Low Cost Object Tracking by Computer Vision Using 8 Bits Communication with a Viper Robot

Computer vision systems have been an area of study for multiple areas of engineering. Vision systems have been studied greatly in the area of robotics, as it is of great interest to better automate processes. Many perception systems have been created to improve object handling capabilities for robots. This article seeks to create a computer vision system capable of detecting objects and communicating with a robot in a cost-effective way. To achieve the development of the project the incremental methodology is used, having as first increment the object detection algorithm, the second increment the communication system and the third increment the implementation of an object classification library. In this project, object segmentation tools and neural networks are used to create an intelligent object classification system. The cost for this project are lower than others solution from the market. Although the costs show notorious differences, no motion detection tests were performed, which will be done as a second phase of the project.