Autonomous Control of an Electric Vehicle by Computer Vision Applied to the Teaching–Learning Process

The electric vehicle is an attractive tool for the study of autonomous vehicles. In education, it can provide students with concepts on the hardware and software aspects of mobile robotics; therefore, this paper proposes the autonomous control of a low-cost car-like vehicle based on real-time image processing. The vehicle is developed to implement an autonomous control algorithm using image processing, detection of horizontal and vertical traffic signs on the road. In the design of the vehicle, the transmission and steering are modified to resemble the behaviour of an electric vehicle. The implementation of artificial vision uses low-cost elements and free software such as Raspberry Pi with its respective camera module and an Arduino. The Haar Cascade classifier is used to classify the vertical signage, which consists of fast object detection based on automatic learning.