Intel announced that he would resume Racer-Sim program DARPA (United States Defense Advanced Research Projects Agency) design autonomous all-terrain vehicles (vehicles for off-road use). To do this, the company will collaborate with the Computer Vision Center in Barcelona and the University of Texas.
In summary, Intel says it will develop simulation solutions to help bridging the gap between virtual environments and the real world for all-terrain vehicles (simulation of autonomous vehicles off road and in simulation transfer learning).
The Defense Advanced Research Projects Agency (DARPA) has granted Intel Federal LLC, with support from Intel Labs, the Computer Vision Center in Barcelona, Spain, and the University of Texas at Austin, the opportunity to develop advanced simulation system solutions for autonomous all-terrain vehicles. The Robotic Autonomy in Complex Environments with Resiliency – Simulation (RACER-Sim) program aims to create the next generation of vehicle simulation platforms off road to significantly reduce development costs and bridge the gap between simulation and the real world.
“Intel Labs has already made strides in advancing autonomous vehicle simulation through several projects, including the CARLA simulator, and we are proud to participate in RACER-Sim to help push the next frontier of all-purpose robotics. terrain and autonomous vehicles.” That’s why we’ve assembled a team of renowned experts from the Computer Vision Center and UT Austin to create a versatile, open platform to accelerate the progression of off-road robots in all types of environments. environments and conditions. said German Ros, Director of Autonomous Agents Lab at Intel Labs
Because it matters: In the field of autonomous driving, the gap between on-road and off-road deployment is still very large. Many simulation environments currently exist, but few are optimized for the development of off-road autonomy in terms of scale and speed. Additionally, live demos remain the primary method of verifying system performance.
Autonomous off-road vehicles face significant challenges, such as the lack of real roads and extreme terrain with rocks and all kinds of vegetation that make development and testing costly and time-consuming. The RACER-Sim program aims to solve this problem by providing advanced simulation technologies to develop and test solutions, reducing the time to deploy and validate AI-driven autonomous systems.
The program: RACER-Sim comprises two phases throughout the 48-month program, which will aim to accelerate the research and development process for the design of autonomous off-road vehicles. In the first phase, Intel’s goal is to create new simulation platforms and mapping tools that mimic complex off-road environments with maximum precision (e.g., physics, sensor modeling, terrain complexity, etc.), at scales never seen before. Building large-scale simulation environments has traditionally been a resource-intensive process, making it one of the biggest challenges in simulation workflows. To address this issue, the Intel Labs simulation platform will enable customization of future maps, including the creation of massive new environments spanning over 100,000 square miles with just a few clicks.
During the second phase, Intel Labs will work closely with RACER partners to accelerate the research and development process by implementing new algorithms without the use of a physical robot. The teams will then validate the robot’s performance in the simulation, which will save a lot of time and resources. In addition, the second phase will also include the development of new sim2real techniques – the concept of training the robot in the simulation to acquire skills and then transferring these skills to the corresponding real robotic system – which enable the training of autonomous vehicles while -plot directly in the simulation.
Intel expects these new simulation tools to dramatically improve the development of autonomous systems through virtual testing, reducing the risks, costs, and delays associated with traditional test and verification protocols. Going forward, the simulation platform will go beyond validation to create AI models ready for deployment in the real world.