Answers are grounded in this paper and cite their sources.
Efficient algorithms enabling collision-free path-planning for quadrotor UAVs with robotic arms could transform precision agriculture, enabling tasks like fruit inspection and manipulation in complex environments. This study introduces a workspace decomposition strategy for cluttered obstacle regions, like tree crowns, to simplify planning. By partitioning the workspace using the convex hull of each tree crown, the proposed method simultaneously addresses two challenges: (1) determining the optimal sequence for visiting target points (e.g., fruits) and (2) planning collision-free, length-efficient paths around branches. In simulations, a planning method based on this decomposition plans significantly shorter paths (up to 30% in some cases) and reaches more targets than a baseline monolithic planner based on PRM* under similar computational effort. While the scope is limited to combinatorial problems and working in simulations, these insights should prove valuable to future work exploring aerial manipulation in fruit orchards.
No takes yet. Share an insight, caveat, or question.
Kroneman et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: