WUR challenge: producing tomato crop remotely using AI

Answering that question is the goal of the International Autonomous Greenhouse Challenge, organised for the second time by Wageningen University & Research (WUR) and Tencent. In this challenge, which begins this autumn, multidisciplinary teams from around the world will use artificial intelligence to remotely produce vegetables.

Produce a cherry tomato crop

The goal of the challenge is to produce a cherry tomato crop within 6 months with high quality, high productivity and high resource efficiency in greenhouses of WUR, the Netherlands, remotely. Teams will get their own greenhouse compartment and make choices with respect to the control settings of greenhouse actuators and crop management in order to control the tomato production and quality remotely. Teams can also add their own sensors/cameras to generate additional information.

_{Text continues underneath tweet}

Join our challenge Autonomous greenhouses 2nd Edition: “Boost greenhouse vegetable production by artificial intelligence”. We would like to invite computer scientists and horticultural experts to bring in their experience, knowledge and passion! https://t.co/GkCLTlYVS7

— WUR Glastuinbouw (@WURglas) 16 May 2019

Machine learning algorithms

Each team will be able to extract data from their greenhouse compartment and couple it to their own machine learning algorithms to decide on the control settings for the next day/period. They will also send the control settings back to the system so it can control the actuators automatically or send instructions for crop handling to reach a pre-defined goal. WUR will continuously measure performance criteria per compartment and share them with each team and the public.

Artificial Intelligence algorithms to control cucumber production

During the first edition of the Autonomous greenhouse challenge in 2018, 5 international teams were challenged to control a greenhouse cucumber production during a 4-month period with their artificial intelligence algorithms. The first edition resulted in a successful benchmark experiment demonstrating that these algorithms can control greenhouse climate, irrigation and crop growth remotely. The winning team outperformed experienced manual growers.

Who can join?

Teams must consist of experts with a proven background in different fields such as artificial intelligence, sensor technology, crop physiology and horticultural production. Companies and start-ups are invited as well as scientists and students. Team must include at least 3 members. At least 1 team member must be a student. WUR and Tencell encourage teams from all countries to participate.

Registration will be open from 22 May – 15 July 2019.