Reward function

wolf_goat_cabbage_qlearning.py

    def get_reward(self, state):
        # GOAT and WOLF cannot be left unsupervised together
        # GOAT and CABBAGE cannot be left unsupervised together
        if state[2] == self.goal_state[2]:
            return 100
        if {'🐐', '🐺'} <= state[0] and '⛵' not in state[0]:
            return -100
        if {'🐐', '🥦'} <= state[0] and '⛵' not in state[0]:
            return -100
        if {'🐐', '🐺'} <= state[2] and '⛵' not in state[2]:
            return -100
        if {'🐐', '🥦'} <= state[2] and '⛵' not in state[2]:
            return -100
        return -1