Abstract: In view of problem of low precision of using UWB positioning system to obtain position coordinates of shearer, a precision improving algorithm of shearer positioning based on UWB was put forward. The algorithm uses ability of analysis of various signal values of information filter algorithm to filter UWB positioning results, and uses ability of optimal solution of neural network algorithm to evaluate positioning information during certain period when the shearer stays at the head position of scraper conveyer, in order to realize accurate shearer positioning. The experiment results show that after using information filter algorithm to process positioning results, the 3D accuracy can reach 7 cm, and after the application of neural network algorithm, the positioning accuracy can reach 2~3 cm.