Fig. 5From: Mobile robots exploration through cnn-based reinforcement learningQ-value of the evaluation set calculated by the model at different iteration stages. a Q-value test for straight corridor, b Q-value test for circular corridorBack to article page