// moves consist solely of routing decisions
// TODO: in mixed autonomy, we may want to give automated vehicles the option of slowing down/stopping
route_choice = Box(low=0., high=1, shape=(self.scenario.num_vehicles,))
stop_go = Box(low=0., high=1, shape=(self.scenario.num_vehicles,))
return Product([route_choice, stop_go])
After Change
Moves consist of routing decisions, as well as accelerations performed by rl vehicles.
lb = [0, - np.abs(self.env_params["max-deacc"])] * self.scenario.num_rl_vehicles
ub = [2, self.env_params["max-acc"]] * self.scenario.num_rl_vehicles
return Box(np.array(lb), np.array(ub))