Driving Simulation for autonomous driving, ADAS, vehicle dynamics and motorsport

Driving Simulation for Virtual Testing

The Top 10 OEMs that were early adopters of rFpro technology have already launched road cars that started their development, not on a test track, but in a virtual environment using rFpro.  By optimising the vehicle’s design before a physical prototype existed the OEMs significantly reduced cost and time delays in the engineering process.

rFpro provides driving simulation software and Digital-Twins for Autonomous Driving, ADAS and Vehicle Dynamics development, testing and validation.

ADAS and supervised Learning Autonomous Driving

rFpro is being used to train, test and validate Supervised Learning systems for ADAS and Autonomous applications. rFpro’s weather and physically modelled atmosphere, delivering real-time reflections, shadows and lighting mean rFpro can save years from your ADAS, sensor and autonomous development projects.

rFpro has the industry’s largest library of Digital-Twins of public roads, proving grounds and test tracks, spanning North America, Asia and Europe, including multi-lane highways, urban, rural, mountain routes, all copied faithfully from the real world using rFpro’s 3D reconstruction process.

Vehicle Dynamics capable

A unique feature of rFpro, compared to traditional driving simulation solutions, is that it allows driving simulation to be used to test the vehicle dynamics of road vehicles.  By delivering a high resolution road surface in real time, while generating accurate realistic graphics without lag, professional test drivers may contribute to the engineering process while the car design is still model based.

This is also critical for the testing of ADAS and Autonomous Vehicles, where the behaviour of the vehicle being simulated must correlate closely with the real-world vehicle.  rFpro’s accurately modelled Digital-Twins ensure that the virtual sensors, such as cameras and LiDAR, are subject to heave, pitch and roll as the simulated vehicle drives over accurately replicated bumps in the road and also while braking and cornering.

Saving Cost, Risk and Time Delays

Several studies have estimated the number of kilometres that must be driven to prove the safety of an autonomous vehicle. The estimates vary as to how many billions of kilometres must be driven, but what is consistent is that the number of km’s testing required is simply too great, too slow and too expensive if it is completed on roads. rFpro can test, and prove safety, quicker and more effectively. Thousand-year ‘5-sigma’ events can be tested every second, and simulation will form part of the regulatory framework for AVs.

When you are ready to start testing with a real human driver, A driving simulator or engineering workstation using rFpro’s simulation software allows you to conduct testing and calibration with your professional test drivers and engineers without the delays and cost associated with waiting for the testing of real cars or prototypes.

The saving in risk and cost can be significant and we have seen full scale systems pay for themselves on the first virtual test project.

Test earlier, reduce cost of change

A driving simulator running rFpro allows you to test changes, with a human driver, earlier in the design cycle than would otherwise be possible when relying on testing real prototypes and test vehicles.

The earlier that you can identify changes, through simulated testing in rFpro, the less costly those changes are to implement.

Making changes while the engineering process is still largely model-based is relatively inexpensive compared to the cost of making changes to a real vehicle, once you have reached the physical prototype stage.

Test more, test earlier, save $millions

rFpro is being used to train supervised learning systems, to test and validate SAE Levels 4 and 5 autonomous vehicles and to develop and test ADAS systems and sensor models.

rFpro is also being used for vehicle dynamics and drivetrain engineering; to calibrate and test passive chassis designs, steering systems, chassis control systems, drivetrain control systems, traction control, stability control, torque vectoring and engine control systems.

The way in which TerrainServer captures every LiDAR scanned data-point within the tyre contact patch, and integrates them all to provide our vehicle model with accurate road input, improves correlation with our measured data and also feels more realistic for the driver.

Giacomo Tortora
Head of Simulation, Ferrari

Our Clients