
Virtual testing
rFpro provides driving simulation software and Digital-Twins for Autonomous Driving, ADAS and Vehicle Dynamics development, testing and validation. You can immerse the complete autonomous vehicle control system into the virtual world using sensor models, your vehicle dynamics model, your controllers and real world locations.
Sensor realistic simulation
Claytex develops camera, radar and LiDAR sensor models based on physics for use with rFpro. These models are representative of the real sensors and include weather and other noise effects.
Scenario based testing
Our Simulation Manager enables scenario based testing with support for OpenSCENARIO and OpenDRIVE. Define the intended trajectory for the ego vehicle, traffic and pedestrians and measure the system performance. Reuse a single scenario in multiple virtual worlds and vary weather and time of day to challenge your system.
Measuring your performance
Simulation Manager for rFpro includes metrics that can be calculated to measure your system performance during a test and quantify the risk. It provides an extendible plugin mechanism to allow you to add your own metrics to measure performance.
Ground truth
In simulation we can generate pixel perfect, labelled, ground truth data to train and test your perception layer. It can also be used to bypass the perception layer and focus on the planning and actuation part of the control system
Use your vehicle model
rFpro has interfaces to a wide range of vehicle dynamics software including Dymola, IPG CarMaker, SimPack, VI-Grade, Carsim, Simulink and also offers an open API enabling support for others
Real time capable
rFpro supports both real-time and “as fast as possible” simulation modes. This enables both HiL testing of your control system as well as non real-time SiL testing of your control logic. It’s also possible to generate high quality training data by running simulations slower than real-time to capture motion blur and shutter effects into each camera frame.
Control system integration
Our sensor models replicate the output messages from the real devices they represent easing the integration of your controller into the simulator. Generic sensor models also output UDP messages and receivers can accept UDP to bring the “driver” demands back into the simulator. A ROS interface is also available.
Related Articles
- AVS #8: Sustainability, and not over-testing
Given the recent energy market, companies that use simulations have a surprise waiting for them on their balance sheets. Computation is getting expensive. Uncomfortably expensive. Engineering excellence to improve it is needed again. The truth is, there always are costs to simulation. 1 hour of full-scale simulation in general generates approximately 1 kg of CO2. Delivering a complete Safety Case […]
Read More » - Animation of Scenarios
Simulation can be a time-consuming process between creating test scenarios, running simulations and verifying the results, especially if something goes wrong. Whether it’s a vehicle not behaving as you would expect or some interaction between vehicles causing one or more to stop following the route they are given, it all takes time to solve. Streamlining this process and detecting problems […]
Read More » - Ray Tracing Algorithms for Radar Modelling
In previous blog posts we have described various aspects of the radar simulator such as the antenna (Modelling of Phased Array Antenna – Claytex), the interaction with matter (Radar Modelling – Reflection of an EM wave – Claytex) and the validation process (Verification & Validation of a Simulation Model – Claytex). We will now talk about the algorithm it is […]
Read More »
How do we prove that autonomous vehicles are safe?











