Problem
Research questions and friction points this paper is trying to address.
robot motion planning
semantic risk
manipulation safety
collision avoidance
context-dependent risk
Foundation-Model-Guided Topology-Aware Semantic Risk Fields for Manipulation
🤖 AI Summary
This study addresses the limitation of conventional robot motion planning, which considers only geometric collisions while lacking semantic risk awareness. To overcome this, we propose a foundation model-guided, topology-aware semantic risk field. The method fuses 3D geometry with semantic priors to generate dense cost maps, incorporating six-directional risk weighting and geodesic attenuation mechanisms. Combined with voxelized scene modeling, topology-aware masking, and GPU-accelerated parallel computation, it enables semantic safety assessment that extends beyond mere collision avoidance. Evaluations in simulated household environments demonstrate that the proposed approach significantly reduces the semantic exposure of planned trajectories, validating its practicality and reliability for robotic manipulator motion planning.
📝 Abstract
Robot motion planning in everyday environments must satisfy hard geometric constraints while accounting for context-dependent semantic risk. We present a foundation-model-guided, topology-aware semantic risk field that extends manipulation safety beyond collision avoidance. For each manipulated-object/scene-object pair, a foundation model provides six directional risk weights and a pair-specific spatial decay scale. The method combines these priors with voxelized 3D scene geometry using topology-aware shielding and geodesic spatial decay. A GPU-parallel backend batches object-level distance and risk computations to construct a dense 3D field that serves as a modular cost for downstream motion planning. We evaluate the field's shielding behavior under full and partial barriers and compare its 3D workspace representation with a pixel-wise semantic-prior baseline. Across three household simulation scenarios, trajectories optimized with the proposed field have lower semantic exposure than collision-only trajectories under the same geometric constraints. We also evaluate the computational practicality and reliability of the supporting pipeline. Together, these results support the proposed field as a practical topology-aware semantic cost representation for manipulation planning beyond collision avoidance.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Foundation Model
Semantic Risk Field
Topology-Aware Shielding
Motion Planning
GPU Parallelization
🔎 Similar Papers
No similar papers found.
