Foundation Models for Autonomous Driving System: An Initial Roadmap
arXiv:2504.00911v2 Announce Type: replace Abstract: Recent advances in foundation models (FMs), including large language models (LLMs), vision-language models (VLMs), and world models, have opened new opportunities for autonomous driving systems (ADSs) in perception, reasoning, decision-making, and interaction. However, ADSs are safety-critical cyber-physical systems, and integrating FMs into them raises substantial software engineering challenges in data curation, system design, deployment, evaluation, and assurance. To clarify this rapidly evolving landscape, we present an initial roadmap, grounded in a structured literature review, for integrating FMs into autonomous driving across three dimensions: FM infrastructure, in-vehicle integration, and practical deployment. For each dimension,
Authors:Xiongfei Wu, Mingfei Cheng, Xiaoning Ren, Qiang Hu, Jianlang Chen, Yuheng Huang, Maxime Cordy, Yao Zhang, Xiaofei Xie, Lei Ma, Yves Le Traon
View PDF HTML (experimental)
Abstract:Recent advances in foundation models (FMs), including large language models (LLMs), vision-language models (VLMs), and world models, have opened new opportunities for autonomous driving systems (ADSs) in perception, reasoning, decision-making, and interaction. However, ADSs are safety-critical cyber-physical systems, and integrating FMs into them raises substantial software engineering challenges in data curation, system design, deployment, evaluation, and assurance. To clarify this rapidly evolving landscape, we present an initial roadmap, grounded in a structured literature review, for integrating FMs into autonomous driving across three dimensions: FM infrastructure, in-vehicle integration, and practical deployment. For each dimension, we summarize the state of the art, identify key challenges, and highlight open research opportunities. Based on this analysis, we outline research directions for building reliable, safe, and trustworthy FM-enabled ADSs.
Comments: To appear in ACM Transactions on Software Engineering and Methodology (TOSEM)
Subjects:
Software Engineering (cs.SE)
Cite as: arXiv:2504.00911 [cs.SE]
(or arXiv:2504.00911v2 [cs.SE] for this version)
https://doi.org/10.48550/arXiv.2504.00911
arXiv-issued DOI via DataCite
Submission history
From: Xiongfei Wu [view email] [v1] Tue, 1 Apr 2025 15:45:31 UTC (692 KB) [v2] Wed, 1 Apr 2026 20:01:20 UTC (820 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
modellanguage modelfoundation model
Democratizing Marketing Mix Models (MMM) with Open Source and Gen AI
A practical system design combining open-source Bayesian MMM and GenAI for transparent, vendor independent marketing analytics insights. The post Democratizing Marketing Mix Models (MMM) with Open Source and Gen AI appeared first on Towards Data Science .
Estimating Absolute Web Crawl Coverage From Longitudinal Set Intersections
arXiv:2603.15416v2 Announce Type: replace-cross Abstract: Web archives preserve portions of the web, but quantifying their completeness remains challenging. Prior approaches have estimated the coverage of a crawl by either comparing the outcomes of multiple crawlers, or by comparing the results of a single crawl to external ground truth datasets. We propose a method to estimate the absolute coverage of a crawl using only the archive's own longitudinal data, i.e., the data collected by multiple subsequent crawls. Our key insight is that coverage can be estimated from the empirical URL overlaps between subsequent crawls, which are in turn well described by a simple urn process. The parameters of the urn model can then be inferred from longitudinal crawl data using linear regression. Applied
Inside Intelligent Enterprises
Taking a proactive approach to managing operational technology (OT) and IoT systems has significant business advantages today and provides foundations for AI led transformation in the future. From ensuring production line uptime, monitoring safety systems and providing data to enterprise systems, all of this data is valuable in the shift to digitalisation. However, this is an area that industries such as manufacturing have had challenges in adoption and integration due to a diverse range of systems, technologies, and infrastructure. Addressing these challenges the WINGS.OTNxT.AI platform is an end-to-end managed service that covers the entire spectrum of OT and IoT systems, covering devices, networks, and applications. A joint development between Wipro and Intel, both companies brought the
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Models
LLM Wiki Revolution: How Andrej Karpathy s Idea is Changing AI
Think about revisiting items you ve saved to Pocket, Notion or your bookmarks. Most people don t have the time to re-read all of these things after they ve saved them to these various apps, unless they have a need. We are excellent at collecting tons of information. However, we are just not very good at making any [ ] The post LLM Wiki Revolution: How Andrej Karpathy s Idea is Changing AI appeared first on Analytics Vidhya .
Estimating Absolute Web Crawl Coverage From Longitudinal Set Intersections
arXiv:2603.15416v2 Announce Type: replace-cross Abstract: Web archives preserve portions of the web, but quantifying their completeness remains challenging. Prior approaches have estimated the coverage of a crawl by either comparing the outcomes of multiple crawlers, or by comparing the results of a single crawl to external ground truth datasets. We propose a method to estimate the absolute coverage of a crawl using only the archive's own longitudinal data, i.e., the data collected by multiple subsequent crawls. Our key insight is that coverage can be estimated from the empirical URL overlaps between subsequent crawls, which are in turn well described by a simple urn process. The parameters of the urn model can then be inferred from longitudinal crawl data using linear regression. Applied
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!