3D IMAGING SENSOR
360° FMCW radar for industrial applications
Project Idea & Initial Situation
In the RadRob project, a 3D imaging radar system was developed to enable high-resolution environmental sensing for robotics and industrial applications. The objective was to generate real-time three-dimensional images independent of lighting and visibility conditions.
Development Steps
Development of a MIMO radar front end with 360° coverage
Design of antenna arrays with dielectric lenses for enhanced angular resolution
Integration of multiple sensor modules via CSI-2 multiplexers
Development of the RadRob 360° Viewer for real-time visualization of range–Doppler maps and 3D target data
Calibration procedures for dynamic scenarios
Results & Demonstrators
The demonstrator developed within the project detected objects at millimeter-level resolution and enabled precise detection and motion tracking, even for obscured or fast-moving targets. The technology was successfully tested in robotics and security applications.
Technological Basis
The system combines 3D MIMO radar, advanced signal processing, and synchronous sensor data fusion. Modular interfaces allow the system to be scaled and integrated into existing platforms. The RadRob Viewer software visualizes all measurement data in real time and serves as a development and analysis tool.
Cooperation
The RadRob sensor was developed in cooperation with TU Berlin.
RadRob 3D-Imaging Sensors: the Pros
There are three different competing sensor technologies on the market today, which are also used in robotic systems (stereo camera systems, LiDAR, radar systems).
| Stereo-camera-System | LiDAR | Radar-Sensor | RadRob | |
|---|---|---|---|---|
Sample system: Microsoft Kinect | sample system: MRS6000 | Sample system: LRR4 | AiM of the project | |
| Auflösung | ||||
| Resolution | 2 mm @ 2 m | > 1 mm | > 10 mm | > 1 mm |
| Latenz | ||||
| Latency | <10 ms | <100 ms | 10 ms | < 1 ms |
| Position Bestimmung von Gegenständen | ||||
| Position determination of objects | Not possible | Not possible | Not possible | Possible with an accuracy of 5 mm |
| Field of view | ||||
| Field of view | 58° | 120° (horizontal) 15°(vertical) | 120° | 360° (horizontal) 45°(vertical) |
| Kosten | ||||
| Costs | 50 € | 6.500 € | 120 € | < 1000 € |
| Quelle | ||||
| Source | [1] | [2] | [3] | |
[1] „researchgate,“ [Online]. Available: https://www.researchgate.net/figure/The-Microsoft-Kinect-3D-Camera-Sensor-System-an-IR-transmitter-3D-Depth-Sensors_fig15_309740491. [Zugriff am 05 03 2021].
[2] P. Boulay und A. Debray, „LiDAR for Automtive and Industrial Applications,“ Yole Developpement, France, 2020.
[3] „bosch,“ [Online]. Available: https://www.bosch-mobility-solutions.com/en/products-and-services/passenger-cars-and-light-commercial-vehicles/driver-assistance-systems/automatic-emergency-braking/long-range-radar-sensor/. [Zugriff am 11 03 2021].
The angular resolution allows the radar to distinguish between objects that are close together.