# Agri-OpenCore > Welcome to Agri-Opencore AGRI-OPENCORE is a bold initiative to resolve major issues faced by the agricultural sector and businesses driving the automation and digitalisation of this crucial industry. --- ## Pages - [AOC Scenario Base](https://agri-opencore.org/software-stacks/aoc-scenario-base/) - [AOC Fruit Detector](https://agri-opencore.org/software-stacks/aoc-fruit-detector/) - [Software Stacks](https://agri-opencore.org/software-stacks/): Available AOC Stacks: - [AOC Tomato Farm Digital Twin](https://agri-opencore.org/software-stacks/aoc-tomato-farm-digital-twin/) - [Getting started](https://agri-opencore.org/getting-started/) - [Agri-OpenCore Architecture Overview](https://agri-opencore.org/getting-started/agri-opencore-architecture-overview/): Agri-OpenCore Architecture Overview The Agri-OpenCore project follows a carefully designed architecture that promotes modularity, reusability, and clear separation of concerns.... - [Video & Media](https://agri-opencore.org/media/): Agri-OpenCore on YouTube Our video playlist, show-casing some of our results. Agri-OpenCore in the News - [About](https://agri-opencore.org/about/): Agri-OpenCore Project Project Summary Agri-OpenCore is an ambitious initiative aimed at revolutionizing the UK horticulture sector through advanced robotics and... - [Partners](https://agri-opencore.org/partners/): Partner Focus APS The APS Group is the UK’s leading supplier of British tomatoes to the high street, growers and suppliers of... - [Cookie Policy (UK)](https://agri-opencore.org/cookie-policy-uk/) - [Welcome to Agri-Opencore](https://agri-opencore.org/): Welcome to Agri-Opencore AGRI-OPENCORE is a bold initiative to resolve major issues faced by the agricultural sector and businesses driving... - [News](https://agri-opencore.org/news/) --- ## Posts --- # # Detailed Content ## Pages AOC Scenario Base Docker Environment This repository provides a Docker base image for scenarios in the Agri-OpenCore (AOC) project. The base image installs all non-ROS2 open source dependencies required to build AOC software. It's based on the L-CAS ROS Cuda Desktop Images: https://github. com/LCAS/ros-docker-images Key Components Base Image Built on top of lcas. lincoln. ac. uk/lcas/ros-docker-images (currently jammy-cuda12. 2-humble-2. 1. 0 tag) Uses Ubuntu 22. 04 (Jammy) with CUDA support ROS2 Humble distribution Main Features ROS2 Humble desktop installation Navigation2 and Gazebo simulation packages MongoDB 7. 0 with Compass GUI ZED SDK 4. 0 with CUDA support Code-server (VS Code in browser) setup MQTT support via paho-mqtt Various ROS2 packages for: Navigation Simulation Image processing Robot localization Mapping The following packages are installed from source: Package Version Repository Description/Features g2o Debian/Humble branch libg2o-release General Graph Optimization framework IPOPT Solver 3. 13 (stable) Ipopt control_box_rst v0. 0. 0 control_box_rst DecompROS humble_dev branch DecompROS kindr v0. 0. 0 kindr GTSAM commit 6c97e4b gtsam OpenVDB v10. 0. 1 openvdb OctoMap v1. 9. 6 octomap Casadi main branch casadi Built with Python, IPOPT, OpenMP, and Thread support Livox SDK2 v0. 0. 0 Livox-SDK2 CI/CD Pipeline Automated Docker image builds via GitHub Actions Builds triggered on: Push to main branch Pull requests Version tags Manual dispatch Images pushed to LCAS private registry Build The image can be build locally via Docker Compose or bake: docker compose build or docker buildx bake --- AOC Fruit Detector using Detectron2 MaskRCNN Instance segmentation of a scene and output Mask-RCNN predictions as images and json message/file (Agri-OpenCore), fully ROS2 integrated system. Installation, Requirements and Running Without Docker Installation The docker container automatically installs all required dependencies. The main dependencises are listed below. However, without dockerisation, the following required packages or check/install required versions should be installed from requirements. txt file. python3 torchvision pickle numpy opencv-python cv-bridge scikit-image matplotlib detectron2 pip install -r requirements. txt Detectron2 package is also automatically installed by Docker container. However, without dockerisation, it can be cloned from GitHub and installed into the workspace. git clone https://github. com/facebookresearch/detectron2. git python3 -m pip install -e detectron2 Dockerised Installation Open in Visual Studio Code: Open the cloned repository in VSCode. VSCode will prompt you to "Reopen in Container. " Alternatively, you can use the command palette (Ctrl+Shift+P) and search for the "reopen in container" command. Accessing the Desktop Interface: Open