In order to improve usability and safety, modern unmanned aerial vehicles (UAVs) are equipped with sensors to monitor the environment, such as laser-scanners and cameras. One important aspect in this monitoring process is to detect obstacles in the flight path in order to avoid collisions. Since a large number of consumer UAVs suffer from tight
Welcome to the advisory board, University of Glasgow University Avenue, Glasgow G12 8QQ www.gla.ac.uk
Welcome to the advisory board, Motion Impossible Unit 12 Brunel Court, Dean Road, Yate, Bristol. BS37 5NR +44 1454 501 010 info@motion-impossible.com motion-impossible.com
Fraunhofer demo TULIPP hardware at the European Machine Vision Association Forum, 5th - 7th September 2018, Bologna, Italy. The demo here has a dual/stereo camera setup. This solution weighs less than a depth sensor, is cheaper and can perform real-time depth estimation for obstacle/collision avoidance. Download a copy of the Poster
On Thursday September 6th, Fraunhofer IOSB will present TULIPP at the 3rd EMVA Forum in Bologna. We will show a live demo of real-time stereo processing on the EMC2-DP board equipped with a Xilinx Ultrascale+ during the poster session. Looking forward to seeing you there! https://www.emva.org/
AI is destined to be the future of agriculture but there are significant challenges involved, the TULIPP Agri Kit will help you productively develop safe, reliable, intelligent and efficient robots by providing: Certifiable hard real time performance Integrated artificial intelligence leveraging custom accelerators A scalable, high performance platform with guaranteed time and space isolation ROS
Welcome to the advisory board, University of Leicester 7-9 Salisbury Road Leicester Leicestershire LE1 7QR United Kingdom www.le.ac.uk/innovation Tel. +44 1162 522 185
The first version of the TULIPP tool-chain is now available to the public. The TULIPP toolchain is called STHEM - an acronym for Supporting uTilities for Heterogeneous EMbedded image processing platforms. The key feature of STHEM is that it automatically and non-intrusively captures fine-grained performance and power consumption profiles. This enables the developer to quickly determine
The performance of the reactive obstacle avoidance was tested on real objects in a hardware in the loop (HIL) setup. In our approach the disparity image is calculated on Xilinx Ultrascale+ FPGA and transformed into U-map and V-map. The images are filtered and waypoints are estimated. The application runs in the DJI Simulator.
Welcome to the advisory board, Erle Robotics Calle Venta de la Estrella, 6 Pab. 130, 01006 Vitoria, Álava, Spain http://erlerobotics.com Tel. 0034 945 30 68 20
