自主行驶铲运机路径跟踪控制

Path following control of autonomous LHD

铲运机是集铲土、运土功能一体化的铰接式车辆,常用于井下物料运输。跟踪控制是实现该类设备自动驾驶的核心技术之一。针对该种车辆特有的惯性大、响应慢、约束条件多、行驶易蛇形摆动等问题,提出了适用于铰接车路径跟踪的模型预测控制算法。引入铰接式车辆运动学模型预测未来车辆行驶状态,以路径跟踪的横向偏差、纵向偏差、航向角偏差、铰接角偏差最小化为目标函数,结合实际车辆的转向、加减速度、车速约束求解车速、转角速度最优控制序列。使用Matlab/MapleSim软件进行仿真试验,结果表明:使用该控制器可以实现y向偏差小于0.06 m,航向角偏差小于0.3°,最大铰接角控制偏差小于0.8°的控制精度。于三山岛金矿-645中段进行实车试验,实现从主巷道至溜进口的自动驾驶,具有实际工业应用前景与参考价值。

The scraper is an articulated vehicle with the functions of shoveling and transporting soil.This kind of vehicle is often used for underground ore transportation.Path tracking control is one of the core technologies to realize the automatic driving of this kind of equipment.Aiming at the problems of large inertia,slow response,many constraints and easy snake swing,a model predictive control algorithm for articulated vehicle path tracking is proposed in this paper.In this paper,the kinematic model is used to predict future vehicle driving status.The objective function is to minimize the lateral deviation,longitudinal deviation,heading angle deviation,and articulation angle deviation of path tracking.The optimal control parameters of vehicle speed and corner speed are solved according to the steering,acceleration and deceleration,and vehicle speed constraints of the actual vehicle.Matlab/Maplesim software is used for simulation tests.The results show that the controller can achieve the control accuracy of Y-direction deviation within 0.06 m,heading angle deviation within 0.3°,and the maximum articulation angle control deviation less than 0.8°.The real vehicle test is carried out in the-645 middle section of the Sanshandao gold mine to realize the automatic driving from the main roadway to the working area.The algorithm has practical industrial application prospects and reference value.