Path planning of greenhouse electric crawler tractor based on the improved A* and DWA algorithms

Abstract

To improve the intelligence level and the navigation efficiency of electric crawler tractors in facility greenhouses, this paper proposes a path planning algorithm based on the fusion of the improved A* algorithm and the DWA algorithm. The weight coefficients are integrated into the heuristic function of the A* algorithm, the key point selection strategy is improved, and the second-order Bessel curves are used to smooth the path trajectories. Besides, the DWA algorithm is integrated, and the key point of global paths planned by the improved A* algorithm is taken as an interpolation point. This addresses the issue that the traditional A* algorithm needs to search many nodes and has a low computational efficiency, with many path turning points and unsmooth paths. The results of simulation experiments proved that the improved A* algorithm is less time-consuming and obtains more smoother path than the Dijkstra, RRT, and traditional A* algorithms. Meanwhile, tests in a facility greenhouse show that the electric crawler tractor can realize autonomous navigation and obstacle avoidance, with a maximum lateral deviation of 11.20 cm and a maximum heading deviation of 13°, which can meet the requirements of actual operation in facility greenhouses.