26 Jun 2019 |
World innovation news |
Information and Communications Technologies
Participatory Urban E-Agriculture
Automation is unavoidable in the agricultural sector. Whether using robots, or sensors to monitor crop conditions in real time, e-agriculture is about improving food quality and managing natural resources. Although the technology-agriculture duo may frighten people who associate it with GMOs, junk food, and bland foods, it is important to remember that, on the contrary, many small-scale farms are working to use technology appropriately, abandoning practices that gave the food industry its bad reputation, such as over-harvesting or using harmful chemicals. In recent years, universities have been following the trend within a framework of scientific research or the design of eco-sustainable buildings. At the Massachusetts Institute of Technology (MIT), a new research laboratory set up a farm in Middleton, Massachusetts, to democratize environmental farming practices.
Smart, Local and Participatory Agriculture
In the footsteps of urban agriculture, the OpenAg group is a cyber-plant laboratory inventing the farming methods of the future, taking climate change into consideration. In addition to food, the group aims to produce more effective medicinal plants. The laboratory uses machine learning, botany and chemistry and hopes to develop open-source technologies that study plant-environment interactions within a small-scale precision-farming context. In this regard, Caleb Harper, senior researcher at the MIT Media Lab, pointed out that data from the agricultural sector are not widely available to the public. By sharing the results of its research, the OpenAg group wishes to contribute to network science. To this end, the laboratory developed an educational and participatory project for young people to share their agricultural data on a web platform created for this purpose. University and high school students from 65 countries use “personal food computers”, designed by the laboratory, to monitor and regulate plant growth, and to send the captured information to the database.
The personal food computer
Basil is one of the cyber-plants grown in the lab’s small farm. The laboratory designed a smart experimental culture system to improve the taste of the aromatic plant. The study demonstrating the feasibility of this approach is entitled “Flavor-cyber-agriculture: Optimization of plant metabolites in an open-source control environment through surrogate modeling.” It was co-authored by Arielle J. Johnson, Elliot Meyerson, John Parra, Timothy L. Savas, Risto Miikkulainen and Caleb B. Harper, and published in Plos One on April 3, 2019. The method combines metabolomic analyses and deep learning to evaluate conditions that optimize plant flavour. This pilot experiment made it possible to study basil growing in a hydroponic farm. The study presented the following steps:
- Finding the best weather conditions, including photoperiod and exposure to ultraviolet light, which can affect the production of aromatic molecules in edible plants.
- Measuring flavour-active volatile levels in each plant using gas chromatography-mass spectrometry.
- Studying environmental influences on the basil phenotype.
The algorithm, developed by the MIT team and researchers at Sentient Technologies, analyzed millions of combinations of exposure time to light and UV, and generated sets of conditions to improve flavours.
The team is currently working on several other projects, while continuing to study basil in order to increase compounds that can be used in the treatment of disease, such as diabetes. The laboratory is also studying other growing conditions, namely temperature, humidity, colour of light, and the effects of adding plant hormones or nutrients. Another ongoing project is the development of a technique for natural insect control using chitosan, a polymer found in insect shells. The team is also working on hazel trees for candy maker Ferrero whose process consumes about 25% of the world’s hazelnuts.