The Improved A* Obstacle Avoidance Algorithm for the Plant Protection UAV with Millimeter Wave Radar and Monocular Camera Data Fusion

To enhance obstacle avoidance abilities of the plant protection UAV in unstructured farmland, this article improved traditional A* algorithms through dynamic heuristic functions, search point optimization, and inflection optimization based on millimeter wave radar monocular camera data fusion. Obstacle information extraction experiments were carried out. The performance between algorithm was compared. Additionally, also results show that maximum error distance measurement fusion method 8.2%. width height 27.3% 18.5%, respectively. is more useful path planning, significantly reduces processing time, grid, turning points. at most increases length by 2.0%, least time 68.4%, grid 74.9%, points 20.7%. trajectory offset proportional to flight speed, with a 1.4 m. inversely minimum 1.6 This can provide new idea for UAV.