This research examines the utilization of Inverse Reinforcement Learning and Deep Q-Networks in traffic signal management to enhance urban mobility through the automation of reward function design. The study illustrates an 81.35% reduction in vehicle waiting time and a 60.15% decrease in queue length, achieved by training an Inverse Reinforcement Learning model with expert trajectory data and incorporating it into a Reinforcement Learning framework. Although throughput slightly decreased 25.11%, this trade-off enabled significant gains in congestion reduction. In comparison to the original Deep Q-Network methodology, the Inverse Reinforcement Learning controller realizes a 17.77% enhancement in waiting time and a 5.37% reduction in queue length, however throughput experiences a minor decline of 3.12%. The findings highlight the sensitivity to reward function weights, with the Inverse Reinforcement Learning controller attaining optimal equilibrium across performance measures, markedly enhancing urban traffic conditions. This study demonstrates the potential of Inverse Reinforcement Learning in optimizing traffic signals by deriving reward functions from expert tactics.