rFpro has developed AV elevate, an award-winning, fully-integrated simulation solution designed to accelerate the development of autonomous vehicles (AVs) and Advanced Driver Assistance Systems (ADAS). Offering unrivalled accuracy, AV elevate enables comprehensive closed-loop perception testing and creates engineering-grade synthetic training data, reducing the reliance on real-world data collection and testing.
Simon George, Director, Embedded Software – Solutions Engineering, AMD
AV elevate enables engineers to tune, train, and test the entire AV technology stack in a single, unified platform. This includes everything from sensor system tuning to training perception algorithms and testing control software, ensuring rapid development cycles while significantly lowering costs.
The platform integrates high-fidelity models for all major AV sensor types, including lidar, radar, and cameras, allowing for sensor fusion testing with unmatched accuracy. AV elevate’s synchronous architecture supports the parallel testing of hundreds of sensors, making it the ideal solution for optimising sensor configurations. rFpro’s extensive library of standard sensor models also includes digital twins of commercially available sensors, such as Velodyne’s HDL-32 lidar, enabling development to proceed before physical hardware exists.
AV elevate’s Simulation Manager simplifies the setup and execution of test scenarios. It supports industry-standard frameworks like OpenSCENARIO® and OpenDRIVE®, allowing for seamless integration with third-party databases. It enables elements, such as sensor configurations, traffic conditions, pedestrians, weather, and street furniture, to be quickly varied. It enables the creation of focused variations of the base scenario to be generated, creating hundreds of edge-case scenarios for testing.
Traditional methods of manually annotating training data are both time-consuming and error-prone. AV elevate revolutionises this process by automating the generation of synthetic training data that is 100% accurate, error-free, and 150 times more cost-effective than manual annotation. This approach not only speeds up training but also reduces the need for extensive real-world testing and data collection.
AV elevate generates and subjects perception and control systems to a massive array of simulated driving scenarios to develop, train and test them. Due to AV elevate’s synchronous architecture, the time to complete tests is directly linked to the available computing power. As the platform is cloud and High-Performance Computing (HPC) compatible, it allows users to flexibly and rapidly scale testing as required without significant investment in internal computing hardware.
At its core, AV elevate leverages rFpro’s industry-leading ray tracing rendering technology and physically modeled virtual environments. With a library of more than 180 real-world digital twins, AV elevate produces high-fidelity data, replicating how sensor systems perceive the world. This includes crucial phenomena, such as motion blur and rolling shutter effects, providing the highest quality training data and testing environment for AVs.
Designed to remove the biggest barriers to AV development, AV elevate ensures faster, safer, and more cost-effective testing. Its powerful synchronous architecture, coupled with cutting-edge sensor models and simulation tools, makes it the most advanced solution for tuning, training, and testing autonomous vehicle systems. Whether you are an OEM, a Tier 1 supplier, or a sensor developer, AV elevate will help you bring safer, more reliable AV technology to market faster.
Get in touch with the rFpro team to learn how AV elevate can accelerate your path to market while slashing your development costs.