Data-enhanced convolution neural networks for wall following robot navigation

Abstract

Machine learning has been used for solving the Robot Navigation Task through the wall-following control. The wall-following control involves the movement of the robot in some directed direction maintaining a constant distance from a given wall. The path of the movement of robot is measured by ultrasonic sensors. Many machine learning methods have been used for this problem, as classifiers, but Convolution Neural Networks (CNN) outperforms them all with almost 98% of accuracy. This study compared the performance of five classifiers SVC, MLR, ANN, CNN-1D, and CNN-2D, which play the part of controller in the navigation work. We have used the ultrasonic sensor data to understand the hidden pattern in the navigation work and classified the actions by robot in terms of different motions performed by robot in response to it. The classification reports of CNN-2D and CNN-1D with Artificial Neural Networks (ANN) have also been presented in this paper. The smart Data-Enhancement used in proposed method significantly improves the classification performance of all classifiers, especially CNN.