The Design Automation Conference (DAC), which takes place in San Francisco this year, will see the culmination of a contest involving more than 100 teams vying to demonstrate the best use of machine learning on embedded hardware in a flying drone.
The contest was created and organized by Chris Rowen of Cognite Ventures, Professor Yiyu Shi, of the University of Notre Dame, Professor Jingtong Hu of the University of Pittsburgh, and Professor Bei Yu of the Chinese University of Hong Kong.
The call for participating teams started in September 2017 with an overwhelming response of 143 teams applying. After review, the organizers selected 114 teams to compete in the contest. Fifty teams chose to develop their drone technology on the PYNQ Platform from sponsor Xilinx. This platform uses Python as the core programming language together with libraries to access the programmable logic and microprocessors on Xilinx’s Zynq embedded-CPU FPGAs. Other teams opted to use a Jetson TX2 from fellow industrial sponsor nVidia, which is based on a GPU computing running the TensorRT software. The third industrial sponsor DJI is the specialist with the drone hardware.
Teams are competing in the six months leading up to the 55th Design Automation Conference, which is being held June 24–28, 2018, at Moscone West Center, San Francisco, CA. The competing teams will post design results on the last Tuesday of each month, starting Tuesday (February 27, 2018). Fellow engineers and embedded design enthusiasts can follow the progress of each team on a dedicated page at the DAC website.
“It’s so rewarding to see industry and innovators come together to enable superb system design,” said Professor Sharon Hu, general chair of the 55th DAC. “I can’t over-state the energy and generosity of the organizers and contest sponsors — nVidia, Xilinx, DJI, and ACM/SIGDA — to bring this contest to DAC.”
“As the premier conference devoted to the design and design automation of electronic circuits and systems, DAC is the perfect venue to have such a contest focused on the latest technology when it comes to artificial intelligence and machine learning,” she added.
The System Design Contest awards will be given to the top three teams in both FPGA and GPU categories. The top winners will be at DAC in San Francisco to discuss the project and present their winning designs.