[Paper Review] Development and Simulation-based Testing of a 5G-Connected Intersection AEB System

Michael Khayyat, Stefano Arrigoni+1 more|arXiv (Cornell University)|Feb 2, 2021

Vehicular Ad Hoc Networks (VANETs)Engineering28 references21 citations

TL;DR

This paper proposes a 5G-connected Intersection AEB system using radar and Pirelli CyberTyre sensors to improve collision avoidance at blind intersections. It introduces two control logics: one integrating connectivity with existing radar-based braking, and a novel adaptive exponential braking scheme using real-time adherence estimation. Simulation results show both systems successfully prevent collisions in low-friction conditions, with the adaptive system offering smoother, safer braking and improved robustness by leveraging 5G-enabled real-time data.

ABSTRACT

In Europe, 20% of road crashes occur at intersections. In recent years, evolving communication technologies are making V2V and V2I faster and more reliable; with such advancements, these crashes, as well as their economic cost, can be partially reduced. In this work, we concentrate on straight path intersection collisions. Connectivity-based algorithms relying on 5G technology and smart sensors are presented and compared to a commercial radar AEB logic in order to evaluate performances and effectiveness in collision avoidance or mitigation. The aforementioned novel safety systems are tested in a blind intersection and low adherence scenario. The first algorithm proposed is obtained by incorporating connectivity information to the original control scheme, while the second algorithm proposed is a novel control logic fully capable of utilizing also adherence estimation provided by smart sensors. Test results show an improvement in terms of safety for both the architecture and high prospects for future developments.

Motivation & Objective

• To develop and test a 5G-connected Intersection AEB system that enhances safety at blind intersections.
• To evaluate the performance of connectivity-based AEB systems using 5G and smart sensor data compared to conventional radar-only AEB.
• To investigate the impact of real-time ground friction estimation on braking system adaptability and collision mitigation.
• To assess the feasibility and effectiveness of simulation-based testing for future connected vehicle safety systems.

Proposed method

• Proposes a hybrid AEB architecture that combines radar-based TTC logic with an additional trigger based on braking distance calculation.
• Introduces a novel adaptive braking logic using an exponential deceleration profile that adjusts based on real-time adherence estimation from Pirelli CyberTyre sensors.
• Employs CarMaker simulation software to model vehicle dynamics, including actuator dynamics and road friction effects.
• Uses a modified TTC threshold equation (TTCthreshold = Vego² / (2µg)) to account for variable ground friction in collision risk assessment.
• Implements a braking distance model incorporating actuator dynamics: dbrake = vx²/(2ax) + vx/(2Jact·ax) - 1/(24Jact²·a³x).
• Tests both systems under low and normal ground friction conditions, with and without accurate friction data, using virtual simulations.

Experimental results

Research questions

• RQ1 How does a 5G-connected AEB system using connectivity and smart sensor data improve collision avoidance at blind intersections compared to conventional radar-only systems?
• RQ2 To what extent can real-time ground friction estimation enhance the robustness and safety of AEB systems in low-adhesion conditions?
• RQ3 How does an adaptive exponential braking profile compare to a stepped braking logic in terms of driving comfort and collision mitigation?
• RQ4 What is the impact of inaccurate friction data from active maps on the performance of a 5G-connected AEB system?

Key findings

• The upgraded system with adaptive exponential braking (System B) successfully avoided collision in low-friction conditions (µ = 0.4) even when the active map overestimated friction (µ = 0.6), demonstrating strong adaptability.
• System B achieved smoother deceleration profiles compared to the stepped braking of System A, improving driving comfort and control.
• In low-friction scenarios, System B stopped the vehicle further from the collision point than System A, indicating enhanced safety margins.
• The system with a priori knowledge of low friction (µ = 0.4) achieved stable and effective braking, avoiding collision even when the vehicle could not reach the requested deceleration due to poor traction.
• When friction was underestimated by the map (µ = 0.6 vs actual µ = 0.4), the system delayed braking but rapidly applied maximum deceleration, successfully avoiding collision.
• The simulation results confirm that integrating real-time adherence estimation with 5G connectivity significantly improves AEB system performance and robustness in challenging road conditions.