基于改进Cartographer算法的位姿估计技术研究  

Research on Pose Estimation Technology Based on Improved Cartographer Algorithm

作　　者：谷林涛 夏青元[1] GU Lintao;XIA Qingyuan(Nanjing University of Science and Technology,Nanjing 210094)

机构地区：[1]南京理工大学,南京210094

出　　处：《计算机与数字工程》2025年第3期708-712,共5页Computer & Digital Engineering

摘　　要：针对Cartographer算法在前端位姿融合算法结果精度不高,导致相关性扫描匹配计算得到的位姿初值不准确,进一步导致基于Ceres的优化扫描匹配陷入局部最优的问题,提出在位姿增量融合算法的基础上改进里程计数据的使用策略,最后在数据集上进行实验分析。实验结果表明,改进算法在建图误差方面有明显改善。Aiming at the low accuracy of the Cartographer algorithm in the front-end pose fusion algorithm,the initial value of the pose calculated by the correlation scan matching is inaccurate,which further leads to the problem that the Ceres-based optimized scan matching falls into the local optimum.A strategy to improve the use of odom data based on the pose incremental fusion algorithm is proposed,and finally the experimental analysis is performed on the data set.The results show that the improved algorithm has obvious improvements in mapping error.

关 键 词：SLAM Cartographer 里程计 位姿估计 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]