Sensor Simulation: Environment Perception for ADAS and AD

The detection of the environment builds the basis for assisted and automated driving (ADAS/AD). For the development and testing of driving functions, the vehicle and environment simulation DYNA4 offers models of lidar, camera, radar and ultrasonic sensors. Depending on the boundaries of your system under test, DYNA4 provides virtual sensor input on different levels, from physics-based raw sensor data to fused object lists. This enables, for example, the following use-cases:

Testing object detection with virtually generated camera pictures or lidar point clouds
Testing sensor fusion algorithms with object lists from multiple sensors such as radar and camera
Testing ADAS/AD functions with a fused object list as an input

Sensor input on different levels from physics-based raw sensor data to fused object lists

Advantages

Real-time capable simulation for development and testing of sensor-based ADAS/AD functions
From early development phases to determine the sensor configuration up to virtual validation
Full sensor set-up with ultrasonic, lidar, camera and radar sensors
Output as raw data, target lists or object lists
Variety of scenarios covering complex static and dynamic environments from parking to autonomous driving in surrounding traffic
Closed-loop system tests from environment perception over actuation to realistic vehicle reaction
Vehicle dynamics for realistic sensor movements
Efficient test coverage through test automation with numerous variants

Object Lists

Object list: Semantic image segmentation

Sensor-specific object lists or fused object list
Idealized ground-truth information or consideration of occlusion
Highly efficient computation based on bounding-boxes or semantic image segmentation with consideration of exact geometries
Provide simulated sensor data in ASAM OSI format

Lidar

Simulation of Lidar point cloud in DYNA4

Reflection intensity based on angle between laser beam and object surface and its material properties
Availability of rotating and non-rotating lidar sensors
Opening angle and signal resolution adjustable
3D point cloud output as ROS Topic via DDS or via UDP in Velodyne format
Video: Vehicle simulation with Velodyne sensor and city traffic in DYNA4

Cameras

Configurable cameras with ultra wide angle view

Configurable cameras with opening angles up to 360°
Distortion parameterization with OpenCV or Scaramuzza parameters
Support of dirt on the lens
Display of RGB image streams on separate screens for image injection
Usage from MiL (algorithm development) to HiL (image injection on ECU)
Video: Configurable cameras for virtual ADAS testing with DYNA4

Ultrasound

Simulation of ultrasonic sensors in a parking scenario

Consideration of propagation and atmospheric attenuation
Absorption and reflection based on object geometry and material properties
Adjustable opening angle and signal resolution
Output of an intensity depth histogram

Radar

Simulation of radar sensors as input for ADAS function development, e.g. ACC, AEB, BSD etc.

Scattered radar waves based on object geometry and material properties
Consideration of different antenna characteristics (Short-, Mid-, Long-Range)
Adjustable opening angle and signal resolution
Output of raw data such as relative velocity and distance to the object as well as the intensity of the electric field or GPU-based Fourier transform to generate range Doppler plots