Infrastructure Safety Support System for Smart Cities with Autonomous Vehicles
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 research addresses the challenge of ensuring safety and mobility for autonomous vehicles (AVs) in mixed-traffic environments, where AVs share roads with human-driven vehicles for decades. While AVs are anticipated to significantly reduce crashes, public hesitation and slow adoption rates necessitate technologies that support AV decision-making in complex, multimodal settings. The study aims to develop an Infrastructure Safety Support System that embeds Vehicle-to-Infrastructure (V2I) enabled sensor networks into transportation infrastructure. This system provides real-time traffic and road condition data to AVs, enabling them to make learned and ethical decisions regarding driving speed and safe following distances. The methodology involves two primary components: the development of an infrastructure-embedded sensor network and the creation of a new car-following algorithm. The sensor network utilizes robust, glass-fiber-reinforced polymer (GFRP) packaged Fiber Bragg Grating (FBG) sensors installed within pavement layers. These sensors monitor seven key parameters: vehicle speed, wheelbase, dynamic weight (weigh-in-motion), vehicle count, vehicle classification, road roughness, and internal pavement cracks. Field testing was conducted at the MnROAD facility in Minnesota, where sensors were installed under the wheel path of an interstate mainline. Data from these sensors was processed using machine learning techniques, including Support Vector Machines (SVM), neural networks, and k-nearest neighbors, to classify vehicles and estimate traffic characteristics. Concurrently, the researchers developed a new Cumulative-anticipative Car-Following (CACF) algorithm designed to integrate this real-time infrastructure data into AV decision-making processes. The findings demonstrate the efficacy of both the sensor network and the algorithm. The GFRP-FBG sensor system achieved speed and wheelbase estimation accuracies of 95% or higher when validated against radar gun measurements. Weigh-in-motion measurements showed errors ranging from approximately 8% to 18%, with longitudinal sensor components providing the highest sensitivity. Vehicle classification using the SVM-based data fusion strategy yielded high accuracy across different vehicle types. Microsimulation results using VISSIM software indicated that the CACF model, when integrated with the infrastructure support system, significantly improved traffic safety and mobility compared to traditional models like Wiedemann 99 and Cooperative Adaptive Cruise Control (CACC). Specifically, the CACF model reduced the average number of traffic conflicts and optimized mean speeds across various speed limits (55–75 mph). The study also included an optimization analysis to balance the effectiveness and affordability of the sensor network layout. The significance of this work lies in its contribution to the development of safe smart cities by providing a reliable, cost-effective infrastructure solution for AV support. By embedding durable sensors directly into the pavement, the system offers continuous, weather-resistant monitoring that enhances AV self-awareness in mixed-driver scenarios. The integration of infrastructure data into car-following algorithms demonstrates a viable path for improving traffic flow and reducing crashes during the long transitional period of AV adoption. Additionally, the project supports educational development by engaging students in smart city technologies, preparing them for future careers in transportation infrastructure and autonomous systems.
Key finding
The infrastructure safety support system with the new cumulative-anticipative car-following algorithm improved traffic safety and mobility for autonomous vehicles in mixed driver conditions as shown by microsimulation results.
Methodology
mixed_methods
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 | — | — | 19 | 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.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: validation psychometrics