Research on Path Planning and Simulation of Intelligent Logistics Vehicle Based on Improved EKF-SALM Algorithm
Ze Li, Lijie Cui, Zhang YanZhong, Chu Peng · 2020 5th International Conference on Mechanical, Control and Computer Engineering (ICMCCE) · 2020
The traditional algorithm of Simultaneous Localization and Mapping (SLAM) based on Extended Kalman Filter (EKF) has two problems. The local linearization method adopted by the traditional EKF causes cumulative errors. The SLAM algorithm is vulnerable to external environment interference. In this paper, an improved EKF-SALM algorithm based on error correction factor is proposed and applied to the path planning of intelligent logistics vehicles. The algorithm focuses on the errors between the predicted values and the observed values of landmark feature points. When any error exceeds the preset threshold value, the prior estimation error covariance will be adjusted through the error correction factor in order to make quick corrections to the error. Simulation results show that the algorithm can effectively improve the accuracy of both vehicle positioning and map building.