Possibilities of Implementing Pre-Trained Feed-forward Neural Networks in Mobile Robotics
Vladimír Stenchlák, Vladimír Tlach, Ivan Kuric, Martin Bohušík, Zuzana Ságová, Dorota Więcek · Transportation research procedia · 2023
This paper is describing how artificial intelligence algorithms can be implemented in controlling development boards like Arduino, ST32, or Raspberry for mobile crawler robots. Especially in the classification of decisions based on the input variables. The artificial intelligence algorithms specifically used in this paper and this example are called feed-forward artificial neural networks. In this paper, there is described how a relatively small neural network that is pre-trained in MATLAB platform can be implemented to the Arduino control board for controlling crawler mobile robot. In this article, there is described the mechanical construction and electrical connections of the mobile robot for a better understanding of this application. The designed crawler mobile robot in this article was primarily used for educational purposes.