Recurrent neural network and extended Kalman filter in SLAM problem

The work presented in this paper is aimed at evaluating of the systematic bias error in mathematical model of mobile robot on the process of simultaneous localization and mapping during navigation in unknown environment. We have developed approach using a recurrent neural network and extended Kalman filter. Simulation results show that the proposed algorithm is very effective compared with standard Kalman filter algorithm and adaptive algorithm based on Kalman filter and feedforward neural network. Moreover, the proposed algorithm enables a non-constant biased error in vehicle model.