April 26, 2022

Intel RACER Sim 400x275The Defense Advanced Research Project Agency (DARPA) has awarded Intel Federal LLC, with support from Intel Labs and other collaborators, the opportunity to develop advanced simulation solutions for off-road autonomous ground vehicles (AGVs). The program, titled the Robotic Autonomy in Complex Environment with Resiliency – Simulation (RACER-Sim), aims to create the next generation of off-road simulation platforms to help developers significantly reduce the cost of development, and bridge the gap between simulation and the real world.

In addition to Intel, the group includes collaborators from the Computer Vision Center Barcelona, and the University of Texas at Austin.

“Intel Labs has already made progress in advancing autonomous vehicle simulation through several projects, including the CARLA simulator, and we’re proud to participate in RACER-Sim to continue contributing to the next frontier of off-road robotics and autonomous vehicles,” said German Ros, autonomous agents lab director at Intel Labs. “We brought together a team of renowned experts from the Computer Vision Center and UT Austin with the goal of creating a versatile and open platform to accelerate progress in off-road ground robots for all types of environments and conditions.”

Intel Labs said the gap between on-road and off-road deployments of autonomous vehicles is still significant. While many simulation environments exist today, few of them are optimized for off-road autonomy development at scale and speed. In addition, real-world demonstrations continue to serve as the primary method to verify system performance.

“Off-road autonomous vehicles must deal with substantial challenges, including a lack of road networks and extreme terrain with rocks and all types of vegetation, among many others,” said Intel Labs. “Such extreme conditions make developing and testing expensive and slow. The RACER-Sim program aims to solve this problem by providing advanced simulation technologies to develop and test solutions, reducing deployment time and validation of AI-powered autonomous systems.

The RACER-Sim program includes two phases over 48 months, with the aim of accelerating the entire research and development process for designing off-road AGVs. In the first phase, Intel said it will focus on creating new simulation platforms and map generation tools that mimic complex off-road environments with the highest accuracy (for example, physics, sensor modeling and terrain complexity) at scales never seen before. “Creating simulation environments at scale is a process that traditionally requires significant resources and is one of the biggest challenges in simulation workflows,” said Intel Labs. “Intel Labs’ simulation platform will enable customization of future maps, including the creation of massive new environments covering more than 100,000 square miles with just a few clicks.”

In the second phase, Intel said it will work with RACER collaborators to accelerate the R&D process by implementing new algorithms without the use of a physical robot. Teams will then validate the performance of the robot in simulation in order to save time and resources. Phase two will also include the development of new sim2real techniques – the concept of training the robot in simulation to acquire skills and then transferring these skills to a corresponding real robotic system – enabling the training of off-road autonomous ground vehicles directly in simulation.

The company added that it expects the new simulation tools to improve the development of autonomous systems using virtual testing, which aims to reduce the risks, costs, and delays associated with traditional testing and verification protocols. “In the future, the simulation platform will go beyond validation to create AI models ready for implementation in the real world,” said Intel Labs.

Related video: