Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Safe and Efficient CAV Lane Changing using Decentralised Safety Shields

Bharathkumar Hegde, Melanie Bouroche

TL;DR

This work tackles safe and efficient lane-changing for Connected Autonomous Vehicles by introducing a decentralised Hybrid Safety Shield (HSS) that combines optimisation-based Control Barrier Functions (CBFs) with a rule-based lateral safety layer. The HSS is integrated with a MARL-based lane-change controller to form MARL-HSS, enabling safe decision-making in dynamic traffic while preserving traffic throughput. Empirical results in an on-ramp merging scenario show zero crashes and competitive speeds across light and moderate densities, with MARL-HSS outperforming a state-of-the-art unsafe MARL baseline in safety and achieving similar efficiency. The approach offers a scalable, safety-guaranteed forward-deployed framework for CAV coordination, with potential extensions to mixed traffic and collaborative shield design.

Abstract

Lane changing is a complex decision-making problem for Connected and Autonomous Vehicles (CAVs) as it requires balancing traffic efficiency with safety. Although traffic efficiency can be improved by using vehicular communication for training lane change controllers using Multi-Agent Reinforcement Learning (MARL), ensuring safety is difficult. To address this issue, we propose a decentralised Hybrid Safety Shield (HSS) that combines optimisation and a rule-based approach to guarantee safety. Our method applies control barrier functions to constrain longitudinal and lateral control inputs of a CAV to ensure safe manoeuvres. Additionally, we present an architecture to integrate HSS with MARL, called MARL-HSS, to improve traffic efficiency while ensuring safety. We evaluate MARL-HSS using a gym-like environment that simulates an on-ramp merging scenario with two levels of traffic densities, such as light and moderate densities. The results show that HSS provides a safety guarantee by strictly enforcing a dynamic safety constraint defined on a time headway, even in moderate traffic density that offers challenging lane change scenarios. Moreover, the proposed method learns stable policies compared to the baseline, a state-of-the-art MARL lane change controller without a safety shield. Further policy evaluation shows that our method achieves a balance between safety and traffic efficiency with zero crashes and comparable average speeds in light and moderate traffic densities.

Safe and Efficient CAV Lane Changing using Decentralised Safety Shields

TL;DR

This work tackles safe and efficient lane-changing for Connected Autonomous Vehicles by introducing a decentralised Hybrid Safety Shield (HSS) that combines optimisation-based Control Barrier Functions (CBFs) with a rule-based lateral safety layer. The HSS is integrated with a MARL-based lane-change controller to form MARL-HSS, enabling safe decision-making in dynamic traffic while preserving traffic throughput. Empirical results in an on-ramp merging scenario show zero crashes and competitive speeds across light and moderate densities, with MARL-HSS outperforming a state-of-the-art unsafe MARL baseline in safety and achieving similar efficiency. The approach offers a scalable, safety-guaranteed forward-deployed framework for CAV coordination, with potential extensions to mixed traffic and collaborative shield design.

Abstract

Lane changing is a complex decision-making problem for Connected and Autonomous Vehicles (CAVs) as it requires balancing traffic efficiency with safety. Although traffic efficiency can be improved by using vehicular communication for training lane change controllers using Multi-Agent Reinforcement Learning (MARL), ensuring safety is difficult. To address this issue, we propose a decentralised Hybrid Safety Shield (HSS) that combines optimisation and a rule-based approach to guarantee safety. Our method applies control barrier functions to constrain longitudinal and lateral control inputs of a CAV to ensure safe manoeuvres. Additionally, we present an architecture to integrate HSS with MARL, called MARL-HSS, to improve traffic efficiency while ensuring safety. We evaluate MARL-HSS using a gym-like environment that simulates an on-ramp merging scenario with two levels of traffic densities, such as light and moderate densities. The results show that HSS provides a safety guarantee by strictly enforcing a dynamic safety constraint defined on a time headway, even in moderate traffic density that offers challenging lane change scenarios. Moreover, the proposed method learns stable policies compared to the baseline, a state-of-the-art MARL lane change controller without a safety shield. Further policy evaluation shows that our method achieves a balance between safety and traffic efficiency with zero crashes and comparable average speeds in light and moderate traffic densities.
Paper Structure (18 sections, 24 equations, 7 figures, 1 table)

This paper contains 18 sections, 24 equations, 7 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Control hierarchies
  4. MARL for on-ramp merging
  5. Vehicle model
  6. Control Barrier Functions for motion planning
  7. Proposed approach
  8. Hybrid Safety Shield
  9. Longitudinal control
  10. Lateral control
  11. Control limits
  12. MARL-HSS
  13. Evaluations and Results
  14. Evaluation scenario and simulation setup
  15. Safety evaluation
  16. ...and 3 more sections

Figures (7)

  • Figure 1: Kinematic bicycle model
  • Figure 2: Topology to identify surrounding vehicles
  • Figure 3: HSS block diagram for longitudinal control
  • Figure 4: MARL-HSS vehicle controller architecture
  • Figure 5: On-ramp merging scenario
  • ...and 2 more figures