Adaptive Traffic Signal Control Model on Intersections Based on Deep Reinforcement Learning

Table 4

Improvements in measures of our model relative to those of other methods for single-intersection case.

SL no.	Method	Reward (%)	Average travel time (%)	Average speed (%)
1	Q-learning	1.6	−2.6	1.3
	LQF	4.3	−3.5	2.2
	Webster	16.3	−7.7	24.5
2	Q-learning	7.2	−8.4	6.9
	LQF	>100	−47.8	83.7
	Webster	>100	−46.4	>100
3	Q-learning	11.8	−4.5	4.2
	LQF	65.7	−26.6	44.4
	Webster	55.9	−30.7	50.6