An advanced simulation system promises to remove one of the biggest barriers to the progress and ultimate adoption of autonomous vehicles, significantly accelerating their development. The reliance on real-world testing and data collection has long been one of the biggest hurdles for creating safe, effective, and efficient autonomous vehicle testing. The painstaking gathering of such data on public roads is a time-consuming, costly, and laborious process.
Also read: Resonate Testing invests in battery testing facility
The Solution: Autonomous Vehicle Testing with Simulation
The advent of a simulation system that enables the tuning of sensor systems, the training of perception and control, and the creation of 100% accurate synthetic training data is, therefore, a significant leap forward for the technology. rFpro’s AV elevate™ is a fully integrated simulation solution designed to accelerate the development of autonomous vehicles. It enables the tuning of sensors, training of perception and control algorithms, and testing of the full AV technology stack.
“AV elevate is a game-changing simulation platform. It is the most advanced solution to enable the complete AV technology stack to be tuned, trained, and tested,” said Matt Daley, Technical Director at rFpro. “For the first time, AV developers can confidently reduce their reliance on real-world testing, instead subjecting systems to AV elevate’s highly-accurate synthetic data. Now, testing and development can be massively and cost-effectively scaled like never before, removing the biggest barrier to the advancement of AVs.”
Key Features of AV elevate
rFpro provides a simulation environment for both the automotive and motorsport industries. It is used for the development and testing of autonomous vehicles, ADAS, vehicle dynamics, and human factor studies. rFpro’s automotive customers include the world’s largest car manufacturers, tier one suppliers, and sensor developers. The platform allows them to simulate, test, and validate new sensors, control systems, and vehicle hardware systems. This simulation environment plays a crucial role in autonomous vehicle testing, ensuring that systems are rigorously tested and validated before real-world implementation.
In motorsport, rFpro is the market leader in professional driver-in-the-loop simulator software, with clients including past and present champions from leading motorsport categories.
Tuning Sensor Systems for Autonomous Vehicle Testing
To create AV elevate, rFpro has integrated several new technologies into its existing platform, including LiDAR, radar, and camera models. It also includes a new Simulation Manager that simplifies the definition of the full vehicle sensor suite, allowing the creation of base test scenarios with thousands of iterations. The platform is also compatible with High Performance Computing (HPC) in the cloud, enabling rapid scaling of testing.
AV elevate integrates high-fidelity sensor models for all major AV sensor types, allowing for installation choices and configurations to be tuned and optimized. The platform’s synchronous architecture enables hundreds of sensors to be tested, allowing sensor fusion testing at an accuracy level not previously possible.
Simulation for Autonomous Vehicles: Synthetic Training Data
Typically, AV developers manually annotate each frame of video, LiDAR point, or radar return to identify objects in the scene to create training data. This approach generally takes 20 minutes per frame and has a 10% error rate. AV elevate automates this process using engineering-grade synthetic training data, making it 100 times faster, 150 times more cost-effective than manual annotation, and completely error-free.
The Simulation Manager in AV elevate provides a user-friendly way to set up and execute test scenarios automatically, creating large-scale simulations with multiple sensor systems. It can automate variations in the vehicle model, environment, and conditions. For example, changes to sensor types, positioning of sensors on the vehicle, traffic, pedestrians, time of day, weather, street furniture, and obstructions can be quickly generated, creating hundreds of edge-case scenarios for testing.
Scalable Testing with AV elevate
AV elevate generates and subjects perception and control systems to a massive array of simulated driving scenarios. This allows for the development, training, and testing of these systems. The platform’s synchronous architecture links the time required to complete tests with the available computing power, allowing testing to scale rapidly without significant investment in internal hardware. This makes autonomous vehicle testing more efficient and cost-effective, helping accelerate the development of autonomous vehicles.
As AV elevate is compatible with cloud computing and HPC, users can flexibly and quickly scale testing as needed. This eliminates the significant barrier of limited internal computational resources, making autonomous vehicle testing faster and more cost-effective.
Virtual Environments and Real-World Accuracy
At the core of AV elevate is rFpro’s industry-recognised, physically modeled virtual environments and ray-tracing rendering technology. Its library of more than 180 real-world digital twins provides a highly accurate and diverse virtual proving ground. Every element in the scene is physically modeled with realistic material characteristics, and the road surface model is accurate to within 1mm.
rFpro’s ray-tracing rendering technology simulates the nuances of how a vehicle’s sensor system perceives the world. This multi-path technique reliably replicates the vast number of light and electromagnetic sources and reflections that happen around a sensor. It generates the highest fidelity data for sensors, including phenomena like motion blur and rolling shutter effects.
Conclusion: Revolutionising Autonomous Vehicle Testing
“The purpose of AV elevate is to help OEMs reduce their reliance on testing and data collection on public roads, significantly cutting costs and accelerating development,” concludes Daley. “We believe our fully integrated simulation solution is the most accurate available, and this gives manufacturers the confidence they need to make that move.”
With rFpro’s AV elevate, the field of autonomous vehicle testing is now equipped with the most advanced simulation platform available, enabling more efficient and cost-effective testing. As the industry moves forward, AV elevate will continue to play a pivotal role in the evolution of autonomous vehicle technologies.
