Neuro – Fuzzy Based Adaptive Traffic Light Management system

摘要

Transportation plays an essential role in human life and all socio-economic interactions are depending on effective transportation system. To control the traffic at intersection point various parameters like vehicle presence, flow rate, traffic intensity, speed, occupancy, density and vehicle type needs to be considered. These parameters help to analyze variations in traffic behavior, design the system where optimal signals can be produced and using distributed approach, neighboring junctions may cooperate to produce optimal control strategies. In this paper, a hybrid neuro-fuzzy based adaptive system has been proposed which can take intelligent decisions depending upon the traffic conditions on the current lane and its adjacent lane. Proposed system is trained, tested and results of the proposed system have been compared with fuzzy inference system and fixed timer based system. Results obtained from proposed system are better as compared to other systems in terms of waiting time of the vehicles and flow rate through intersection.