融合粒子滤波与蓝牙地标矫正的定位算法

Fusion Particle Filter and Bluetooth Landmark Correction Localization Algorithm

由于智能手机中的传感器精度不高,导致行人航位推算(PDR)算法中的方向数据存在较大误差,且随着步数的积累,误差会不断增大。针对此问题提出利用粒子滤波算法,在每个粒子中设置方向属性,并添加方向角偏差,来矫正方向误差;同时,针对PDR算法中产生的漂移误差,采用iBeacon作为地标对该误差进行矫正,利用其信号强度值设置地标矫正条件,来判断行人是否进出地标,使iBeacon地标进行合理的矫正。实验结果表明,粒子滤波算法估计的位置,经过条件地标矫正后,定位精度与合理性均得到了进一步提高。

Due to the low accuracy of sensors in smart phones,there is a large error in the direction data of the Pedestrian Dead Reckoning(PDR)algorithm,and the error will increase with the accumulation of steps.A particle filter algorithm is proposed to correct the direction error by setting the direction attribute in each particle and added the direction angle deviation.Meanwhile,in view of the drift error generated in the PDR algorithm,this paper uses iBeacon as a landmark to corrected the error,and uses the signal strength value correction condition set landmarks to determine whether pedestrians out landmarks,make iBeacon landmark conduct reasonable correction.The experimental results show that the accuracy and rationality of the position estimation by the particle filter algorithm are further improved after the conditional landmark correction.