李天成, 孙树栋, 司书宾, 王军强. 基于粒子聚合重采样的移动机器人蒙特卡洛定位[J]. 机器人, 2010, 32(5): 674-680.
引用本文: 李天成, 孙树栋, 司书宾, 王军强. 基于粒子聚合重采样的移动机器人蒙特卡洛定位[J]. 机器人, 2010, 32(5): 674-680.
LI Tiancheng, SUN Shudong, SI Shubin, WANG Junqiang. Particle Merging Resampling Based Monte Carlo Localization for Mobile Robot[J]. ROBOT, 2010, 32(5): 674-680.
Citation: LI Tiancheng, SUN Shudong, SI Shubin, WANG Junqiang. Particle Merging Resampling Based Monte Carlo Localization for Mobile Robot[J]. ROBOT, 2010, 32(5): 674-680.

基于粒子聚合重采样的移动机器人蒙特卡洛定位

Particle Merging Resampling Based Monte Carlo Localization for Mobile Robot

  • 摘要: 提出了一种基于粒子聚合重采样的移动机器人聚合蒙特卡洛定位(Merge Monte Carlo localization,Merge-MCL)方法.首先将移动机器人作业空间划分为离散栅格,建立栅格集,然后提出一种基于粒子空间相近性的粒子聚合技术,在保证粒子空间分布合理性的同时自适应调整粒子集规模.提出的粒子聚合重采样方法能够缓解粒子权值退化问题,并避免了传统重采样方法导致的多样性匮乏问题.仿真结果表明,粒子聚合重采样方法能够有效控制粒子集规模,聚合蒙特卡洛定位方法是鲁棒、有效的.

     

    Abstract: A merge Monte Carlo localization(Merge-MCL) method for mobile robot based on particle merging resampling is presented.Grid cells and grid sets are established to represent the workspace of mobile robot,and then a particle merging technique based on the particles' spatial similarity is proposed.The technique adapts the number of particles according to the rational distribution of spatial particles.Resampling using the particle merging technique mitigates the weight degeneracy problem of particles and avoids diversity impoverishment caused by the traditional resampling methods.Simulation results illustrate that the particle merging resampling can adapt the number of particles efficiently and the Merge-MCL method is efficient and robust.

     

/

返回文章
返回