Fully autonomous control of greenhouse lettuce production
The winners of the first stage of the Autonomous Greenhouse Challenge will be announced during an open access webinar on July 14. The first part of the 3rd edition of the Autonomous Greenhouse Challenge takes place in the form of an open online challenge. It aims to test the machine learning and computer vision skills of participants in the artificial intelligence community. During the webinar, the winners of the Online Challenge will be announced, the best teams will present their strategy and a panel of experts will discuss the future of Autonomous Greenhouses.
The Autonomous Greenhouse Challenge is an initiative of Wageningen University and Research (WUR) and Tencent. By organizing the challenge, they want to connect the world of artificial intelligence (AI) and food production, create more knowledge, make this knowledge accessible to the public and thus contribute to global efforts to make our food systems more sustainable.
The first part of the 3rd edition of the Autonomous Greenhouse Challenge takes place in the form of an open online challenge, aimed at testing the machine learning and computer vision skills of AI experts from companies and start-ups , students and researchers from universities and research institutes. In the online challenge, they are rated based on how they virtually grow lettuce and judge images of lettuce at different stages of growth.
The online challenge runs throughout the month of June. More than 50 teams and 250 participants are registered.
In Part A – The Computer Vision Challenge – teams must develop computer vision algorithms based on training images and must use the algorithm to detect plant characteristics (diameter, height, weight and leaf area of the plant) on invisible images. Photos are provided by WUR and will be made public after the online challenge.
In Part B – The Machine Learning Challenge – teams must develop machine learning algorithms to autonomously control ventilation, heating, lighting, etc. in a virtual greenhouse to stimulate the growth of a virtual crop and maximize the net profit. WUR provides access to a twin digital model of a lettuce greenhouse (simulator) as a play environment.
On July 14 from 2:00 p.m. to 3:30 p.m. CET will be the final event of the online challenge. During a webinar, the winners of the Online Challenge will be announced, the best teams will present their strategy and a panel of experts will discuss among themselves and with the other participants the future of Autonomous Greenhouses. The webinar will be broadcast publicly via the WUR Glasshouse Horticulture Business Unit’s YouTube channel:
More information on the Autonomous Greenhouse Challenge and the July 14 program is available at www.autonomousgreenhouses.com.
Wageningen University & Research and Tencent
Wageningen University & Research (WUR) is ranked as the best agricultural university in the world and works on, among others, stand-alone greenhouses, climate control, crop modeling, use of sensors, data, intelligence, computer vision and robotics. Tencent is the sixth largest publicly traded company in the world. Its mission is to improve the quality of human life through Internet services.
Together, they want to connect the world of artificial intelligence (AI) and food production, create more knowledge, make this knowledge accessible to the public and thus contribute to global efforts to make our food systems more sustainable. The ever-growing world population is creating an increased demand for fresh and healthy food. Self-contained greenhouses and remote digital agriculture can help feed more people with products rich in vitamins and minerals, increase food security, and produce more vegetables with fewer resources such as water and energy. The potential has been successfully demonstrated in previous editions of the Autonomous Greenhouse Challenge.
The other sponsors of the Autonomous Greenhouse Challenge are Fluence by Osram, Gebr. Geers bv, Municipality of Lansingerland, Glastuinbouw Nederland, Hortiplan, Kas als Energiebron, LetsGrow, Ridder Climate Screens and Sigrow.