Learning from Mistakes: Loss-Aware Memory Enhanced Continual Learning for LiDAR Place Recognition

2025-11-19

AI Summary

To address catastrophic forgetting in LiDAR-based place recognition under continual learning, this paper proposes the KDF+ framework. Methodologically, it introduces a loss-aware sampling strategy that dynamically selects high-difficulty replay samples based on their prediction losses on historical tasks; additionally, it designs a replay enhancement mechanism that explicitly optimizes the loss on historical samples during new-task training to strengthen long-term memory retention. KDF+ seamlessly integrates with mainstream LiDAR point cloud feature extraction networks without modifying the backbone architecture. Evaluated on multiple standard benchmarks, KDF+ consistently outperforms existing continual learning methods, achieving an average accuracy improvement of 3.2–5.8% while maintaining strong stability and scalability. The framework thus provides a practical, robust continual learning solution for place recognition in dynamic environments.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

LiDAR place recognition plays a crucial role in SLAM, robot navigation, and autonomous driving. However, existing LiDAR place recognition methods often struggle to adapt to new environments without forgetting previously learned knowledge, a challenge widely known as catastrophic forgetting. To address this issue, we propose KDF+, a novel continual learning framework for LiDAR place recognition that extends the KDF paradigm with a loss-aware sampling strategy and a rehearsal enhancement mechanism. The proposed sampling strategy estimates the learning difficulty of each sample via its loss value and selects samples for replay according to their estimated difficulty. Harder samples, which tend to encode more discriminative information, are sampled with higher probability while maintaining distributional coverage across the dataset. In addition, the rehearsal enhancement mechanism encourages memory samples to be further refined during new-task training by slightly reducing their loss relative to previous tasks, thereby reinforcing long-term knowledge retention. Extensive experiments across multiple benchmarks demonstrate that KDF+ consistently outperforms existing continual learning methods and can be seamlessly integrated into state-of-the-art continual learning for LiDAR place recognition frameworks to yield significant and stable performance gains. The code will be available at https://github.com/repo/KDF-plus.

Problem

• Addresses catastrophic forgetting in LiDAR place recognition during new environment adaptation
• Enhances continual learning through loss-aware sampling of difficult discriminative samples
• Improves long-term knowledge retention via rehearsal mechanism refinement during training

Innovation

• Loss-aware sampling strategy for replay selection
• Rehearsal enhancement mechanism for knowledge retention
• Seamless integration into existing continual learning frameworks