Autonomous vegetation management in unstructured forest environments imposes a conflicting requirement: maximizing wide-area coverage while maintaining close-proximity safety around irregular obstacles. Conventional repulsion-based avoidance methods often fail to meet mowing requirements by prematurely steering robots away from target trees, resulting in significant unmowed gaps. To address this limitation, this paper proposes a Hybrid Path Planning (HPP) framework that combines a shared global Boustrophedon coverage scaffold with a local orbital maneuvering strategy inspired by celestial two-body dynamics. Rather than redefining the full environment model, the proposed method treats the currently active tree as the dominant local interaction center and generates orbit-like trunk-proximal motion around it. A variable virtual mass model is introduced so that the local attraction weakens as mowing progresses, thereby supporting transition to a rejoining phase governed by a finite state machine (FSM). MATLAB simulations indicate that the proposed framework can improve the trade-off among near-tree coverage, clearance preservation, and trajectory continuity relative to repulsion-centered local-avoidance baselines under the same global traversal scaffold.
Jang et al. (Wed,) studied this question.