Adaptive Traffic Signal Control for Energy Efficiency Using Deep Learning and Consumer Electronics

This paper addresses the problem of energy inefficiency in conventional traffic signal control systems, which fail to adapt to dynamic traffic conditions. We propose a novel hybrid approach that integrates Deep Reinforcement Learning (DRL) and Genetic Algorithm (GA) to optimize both energy consumption and traffic flow. The DRL model dynamically adjusts traffic signal timings in real-time based on traffic conditions, while the GA optimizes signal schedules at a global level for energy efficiency. This combination enables continuous adaptation to varying traffic patterns, reducing congestion and idle time while minimizing energy usage. Simulation results demonstrate that the DRL-GA hybrid system achieves significant improvements over traditional fixed and adaptive signal control methods, reducing energy consumption by 15-25%, improving traffic flow by 10-20%, and decreasing vehicle delay by 20-30%. These findings highlight the scalability and effectiveness of our approach in enhancing urban transportation sustainability and efficiency, contributing to smarter and greener cities.