Developing Cost-Effective Solutions For Autonomous Vehicle Software Testing Using Simulated Environments Using AI Techniques

Abstract

Autonomous vehicle systems are expected to significantly impact society, increasing safety and providing mobility for a broader population. However, considerable technical and certification challenges are being identified, and several impediments must be overcome. Software orchestration, user interface, security, and the impact of vulnerabilities are key issues. Due to the wide range of operational scenarios and environmental conditions in which an autonomous vehicle must operate safely, stakeholders face the challenge of increasing the cost and complexity of real physical testing. This paper proposes creating, developing, and using cost-effective testing solutions for autonomous vehicle software using artificial intelligence and parallel computing-based tests. The breakthrough prototype for the developed proposal has been tested, and results for specific hardware and software configurations have been presented and compared. The results conclude that it provides enormous economic and operational advantages and fulfills the proposed intelligent automated tests for autonomous vehicle software needs.