# ensures agent does not get stuck for prolonged periods
if len(history) == 5:
if history[0][0] == history[4][0] and history[0][
1] == history[4][1]:
hfo.act(REORIENT)
history = []
continue
hfo.act(MOVE)
else:
state, valid_teammates = state_representer.get_representation(
features, args.numTeammates)
print("Valid Teammates: ", valid_teammates)
if 0 in valid_teammates:
q_learner.set_invalid(state, valid_teammates)
if action is not None:
reward = get_reward(status)
reward_printer(state, action, reward)
q_learner.update(state, action, reward)
action = q_learner.get_action(state, valid_teammates)
if action == 0:
print("Action Taken: DRIBBLE \n")
hfo.act(DRIBBLE)
elif action == 1:
print("Action Taken: SHOOT \n")
hfo.act(SHOOT)
elif args.numTeammates > 0:
