Sensor Models for a more Realistic Simulation

BASELABS Models for virtual testing of environmental models and ADAS

Virtual test and validation of ADAS ADASsystems is an established procedurepractice to reduce the effort for test drives with real vehicles. BASELABS Models enhances established simulation environments by more realistic sensor models. It injects typical errors of ADAS sensors into the simulation to increase the degree of realness and validity of the virtual test.

OVERVIEW

Sensor Models for Virtual ADAS Sensors

BASELABS Models extends simulation tools by more realistic sensor models. The models increase the degree of realness of the simulated sensors to enhance the validity of the virtual test of ADAS and automated driving systems. BASELABS Models contains generic sensor models, which mimic important characteristics of typical ADAS sensors, e.g. Continental ARS308 or Mobileye. Among others, these characteristics contain measurement inaccuracies, false alarms and delays. All models are real-time capable and can be used in closed loop simulations. BASELABS Models is available for PreScan, CarMaker and CarSim.

Overview

Integration in Validation Environments

BASELABS Models can be used in in-the-loop setups. This includes Model-in-the-Loop (MiL), Software-in-the-Loop (SiL), Hardware-in-the-Loop (HiL) and Vehicle-in-the-Loop (ViL). The integration into evaluation tool chains based on Matlab Simulink is possible. The implementation for PreScan supports the integration into Simulink out of the box.

Overview

Reproducible Results

Simulations can be run either in an explorative way to cover the broadest variety of possible sensor errors or they can be repeated with the same modelling of sensor input with the same distribution of injected errors to evaluate a critical constellation in a reproducible way.

Loading the player ...

Usage of BASELABS Models in PreScan with simulation of true and false detections.

Working Principle

BASELABS' expertise originates from the development of data fusion systems which can handle the error-prone detection characteristics of real ADAS sensors. For BASELABS Models, we invert this knowhow to inject errors into the simulation that are comparable to the characteristics of real sensors. The user of BASELABS Models can modify this characteristics by setting parameters for the injected errors to adjust the behavior of the simulated sensor.

Sensor Models

Radar Detector Model

BASELABS Models contains a radar sensor model that simulates typical automotive radars, e.g. Continental 308/408 (targets). The Radar Detector Model is suitable for radars that deliver detections, i.e. measurements of objects that have not been filtered or accumulated over time. The model simulates noisy measurements, missed detections or false negatives, clutter or false positives and latency.

Sensor Models

Smart Radar Model

BASELABS Models contains a radar sensor model that simulates typical automotive smart radars, e.g. Continental 308/408 (tracks). The Smart Radar Model is suitable for radars that deliver tracks, i.e. objects whose measurements have been filtered or accumulated over time. The model simulates noisy measurements, missed detections or false negatives, clutter or false positives and latency including latencies among the provided values of a track, e.g. the velocity and acceleration are delayed compared to the position.

Sensor Models

Infusion of Typical Errors of Radar Sensors

True Positives with Noise

False Negatives

False Positives

Latency

Sensor Models

Smart Camera Model

To simulate smart cameras such as Mobileye devices, BASELABS Models contains a Smart Camera Model that statistically models typical effects of these devices, e.g. errors due to pitching and braking and delays of individual values of the provided objects such as velocity and acceleration.

Sensor Models

Infusion of Typical Errors of Camera Sensors

A camera sensor naturally may determine only the position of objects in the image in the camera directly. Based on the observation of those object positions over time dynamic properties of the objects are determined. However, these methods are subject to typical errors, which are modelled by the Smart Camera sensor model:

  • The filter algorithms for the estimation of the object dynamics based on a sequence of object positions, which are applied inside Smart Cameras, are subject to statistical errors as they are based on model assumptions.
  • The motion of the vehicle, such as pitching on braking, affects the position of objects in the image of the Smart Camera and therefore also the output of the filter algorithms.

Contact & Downloads

Top of page
Evaluate now!
Close

evaluate now!

Please contact us for an evaluation (+49 (0)371 3371 51 51) or leave your contact data below.

Submit
* Required fields