基于随机有限集的SLAM算法

提出一种基于随机有限集的同步定位与地图创建算法，该算法利用随机有限集对环境地图和传感器观测信息建模，建立联合目标状态变量的随机有限集。依据Bayesian估计框架，利用概率假设密度滤波的粒子滤波实现对机器人位姿和环境地图进行同时估计。新算法避免了数据关联过程，并能更加自然有效地表达同步定位与地图创建(simultaneous localization and mapping, SLAM)问题中多特征-多观测特性及多种传感器信息。在仿真实验中,利用FastSLAM2.0算法和新算法进行对比，实验结果验证了新算法的优越性。

SLAM algorithm based on random finite set

A novel simultaneous localization and mapping(SLAM) algorithm based on the random finite set(RFS) theory is proposed,it models environmental map and sensor observations with RFS,and establishes the RFS of joint target state variable.The algorithm framework is Based on Bayesian estimator,uses a probability hypothesis density filter which is realized by particle filter to estimate robot’s poses and environmental map simultaneously.The new algorithm avoids the data association and describes the multifeature-multiobserve characteristics more accurately and naturally as well as multiple sensor information.Simulations are presented to compare the performance of the new algorithm with that of the FastSLAM 2.0,the simulation results verify the superiority of the new algorithm.