摘要
针对目前基于栅格地图的路径规划算法,机器人导航无法同时满足实时性与精确性要求,以及原始Voronoi图算法生成的全局路径过于臃肿和现有骨架提取的方式生成的路线过于冗余,导致过多占用机器人存储量的问题,提出一种改进的骨架提取算法,先对已知先验信息的栅格地图进行二值化、腐蚀、膨胀预处理,再生成全新的精简骨架,之后对每一次导航规划的路径进行基于改进的三次样条平滑处理,在此基础之上结合局部路径规划器DWA算法,实现机器人导航。通过多次仿真试验和真实环境试验,结果表明,利用所设计的算法得到的精简骨架比原始Voronoi图以及现有骨架更为简洁,优化之后的路径也更加平滑,提高了移动机器人运动的稳定性及移动效率,并以最安全的方式指导机器人移动。
An improved skeleton extraction algorithm is proposed to solve the problems that raster map-based path planning algorithms the robot navigation cannot meet the requirements of real-time and accuracy simultaneously, and robot storage is occupied due to the globle path generated by the original Voronoi diagram algorithm is too bloated and the route generated by the existing method of skeleton extraction is too redundant. Firstly, binarize the raster map with known prior information and carry out corrosion and expansion pretreatment, regenerating a new leaner skeleton. Then, the path of navigation planning with improved cubic spline is smoothed;the local path planner DWA algorithm is combined to realize robot navigation. The simulation and real environment experiments results show that the simplified skeleton is more concise than the original Voronoi diagram and the existing skeleton, the optimized path is smoother, which improves the stability and efficiency of the motion of the mobile robot, and guides the robot to move in the safest way.
作者
蒋林
李峻
马先重
聂文康
朱建阳
雷斌
JIANG Lin;LI Jun;MA Xianzhong;NIE Wenkang;ZHU Jianyang;LEI Bin(Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081;Institute of Robotics and Intelligent Systems,Wuhan University of Science and Technology,Wuhan 430081)
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2020年第13期138-148,共11页
Journal of Mechanical Engineering
基金
国家重点研发计划(2019YFB1310000)
湖北省自然科学基金(2018CFB626)
武汉市应用基础前沿项目(2019010701011404)
机器人与智能系统研究院开放基金(F201803)资助项目。
