A CAD-native hybrid path-planning framework for dynamic assembly layouts

Abstract Assessing robotic accessibility during assembly-layout design helps identify geometric conflicts before robot programming and physics-based validation. This paper presents a Computer-Aided Design (CAD) native framework for three-dimensional geometric path planning in assembly environments containing static and moving obstacles. The framework integrates global A* search with local Artificial Potential Field (APF) adjustment within FreeCAD. Geometry is extracted from the CAD model, obstacles are represented using inflated axis-aligned bounding boxes, and a three-dimensional voxel grid is constructed for collision-aware planning. The initial A* path is refined through APF-based local adjustment, followed by collision, clearance, and path-quality verification using line-of-sight pruning and optional spline smoothing. The framework was evaluated using controlled planning scenarios with different obstacle motions, workspace complexities, and blocked passages, together with parameter-sensitivity analysis, and industrial demonstrations in the FreeCAD Robot Workbench. Across 20 independent trials per scenario, the framework achieved success rates ranging from 75 to 90%, depending on scenario complexity. The results demonstrate that the proposed framework supports geometric accessibility assessment during CAD-based assembly-layout design.