Contributions of autonomous driving technologies to the generation of smart electromobility ecosystems in emerging countries

Abstract

The development of large cities and the increased demand for mobility in them has required improvements and changes in the way we transport ourselves. This change has generated new urban mobility environments, some more efficient than others, depending on the country, culture, and society in which it develops. Besides, in these times of rapid exponential changes, it is useful to have informal and unstructured organizations that help generate an open and interactive ecosystem for the exchange of ideas and the flow of information that founded and strengthened innovation projects. Investment in Intelligent Transport Systems (ITS) has been increasing in recent years, where only developed countries have the necessary infrastructure to integrate autonomous vehicles effectively into their roads. For countries with a lower quality of infrastructure, such as Latin American countries, it becomes a major challenge in terms of technology, algorithms, and investment. The characterization of urban circuits for autonomous navigation is a crucial task for the fast integration of this technology in the near future. The development of these vehicle technologies has been increasing too, which has allowed to improve their capabilities in autonomous driving systems; many of these features are related to Advanced Driving Assistance Systems (ADAS) and autonomous driving systems (ADS). This capability improvement has been achieved because of recent developments in automationoriented software and hardware. Such improvements, allowed the vehicle to achieve a more precise perception of its working environment. The open literature contains a variety of works related to the subject. They employ a diversity of techniques ranging from probabilistic to ones based on Artificial Intelligence. The increase in computing capacity, well known to many, has opened plentiful opportunities for the algorithmic processing needed by these applications, making way for the development of autonomous navigation, in many cases with astounding results. An important part of the development and testing of these algorithms is through the generation of prototype platforms with the necessary instrumentation to implement autonomous driving functionalities in different environments. For this reason, this thesis work contributes by presenting a general framework to address this problem by considering urban mobility as an Interactive Ecosystem of Research, Innovation, Engineering, and Entrepreneurship that brings together the different actors of this ecosystem (technical and non-technical) to generate systematic and articulated actions to alleviate this challenge in emerging countries. In a technical way, this work addresses the problem by proposing a route and infrastructure analyzer for smart electromobility, thus allowing the evaluation of the effective integration of smart routes for autonomous vehicles(AVs) in megacities. The overall objective of this approach focuses on the acquisition of data from various sensors installed on the vehicle. These data are post-analyzed through computer vision systems and data analysis tools (Machine Learning) in order to generate metrics that provide an accurate mapping and smart characterization of the proposed routes. Part of this analysis is complemented by simulation tools (of routes and vehicles) and the generation of navigation models for autonomous vehicles. Finally, the results of the driving automation of two platforms, each developed by different automation methods, are reported. First, we present the results obtained from the automation of the navigation of a modular vehicle platform generated at Tecnologico de Monterrey, with the intention of generating a low-cost modular AGV. The second platform presents the driving automation of the e.Go Life electric vehicle prototype developed at RWTH Aachen University. Its capabilities and functionalities within controlled industrial environments are shown. Both platforms aim to test and analyze the different algorithms needed to generate successful navigation of autonomous vehicles within established routes.