An Improved UFastSLAM With Generalized Correntropy Loss and Adaptive Genetic Resampling

The simultaneous localization and mapping (SLAM) is a research hotspot in robot navigation. In this paper, an improved UFastSLAM with generalized correntropy loss and adaptive genetic resampling is proposed. Specifically, the unscented Kalman filter algorithm with generalized correntropy loss is improved as the importance sampling in particle filter. Then, an adaptive genetic algorithm is employed to complete the resampling of particle filter. Finally, the improved UFastSLAM with generalized correntropy loss is presented to complete robot tracking. The proposed algorithm can complete robot tracking with high accuracy performance, and obtain reliable state estimation under the non-Gaussian measurement noise in SLAM. Simulation and experiment results exhibit the availability of the proposed SLAM algorithm.