the user interface by navigating to the PORTS tab in VSCode, selecting port 6080 (or port 5801 for the CUDA-OpenGL version), and opening it in the browser. Getting Started: Build Packages Build package with cd ${your_ws} && colcon build source install/setup. bash Running Run following to publish/save annotations detected by aoc_fruit_detector package. ros2 launch aoc_fruit_detector fruit_detection. launch. py Parameters The config folder contains two parameter files for specifying system characteristics: ros_params. yaml allows tuning of ROS framework parameters. non_ros_params. yaml contains parameters for the fruit detection module. Key ROS2 parameters min_depth, max_depth: Define the minimum... --- Available AOC Stacks: RepositoriesAOC Fruit Detector --- Tomato Farm/Glasshouse Generator/Simulator This package is a tomato farm/glasshouse generator/simulator compatible with both Gazebo and Unity in ROS2 as part of the Agri-open Core (AOC) project. This repository consists of four key parts: Random tomato farm generator in Unity Random tomato glasshouse generator in Unity Random tomato glasshouse generator in Gazebo Simulator for the generated tomato farms in Unity and Gazebo Gazebo Glasshouse Generator Unity Glasshouse Generator Unity Tomato Farm Generator Video A group of tomato farm environments generated with these packages can be seen here. References If you use this project in your research or work, please cite the following paper: @inproceedings{flores2024, author = {Espejel Flores, Juan Pablo and Yilmaz, Abdurrahman and Soriano Avendaño, Luis Arturo and Cielniak, Grzegorz}, title = {Comparative Analysis of Unity and Gazebo Simulators for Digital Twins of Robotic Tomato Harvesting Scenarios}, booktitle = {Towards Autonomous Robotic Systems (TAROS 2024)}, year = {2024}, doi = {10. 1007/978-3-031-72059-8_2} } Installation Instructions Prepare a desktop PC with Ubuntu 22. 04 and install the ROS2 Humble release. Install Git: sudo apt install git Clone the repository and build it mkdir -p ${your_ws}/src cd ${your_ws}/src git clone --branch main git@github. com:LCAS/aoc_tomato_farm. git git checkout main cd ${your_ws} && colcon build source install/setup. bash Before generating Unity-ROS2 simulations, set the ROS 2 middleware and the localhost-only mode in the ~/. profile file (or in ~/. bash_profile or ~/. bash_login if either of those exists): export ROS_LOCALHOST_ONLY=1 export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp Tomato Glasshouse in Gazebo Generation of a New Tomato Glasshouse Run Jupyter... --- Agri-OpenCore Architecture Overview The Agri-OpenCore project follows a carefully designed architecture that promotes modularity, reusability, and clear separation of concerns. This architecture enables seamless deployment of robotic solutions across different agricultural scenarios, supporting both real-world and simulation environments. Core Components Scenarios At the top level, our architecture is organised around Scenarios. These represent complete agricultural use cases and orchestrate the interaction between platforms, environments, and algorithmic stacks. Importantly, scenarios are designed to be parameterised, allowing the same setup to be used in both real-world deployments and simulations. Platforms The Platform layer defines the robotic hardware configurations. This includes: Robot base configuration Sensor arrays (from individual sensors to complex sensing solutions) ROS control interfaces Platforms are primarily configuration-focused, ensuring clean separation between hardware specifications and algorithmic implementations. Environments The Environment component encapsulates all environment-specific configurations and parameters. Using standardised templates, we ensure consistent environment representation across different deployment scenarios. Stacks Stacks contain the core algorithmic implementations and form the backbone of our agricultural robotics solutions. Multiple stacks can be combined to create complex behaviours, with each stack focusing on specific functionalities such as navigation, perception, or manipulation. Deployment Our architecture features automated deployment processes, including: Docker image generation for each scenario Deployment via the LCAS Lincoln registry (lcas. lincoln. ac. uk) Continuous Integration/Continuous Deployment (CI/CD) pipelines Template System The architecture leverages a comprehensive template system: AOC Robot Templates: Standardised structure for robot configurations Environment Templates: Consistent environment setup and parameterisation Repository Templates: Common structure across all component repositories Design Principles Clear... --- Agri-OpenCore on YouTube Our video playlist, show-casing some of our results. https://youtube. com/playlist? list=PLnS6TQ_QsDUxGhP6Ov4KUG8d2FSH2dgRX Agri-OpenCore in the News University of Lincoln: Project Launched to Accelerate Crop Harvesting DEFRA, UK Government: £9. 