Search papers, labs, and topics across Lattice.
This paper introduces Score-Induced Guiding Vector Field (SGVF), a novel framework that uses score-based diffusion models to generate guiding vector fields (GVFs) directly from data distributions, overcoming limitations of classical GVF methods that require smooth, ordered curves. SGVF learns tangent fields by incorporating unit-norm, orthogonality, and directional-consistency losses, enabling effective guidance along complex, unordered paths and branching topologies. Experiments demonstrate SGVF's superior performance in robotic path following compared to classical GVFs, particularly in scenarios with complex path geometries.
Score-based diffusion models can now generate robust guiding vector fields for robotic path following, even when traditional methods stumble on unordered, branching, or probabilistically-generated paths.
Guiding Vector Fields (GVFs) are a powerful tool for robotic path following. However, classical methods assume smooth, ordered curves and fail when paths are unordered, multi-branch, or generated by probabilistic models. We propose a unified framework, termed the Score-Induced Guiding Vector Field (SGVF), which leverages score-based generative modeling to construct vector fields directly from data distributions. SGVF learns tangent fields from point clouds with unit-norm, orthogonality, and directional-consistency losses, ensuring geometric fidelity and control feasibility. This approach removes the reliance on ad-hoc path segmentation and enables guidance along complex topologies such as branching and pseudo-manifolds. The study establishes a correspondence between score vanishing in diffusion models and GVF singularities and highlights representational capacity near sharp path curvatures. Experiments on robotic navigation in planar environments demonstrate that SGVF achieves reliable path following in scenarios where classical GVFs fail, underscoring its potential as a bridge between generative modeling and geometric control. Code and experiment video are available at https://github.com/czr-gif/Guiding-Vector-Field-Generation-via-Score-based-Diffusion-Model.
