ANFIS Based UKF-SLAM Path Planning Method

Reducing error in each step of robot motion is one of the important issue in the Simultaneous Localization and Mapping algorithms. Hence, an accurate estimation method can enhance the behavior of the SLAM algorithm in the unknown environment. A hybrid estimation method is presented in this work which is consisted of ANFIS with Unscented Kalman Filter (UKF) to estimate the estate of robots and landmarks. This estimation will be reduced the localization errors of the landmark and robot pose. So, the accuracy of the system will be improved in the planar system. The simulation results based on different environment revealed that the proposed algorithm is adaptable to the different environment with high accuracy, and efficiency in comparison with the previous works such as UKFSLAM and Neuro based UKFSLAM.