Continual Cross-Dataset Adaptation in Road Surface Classification

TL;DR

This paper presents a continual learning approach for road surface classification in autonomous vehicles, enabling efficient cross-dataset adaptation without catastrophic forgetting, thus improving generalization and reducing retraining costs.

Contribution

It introduces a continual learning finetuning method tailored for AV road classification, addressing dataset shifts and avoiding catastrophic forgetting.

Findings

Achieves performance close to retraining from scratch.

Outperforms naive finetuning in cross-dataset adaptation.

Reduces computational and economic costs in AV training.

Abstract

Accurate road surface classification is crucial for autonomous vehicles (AVs) to optimize driving conditions, enhance safety, and enable advanced road mapping. However, deep learning models for road surface classification suffer from poor generalization when tested on unseen datasets. To update these models with new information, also the original training dataset must be taken into account, in order to avoid catastrophic forgetting. This is, however, inefficient if not impossible, e.g., when the data is collected in streams or large amounts. To overcome this limitation and enable fast and efficient cross-dataset adaptation, we propose to employ continual learning finetuning methods designed to retain past knowledge while adapting to new data, thus effectively avoiding forgetting. Experimental results demonstrate the superiority of this approach over naive finetuning, achieving performance close to fresh retraining. While solving this known problem, we also provide a general description of how the same technique can be adopted in other AV scenarios. We highlight the potential computational and economic benefits that a continual-based adaptation can bring to the AV industry, while also reducing greenhouse emissions due to unnecessary joint retraining.