, and the estimation error is kept as of least mean square. Experimental result shows that the algorithm can deal with the conflict in information from sensors and transact the information from a faulty sensor to ensure the reliability and accuracy of final result.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2005年第8期1019-1021,1062,共4页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助项目(50105002)
哈尔滨工业大学基金资助项目(HIT.2001.14)
