An Improved RRT Algorithm for Path Planning of Unmanned Spraying Device

In the hilly fruit orchards of Sichuan, fruit trees are planted in evenly spaced straight rows, dividing the workspace into long, narrow spaces. In these areas, the convergence efficiency of the Rapidly Exploring Random Tree (RRT) algorithm is low, resulting in discontinuous path planning. To address these issues, an improved RRT algorithm is proposed by this paper. For reducing the randomness and enhancing convergence speed, the watershed algorithm is used to extract guiding nodes from the grid map. Additionally, in order to meet dynamic constraints of practical applications, angle constraints and path smoothing algorithms are incorporated. The experimental results indicate that, compared with other algorithms, the proposed method reduces the time consumption by ${8 8. 3 5 \%}, {8 5. 5 6 \%}, 55.32 \%$ and ${9 5. 4 4 \%}$, and the number of iterations by ${9 0. 9 4 \%} {9} {9 0. 3 9 \%}, {8 9. 1 4 \%}$ and 91.68%, respectively. These results show the significant advantages of the proposed algorithm in improving the planning speed and the path quality.