摘要
提出一种基于bug算法、适用于具有最小转弯半径的非360探测范围车辆的避障导航算法,使四轮车辆能够以连续而平滑的路径绕行障碍物边缘,最终完成定目标导航.算法使用"虚拟触角"直观表示前方的通行状态;参考"触须法"分析车辆转弯状态,根据周围环境不断改变车辆的最小转弯半径,实时修正转弯路线.轮式机器人Pioneer3-AT上的试验结果对比显示,机器人可以利用该算法顺利完成未知环境下定目标无碰撞导航,生成的路径连续而平滑.本算法解决了这类车辆不能应用bug类算法的问题.
An obstacle-avoidance navigation algorithm based on bug algorithm is proposed for the vehicle using non-360°range sensors and having minimum turning radius. Then the vehicle can bypass obstacle edge along a continuous and smooth path and complete the fixed-goal navigation task. This algorithm uses ＂virtual antenna＂ to intuitively represent passable state in the front. The ＂tentacle algorithm＂ is used to analyze turning state, and the minimum turning radius of the vehicle is changed according to the surrounding environment to correct the turning route in real time. The test result on a Pioneer 3-AT wheeled robot show that the robot using the proposed algorithm can reach the fixed goal in an unknown environment without collision, and the trajectory is continuous and smooth. The algorithm solve the problem that bug algorithm cannot be used in this kind of vehicles.
出处
《机器人》
EI
CSCD
北大核心
2013年第5期527-534,共8页
Robot
基金
国家自然科学基金资助项目(61175081)
高等学校博士学科点专项科研基金资助项目(20114404110003)
农业部948计划资助项目(2011-G32)
