Multi-Object Tracking for Collision Avoidance Using Multiple Cameras in Open RAN Networks

2025-04-09

AI Summary

To address low detection-and-tracking accuracy and high latency in vehicular multi-object detection and tracking within Open Radio Access Network (Open RAN) environments—leading to untimely collision warnings—this paper proposes an end-to-end framework leveraging multi-camera collaborative perception and edge-intelligent agents. Methodologically, it is the first to integrate distributed visual fusion, a lightweight trajectory prediction model, and standardized communication protocols within a real-world Open RAN architecture, enabling a closed-loop pipeline spanning detection, fusion, prediction, and decision-making. The core contribution is the establishment of the first Open RAN–oriented joint optimization paradigm for multi-source vision and edge computing, significantly enhancing tracking robustness and real-time performance in complex traffic scenarios: end-to-end latency is <120 ms, and MOTA reaches 78.3%, enabling highly reliable collision avoidance.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

This paper deals with the multi-object detection and tracking problem, within the scope of open Radio Access Network (RAN), for collision avoidance in vehicular scenarios. To this end, a set of distributed intelligent agents collocated with cameras are considered. The fusion of detected objects is done at an edge service, considering Open RAN connectivity. Then, the edge service predicts the objects trajectories for collision avoidance. Compared to the related work a more realistic Open RAN network is implemented and multiple cameras are used.

Problem

• Multi-object tracking for collision avoidance in vehicles
• Using multiple cameras in Open RAN networks
• Edge-based trajectory prediction for realistic scenarios

Innovation

• Distributed intelligent agents with cameras
• Edge service for object fusion
• Open RAN network for connectivity