Algorithm 1: Predict the optimal route

Input: Start state; Result: Optimal route; initialization; Initialize Q(s,a); Initialize state ’s’; Choose an action ’a’ using epsilon-greedy approach; for each time step do  Take a;  Observe the reward r(t+1) and the state s(t+1);  Update Q(s(t),a(t));  s(t) ← s(t+1);  a(t) ← a(t+1) end