Multilane Spatiotemporal Trajectory Optimization Method (MSTTOM) for Connected Vehicles

Algorithm 1

Q-learning algorithm.

	Q-learning algorithm
(1)	Initialize q (s, a);
(2)	While ()
(3)	{Select the initial state s0 and action a0 according to the ε-greedy strategy;
(4)	While ()
(5)	{Select the action at the state st according to the ε-greedy strategy, get reward rt and the next state st+1;
(6)	;
(7)	; } }
(8)	Get the optimal strategy