基于特征地图的煤矿辅助运输车辆定位方法

Localization method for auxiliary transport vehicles of coal mine based on feature map

针对煤矿辅助运输车辆无人驾驶技术所包含的车辆自主实时定位问题,考虑井下巷道的环境特点和车辆作业行驶要求,提出了一种基于特征地图的车辆定位方法,建立了车辆在井下巷道行驶的运动方程及测量方程。采用扩展卡尔曼滤波讨论了定位算法,由一个高斯分布描述车辆位置状态的置信度。仿真结果表明:以车辆位置状态置信度的均值作为真实状态的一个估计,由状态置信度的均方根值描述的不确定性与实际估计误差的统计分析结果相一致,在同一时刻车辆观测到的特征数量和采样周期数值对状态置信度不确定性产生影响。基于特征地图的定位方法给出了煤矿辅助运输车辆在井下巷道实时位置状态的置信度表示,同时得出了车辆位置的实时估计和定位的不确定性。

In view of the problem of vehicle autonomous real-time positioning included in the unmanned driving technology of coal mine auxiliary transport vehicles,considering the environmental characteristics of the underground roadway and the driving requirements of the vehicle,a vehicle positioning method based on feature map is proposed. The equation of motion and the measurement equation for the vehicle running in the down-hole roadway are established. The positioning algorithm is discussed using extended Kalman filtering,and the result of the vehicle location state is represented by a Gaussian distribution. The simulation results show that the mean value of the vehicle position state confidence is used as an estimate of the real state,and the uncertainty described by the root mean square value of is consistent with the statistical analysis result of the actual estimation error,and that both the quantity of features measured simultaneously by the vehicle and the value of the sampling time have effect on this uncertainty. The feature map positioning method gives the confidence representation of the real-time position of the coal mine auxiliary transport vehicle in the underground mine roadway,and obtains the real-time estimation of the vehicle position and the uncertainty of the positioning.