Three are some Python Point Error in your code at render() and step()

[herywang]

for example:

def reset(self):
        self.maze_view.reset_robot()
        self.state = np.zeros(2)
        self.steps_beyond_done = None
        self.done = False
        return self.state

in the code, you should return copy.deepcopy(self.state) or return np.array(self.state) rather than return self.state