Research on AGV steering control algorithm based on improving pure tracking model

Abstract

Aiming at the problem of low initial accuracy of AGV after steering, we design a vehicle steering control algorithm based on an improved pure tracking model. Firstly, in order to improve the adaptive ability of the pure tracking model, we estimate the look-ahead distance of the pure tracking model in real time through the PSO algorithm. We use the IWO algorithm to optimize the ability of the particle swarm finding fitness, so as to avoid the particle swarm easily falling into local convergence during the working process. Secondly, in order to meet the requirements of the improved pure tracking model for continuous curvature, we add an easing curve to the traditional fishtail U-turn trajectory, and design a non-tangential round fishtail U-turn. Finally, we carry out a simulation test of the algorithm. The test results show that: using the IWO-PSO-PTM algorithm, when the vehicle speed is 0.75m/s for U-turn, the maximum lateral error is less than 0.42m, and the root mean square error is 0.18m. And when the straight line travel distance exceeds 4m after line change, the maximum lateral error is less than 0.02m. The pure tracking algorithm improved by IWO-PSO can effectively improve the initial accuracy of the AGV after steering.