Could artificial intelligence (AI) stop a tiny South American moth from ravaging the worlds’ crops?
Also known as the ‘tomato leaf miner’, the Tuta absoluta moth loves the warm conditions of a greenhouse, and it also loves nibbling on tomatoes, aubergines, bell peppers and more. Outbreaks can be devastating. This summer, Nigeria announced a state of emergency after moths destroyed 90% of tomato crops and sent prices surging from $1.50-7.50 per basket to as high as $212.
Once a farm is infested, these tiny bugs are unstoppable. But Greek business IKnowHow has developed a solution that could stop them from taking hold.
GreenSprayer is an AI-powered platform that detects plant anomalies and instructs robot sprayers to treat them.
“It operates in greenhouse environments and works as a decision support system to the farmer,” says Christos Klaridopoulos, a computer vision and machine learning engineer at IKnowHow. “We have been developing it over the past two years.”
The cutting Edge of AI
But to really help farmers, IKnowHow needed outside support. “[GreenSprayer] required extensive computational resources and was not adequate for scaling-up,” says Klaridopoulos, adding that it would need to detect more pests than just the tomato moth to be commercially useful.
Searching for funding calls, IKnowHow came across BonsAPPs, a European Commission-funded programme supporting “Edge” AI solutions that are implemented close to where data is being collected — such as greenhouses, in GreenSprayer’s case.
Startups taking part were asked to develop scalable, low-cost AI solutions using tools available via Bonseyes, an AI Marketplace that has been developed by a consortium of European SMEs, academic institutions and corporates.
AI is one of the most important technologies being developed now.
More than 300 individuals and SMEs applied to three open calls, which took place during the three years of the project (2021 to 2023). In the first open calls, there were 41 successful applicants who received up to €74k in funding and six months of technical support, with experts on hand to help develop, scale and integrate new solutions.
Other startups that took part included Whispp, a Dutch startup that makes an app that can convert whispered or vocal cord impaired speech into a clear voice, and VidaApp, a Spanish startup which makes portable heart and breathing monitoring devices.
“AI is one of the most important technologies being developed now,” says Yolanda Moreno, project manager at FundingBox, the accelerator in charge of distributing funds and managing open calls for BonsAPPs. “For FundingBox, it’s really important to be involved in the ecosystem. At the moment, we’re participating in 12 projects in AI technology and we have close to €14m to distribute via Open Calls in the upcoming years.”
Fostering AI adoption in Europe
The BonsAPPs project is part of a wider EU push, known as the AI-on-Demand Platform, to bring AI technologies and resources to innovators in all sectors and to foster adoption of AI without having to invest huge sums of money.
All of these ICT-49 projects are developing services that interoperate with an AI on-demand platform and ecosystem, which is the main goal of the European Commission.
Just 2% of SMEs say they are widely using AI within their organisations, compared to 10% of big businesses. Under the Horizon 2020 programme, the European Commission has invested €30m in six projects, including BonsApps, which contribute to the development of a central European AI-on-demand platform.
The AI-on-Demand initiative includes other European projects, such as StairwAI, which aims to help low-tech businesses start using AI tools and performing a quick matchmaking with available AI assets and solutions, and I-NERGY, which focuses on reshaping the energy sector value chain with AI. These AI-on-Demand Platform projects are all commonly referred to ICT-49 projects, after the European Commission topic number assigned to the initiative.
“All of these ICT-49 projects are developing services that interoperate with an AI on-demand platform and ecosystem, which is the main goal of the European Commission,” says Nuria Pazos, a professor at the University of Applied Sciences and Arts of Western Europe (HES-SO), based in Switzerland.
“[We want] to have a platform supporting SMEs trying to innovate in the AI domain but who do not have the resources in house. We can provide the tooling and workflows, and try to speed up the integration of AI technology among SMEs in Europe.”
As the coordinator of the programme, HES-SO’s role was to ensure that each selected proposal was able to move through the programme, where AI talents were matched up with SMEs in order to develop, train and optimise Edge AI solutions for specific industry applications.
Through this process, GreenSprayer saw that its solution would not only help farmers improve yield, but they could also help them to reduce pesticide usage and reduce power consumption, by taking decisions based on their robots’ insights.
At the end of the programme, the solutions were put to the test with a pitch competition. More than 50 attendees voted on various awards — and GreenSprayer came away crowned as the ‘best exploitation solution’.
Klaridopoulos credits the BonsAPPs programme with solidifying the business case for GreenSprayer and creating a solution that is ready to go out onto the market.
“The programme’s ability to connect SMEs with top-tier AI talents and bridge the gap between solutions and market viability was its standout feature,” he says.