Implementation of Autoware Application to real-world Services Based Adaptive Big Data Management System for Autonomous Driving

Abstract

World Connected autonomous driving not only enables transportation but also improves passenger safety, comfort and entertainment. Nowadays, fully autonomous driving is required. Passengers can achieve their goals and turn dead hours of driving into fruitful hours of learning, work, engagement, and relaxation. This must be done in real-time. Our goal in this project is to master connectivity in order to provide V2X (Vehicle-to-Everything) functions for fully autonomous driving. Big data services are changing the way we drive. With autonomous driving, information from various integrated sensors is processed and analyzed in milliseconds. In this case, machine vision in an autonomous vehicle means that all input data is captured by the vehicle's sensors or cameras, and this data is analyzed in real-time during operation. The purpose of our study is to design and implement by expanding the Autoware software stack to enable embedded computer functions. Based on high-performance GPU, from a functional safety perspective, this project focuses on the application of autoware to develop autonomous vehicles by converting window-based sources to ROS-based sources when starting point and point are specified of arrival. Another goal of this study is that the autonomous driving component is software that can be used to analyze autonomous driving with large amounts of data. By connecting to a network, smart cars not only pass information from all their sensors to the cloud but also react immediately to conditions. And the interfaces required at this layer to connect the standalone software defined service platform with the framework for physical management, connectivity, and spatial data in the standalone Oracle Cloud database. This paper proposes a UML diagram including the design of the Deep Learning Process Model Autonomous Driving Based on Big Data and software in it. The software is called embedded software. Model-based testing is a resolution for testing embedded software.