Vehicular ad-hoc networks (VANET) enable vehicles to exchange information on traffic conditions, dynamic status and localization, to enhance road safety and transportation efficiency. A typical VANET application is platooning, which can take advantage of exchanging information on speed, heading and position to allow shorter inter-vehicle distances without compromising safety. However, the platooning performance depends drastically on the quality of the communication channel, which in turn is highly influenced by the medium access control protocol (MAC). Currently, VANETs use the IEEE 802.11p MAC, which follows a carrier sense multiple access with collision avoidance (CSMA/CA) policy that is prone to collisions and degrades significantly with network load. This has led to recent proposals for a time-division multiple access (TDMA)-based MAC that synchronize vehicles’ beacons to prevent or reduce collisions. In this paper, we take CSMA/CA and two TDMA-based overlay protocols, i.e., deployed over CSMA/CA, namely PLEXE-slotted and RA-TDMAp, and carry out extensive simulations with varying platoon sizes, number of occupied lanes and transmit power to deduce empirical models that provide estimates of average number of collisions per second and average busy time ratio. In particular, we show that these estimates can be obtained from observing the number of radio-frequency (RF) neighbours, i.e., number of distinct sources of the packets received by each vehicle per time unit. These estimates can enhance the online adaptation of distributed applications, particularly platooning control, to varying conditions of the communication channel.
Road traffic accidents remain a leading cause of fatalities worldwide, and the consequences are considerably worsened by delayed detection and emergency response. Although several machine learning-based approaches have been proposed, accident detection systems are not widely deployed, and most existing solutions fail to handle the growing complexity of modern traffic environments. This study introduces Collaborative Intelligence for Road Safety (CIRS), a novel, multi-agent, machine-learning-based framework designed for real-time accident detection, semantic scene understanding, and coordinated emergency response. Each agent in CIRS is designed for a distinct role perception, classification, description, localization, and decision-making, working collaboratively to enhance situational awareness and response efficiency. These agents integrate advanced models: YOLOv11 for high-accuracy accident detection and VideoLLaMA3 for contextual-rich scene description. CIRS bridges the gap between low-level visual analysis and high-level situational awareness. Extensive evaluation on a custom dataset comprising (5200 accident, 4800 nonaccident) frames demonstrates the effectiveness of the proposed approach. YOLOv11 achieves a top-1 accuracy of 86.5% and a perfect top-5 accuracy of 100%, ensuring reliable real-time detection. VideoLLaMA3 outperforms other vision-language models with superior factual accuracy and fewer hallucinations, generating strong results in the metrics of BLEU (0.0755), METEOR (0.2258), and ROUGE-L (0.3625). The decentralized multi-agent architecture of CIRS enables scalability, reduced latency, and the timely dispatch of emergency services while minimizing false positives.