a profetional scientefic presentation for international conferance

Created using ChatSlide

This research explores the integration of IoT data with Deep Reinforcement Learning (DRL) to enhance traffic management. It identifies gaps in existing methodologies, particularly in DQN performance under varying traffic conditions. The DQN-Softmax approach utilizes SUMO for modeling and employs temperature decay to optimize action selection. Results indicate significant improvements in traffic flow, with a 55.5% reduction in average queue length and a 75.5% increase in rewards during heavy...