摘要
定位和全覆盖是清扫机器人路径规划的最基本问题。研究了机器人在环境中的相对定位,并采用卡尔曼滤波进行滤波处理,减小误差。研究了基于生物激励神经网络的路径规划算法,通过仿真,发现在障碍物多的情况下重复率较高的问题。对算法进行了改进,提出了一种基于模板的生物激励神经网络的路径规划算法。通过仿真实验,发现算法在减少重复率方面是有效可行的。
Location and complete coverage are the most Jhndamental problems in path planning of cleaning robots.The relative location of the robot in the environment is studied,and kalman filter is usedto filter processing to reduce errors.Path planning algorithm based on biologically inspires neural network ,and simula- tion shows the problem of high repetition rate when there are many obstaclesA Igarithm has been improved,bio- logically inspired neural network path planning algorithm based on the template is proposed here.Through sim- ulation experiment,it is found that the algorithm is feasible and effective to reduce the repetition rate.
出处
《机械设计与制造》
北大核心
2012年第12期160-162,共3页
Machinery Design & Manufacture
基金
机器人技术与系统国家重点实验室自主课题高端智能服务机器人技术研究平台(SKLR201201B)
关键词
全覆盖
生物激励神经网络
路径规划
清扫机器人
Complete Coverage
Biologically Inspired Neural Network
Path Planning
Cleaning Robot