Research on Ranging/GNSS Localization Based on Pollution Collaborative Positioning via Adaptive Kalman Filter

摘要

Collaborative positioning in many applications has broad prospects especially in the complex and weak environment. However, complicated and changeable environment has brought challenges to robust and precision fusion filter methods. To this end, this paper put forward the collaborative positioning algorithm based on adaptive Kalman filtering (CPAKF) according to the maximum likelihood criterion which can adaptively adjust process noise covariance and observation noise covariance, make the fusion filtering adapt to the changeable and complex noise environment, and have a certain anti-interference performance. Then, the pollution collaborative positioning algorithm (PCP) is presented which can achieve isolation of pollution nodes, make the other nodes clear by collaborative positioning and improve the accuracy of all peer nodes in the network ultimately. Simulation analysis of multi-use standalone as well as collaborative positioning based on the traditional kalman and adaptive kalman filtering. Compared to the traditional standalone kalman-based positioning algorithm (SKF), the collaborative positioning algorithm based on adaptive kalman filtering (CPAKF) is much better. Besides, the PCP with much smother curve can avoid pollution nodes affecting others which performs best among three positioning algorithms.