A new research paper proposes a simulation framework to evaluate deep reinforcement learning (DRL) algorithms for connected and automated vehicle (CAV) platoon joining maneuvers in mixed traffic. The study compares Deep Q-Network (DQN), Double Deep Q-Network (DDQN), and Proximal Policy Optimization (PPO), finding that PPO achieves a 98% success rate with a collision rate below 1% when incorporating risk penalties into its reward function. The research also highlights a trade-off between safety, efficiency, and decision speed, and notes that external safety controllers can prevent collisions but may reduce overall efficiency. AI
IMPACT This research could lead to safer and more efficient autonomous vehicle navigation in complex, real-world traffic scenarios.
RANK_REASON The cluster contains a research paper detailing a new simulation framework and evaluation of reinforcement learning algorithms for a specific application.
Read on Hugging Face Daily Papers →
- Connected and automated vehicle (CAV)
- Deep Q-Network (DQN)
- Deep Reinforcement Learning (DRL)
- Double Deep Q-Network (DDQN)
- Proximal Policy Optimization (PPO)
- reinforcement learning
- Simulation of Urban MObility (SUMO)
- deep reinforcement learning
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
