Novel architecture for cloud based next gen vehicle platform — Transition from today to 2025

Abstract

The sole idea of the autonomous driving is to reduce the number of accidents which will be caused by human errors. There will be a phase where autonomous vehicles will have to take part with the human driven vehicles. In this phase autonomous vehicle and vehicle drivers must be careful in driving maneuvers, since it is impossible to just have autonomous vehicles on the road while the transition to fully autonomous driving takes place. To make the system more reliable and robust, authors feel that it must be substantially improved and to achieve this 3 methodologies have been proposed. The use cases are derived from the camera placed in the suitable positions of every vehicle. This creates several use cases with different driving behaviors, which truly reflects the conditions which the autonomous car needs to undergo in series. Authors propose several architectures to address the safety related issues. Since, the number of use cases are so random that it becomes practically impossible to manage data, so we propose to have a secured cloud storage. The cloud provides a secure and reliable data management system to the machine learning algorithm computation. The data management till now is effectively done via trained unique use cases. This make a big data which is encapsulation of different unique use cases under different driving conditions. So we conclude that the usage of ML and cloud based data management is the way forward for handling autonomous vehicle reliably. This may not be the complete solution, but nevertheless we are one step closer towards 2025. This concept can be implemented only in co-operation of the OEMs and Public authorities.