German Ros is the Director of the Autonomous Agents Lab at Intel Labs where he focuses on creating AI-driven solutions to address problems in the spaces of autonomous systems, simulation, digital twins, and 3D generative AI.
Highlights:
- CARLA, a free, open-source simulator designed to support the development of autonomous driving systems, continues to advance the industry.
- AVL engineers enhance CARLA for simulation in rough terrain by coupling CARLA with AVL’s advanced vehicle dynamics software.
Intel has always been at the forefront of technological innovation. True to this legacy, we were part of the collaborative team behind CARLA – an open-source simulator specifically designed to develop and enhance autonomous driving systems. This initiative, comprising Intel, the Computer Vision Center, the Embodied AI Foundation, and the open-source community, has one goal in mind: to advance the possibilities of autonomous vehicles.
The inception of CARLA was largely driven by the idea of complete transparency and community access. We wanted everyone to have access to not only the engine's source code but also the source code of our platform. This open philosophy is rooted in our belief that an accessible platform facilitates adaptation, inviting diverse use cases and scenarios. Launched in 2017 with Intel as the key sponsor, CARLA was created to inspire community collaboration, encouraging academics and smaller companies to participate in the expedition of autonomous driving technologies.
Since its release, CARLA has been incredibly well-received by the community. Today, it is utilized by virtually every university engaged in autonomous driving research, and almost all major corporations with a research and development unit. We are immensely proud to see our vision come to fruition and delighted to see CARLA continuing to play an instrumental role in the research and development process.
Building on the success of CARLA, we are excited to announce a ground-breaking contribution to the off-road applications of autonomous vehicles, made possible by the great work of AVL. The AVL team has enhanced CARLA's capabilities by synergizing it with AVL's advanced vehicle dynamics software, AVL VSM. This union allows for precise simulations in unpaved, challenging terrains and opens new possibilities in sectors such as agriculture, construction, and logistics.
By integrating CARLA with AVL's advanced vehicle dynamics software and open co-simulation platform, Model.CONNECT, AVL has taken CARLA's sophisticated simulation system to a new level. This combination both accurately mirrors vehicle interactions with various surfaces and ensures realistic off-road driving behavior in the simulations, even those in challenging terrains.
AVL has agreed to contribute their CARLA code modifications and technical documentation back to the CARLA project in a significant move to encourage progress and knowledge sharing within the community. Further adding to the excitement, AVL's tools will be available free of charge via the University Partnership Program.
This landmark collaboration marks the beginning of an exciting new era in autonomous vehicle simulation. As a key player and sponsor, Intel is looking forward to seeing how these developments will revolutionize off-road applications. For more information about this project or to get involved, contact AVL directly at info@avl.com.
Stay connected for more updates as we continue breaking ground in the realm of autonomous driving systems.
German Ros serves as director of the Autonomous Agents Lab at Intel Labs. German’s obsession is to create autonomous systems that can operate in the physical world at scale, showing intelligence, dexterity, and safe behaviors. In order to fulfill this goal, German focuses on problems at the intersection of machine learning, computer vision, computer graphics, simulation for AI, 3D geometry, and autonomous systems. German is a strong open-source advocate, and as part of his research, he has led top-tier open-source projects such as the CARLA Autonomous Driving simulator, and the Open3D library. Before joining Intel Labs, German served as a Research Scientist at Toyota Research Institute (TRI), where he conducted research in the area of Simulation for Autonomous Driving, Scene Understanding, and Domain Adaptation, in the context of Autonomous Driving.