基于粒子群优化的移动机器人全局定位算法

为了提高图像特征匹配过程中地图库的搜索速度,提出了一种局部子地图库搜索方法。并针对机器人自定位过程中会产生多个候选位姿,提出了利用粒子群优化算法来优化候选位姿以得到定位精度较高的优化解的全局定位新方法。最后通过Pioneer 3DX在室内环境的全局定位实验分析了粒子群优化前后定位精度和定位时间的变化。实验结果表明,新算法以牺牲较少的计算时间获得了较高的定位精度。

Global localization algorithm based on particle swarm optimization for mobile robot

In order to increase the searching speed of map database for image feature matching, a local sub-map searching algorithm was introduced. Since many pose candidates will be generated in mobile robot localization for itself, a global localization approach with pose candidate optimization and high localization precision by using particle swarm optimization algorithm was proposed. Experiment on mobile robot Pioneer 3DX was conducted in real-world indoor environment for the analysis of the localization precision and computation time before and after pose candidate optimization. The experiment results show that the proposed approach can improve localization accuracy with a little computational cost.