Foundation CAN LM: A Pretrained Language Model For Automotive CAN Data

TL;DR

This paper introduces a large-scale pretrained model for automotive CAN data, enabling multi-task learning and improved generalization across various automotive AI applications by treating CAN signals as a language.

Contribution

It presents the first foundation model for CAN data, with a unified tokenization scheme and multi-task fine-tuning, bridging the gap between NLP paradigms and automotive signal processing.

Findings

Pretrained CAN model adapts effectively to diverse tasks

Unified tokenization scheme handles mixed signals

Model demonstrates improved cross-task generalization

Abstract

The Controller Area Network (CAN) bus provides a rich source of vehicular signals increasingly leveraged for applications in automotive and auto insurance domains, including collision detection, predictive maintenance, and driver risk modeling. Despite this potential, existing pipelines largely train isolated task-specific models on raw CAN data, with only limited efforts exploring decoded signals. Such fragmentation prevents shared representation learning and limits cross-task generalization. By contrast, natural language processing (NLP) and computer vision (CV) have been transformed by the foundation model paradigm: large-scale pretraining followed by task-specific adaptation. In this work, we introduce the foundation CAN model that demonstrates multi-objective downstream generalization using a single pretrained backbone. Our approach treats CAN data as a language: we pretrain on large-scale, unlabeled decoded CAN signals and fine-tune across heterogeneous auto insurance tasks. To enable this, we propose a unified tokenization scheme for mixed discrete-continuous signals and address challenges of temporal complexity and trip-specific variability. Our results show that one pretrained CAN model can adapt effectively to diverse predictive tasks, validating that the foundation modeling paradigm, proven in NLP and CV, also holds for CAN data. This establishes a new direction for generalizable representation learning in automotive AI.