Habari protecting crops from climate change
Protecting crops from climate change so they don’t go to waste, but without the limiting factor of humans in the loop or ordinary money. A huge effort, taken up by Team Habari at the Odyssey Hackathon in Groningen. Their solution: connect farmers all over the world using shared knowledge to optimally protect and grow crops using artificial intelligence.
Habari wants to do this by anticipating pro actively on climate change by using neural networks and artificial intelligence, combined with models on how to allocate food or water. So their solution is not only trying to thwart frost damage, but all weather effects.
Climate change will have a devastating effect on world food rations, unless we know how to protect our crops. One of the big problems is loss of crops by frost and not only in areas you’d expect frost to be a problem, therefore the team chose frost as its first use case.
For example, in parts of Africa frost is a problem which returns every year, says Nasa. The thin layer of white frost causes millions of dollars in damage to crops every year to crops such as tea or coffee. Farmers need to know under what conditions frost forms, how often it occurs and where it will occur next and what they can do to avoid heavy damages.
Habari’s solution uses images from satellites and radar, combined with information on the ground which is then monitored and analyzed by artificial intelligence. To make sure the information is stored, shared and used all over the world, Habari builds a blockchain system to combine all this shared knowledge and put it to work for all farmers over the world.
Apart from the digital solutions, Habari also describes how to solve more physical solutions, like by providing design blueprints for physical crop protection in form of novel bio-inspired automated plant covers. The covers should be affordable, sustainable, non-disruptive in its environment and made by locally abundant materials and waste.
The underlying system will consist of multiple parts, think of the ‘digital twin’ of the machinery which is used and the same goes for the crops. This digital twin is then monitored and analyzed by the systems in the background, including information from sensors in the vicinity of the crops and the fields.
All data combined, creates a data flow with which autonomous agents, like drones, 3D printers or anything you can think of necessary for farming crops, can perform actions. Because of the use of digital twins, the system also knows when parts are broken and which need fixing, which need replacement and which can be reused as a source of materials for new products.
For anyone reading about this for the first time: it sounds like a grand effort. Too much for only 48 hours of building! And that’s correct, this is the starting point to build a minimal viable product. The view of the team is to deliver a neural network for crop loss prediction, an AI based crop damage assessment system after a catastrophic event, a simulation of local material supply chain, a blockchain containing participants and models, an endpoint where internet of things-devices can connect to drones, 3D-printers etc. All that is monitored by a dashboard with transaction and crop data and a login for farmers and contributors.
