TM-APR: Thermal Temporal-Memory Localization via Analytic Online Adaptation

arXiv:2609.26766v1 Announce Type: new Abstract: Thermal Visual Place Recognition (Thermal VPR) maps camera observations to metric poses within a mapped environment, serving as a prerequisite for autonomous navigation. However, thermal VPR suffers from severe environmental dependence, heavy online retraining overheads, and an inability to model dynamic non-linear shifts, causing existing frameworks to fail during online deployment. To achieve robust domain-invariant place recognition, we bridge Analytic Class-Incremental Learning (ACIL) with domain-invariant VPR for the first time, revealing that its gradient-free matrix updates construct a surprisingly strong baseline that outperforms conventional fine-tuning. Nevertheless, standard ACIL exhibits a critical vulnerability to extreme non-linear thermal fluctuations due to its structural linear assumptions. To overcome this limitation, we exploit a novel algebraic equivalence between ACIL and modern control theory, proposing a framework which embeds Unscented propagation (U-ACIL), Gaussian Mixture partitioning (GMM-ACIL), and minimax H_infty optimization (H_infty-ACIL) directly into the update loop. Our formulation guarantees exact closed-form matrix updates within O(1) computational complexity, bypassing backpropagation to ensure that the online update latency (Delta t_{learn}) remains strictly bounded below the sensor acquisition interval (Delta t_{acquire}), thereby eliminating trajectory jumps in real-time SLAM pipelines.

Source: arXiv cs.RO | 2026-09-23