SecForCARs logo

Generation of Artificial Radar Targets

Ulm University

In order to guarantee the safety and security of autonomous vehicles, rigorous tests and verifications are required. Analogously to a crash test center, to ensure the integrity specifically of the sensors employed in autonomous cars, an environment suited for intense testing of the sensors employed in autonomous cars can be envisioned. In such a test center the effectiveness of the security measures and safety-relevant features is investigated.

 

While there is a multitude of different sensors like cameras, lidar-, sonar- or radar sensors, the focus of this video is on radar sensors. Due to their robustness, radar sensors are commonly used in advanced driver assistance systems (ADAS) like adaptive cruise control or advanced emergency braking. Therefore, in the project SecForCARs, a radar target simulator was developed that could be used for safety- and security testing, by generating artificial radar targets.

The demo video shows the radar target simulator in action, capable of generating multiple targets with individual ranges and velocities, allowing for high versatility. The radar target simulator is suited for chirp-sequence frequency modulated continuous waveform radars, which are commonly used in ADAS. The demo video presents briefly the basic principle of operation first. Afterward, a simulation with the target simulator is presented in which a car is approaching the radar under test laterally. Modulating the target signal onto the radar transmit signal allows for the simulation of highly complex extended targets. This approach allows to generation of targets without prior knowledge about the radar sensor by estimating the radar’s waveform parameters.