European Commission logo
English English
CORDIS - EU research results
CORDIS

AI Adviser for Agronomy and Food Safety

Periodic Reporting for period 1 - ROBI (AI Adviser for Agronomy and Food Safety)

Reporting period: 2022-06-01 to 2023-05-31

Each year, a significant portion of global food crops, approximately 40%, is lost due to the detrimental effects of pests and plant diseases, leading to an alarming wastage of 1.3 billion tons of food. The European Union (EU) and individual countries have imposed regulations that restrict the quantity of chemicals farmers can utilize to combat crop diseases. However, this presents a challenge as farmers face limitations in terms of time and expertise, making it difficult for them to navigate and adhere to the complex set of rules while making informed decisions regarding the most effective strategies to address the issue.

With this project, AGRIVI is focused on developing an innovative AI-driven digital advisory tool for farmers. The main objective of the project is to provide guidance for SME farmers through the process of when to spray based on AI pest risk detection, which products to use, which rules must be respected and when is it safe for harvest while taking into consideration regulations and making sure food safety standards are met, while utilizing the power of artificial intelligence (AI) and mobile technology.

The motivation behind this project comes from our understanding and knowledge of the challenges faced by farmers in optimizing their agricultural practices and ensuring the safety and quality of their produce. Farmers often encounter difficulties in making informed decisions regarding crop spraying and choosing the appropriate crop protection products. These decisions significantly impact yield maximization, waste reduction, and compliance with safety and quality standards.

The project's impact pathway includes developing a WhatsApp-based chatbot as the channel for the AI-driven digital advisory tool. The chatbot will be based on AGRIVI's proprietary AI pest risk detection algorithm, which analyzes environmental factors and crop data. By utilizing this algorithm, the chatbot will predict pest infestation risks, provide real-time spraying information, and offer personalized recommendations to farmers. The intuitive conversation-interface of the chatbot will enable farmers to ask questions and receive tailored advice that addresses their specific needs.

The expected impacts of the project are significant. As we provide users with access to cutting-edge technology, the tool aims to improve their yields, reduce waste, and ensure the safety and quality of their produce. It enforces making informed decisions based on accurate data and expert recommendations. Moreover, by optimizing crop spraying and harvest planning, the tool contributes to sustainable agricultural practices.

Overall, this project impacts encompass improved yields, reduced waste, and enhanced safety and quality standards by utilizing AI and mobile technology to provide farmers with timely and accurate advice on crop spraying and harvest planning.
In this project, several significant activities have been performed, leading to notable achievements on the technical aspects of the project.

The technical solution design phase was successfully executed, resulting in development of software specification which serves engineering teams as development blueprint throughout the execution phase. Software specification contains thorough information on the overall technical solution for ROBI that entails details about infrastructure, development, and quality assurance requirements for seamless delivery of the product.

All development activities were successfully initiated and are well underway with a focus on integrating the CPP databases, advancing algorithm development for spraying recommendations, chatbot, integrations and testing. A significant milestone has been reached with the development of the first chatbot prototype, which is currently undergoing rigorous testing. This testing phase aims to ensure the chatbot's functionality and effectiveness in meeting the project's objectives.

Simultaneously, the internal agronomy expert team has been actively involved in data collection and labeling for AI model development. This ongoing process played and plays a vital role in enhancing the chatbot's capabilities and optimizing its performance by utilizing high-quality data.
The project has initiated early market testing activities, which are centered around gaining a deep understanding of the market and defining the most effective go-to-market approach for the product. These activities involve extensive research and analysis to identify the target audience, assess market needs, and determine the optimal strategies for introducing and promoting the product.

As part of the piloting phase, the project aims to refine the product iteratively based on valuable feedback collected from real end-user representatives. These representatives are selected from agricultural cooperatives and growing organizations, ensuring that the feedback is directly from those who will benefit from and interact with the product daily.

The feedback received from the end-user representatives will be carefully analyzed and used to drive iterative improvements in the product. This iterative process allows for the incorporation of valuable insights, addressing any issues or gaps and enhancing the product's functionality, usability, and overall performance. By actively engaging with the end-users during the piloting phase, the project can gather firsthand feedback, validate the product's value proposition, and make necessary adjustments to meet the specific needs and preferences of the target market. This iterative approach ensures that the final product aligns closely with market requirements, leading to greater user satisfaction and increased chances of successful market adoption. To address the potential language barrier associated with introducing a new product, our focus is on developing a solution that can be effortlessly localized and adapted to suit the specific needs of the target market. This approach maximizes the product's potential for success in diverse linguistic environments, facilitating seamless communication and engagement with users from various cultural backgrounds once the new product is commercially available.