Vision-Language Work Zone Intelligence for Safety-Critical Speed Regulation of Mixed-Autonomy Vehicles in Dynamic Environments

2026-06-07

AI Summary

Temporary speed limit signs in active construction zones are often missing or inconsistent, posing safety risks to both human-driven and autonomous vehicles. This work proposes the first purely onboard vision-language perception system that operates without reliance on digital maps or roadside infrastructure. By jointly leveraging object detection and semantic validation, the system identifies active construction zones and their associated temporary speed limits. Temporal smoothing and hysteresis mechanisms are integrated to enhance state stability. Implemented on a low-cost embedded platform, the system runs in real time and achieves a 96.5% recall (68.7% precision) for construction zone event detection and 95.45% precision (53.85% recall) for speed limit recognition on the ROADWork dataset, with zero misclassifications and only one false positive.

Abstract

Temporary work-zone speed limits are communicated through visually inconsistent signage and are often missing from digital maps, creating safety risks for human drivers and automated vehicle systems. We present a real-time, onboard perception pipeline that detects active work zones, recognizes associated temporary speed limits, and outputs a law-aware work-zone state and speed value suitable for driver alerts or downstream automated control. The system fuses object detections with semantic verification and temporally smoothed, hysteresis-based state transitions to reduce false activations and flicker in dynamic scenes, and runs fully on low-cost embedded hardware. Evaluated manually on a annotated subset of the ROADWork dataset (490 sequences), the system achieves inside-work-zone event-level recall of 96.5% and event-level precision of 68.7%. Speed-limit recognition evaluated on 35 minutes of in-house driving data attains 95.45% precision and 53.85% recall, with no incorrect speed classifications and a single false positive. These results demonstrate a practical, scalable approach for grounding work-zone speed awareness directly in onboard perception rather than maps or infrastructure. We release our source code for the proposed system pipeline on our GitHub repository: https://github.com/Mi3-Lab/workzone

Problem

• work zone
• speed regulation
• vision-language
• mixed-autonomy vehicles
• temporary speed limits

Innovation

• vision-language perception
• work-zone detection
• temporary speed limit recognition
• embedded real-time system
• hysteresis-based state smoothing