基于多传感器的室内三维定位算法研究

Indoor Three-Dimensional Positioning Algorithm Based on Multi-Sensors

针对消防员在日常救援过程中难以准确获知自身位置坐标的问题,采用姿态传感器和气压计,提出一种室内三维定位算法(ITPA)。在ITPA算法中,根据加速度的幅值均方根、加速度的幅值方差和角速度的幅值均方根,实现过零监测,从而获知消防员的行走步数。根据气压计的数据,采用高度获取,卡尔曼滤波和异常数据处理等操作,获知消防员的高度,并判断其移动行为。通过二维移动距离计算和位置获取操作获知二维坐标。根据消防员的行为对其三维坐标观测值进行修正,并采用Kalman融合算法估计消防员的当前三维位置坐标。实验结果表明:在直行行走、楼梯行走和综合行走下,ITPA算法都能获得较接近真实路线的消防员室内三维位置,降低了算法的步数误差、距离误差和漂移误差,比FINS,IPNS和IPA3D算法更优。

Firefighters can’t accurately know their own position coordinates during daily rescue process. In order to solve the problem,attitude sensor and barometer are used and an indoor three-dimensional positioning algorithm(ITPA)is proposed. In the ITPA algorithm,zero-crossing detection is realized according to root mean square of acceleration,amplitude variance of acceleration and root mean square of angular velocity. And the number of firefighter’s walking steps is obtained. The operations,such as height acquisition,Kalman filter and abnormal data process,are used to obtain the height of firefighter and judge the movement behaviors according to the data of barometer. The two-dimensional coordinates are obtained by two-dimensional moving distance calculation and position acquisition operation. The observation values of three-dimensional coordinates are modified according to the behaviors of firefighter. The current three-dimensional position coordinates of firefighter is estimated by Kalman fusion algorithm. The experimental results show that in the straight walking,stair walking and integrated walking,ITPA algorithm can obtain the indoor three-dimensional position coordinates of firefighters which are closer to the real path. ITPA algorithm reduces the step error,distance error and drift error. It outperforms FINS,IPNS and IPA3 D algorithms.