A Forward Collision Warning System Using Driving Intention Recognition of the Front Vehicle and V2V Communication

Yang, Wei; Wan, Bo; Qu, Xiaolei · 2020 · DOAJ

DOI: 10.1109/ACCESS.2020.2963854

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the limitations of existing Forward Collision Warning (FCW) systems, which often provide warnings too late for drivers to react smoothly, thereby increasing the risk of rear-end collisions. The authors propose a novel FCW system that leverages Vehicle-to-Vehicle (V2V) communication to transmit the driving intention of the front vehicle to the following vehicle. By recognizing whether the front vehicle intends to maintain speed, accelerate, brake normally, or brake in an emergency, the system can issue earlier and more accurate warnings than traditional systems relying solely on fixed Time-to-Collision (TTC) thresholds. The proposed system comprises two primary modules: a driving intention recognition module and an FCW module. The recognition module utilizes a double-layer Hidden Markov Model (HMM). The first layer classifies driving behaviors (braking and acceleration) based on sensor data such as pedal force and speed, categorizing them into five distinct states. The second layer uses these behavior classifications along with speed data to infer the driver’s specific intention. This information, along with other driving parameters, is transmitted to the following vehicle via Dedicated Short-Range Communication (DSRC) using the IEEE 802.11p protocol. The FCW module in the following vehicle employs a kinematic-based model that calculates a critical distance considering the front vehicle’s recognized intention, the following vehicle’s speed, and driver reaction times. If the actual distance falls below this calculated critical distance, a warning is triggered. To evaluate the system, the authors conducted simulation tests using PreScan software and real-world road tests. The simulation involved ten experienced drivers generating 1,400 data samples across four driving scenarios. The results demonstrated that the proposed system achieved a correct warning rate of 97.67%, which was 6.34% higher than a comparative system using a fixed TTC threshold. Real vehicle tests confirmed that the intention-based system provided earlier warnings than the TTC-based system. Specifically, the timely warning rate, defined as the ratio of warnings issued at the beginning of braking to total warnings, was 93.33%. The study concludes that integrating driving intention recognition with V2V communication significantly enhances the effectiveness of FCW systems. By predicting the near-future actions of the front vehicle, the system provides following drivers with additional time for smooth braking, thereby improving road safety and reducing the likelihood of rear-end collisions. This approach offers a robust alternative to static threshold models, adapting to varying traffic conditions and driver behaviors.

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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• time to collision looming