arXiv:2606.18864v1 Announce Type: cross Abstract: This paper studies how to scale learning-based automatic emergency braking (AEB) with massive unlabeled fleet data under production constraints. Our approach is based on meta-feedback semi-supervised learning (MF-SSL), where a teacher generates pseudo labels for unlabeled driving data and is updated using a small labeled anchor set as safety-critical feedback. In production, anchor ambiguity and labeled-unlabeled mismatch can amplify systematic pseudo-label errors, leading to spurious triggers. We propose a stabilized MF-SSL framework with (i)
The paper leverages massive unlabeled fleet data, a resource now widely available from autonomous vehicle testing, to address known challenges in learning-based AEB systems.
Improving the reliability and scalability of learning-based autonomous emergency braking is critical for the widespread adoption and safety of autonomous driving technologies.
The proposed meta-feedback semi-supervised learning framework offers a robust method to integrate large datasets for AEB, potentially accelerating deployment and reducing safety incidents.
- · Autonomous vehicle manufacturers
- · AI safety researchers
- · Sensor manufacturers
- · Traditional rule-based AEB systems
- · Car insurance companies (long term)
Enhanced safety and reduced accident rates in vehicles equipped with advanced AEB systems.
Accelerated public trust and regulatory approval for higher levels of autonomous driving.
Shift in liability and actuarial models as AI takes on more critical driving functions.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI