Optimizing the Junction-Tree-Based Reinforcement Learning Algorithm for Network-Wide Signal Coordination

Table 5

Summary of improvement effect on the JTA-based RL algorithm.

Grouping	Features	Improvement
Group 1	Existing algorithm	The control group
Group 2	Optimizations of basic parameters and the information transmission mode	(i) The green light of each phase was more flexible; (ii) The system efficiency of signal coordination improved; (iii) The operations of some intersections were still poor which need to be improved
Group 3	Optimizations of the information transmission rule and the return	(i) The maximum queue length of the system was reduced; (ii) The fluctuation of the queue length at the intersection was optimized