Multi-agent Traffic Light Controls with SUMO

Computer Science

• Adaptive Agent 100%
• Control Domain 33%
• Control Method 33%
• Control Signal 66%
• Deep Q-Network 100%
• Deep Reinforcement Learning 33%
• multi agent 100%
• Potential Benefit 33%
• Reactive Agent 66%
• Reinforcement Learning 33%
• Traffic Control 33%

INIS

• algorithms 25%
• comparative evaluations 25%
• control 100%
• control systems 50%
• demand 25%
• increasing 25%
• learning 50%
• length 25%
• optimization 25%
• performance 50%
• policy 25%
• queues 25%
• signals 50%
• simulation 50%
• traffic control 25%

Engineering

• Agent System 50%
• Control Signal 100%
• Control Systems 100%
• Deep Reinforcement Learning 50%
• Optimal Policy 50%
• Potential Benefit 50%
• Reinforcement Learning 50%
• Traffic Control 50%