摘要
为了解决移动机器人在未知环境下的避障问题,提出了一种从障碍物检测、预测到避撞的避障方法。设定圆形窗口作为机器人有效扫描区域,利用激光传感器采集到的数据结合聚类、匹配和分类算法确定障碍物类型和动态障碍物运动信息,绘制窗口内的动态局部地图来预测动态障碍物与机器人的碰撞关系,结合Morphin算法实现有效的避障。仿真实验表明,在该算法下移动机器人能够有效地检测出障碍物,进行碰撞预测,并做出合理地避障措施。
In order to solve the problem of obstacle avoidance for mobile robot in unknown environment,a method of obstacle avoidance including obstacle detection and prediction is proposed. Setting the circular window as an effective scanning area of robot,combing with clustering,matching and classification algorithm to determine the motion types of obstacles and motion information of dynamic obstacles by laser sensor information collected data,draw a map of local dynamic collision window to predict the collision relation between dynamic obstacles and the robot,and using the Morphin algorithm to achieve obstacle avoidance effectively. The simulation results show that the mobile robot can detect obstacles effectively,predict the collision and make reasonable obstacle avoidance measures.
引文
[1]王蛟龙,周洁,高慧,等.基于局部环境形状特征识别的移动机器人避障方法[J].信息与控制,2015,44(1):91-98.
[2]罗京,刘成林,刘飞.多移动机器人的领航-跟随编队避障控制[J].智能系统学报,2017,12(02):1-10.
[3]张启彬,王鹏,陈宗海.基于速度空间的移动机器人同时避障和轨迹跟踪方法[J].控制与决策,2017,32(02):358-362.
[4]张燕,许京,陈玲玲,等.基于激光距离传感器的路况识别系统的设计[J].激光与红外,2016,46(03):265-270.
[5]张德龙,李威凌,吴怀宇,等.基于学习机制的移动机器人动态场景自适应导航方法[J].信息与控制,2016,45(05):521-529.
[6]黄如林,梁华为,陈佳佳,等.基于激光雷达的无人驾驶汽车动态障碍物检测、跟踪与识别方法[J].机器人,2016,38(4):437-443.
[7]刘杰,闫清东,唐正华.基于激光雷达的移动机器人避障规划仿真研究[J].计算机工程,2015,41(4):306-310.
[8]杨月全,韩飞,曹志强,等.基于激光传感器的动态拟合避障控制与仿真[J].系统仿真学报,2013,25(4):118-122.
[9]祝继华,周颐,王晓春,等.基于图像配准的栅格地图拼接方法[J].自动化学报,2015,41(2):285-294.
[10]诸葛程晨,唐振民,石朝侠.基于多层Morphin搜索树的UGV局部路径规划算法[J].机器人,2014,04:491-497.
[11]晓凤,胡伟,郑博嘉,等.基于改进蚁群算法与Morphin算法的机器人路径规划方法[J].科技导报,2015,33(3):84-89.