Position Selection and Collision-Free Path Planning for Fruit Picking Robots

As the population continues to grow, the need for increased food production has become increasingly evident. However, the agricultural workforce is decreasing due to industrialization. To address this gap, numerous research efforts are underway to adopt robots for harvesting. While significant progress has been made, especially in fruit harvesting using robotic systems, there are still inefficiencies and areas where human intervention is necessary. This paper proposes an algorithm for selecting the position of a mobile manipulator for fruit harvesting using an inverse reachability map to achieve automation and enhance efficiency. Additionally, a method for collision-free path planning in joint space using the model predictive artificial potential field (MPAPF) algorithm is proposed. The effectiveness of the proposed position selection and path planning algorithms was validated through harvesting simulations and a collision-free path was validated in experiments.