Path planning of apple tree pruning robot based on improved RRT algorithm

Abstract

<b><sc>Abstract.</sc></b> The robot technology is widely used in many agricultural operations. However, apple tree production operations are usually performed manually. Apple tree pruning is labor-intensive, accounting for about 20% of total labor costs. The application of apple tree pruning robots can reduce the cost of fruit production. To improve the navigation efficiency of apple tree pruning robot, an improved RRT algorithm was proposed. To address the issue of blind search of the RRT algorithm, a goal-biased strategy was introduced to enhance the directionality of the algorithm in the expansion process. Then, a greedy strategy was used to speed up the search speed of the algorithm. Finally, the path was pruned to remove unnecessary nodes of the path and further reduce the path length. Navigation experiments were conducted in the apple orchard map, and the experimental results show that the improved algorithm proposed in this paper can cover the whole apple orchard with a path length of 2411.81, which is 14% shorter than the RRT algorithm. The search time is 5.74s, which is 55% shorter than the RRT algorithm. This indicates that the improved RRT algorithm has better adaptability and planning efficiency in the apple orchard environment.