Search papers, labs, and topics across Lattice.
This paper introduces SLAM, a novel framework for Visual Place Recognition (VPR) that enables continuous adaptation to new environments while preventing catastrophic forgetting. By integrating uncertainty-aware smoothing, topological space partitioning using Gaussian Mixture Models, and $H_\infty$ robust bound optimization, SLAM achieves a state-of-the-art nominal accuracy of 27.5% without requiring architectural changes. The framework allows for a mathematically guaranteed minimax robust bound, facilitating a dynamic balance between placement precision and disturbance attenuation through a single regularization parameter.
SLAM achieves a breakthrough in Visual Place Recognition by ensuring robust performance in new environments without sacrificing accuracy or requiring complex architectural changes.
Visual Place Recognition (VPR) in lifelong deployment requires continuous adaptation to new environments without catastrophic forgetting. In this paper, we propose SLAM, a Structured and Localized Analytic Manifold adaptation framework. Our framework elegantly unifies uncertainty-aware smoothing via Unscented transformation, topological space partitioning through a Gaussian Mixture Model (GMM), and $H_\infty$ robust bound optimization into a singular, unified closed-form analytical recursion. Exhaustive ablation studies demonstrate that while the synergistic combination of uncertainty smoothing and localized mapping (U+G configuration) achieves the state-of-the-art nominal accuracy of 27.5%, the full deployment of the $H_\infty$ bound does not require an architectural split; rather, it introduces a mathematically guaranteed minimax robust bound. This formulation enables the system to seamlessly modulate the intrinsic trade-off between nominal placement precision and worst-case disturbance attenuation through a single regularization parameter.
