loader image

The Best Practice Portfolio

The Blueprint for Sustainable, AI-Driven Retrofitting
in Apple Orchards and Wheat Fields.

From Prototype to Standard

Welcome to the Smart Droplets Best Practice Portfolio. Over the past four years, our consortium has moved beyond theoretical research to validate a fully autonomous retrofitting solution in real-world environments.

This portfolio represents our legacy to the agricultural community. It collects our open-source code, field-validated operational protocols, safety recommendations, and scientific findings. Whether you are a developer looking to integrate our AI, a farmer considering retrofitting, or a policymaker shaping the future of AgTech, these resources provide the blueprint for the next generation of sustainable agriculture.

Pillar 1

The Developer Toolkit
(Technical Practices)

For System Integrators, AI Engineers, and OEMs.

The Smart Droplets ecosystem is built on a foundation of open-access, stable software. We have released 9 public repositories (Version 1.0) that serve as the technical blueprint for the next generation of autonomous agricultural systems. These resources are maintained as part of our five-year commitment to project legacy.

The Interoperability Layer
(Connecting Tractors to Cloud)

Reference Implementations for NGSI-LD & Cloud Integration.

  • The Tools: The smartDropletsDataAdapters, retrofitted_tractor_data_adapters and contextBrokerExamples.
  • What it does: This suite of tools provides Python-based adapters to bridge the gap between physical robotic hardware (tractors), recommendation engines, and cloud-based data management platforms. It includes ready-to-use templates for the Orion Context Broker.
  • Why it matters: It provides a field-validated reference for using NGSI-LD Smart Data Models in robotics. This ensures that any ROS2-based machine can achieve full interoperability within the FIWARE ecosystem, preventing vendor lock-in and streamlining data exchange.

The Digital Twin Framework

Physics-Based Virtual Validation
  • The Tool: The digitalTwinFramework repository.
  • What it does: A comprehensive simulation environment and documentation for modeling complex orchard geometries and testing autonomous behavior in a virtual space.
  • Why it matters: It allows developers to stress-test navigation and spraying algorithms in a high-fidelity physics environment before moving to expensive and sensitive field hardware. This reduces R&D costs and significantly improves deployment safety.

Cloud & AI Services

Scalable Infrastructure & Edge Intelligence
  • The Tools: sd-cloud-k8s, edgeAI, and weatherForecastDataService.
  • What it does: These repositories provide the underlying Kubernetes (k8s) infrastructure for cloud deployment, alongside specialized services for edge-based AI processing and localized weather forecasting.

Pillar 2​

Field Validation
(Operational Excellence and Proven Outcomes)

For Agronomists, Farmers, and Cooperatives.

The Smart Droplets technology has been rigorously validated in the apple orchards of Spain and the wheat fields of Lithuania. These results prove that our autonomous retrofit solution provides immediate, measurable benefits across different cropping systems.

The Impact: Field-Proven Results

Our dual-pilot campaign delivered significant reductions in input use while maintaining high-fidelity crop protection:

  • Pesticide Reduction: Achieved 15–21% savings through canopy-optimized spraying.
  • Nutrient Efficiency: Validated 19–26% reduction in Nitrogen application via high-speed Direct Injection.

Water Conservation: Recorded up to 24% water savings in irrigation-intensive zones.

The Retrofit Performance Report (D5.2)

Evidence-Based Validation This report summarizes the data collected during our final pilot phases. It provides the comparative analysis of traditional vs. autonomous spraying, proving the system’s reliability and efficiency in diverse European agricultural contexts.

Tech Briefs for Integrators & Farmers

Practical Guides for the Modern Farm A series of concise one-pagers designed to help farmers and system integrators manage the hardware effectively. These guides strip away the technical jargon to focus on field readiness.

  • Topics:  Installing the Retrofit Kit, Calibrating LiDAR Sensors, and Daily Maintenance Checklists.

Direct Injection & Spraying Protocols

Best practices for managing the Direct Injection System. Learn how to configure tank pressures and manage real-time mixing to ensure precise “spot spraying” without clogging or contamination.

Pillar 3

Policy & Safety
(Regulatory Frameworks and Industry Standards)

For Regulators, Safety Inspectors, and Standardization Bodies.

Retrofitting legacy machinery presents unique safety and sustainability challenges. Through our work with Working Groups such as the ROBAGRI Safety and Technical Committee, Smart Droplets is actively shaping the evolution of European standards to ensure a safe, green transition for autonomous farming.

Strategic Industry Recommendations

To support the adoption of autonomous technologies, the consortium advocates for the following pillars within international standardization frameworks:

  • Data Interoperability: Adopt NGSI-LD Smart Data Models to prevent vendor lock-in and enable seamless communication across multi-vendor robotic fleets.
  • Autonomous Safety: Revise ISO 18497 to include a certification pathway for non-invasive retrofits on legacy tractors, supported by software-independent failsafe mechanisms.
  • AI Policy Impact: Implement Semantic Obstacle Classification under the EU AI Act to replace binary detection with risk-based perception, reducing false emergency stops.

Access to Technical Documentation

Detailed safety analyses and technical proposals related to these recommendations are currently being channeled through formal industry committees.
If you represent a certified public body or a standardization working group and wish to discuss these technical findings, please Contact Us.

Pillar 4

The Knowledge Hub
(Science & Education)

For Researchers, Students, and the General Public.

Ensuring the longevity of our results through education and scientific rigor.

The Smart Droplets Podcast

Go behind the scenes with the engineers, agronomists, and data scientists who built the system. Listen to candid discussions about the real-world challenges of GPS-denied navigation and the future of AI in farming.

Scientific Publications

The peer-reviewed science that validates our approach. Access our library of Open Access papers published by leading institutions, including Wageningen University (WU) and the Agricultural University of Athens (AUA).

  • Key Topics: Deep Learning for Canopy Characterization, SLAM Algorithms in Orchards, and Digital Twin Architectures

The Smart Droplets Academy

The technology is only as good as the operator. Access our permanent suite of open training modules designed to upskill the agricultural workforce in handling and maintaining autonomous systems.