Perception Sharing for Cooperative Driving Automation (CDA)
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This report addresses the safety challenge posed by pedestrian and bicyclist fatalities, which constitute approximately 19 percent of all traffic deaths in the United States. These incidents are often exacerbated by visual limitations, such as blind spots or obstructed sight lines, that prevent drivers and automated vehicles from detecting vulnerable road users. To mitigate these risks, the study evaluates Cooperative Driving Automation (CDA) through perception sharing, a system where vehicles and infrastructure exchange location data and object information to enhance situational awareness. This approach aims to improve immediate collision avoidance, support better path planning for automated systems, and optimize mobility and energy performance. The research team conducted experimental tests using the CARMA cooperative perception system to evaluate its efficacy in vulnerable road user scenarios. The experimental design involved a stationary vehicle approaching a crosswalk where a pedestrian walked at multiple angles. Some scenarios included obstructed lines of sight for the vehicle, while others did not. The infrastructure component utilized a roadside unit equipped with a thermal sensor for object detection, while the CARMA Streets and Vehicle-to-Everything (V2X) Hub handled information processing and communication. The stationary vehicle, running the CARMA Platform, served as the receiver for the cooperative perception data. Independent evaluators from the Volpe National Transportation Systems Center assessed the system based on the consistency and frequency of information transfer, latency between detection and message transmission, and the accuracy of position estimates compared to LiDAR data when clear sight lines were available. The results demonstrated that the CARMA cooperative perception system successfully transferred and processed information with an average latency of less than 6.5 milliseconds, from the moment a vulnerable road user was detected to when the data was communicated to the cooperative automated driving system. This low latency enables faster safety decision-making. The evaluation also analyzed the difference between projected and actual distances of vulnerable road users from infrastructure sensors, highlighting the importance of accurate detection. The findings indicate that cooperative perception significantly enhances the ability of automated systems to perceive hazards beyond their direct line of sight. However, the report identifies critical areas for future development to maximize system efficacy. These include the careful calibration of infrastructure-based sensors, which was found to be determinative for performance. Additionally, the standardization of sensor outputs, particularly uncertainty measures, and the development of standardized application programming interfaces are recommended to facilitate effective data fusion and interoperability across different vendors. Future work will also focus on testing sensor fusion in simulation and developing adaptive communication protocols that adjust message content based on traffic context.
Key finding
The CARMA cooperative perception system successfully transferred and processed vulnerable road user detection information in less than 6.5 milliseconds on average.
Methodology
field_study
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 24 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.