SRA-CP: Spontaneous Risk-Aware Selective Cooperative Perception

2025-11-21

AI Summary

Existing cooperative perception methods face two key bottlenecks: excessive redundant data transmission exceeding available bandwidth, and reliance on predefined communication partners—limiting adaptability in dynamic vehicular networks. This paper proposes a decentralized, risk-driven, on-demand cooperative perception framework. It introduces a lightweight perception coverage summary broadcasting mechanism; a spontaneous blind-spot risk identification module that triggers collaboration only when safety-critical blind spots emerge; and a safety-priority-weighted fusion strategy coupled with bandwidth-adaptive transmission. Experiments on public benchmarks demonstrate that our approach achieves only a 1% average accuracy degradation compared to generic cooperative baselines, while reducing communication overhead by 80% (i.e., retaining only 20% of baseline traffic). Moreover, it improves perception performance by 15% over state-of-the-art selective methods, significantly enhancing both communication efficiency and safety-aware adaptability.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Multiagent Systems: Coordination and Collaboration

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Security and Privacy: Data transparency and provenance

Abstract

Cooperative perception (CP) offers significant potential to overcome the limitations of single-vehicle sensing by enabling information sharing among connected vehicles (CVs). However, existing generic CP approaches need to transmit large volumes of perception data that are irrelevant to the driving safety, exceeding available communication bandwidth. Moreover, most CP frameworks rely on pre-defined communication partners, making them unsuitable for dynamic traffic environments. This paper proposes a Spontaneous Risk-Aware Selective Cooperative Perception (SRA-CP) framework to address these challenges. SRA-CP introduces a decentralized protocol where connected agents continuously broadcast lightweight perception coverage summaries and initiate targeted cooperation only when risk-relevant blind zones are detected. A perceptual risk identification module enables each CV to locally assess the impact of occlusions on its driving task and determine whether cooperation is necessary. When CP is triggered, the ego vehicle selects appropriate peers based on shared perception coverage and engages in selective information exchange through a fusion module that prioritizes safety-critical content and adapts to bandwidth constraints. We evaluate SRA-CP on a public dataset against several representative baselines. Results show that SRA-CP achieves less than 1% average precision (AP) loss for safety-critical objects compared to generic CP, while using only 20% of the communication bandwidth. Moreover, it improves the perception performance by 15% over existing selective CP methods that do not incorporate risk awareness.

Problem

• Overcoming bandwidth limitations by transmitting only safety-relevant perception data
• Eliminating dependency on pre-defined communication partners in dynamic environments
• Addressing occlusion risks through selective cooperation triggered by blind zones

Innovation

• Decentralized protocol using lightweight perception coverage summaries
• Risk identification module triggers cooperation for blind zones
• Selective information exchange prioritizes safety-critical content