When traffic lights make a difference
Problem being addressed
Traffic congestion has been an increasingly critical matter and the intelligent control of traffic signal is critical to the optimization of transportation systems. Unreasonable signal control significantly leads to waste of traffic resource and traffic delays and the key to solving urban congestion is to dredge the intersection, where almost all collisions and delays in urban traffic are concentrated.
A novel multi-agent interaction mechanism which uses the space-time information for traffic signal control based on deep reinforcement learning method. The method considers distant intersections and uses this information as a penalty to correct the calculation of current rewards, so that the agents have the ability of communication and collaboration. Due to the various traffic volume of each road, the influence of the surrounding intersections may be different. Therefore, the suggested attention mechanism corrects the influence weight of the surrounding intersections on the current intersection.
Advantages of this solution
Simulation and experiment results demonstrate that the proposed model can get better performance than previous studies, by amending the reward.
Solution originally applied in these industries
Possible New Application of the Work
Chemical and Materials Industry
Reinforcement learning can be applied in optimizing chemical reactions to can reduce time-consuming and trial-and-error work in a relatively stable environment.
Reinforcement learning is the mainstream algorithm used to solve different games and sometimes achieve super-human performance.
Source URL: #############