Title:

A Passive Detection System for Localizing Vulnerable Road Users

Compared to other road users, vulnerable road users (VRUs) such as pedestrians, bicyclists, and motorcyclists are at higher risk of severe injuries or fatalities in road accidents due to their limited protection. Developing a practical and cost-effective VRU detection system is important to help both drivers and VRUs avoid accidents. Current methods to reduce accidents involve using vehicle-based sensors like infrared, cameras, radar, or lidar to detect VRUs and prevent collisions. However, these methods have limitations such as restricted range, obstructed views, and difficulty in adverse weather conditions. A promising solution involves cooperative collision avoidance systems, where VRUs use wireless technologies to share their location data with nearby vehicles to assess collision risks and provide timely alerts to prevent potential collisions. However, if not all VRUs use a cooperative system, should vehicles ignore their safety? The answer to this question lies in passive VRU detection. By utilizing sensor networks to detect opportunistic signals such as Wi-Fi, Bluetooth, and Cellular from VRU devices like smartphones, smartwatches, and headphones, and then applying signal processing and wireless localization methods, it is possible to detect and track VRUs even without their full cooperation. Therefore, the goal of this research is to passively localize VRUs by sensing opportunistic signals. Range-based RF localization technique has been used by incorporating the Least Square Estimation (LSE) and Kalman Filter (KF). Furthermore, the data fusion method has been adopted to fuse roadside infrastructure-mounted camera data. A preliminary simulation shows promising results of the proposed passive VRU detection and localization system.

Name

Comment*

Comments are viewable only by submitter

Captcha*