Subscribe to Research Vehicle Tech to read this report or purchase it separately. Subscribe Now
The long-tail issue of AI recognition technologies is expected to be resolved by combining generative AI with autonomous driving.
Abstract
Generative AI applications are bringing new opportunities to the development of autonomous driving systems. Generative AI can generate data, such as virtual simulated traffic scenes and driving behaviors, that can be used to train the autonomous driving system in functions such as perception, prediction, planning, and decision-making.
In addition, it can generate various special situations to solve the long tail of identification technology. Nvidia, Tesla, Wayve and other companies are actively developing autonomous driving simulation scene generation technology to seize market opportunities.
Autonomous driving aims to achieve goals of high safety, efficiency and comfortable driving to improve the driving experience for drivers and passengers. Autonomous driving systems are divided into six levels by SAE International, and unified standards have been established to ensure that all parties have a better understanding of the development and application of the technology, while also helping to maintain driving safety and regulatory supervision.
Table of contents
Table 1: Assistances autonomous driving perform during vehicle traveling
Chart 3: Variety of autonomous driving vehicles and features
Table 4: Advantages of combining generative AI with autonomous driving
Table 5: Demonstration of Nvidia Drive Sim and Strive algorithm
Table 9: Summary on generative AI's support on autonomous driving
Price: NT$22,400 (approx. US$700)
