Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Collaborative Safety Shield for Safe and Efficient CAV Lane Changes in Congested On-Ramp Merging

Bharathkumar Hegde, Melanie Bouroche

TL;DR

The paper tackles safety and efficiency of lane changes for connected autonomous vehicles in congested on-ramp merging. It introduces MASS, a Multi-Agent Safety Shield based on Control Barrier Functions, and integrates it with a MAPPO-like MARL lane-changing controller to form MARL-MASS, guided by an interaction-topology graph. A customised reward prioritises speed and timely merging while safety constraints are strictly enforced, improving training stability. In dense traffic simulations, MARL-MASS delivers safety guarantees (time headway around $0.5$ s) and competitive efficiency (higher speeds, increased merging rates) compared with baselines, with open-source code provided for replication.

Abstract

Lane changing in dense traffic is a significant challenge for Connected and Autonomous Vehicles (CAVs). Existing lane change controllers primarily either ensure safety or collaboratively improve traffic efficiency, but do not consider these conflicting objectives together. To address this, we propose the Multi-Agent Safety Shield (MASS), designed using Control Barrier Functions (CBFs) to enable safe and collaborative lane changes. The MASS enables collaboration by capturing multi-agent interactions among CAVs through interaction topologies constructed as a graph using a simple algorithm. Further, a state-of-the-art Multi-Agent Reinforcement Learning (MARL) lane change controller is extended by integrating MASS to ensure safety and defining a customised reward function to prioritise efficiency improvements. As a result, we propose a lane change controller, known as MARL-MASS, and evaluate it in a congested on-ramp merging simulation. The results demonstrate that MASS enables collaborative lane changes with safety guarantees by strictly respecting the safety constraints. Moreover, the proposed custom reward function improves the stability of MARL policies trained with a safety shield. Overall, by encouraging the exploration of a collaborative lane change policy while respecting safety constraints, MARL-MASS effectively balances the trade-off between ensuring safety and improving traffic efficiency in congested traffic. The code for MARL-MASS is available with an open-source licence at https://github.com/hkbharath/MARL-MASS

A Collaborative Safety Shield for Safe and Efficient CAV Lane Changes in Congested On-Ramp Merging

TL;DR

The paper tackles safety and efficiency of lane changes for connected autonomous vehicles in congested on-ramp merging. It introduces MASS, a Multi-Agent Safety Shield based on Control Barrier Functions, and integrates it with a MAPPO-like MARL lane-changing controller to form MARL-MASS, guided by an interaction-topology graph. A customised reward prioritises speed and timely merging while safety constraints are strictly enforced, improving training stability. In dense traffic simulations, MARL-MASS delivers safety guarantees (time headway around s) and competitive efficiency (higher speeds, increased merging rates) compared with baselines, with open-source code provided for replication.

Abstract

Lane changing in dense traffic is a significant challenge for Connected and Autonomous Vehicles (CAVs). Existing lane change controllers primarily either ensure safety or collaboratively improve traffic efficiency, but do not consider these conflicting objectives together. To address this, we propose the Multi-Agent Safety Shield (MASS), designed using Control Barrier Functions (CBFs) to enable safe and collaborative lane changes. The MASS enables collaboration by capturing multi-agent interactions among CAVs through interaction topologies constructed as a graph using a simple algorithm. Further, a state-of-the-art Multi-Agent Reinforcement Learning (MARL) lane change controller is extended by integrating MASS to ensure safety and defining a customised reward function to prioritise efficiency improvements. As a result, we propose a lane change controller, known as MARL-MASS, and evaluate it in a congested on-ramp merging simulation. The results demonstrate that MASS enables collaborative lane changes with safety guarantees by strictly respecting the safety constraints. Moreover, the proposed custom reward function improves the stability of MARL policies trained with a safety shield. Overall, by encouraging the exploration of a collaborative lane change policy while respecting safety constraints, MARL-MASS effectively balances the trade-off between ensuring safety and improving traffic efficiency in congested traffic. The code for MARL-MASS is available with an open-source licence at https://github.com/hkbharath/MARL-MASS
Paper Structure (15 sections, 14 equations, 8 figures, 1 table, 1 algorithm)

This paper contains 15 sections, 14 equations, 8 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background
  3. Vehicle control
  4. Multi-Agent Reinforcement Learning for lane changing
  5. Control Barrier Functions for motion planning
  6. Proposed Approach
  7. Multi-Agent Safety Shield
  8. Interaction Topology for CAVs
  9. Customised Reward for On-ramp Merging with MASS
  10. Evaluation and Results
  11. Evaluation scenario and simulation setup
  12. Safety evaluation
  13. MARL training stability
  14. MARL policy evaluation
  15. Conclusion

Figures (8)

  • Figure 1: Illustration of different traffic levels in a highway merging section
  • Figure 2: Kinematic bicycle model
  • Figure 3: The MARL-MASS vehicle controller architecture
  • Figure 4: Actor dependency for CAVs with decentralised CBF shield
  • Figure 5: Illustration of the headway reward for a vehicle moving with a velocity of $25~\mathrm{m/s}$
  • ...and 3 more figures