The VI Sensor is a dedicated, high performance hardware platform for visual-inertial odometry applications. It features a high-quality global shutter HDR stereoscopic camera and a tactical-grade inertial measurement system that are thoroughly calibrated and time synchronized. The embedded FPGA and ARM Dual Core Cortex-A9 platform provides enough power for your real-time localization and navigation tasks in- and outdoors.
Embedded Feature Detection and Extraction
In a computer vision pipeline, feature detection is one of the most computationally expenstive subtasks. The XILINX based FPGA allows to outsource Fast Harris Corner and BRISK feature detection and extraction to dedicated hardware in order to save high-level computing power for other tasks.
The VI Sensor can be connected to high-level computing platforms using the USB3 or Ethernet interface. Output data is provided at fast rates (Images 20 Hz, BRISK features 20Hz and IMU 1000 Hz) to enable highly dynamic applications. Its small footprint (130 x 30 x 65 mm), low weight (< 200g) and power consumption below 10 W make it ideal for applications in mobile robotics – even on UAV’s!
Ready for your Research!
The VI Sensor is fully ROS and OpenCV compatible: In combination with state-of-the-art visual odometry frameworks, you gain easy access to accurate real-time pose estimation without having to design your own hardware.
Don’t miss any news about the upcoming product launch!Sign Up for the Skybotix VI Sensor Newsletter!
Related Papers & Work
The sensor technology is developed in collaboration with the Autonomous Systems Lab, ETH Zurich where prototypes of the sensor have already been extensively field-tested. Refer to the following extract of realized projects & published papers:
Within the ECHORD TUAV project the VI-Sensor has been used for dense stereo. The resulting data and the sensor pose output has been used to build up an octomap.
Work done by:
R. Voigt, J. Nikolic, C. Hürzeler, S. Weiss, L. Kneip, R. Siegwart, “Robust Embedded Egomotion Estimation“, Proc. of The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), San Francisco, September 2011.
Burri, M.; Nikolic, J.; Hurzeler, C.; Caprari, G.; Siegwart, R., “Aerial service robots for visual inspection of thermal power plant boiler systems“, Applied Robotics for the Power Industry (CARPI), 2012 2nd International Conference on , vol., no., pp.70,75, 11-13 Sept. 2012