Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Highlights

We propose an improved RRT-Connect algorithm, combining a target gravity idea, the ant colony algorithm, and a path smoothing method.

Abstract. An improved RRT-Connect algorithm with a greedy strategy was proposed to solve the obstacle avoidance problem of apple tree pruning robots in unstructured environments. The Denavit-Hartenberg parameter method was used to establish the kinematic model of the manipulator. Obstacle avoidance path planning was carried out by setting a collision detection model between the manipulator and the obstacles. The idea of target gravity in the artificial potential field was incorporated. It strengthens the target orientation for the selection of expansion points and expansion directions during the expansion process of the RRT-Connect algorithm. It accelerates the convergence speed of the algorithm. At the same time, the ant colony algorithm (ACA) was used to obtain a better path. A smoothing process was proposed to optimize the path generated by RRT-Connect. Performance tests in a virtual environment show that the optimal combination of step size and gravitational coefficient in path search is a step size ρ of 1.5° and a gravitational coefficient k of 1.8. The success rate of path planning is 100%. The optimized path length can be shortened by up to 73.5% compared to the path length before optimization. A 6-degree-of-freedom manipulator was used to conduct pruning experiments on dormant apple trees in an actual laboratory environment. The results show that the path planned by the improved algorithm can successfully make the manipulator move from the initial position to the target position without collision. The study is expected to provide a valuable reference for the path planning of robotic pruning systems.