AgrAInno: Human-Centric Generative AI-Based Smart App for Weed Detection and Species Recognition

Overview

AgrAInno is funded through the TUM ForTe Bridge-to-Innovation Grant and focuses on developing a human-centric generative AI-powered smart application for weed detection and plant species recognition in agricultural fields. The project bridges cutting-edge AI models with practical farm usability, creating an intuitive tool that supports farmers, advisors, and agronomists in identifying and managing crops and weed problems efficiently.

Objectives

  • Develop a generative AI-based mobile app for real-time weed detection and species identification in field conditions.
  • Integrate human-centric design principles to ensure usability for non-expert end users including farmers and field advisors.
  • Train and evaluate deep learning models for robust weed and species recognition across diverse crop systems and growth stages.
  • Bridge innovation to application by translating state-of-the-art crop health monitoring and AI capabilities into a deployable, field-ready tool.

Funded by: TUM ForTe Bridge-to-Innovation Grant

Duration: 06.2024 – 12.2025

Agriweed App Prototype

https://apps.apple.com/de/app/agriweed/id6752901076

From Research to Innovation

The AgrAInno concept is growing beyond its academic roots. Building on the research outcomes, the project is expanding into a broader innovation and product development effort — with a dedicated platform at agrainno.com emerging as a hub for the next phase of development. This transition from university grant to real-world agri-tech innovation reflects the team’s ambition to bring AI-powered agricultural intelligence directly into the hands of farmers and advisors at scale.

Chenghao Lu
Chenghao Lu
Research assistant & Joint Phd student

My research interests are Agricultural Digital Twin and Smart Breeding. My project focuses on the construction of a Digital Twin system for maize, aiming to fuse Genetic x Environmental x Management data to construct a high-precision and dynamically updated maize growth virtual twin. By introducing a DNA language model with a multimodal fusion approach, I hope to achieve accurate simulation of maize phenotypes and breeding to support sustainable agricultural development.

Prof. Dr. Kang Yu
Prof. Dr. Kang Yu
Professor of Precision Agriculture

My research interests include precision crop farming, hyperspectral remote sensing, and AI in agriculture.