ADAS & Autonomous Vehicle Technology Expo and Conference
San Jose, CA
Topic: “Applying modern AI to advanced autonomous driving”
Dive into the application of latest AI models in developing autonomous driving technology
Discuss the transformative impacts of modern AI in building more robust and adaptive AD systems
Share Plus’s real-world commercialization experience across the U.S., Europe and Australia
By: Anurag Ganguli, VP of R&D at Plus
E-interview with the ADAS & Autonomous Vehicle Technology Conference team
1. What are top 3 developments in ADAS/AD technologies you expect to see in the next 5 years.
Large AI models will continue to evolve, enabling vehicles to make more complex decisions in real-world scenarios.
ADAS features will become more popular, and come standard in more vehicles.
Series production of autonomous vehicles like autonomous trucks will start.
2. When do you expect to see full autonomy on US roads? When should we expect commercialization of AV technologies at scale?
The path to full autonomy on US roads is progressing, but it’s a complex journey. We’re already seeing driverless robotaxis operate in certain areas. However, widespread deployment of fully autonomous vehicles across diverse environments is still some years away.
When it comes to autonomous trucks — There are currently no fully autonomous trucks commercially available for purchase or widely deployed across the country. We expect fully autonomous trucks to be on the road in the next several years. As we continue to work with our partners, including Scania/MAN/Navistar of the TRATON GROUP, Hyundai Motor Company, and IVECO, to accelerate the commercialization of our Level 4 autonomous driving technology, the exact timing will depend on regulations, technology, and customer readiness.
That said, we at Plus have a unique commercialization approach where we have applied our AI model based Level 4 autonomous driving software to different solutions and use cases that cover all levels of autonomy. These solutions are being integrated into next-gen safety systems, as well as driver-assist features and fully autonomous vehicles.
3. What are the main 3 obstacles in the road to full deployment? How can they be overcome?
A key challenge is to safely handle edge cases. Conventional rule-based approaches are insufficient. To address these we must develop systems that can emulate the nuanced decision-making of human drivers, while also possessing the ability to adapt to new and unforeseen circumstances. That’s why we use large-scale AI models capable of processing vast amounts of data which underlies our end-to-end system that has the ability to reason.
A National framework is needed in order to address the patchwork of radically different regulations that exist from state to state. Different states have varying regulations and standards for autonomous vehicles, complicating deployment. We are actively collaborating with regulators, industry peers, and various stakeholders to develop unified legislation to support the safe operation and deployment of autonomous vehicles in the U.S.
Building public trust in the safety and reliability of AVs is crucial for widespread adoption. This takes time and data. That’s why we believe our progressive approach of commercializing our self-driving technology today in driver-in solutions and using those driver-in applications and deployments to accumulate data and real-world miles is critically important to demonstrate the capability, maturity and reliability of our self-driving technology, before removing the driver from the vehicle.
