Accelerating AI in Agriculture Using farm-ng Amiga Robots and NVIDIA Robotics


The Case for Automating Specialty Crop Farming

Farm-ng is a California-based robotics company founded to address the unique challenges of specialty crop farming. Unlike large-scale monocrops that have become highly mechanized, specialty crops like fruits, nuts, and vegetables still rely on manual labor with significantly higher operational costs due to the complexity of different tasks and terrains. To mitigate these challenges, farm-ng has integrated NVIDIA accelerated computing into the Amiga, our modular wheeled farming robot.

NVIDIA technologies such as the NVIDIA Isaac and Jetson platforms, which facilitate the development and deployment of AI-powered robots, are already relied on by more than 1.2 million developers and 10,000 customers and partners. The goal of this integration is to improve operational efficiency for farms while reinventing food production systems using the capabilities of AI and robotics.

Amiga Development Kit (ADK): The Easiest Way to Develop and Deploy Robotics Applications

The ADK is a complete toolkit designed to make real-world robotics applications easy to program and deploy. It combines a range of hardware components and a versatile software suite, making it accessible to both farmers and developers. This simplifies the programming process, enabling the efficient deployment of custom solutions in complex agricultural settings.

Integrating NVIDIA Jetson Platform for Edge AI and Robotics

The integration of NVIDIA Jetson into each farm-ng robot enhances the ADK in several key areas:

  • Enhanced Data Processing: NVIDIA Jetson enables the ADK to process large volumes of data efficiently. This is crucial for real-time analytics and decision-making in complex agricultural environments.

  • Advanced Sensor Integration: Improved sensor performance, facilitated by NVIDIA Jetson, allows for more accurate and detailed data collection. This includes soil analysis, crop health monitoring, and environmental sensing.

  • AI-Driven Functionalities: With NVIDIA Jetson, the ADK supports more sophisticated AI operations. This includes machine learning models for predictive analysis, autonomous navigation, and pattern recognition or anomaly detection in agricultural settings.

Evolution of AmigaOS: Amplifying the Capabilities of ADK

Each AmigaOS release is purpose-built to expand the limits of what the ADK can do:

  • AmigaOS Artichoke: This initial version introduced basic autonomous features and AI functionalities. It laid the foundation for robotic movement and simple decision-making tasks in agricultural settings.

  • AmigaOS Barley: Provides advanced navigation tools, such as RTK-GPS integration for precise geolocation with up to 2 cm accuracy. It also improved data management capabilities with a robust track-and-repeat workflow for automating navigation tasks and remote teleoperation from any browser.

  • AmigaOS Clementine: Introduces an advanced Camera app for high-resolution imaging and live streaming, a revamped File Manager for efficient data handling, and enhanced AI capabilities for complex farming tasks.

Expansive API Suite

Customizable solutions for a variety of farming tasks in a fraction of the time.

Custom Web Apps

No-code farming applications on the Amiga Brain.


Clementine offers a suite of features that build on the advances from AmigaOS Barley.

Enhanced Camera App

Offers sophisticated imaging capabilities, essential for in-depth agricultural analysis, including crop monitoring and pest detection.

AI & Machine Learning

Supports a range of algorithms and machine learning models for predictive crop analytics, autonomous navigation, and real time decision-making.

Detailed Features of Most Recent Releases


AmigaOS Barley

AutoPlot App

Self-guided navigation and mapping available standard on each unit.

W-A-S-D Teleoperation

Keyboard-based controls that make farming feel like gaming.

AmigaOS Clementine

Barley was a milestone release for the robotics community that includes built-in autonomy and cloud-based support for rapid deployment of custom apps directly onto any Amiga robot.

Secure Remote Access

Encrypted connection to any robot from any location.

Data Management Tools

These tools allow for efficient organization, storage, and processing of vast amounts of agricultural data, facilitating easier access and analysis.

Real-World Applications and Impact


Autonomy Plant Line Following and Weeding

UC SANTA CRUZ CENTER FOR AGROECOLOGY

The ADK platform is used for a variety of purposes, from the detection of invasive pests, mapping plant species, autonomous spraying, targeted weed eradication, yield estimation, compost spreading, and other cultivation tasks.

Image Data Collection for Deep Learned Farming

WESTERN GROWERS ASSOCIATION AND UNIVERSITY OF ARIZONA

Autonomous Seeding

EARTHBOUND FARMS

Hyper-Precise Herbicide Application

NORTH CAROLINA PLANT SCIENCES INSTITUTE

Laser Weeding

LAUDANDO & ASSOCIATES


Shaping the Future of Robotics and Sustainable Food Production

The integration of NVIDIA Jetson within the ADK is a significant advancement in embodied AI and in-field robotics, enhancing the capabilities of specialty crop automation and laying the foundation for a sustainable food supply chain resilient to climatic and economic shocks.

Visit the ADK repository for more information on AmigaOS, the impact of NVIDIA Jetson in agriculture, or to start building your own robotics application today. https://amiga.farm-ng.com/docs/getting-started/