Development of a computer vision module for autonomous vehicles

Abstract

The favorable geopolitical position and very large transit potential of the Republic of Kazakhstan in the field of land freight traffic between China and Europe makes the transport logistics industry one of the most promising areas for the development of the country’s economy. In this context, deployment of unmanned cargo vehicles to minimize the costs of fuel consumption and use of human labor in labor-intensive and routine operations of logistic processes both inside warehouses and during freight transportation on public roads seems natural and efficient as ever. This paper describes the results of a research work on development of a computer vision module for an autonomous truck prototype. The performed project stages include installation of the necessary equipment, training of computer vision models and development of a mapping between cameras and LIDAR sensor for object classification and localization purposes. Key words: Computer vision, autonomous vehicle, vehicle trajectory planning, real-time trajectory planning, unmanned solution.