基于改进A*和人工势场算法的无轨胶轮车井下无人驾驶路径规划  

Underground driverless path planning of trackless rubber tyred vehicle based on improved A*and artificial potential field algorithm

作　　者：鲍久圣[1] 张牧野 葛世荣 刘琴 袁晓明 王茂森 阴妍[1] 赵亮[1] BAO Jiusheng;ZHANG Muye;GE Shirong;LIU Qin;YUAN Xiaoming;WANG Maosen;YIN Yan;ZHAO Liang(School of Mechatronic Engineering,China University of Mining and Technology,Xuzhou 221116,China;School of Mechanical,Electronic & Information Engineering,China University of Mining and Technology-Beijing,Beijing 100083,China;Taiyuan Institute of China Coal Technology & Engineering Group Co.,Ltd.,Taiyuan 030006,China)

机构地区：[1]中国矿业大学机电工程学院,江苏徐州221116 [2]中国矿业大学(北京)机电与信息工程学院,北京100083 [3]中国煤炭科工集团太原研究院有限公司,山西太原030006

出　　处：《煤炭学报》2022年第3期1347-1360,共14页Journal of China Coal Society

基　　金：国家煤矿安全监察局科研基金资助项目(2019-行管司-022-02);中国煤炭科工集团重点基金资助项目(2018-TD-ZD011);江苏高校优势学科建设工程资助项目。

摘　　要：经过多年发展无人驾驶技术逐渐成熟,已经开始在地面车辆得到了广泛应用,同时也将成为解决无轨胶轮车在井下实现高效、安全、智能运输的重要途径,但与地面车辆成熟的无人驾驶技术不同,无人驾驶技术应用到井下仍存在许多问题亟待解决。针对井下人工驾驶的无轨胶轮车效率低且事故频发等问题,以无人驾驶无轨胶轮车为研究对象,通过仿真和试验开展井下路径规划方法研究。首先,通过分析比较常见路径规划算法的优缺点,分别选择采用最优性A*算法和人工势场算法作为无人驾驶无轨胶轮车全局和局部路径规划的基本算法;其次,针对全局路径规划传统A*算法搜索节点多和路径不平滑问题,分别使用指数函数加权和三次样条插值的方法对其进行改进,将改进前后算法在井下巷道内进行全局路径规划仿真,结果发现:改进后的算法搜索节点数减少了50%,在同一场景下规划路径所需时间仅为传统A*算法的20%,路径规划的效率得到了大幅度提升,生成的路径更加平滑且连续性也较好;同时,针对人工势场算法存在目标不可达、局部最优解等问题,分别引入斥力势场修正因子和出逃力并建立相对速度势场对其进行改进,对比改进前后算法在井下巷道内的局部路径规划仿真结果发现:改进后的人工势场算法可以在各种场景规划出更为合理的行车路径,安全性得到了保障;最后,利用微型无人车试验平台,按照8∶1的缩比搭建模拟井下巷道环境,对无人驾驶无轨胶轮车改进前后的路径规划算法分别进行了无障碍物和有障碍物的路径跟踪试验,结果表明:基于改进A*-人工势场联合算法规划出来的路径更加合理,在8∶1的缩比搭建模拟井下巷道环境中进行的无障碍物规划路径最大跟踪误差仅为0.0312 m,在巷道内会车和避障时能够规划合理的行驶路径,最大偏差仅为0.0353 m,能够满足无After many years’development,the driverless technology has gradually matured and has begun to be widely used in vehicles.At the same time,it will also become an important way to realize an efficient,safe and intelligent transportation for trackless rubber tyred vehicles underground.However,being different from the mature driverless technology of surface vehicles,there are still many problems to be solved when driverless technology is applied underground.Aiming at the problems of low efficiency and frequent accidents of underground manually driven trackless rubber tyred vehicles,taking unmanned trackless rubber tyred vehicles as the research object,the underground path planning method is studied through simulation and test.Firstly,by analyzing and comparing the advantages and disadvantages of common path planning algorithms,the optimal A*algorithm and artificial potential field algorithm are selected as the basic algorithms for the global and local path planning of unmanned trackless rubber tyred vehicles.Secondly,aiming at the problems of many search nodes and uneven path of the traditional A*algorithm for global path planning,the methods of exponential function weighting and cubic spline interpolation are used to improve it respectively.The improved algorithm is simulated in the underground roadway.The results show that the number of search nodes of the improved algorithm is reduced by 50%.In the same scenario,the time required to plan the path is only 20%of that of the traditional A*algorithm,the efficiency of path planning has been greatly improved,and the generated path is smoother and has better continuity.At the same time,aiming at the problems of unreachable target and local optimal solution in the artificial potential field algorithm,the repulsion potential field correction factor and escape force are introduced respectively,and the relative velocity potential field is established to improve it.The simulation results show that the improved artificial potential field algorithm can plan a more reasonable

关 键 词：无轨胶轮车 井下无人驾驶 路径规划 A*算法优化 人工势场算法 

分 类 号：TP525[自动化与计算机技术]