In order to improve the positioning precision of ultra-wideband(UWB)indoor positioning system
an ultra-wideband indoor positioning algorithm based on a double-layer long-short term memory(DL-LSTM)neural network is proposed.The indoor positioning data is collected by the UWB and inertial navigation system(INS).According to its depth characteristics in the non-line-of-sight environment
the DL-LSTM model is established.The first network layer of DL-LSTM model is used to reduce the influence of non-line of sight(NLOS)error on system positioning precision.Then UWB and INS signal is combined in the second network.The second network layer is used to predict the position of UWB/INS
further to improve the positioning precision.The experimental results show that the influence of NLOS error can be reduced effectively by the proposed algorithm
and the location-solving trajectory is closer to the real trajectory
which can meet the requirements of centimeter-level positioning.