自由空间中的机器人自主避障方法研究  

Research on robot autonomous obstacle avoidance method in free space

作　　者：王正家[1] 吴顺尧 何涛[1] 王若 WANG Zhengjia;WU Shunyao;HE Tao;WANG Ruo(School of Mechanical Engineering,Hubei University of Technology,Wuhan 430068,China)

机构地区：[1]湖北工业大学机械工程学院,湖北武汉430068

出　　处：《传感器与微系统》2023年第6期54-57,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(51275158)

摘　　要：移动机器人在运动过程中,针对路面凹坑和突起的大小与位置存在难以检测的问题,影响障碍定位与避障运动的规划,提出了一种路面障碍检测与规避方法。通过在移动机器人上安装激光测距雷达和超声波测距传感器,能够有效获取路面凹陷和突起的特征信息;通过改进的多传感器信息融合算法,实现路面凹陷的大小与位置的分析计算;在避障控制中引入反向传播(BP)神经网络,通过改变BP神经网络的学习速率,提高移动机器人的避障精度。根据该方法,建立了移动机器人避障模型,并进行了仿真验证。结果表明:所提方法可在检测到路面凹陷和突起的同时实现有效避障,为移动机器人应用提供参考。Aiming at the problem that during movement of mobile robot,it is difficult to detect the size and position of the road pits and protrusions,which affect the obstacle positioning and obstacle avoidance movement planning,a road obstacle detection and avoidance method is proposed,through installing laser ranging radar and ultrasonic ranging sensor on mobile robot,which can effectively obtain the characteristic information of the road depression and protrusion.Through the improved multi-sensor information fusion algorithm,the analysis and calculation of the size and position of the road depression are realized.Back propogation(BP)neural network is introduced in the obstacle avoidance control.Obstacle avoidance precision of mobile robots is improved by changing the learning rate of the BP neural network.According to this method,obstacle avoidance model for mobile robot is established and verified by simulation.The results show that the proposed method can effectively avoid obstacles at the same time,detecting depressions and protrusions on the road surface,which can provide a reference for mobile robot applications.

关 键 词：移动机器人 多传感器信息融合 反向传播神经网络算法 避障 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]