An Integrated Lateral and Longitudinal Decision-Making Model for Autonomous Driving Based on Deep Reinforcement Learning

Table 6

Comparison of convergence episodes with different ADV penetration.

ADV: MDV	DQN	DDQN	Dueling DQN
1 : 100	823	792	691
1 : 50	1767	1743	1531
1 : 30	>2000	>2000	1958
1 : 10	>2000	>2000	>2000