Fig. 6From: Mobile robots exploration through cnn-based reinforcement learningAverage score of 5 times evaluation by using the trained model of every 1000 iteration stepsBack to article page