摘要
无论对于单机器人还是多机器人系统,定位能力是其完成各项任务的前提条件。针对常用定位技术难以满足移动机器人群体协同定位精度高、实时性强等要求,提出了一种基于扩展卡尔曼滤波(EKF)和粒子滤波(PF)的混合定位技术。首先,对机器人进行运动建模,然后分别讨论基于扩展卡尔曼滤波和粒子滤波的机器人协同定位的基本原理、优缺点,在此基础上提出基于扩展卡尔曼滤波和粒子滤波相结合的协同定位方法。结果表明,该算法在满足一定条件下可有效解决定位精度与运算量之间的矛盾,可实现在初始条件未知或存在较大误差情况下多机器人快速、精确地协同定位。
The positioning capacity is a prerequisite to complete the tasks, no matter for single-robot or multi-robots sys- tems. It is difficult for common location technique to meet the requirements of mobile robot groups in high co-localization accura- cy and strong real-time. A hybrid location technique based on Extended Kalman Filter （EKF） and Particle Filter （PF） is pro- posed. The multi-robots are motion modeling, then the basic principle of robot co-localization based on EKF or PF and their ad- vantages and disadvantages are discussed respectively. On this basis, another co-localization based on the combination of EKF and PF is proposed. The experiment result shows that this method is effectively in solving the contradiction between positioning accuracy and calculation under some condition. And when the initial condition is unknown or the errors is large, the multi-robots co-localization can also be rapidly and precisely.
出处
《现代电子技术》
2013年第23期95-98,共4页
Modern Electronics Technique
基金
甘肃省工信委专项资金资助([2012]110号)
