Optimization of vehicle actuation and multiplan algorithms for urban traffic control systems

The exponential growth of vehicular volume causes inevitable traffic congestions particularly in urban road networks. Road expansion to accommodate such situations can be very difficult especially when demolishing adjacent residential and commercial buildings is not possible. Effective traffic control systems are pivotal in ensuring progressive throughput of traffic by controlling the volume and sequence of vehicles traveling through intersections and avoiding motional conflicts. Relying on conventional fixed time plan systems deteriorates level of service and increases congestion due to adaptability deprivation to put up with random and continuous variation of traffic flows. The system can be too rigid, lacks the intelligence, not real-time adaptable and does not possess tunable decision-making capabilities. In this work, novel algorithms which transform simple vehicle actuation and multiplan algorithms into highly adaptable, flexible and intelligent systems are presented. These optimizations exhibit continuous learning of the flow variation using vehicle sensing data, making traffic control systems more resilient to random traffic patterns and potential sensor failures. Result when Auto-Revised VA Maximum Time plan in place is presented and discussed. It has shown an improvement in overall of 11.5% in green time allocation at an intersection while ensuring total waiting time is within the limits, thus increasing the efficiency of traffic flow offering road users potential travel time reduction.