13 million awarded to develop cutting-edge farming technology GreenTech NL: Agri-OpenCore robotics project to ease labour crisis --- Agri-OpenCore Project Project Summary Agri-OpenCore is an ambitious initiative aimed at revolutionizing the UK horticulture sector through advanced robotics and automation. This project addresses the critical labor shortage faced by the industry due to Brexit, COVID-19, and changing demographics, which has led to significant crop waste and economic challenges. Objectives: Develop the world's first open development platform for agri-robotic crop harvesting. Reduce the time and cost to develop robotic harvesting systems for various crops. Achieve human-picking-cost-parity for robotic harvesting within two years. Create an enduring legacy to accelerate the development of agri-robotic systems for all crops. Demonstrate trusted, assured, and safe operation of robotic systems in farm environments. Integrate robotics with existing farm infrastructure and human resources. Motivation: The UK horticulture sector, valued at £3 billion, is facing an "absolute crisis" due to labor shortages. With limited access to seasonal migrant workers, crops are going unharvested, leading to significant food waste. The industry urgently needs innovative technologies to transform labor productivity and ensure a resilient, economically efficient farming system. Approach: Open Ecosystem: Create an open development platform using ROS2. 0, allowing for private exploitation while providing core software and interfaces. Collaborative Development: Foster unprecedented cross-sector collaboration between farms, academia, and robotics companies. On-Farm Testing: Conduct extensive knowledge exchange and on-farm testing to remove barriers to robot adoption. Standardization: Work with BSI (UK's National Standards Body) to develop standards for trusted application of agri-robots. Proof of Concept: Demonstrate the effectiveness by developing robotic systems for strawberry and tomato harvesting that achieve... --- PartnerFocusAPSThe APS Group is the UK's leading supplier of British tomatoes to the high street, growers and suppliers of tomatoes and other produceUniversity of Lincoln - Lincoln Agri-RoboticsLincoln Agri-Robotics (LAR) is ‘the world’s first global centre of excellence in agricultural robotics’ (UK Innovation Strategy, July 2021), funded by UKRI’s Research England as part of their Expanding Excellence in England (E3) fund. This centre bridges and expands the strong collaborations that exist between two leading research groups at the University of Lincoln: the Lincoln Institute for Agri-Food Technology (LIAT) and the Lincoln Centre for Autonomous Systems (L-CAS). DOGTOOTH TECHNOLOGIESDogtooth have developed and manufactured fruit harvesting robots since 2015, and is leading the development in commercially viable crop harvesting. EXTEND ROBOTICSExtending human capability to amplify the productivity of labour worldwide. Discover how we are leveraging user-friendly immersive VR, robotics and artificial intelligence to reimagine the way we work. Founded in 2019, our aim at Extend Robotics is to build a future where humans and robots work together to enhance productivity, safety and efficiency across a range of industries. --- Welcome to Agri-Opencore AGRI-OPENCORE is a bold initiative to resolve major issues faced by the agricultural sector and businesses driving the automation and digitalisation of this crucial industry. It is focussed to cut the time and cost to develop a robotic harvesting system for any farm/crop with human-cost-picking-parity performance, and leave a legacy to accelerate development of any agri-robotic system for all crops. AGRI-OPENCORE delivers these objectives through unprecedented cross sector collaboration, the creation of an open ecosystem for innovation, and scaled cocreation and demonstration of technologies, trialled initially on English farms. To deliver these objectives, AGRI-OPENCORE will create the world's first open development platform (software and hardware) for agri-robotic crop harvesting. AGRI-OPENCORE provides open access to core software and interfaces that are hitherto unavailable to SME's, but when adopted can be privately exploited by robotics companies. AGRI-OPENCORE's open ecosystem creates an enduring legacy available for any robotic company and farming system. --- --- ## Posts ---