Research on Track Vehicle Path Tracking Algorithm Based on Improved PSO

Addressing the issues of inadequate tracking precision and excessive steering manipulations in the current unilateral braking tracked vehicle control algorithm, we introduce an adaptive path-following algorithm for such vehicles, leveraging particle swarm optimization (PSO). Utilizing the preview tracking model, an investigation into the path-following technique for track vehicles is conducted. To enhance the adaptability of this model, a fitness function is formulated, taking into consideration tracking precision and the frequency of steering adjustments. Lateral error serves as the primary determinant, while the lookahead distance within the preview tracking framework is dynamically ascertained using the PSO algorithm. To expedite the computation process of PSO and initiate local search promptly, enhancements are made to the inertia weight coefficient and the particle state updating mechanism, alongside the integration of a chaos factor. In this paper, the tracking accuracy and steering control times of the algorithm are comprehensively evaluated through simulation and actual tests on the test platform of the modified 3b55 tracked transport vehicle. When compared to the SSA algorithm, the enhanced PSO algorithm exhibits a quicker convergence rate, superior tracking precision, and a reduced number of steering adjustments.