Connectivity-Aware Representations for Constrained Motion Planning via Multi-Scale Contrastive Learning

arXiv cs.RO / 3/27/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper targets constrained motion planning, where start and goal configurations may fall in disconnected regions (EMD components) and where constraints plus redundancy complicate feasibility.
It learns connectivity-aware latent representations by performing multi-scale manifold learning from local to global neighborhoods and using clustering to generate pseudo-labels for contrastive supervision.
The learned connectivity metric biases the choice of start/goal configurations toward connected regions, explicitly avoiding EMDs before running the planner.
Experimental results across multiple manipulation tasks report a 1.9× improvement in success rate and a 0.43× reduction in planning time versus baseline methods.

Abstract

The objective of constrained motion planning is to connect start and goal configurations while satisfying task-specific constraints. Motion planning becomes inefficient or infeasible when the configurations lie in disconnected regions, known as essentially mutually disconnected (EMD) components. Constraints further restrict feasible space to a lower-dimensional submanifold, while redundancy introduces additional complexity because a single end-effector pose admits infinitely many inverse kinematic solutions that may form discrete self-motion manifolds. This paper addresses these challenges by learning a connectivity-aware representation for selecting start and goal configurations prior to planning. Joint configurations are embedded into a latent space through multi-scale manifold learning across neighborhood ranges from local to global, and clustering generates pseudo-labels that supervise a contrastive learning framework. The proposed framework provides a connectivity-aware measure that biases the selection of start and goal configurations in connected regions, avoiding EMDs and yielding higher success rates with reduced planning time. Experiments on various manipulation tasks showed that our method achieves 1.9 times higher success rates and reduces the planning time by a factor of 0.43 compared to baselines.