Research On Location And Tracking Algorithm Of Moving Target In Mine NLOS Scene

The mine system is a complex system that covers multiple processes such as coal mining,tunneling,transportation,and drainage.Accurate positioning of personnel and mechanical equipment in the mine is a key guarantee for safe and effective production in the mine.However,due to the complex and variable mine environment and numerous obstacles,the problem of signal occlusion is very common.The positioning signal is often in the NLOS(Non Line of Sight)propagation state,seriously reducing the positioning accuracy of the mine positioning system.Based on the principle of mine location technology,this thesis studies the location and tracking methods of moving targets in mine NLOS scenes,with the main research focuses on:(1)Aiming at the problem of positioning accuracy degradation caused by NLOS propagation in coal mines,this thesis proposes a positioning algorithm NIA-ICHAN based on non line-of-sight recognition and suppression.Firstly,based on CMD values displayed in different orders of magnitude in LOS/NLOS channels,a NIA non line-of-sight recognition algorithm based on CMD and statistical hypothesis testing is proposed.Then,using the recognition results of the NIA algorithm,by introducing a correction factorα_i.Combining the Chan algorithm and the steepest gradient method to correct measurements with NLOS errors,a NIA-ICHAN localization algorithm based on prior information and measurement correction is proposed.Finally,simulation results show that the proposed NIA algorithm has a high NLOS recognition accuracy,and the NIA-ICHAN positioning algorithm can effectively suppress NLOS errors,improve positioning accuracy and robustness.(2)In order to further improve the positioning accuracy of moving targets in coal mines and ensure real-time positioning,particle filter(PF)iteration is used to improve the positioning accuracy based on the position results calculated by NIA-ICHAN algorithm.Aiming at the poor tracking performance of the particle filter algorithm in coal mines,a CTPF(Selective Particle Filter)algorithm suitable for tracking moving targets in coal mines is proposed.In this thesis,two improvements have been made to ordinary particle filtering.One is to introduce the concept of effective particle number,which determines whether to perform the resampling process by setting a threshold value,thereby reducing the amount of computation and improving operational efficiency;The second is to introduce Chopthin resampling into particle filtering algorithms.Through the"Chop"and"Thin"processing stages,a group of particles with unequal weights are generated,ensuring the number of effective particles and particle diversity,thereby mitigating the impact of particle diversity deficiency on filtering performance and tracking stability.Finally,simulation results show that the proposed CTPF algorithm has smaller tracking error and more stable performance than traditional resampling filtering algorithms,and can be better applied to mine environments.This thesis has 30 diagrams,8 tables,and 96 references.