摘要
针对传统的激光雷达标准ICP定位算法定位精度依赖初始值、累积误差大的问题,提出了一种基于轮式里程计、IMU和激光雷达的多传感器融合定位方案。首先通过引入激光点云单点误差模型、单个点云误差概率分布来改进标准ICP定位算法;然后基于扩展卡尔曼滤波融合轮式里程计和IMU得到较为精确的里程计数据;最后将该里程计数据作为改进ICP定位算法的解算初值。实验结果表明,该融合定位方案较单一传感器定位在准确性上有较大的提高,减少了定位误差的累积。
According to the traditional standard ICP localization algorithm,the positioning accuracy depends on the initial value and the accumulated error is large.A multi-sensor fusion localization scheme based on wheel odometer,IMU and Lidar are proposed.Firstly,the standard ICP localization algorithm is improved by introducing the single point error model of laser point cloud and considering the probability distribution of single point cloud error.Then,based on extended Kalman filter,the wheel odometer and IMU are fused to obtain more accurate odometer data.Finally,the odometer data is used as the initial value of the improved ICP localization algorithm.The experimental results show that the fusion positioning scheme improves the accuracy and reduces the accumulation of positioning errors.
作者
杨威
武星
楼佩煌
钱晓明
宋阳
Yang Wei;Wu Xing;Lou Peihuang;Qian Xiaoming;Song Yang(College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Jiangsu Nanjing,210016,China)
出处
《机械设计与制造工程》
2024年第6期55-60,共6页
Machine Design and Manufacturing Engineering
