Research on algorithm of personal navigation system based on adaptive vector particle filter

As the representative of many information technologies, navigation technology is quietly entering the details of human life. How to adapt to complex environment and fuse multi-sensor information to achieve more accurate positioning has become the key of navigation technology. Facing the requirements of high-precision and reliable positioning technology, aiming at the problems of low positioning accuracy and poor robustness caused by incomplete information of single source positioning technology, this paper focuses on the fusion positioning method of wireless signal, pedestrian dead reckoning (PDR) and map information based on particle filter (PF), and proposes an adaptive vector particle filter algorithm, A personal navigation (PND) system based on adaptive vector particle filter is designed, which can effectively improve the positioning performance. The experimental results show that the navigation results can correctly reflect the changing process of the attitude, speed and position of the pedestrian's foot. The position error of the navigation algorithm is positively correlated with the walking distance and the number of walking steps, and the relative error between the position error and the traveling distance is about 5%.