Publication Potential
The platform contains distinct technical contributions that could support multiple peer-reviewed publications. Below is an analysis of what's publishable, target venues, and possible paper structures.
Opportunity 1 — Systems (Flagship)
Farmerly AI: An Integrated Platform for Proactive AI-Driven Precision Agriculture with Autonomous IoT Sensor Networks
Computers & Electronics in Agriculture (IF: 8.3)
Smart Agricultural Technology (IF: 6.3)
IEEE IoT Journal (IF: 8.2)
Full system architecture — the integration of proactive AI, IoT sensor fusion, autonomous navigation, and RAG-based knowledge retrieval into a unified platform for NC sweet potato production.
- System architecture (sensor layer → edge AI → actuation)
- Proactive rule engine with condition-based anticipatory alerts
- RAG knowledge retrieval for agricultural decision support
- Real-time WebSocket telemetry and visualization
- Field evaluation with simulated NC Coastal Plain conditions
12–15 pages
Implementation complete
Opportunity 2 — Proactive AI
From Reactive to Proactive: Anticipatory AI Agents for Agricultural Sensor Networks
Frontiers in Plant Science (IF: 5.6)
AI in Agriculture (IF: 8.2)
AAAI Workshop on AI for Agriculture
The shift from reactive query-response AI to proactive, condition-triggered intelligence. Formalizes the proactive rule engine, priority-based insight delivery, and floating tooltip UX pattern.
- Taxonomy of reactive vs. proactive AI in agriculture
- Condition-based rule engine design (threshold + temporal + growth-stage)
- Priority classification and alert fatigue mitigation
- User study: tooltip-based proactive insights vs. traditional alerts
- Comparison with existing agricultural DSS
8–10 pages
Implementation complete
Opportunity 3 — IoT Sensor Network
Design and Simulation of a Multi-Modal IoT Sensor Network for Sweet Potato Production in North Carolina
Sensors MDPI (IF: 3.9)
IEEE Sensors Journal (IF: 4.3)
Precision Agriculture (IF: 6.2)
The 15-sensor-type network architecture, LoRaWAN topology, zone-based deployment strategy, and simulation framework for NC sweet potato fields.
- Sensor selection rationale (moisture, temp, pH, EC, NPK, NDVI, etc.)
- Deployment topology: zone-based placement for sandy loam variability
- Simulation engine: realistic noise models, diurnal cycles, event injection
- Data fusion: multi-depth moisture aggregation, cross-sensor validation
- Cost-benefit analysis: sensor density vs. yield improvement
10–12 pages
Implementation complete
Opportunity 4 — Autonomous Navigation
Pure Pursuit Path Tracking with Boustrophedon Coverage Planning for Autonomous Agricultural Tractors
Journal of Field Robotics (IF: 8.3)
Biosystems Engineering (IF: 5.1)
IEEE ICRA (~40% acceptance)
Ackermann bicycle-model kinematics, pure pursuit controller, AB-line guidance, headland turns, and coverage planning for row-crop agriculture.
- Bicycle-model Ackermann kinematics with realistic constraints
- Pure pursuit controller tuning for agricultural speeds (0–8 km/h)
- Boustrophedon decomposition for irregular field boundaries
- GPS noise models (RTK/DGPS/standalone) and cross-track error
- Geofence enforcement and safety boundary monitoring
- Coverage efficiency, path smoothness, fuel consumption results
8–10 pages
Implementation complete
Opportunity 5 — RAG for Agriculture
Retrieval-Augmented Generation for Context-Aware Agricultural Decision Support: A Sweet Potato Case Study
Expert Systems with Applications (IF: 8.5)
Knowledge-Based Systems (IF: 8.8)
ACL Workshop on NLP for Agriculture
Lightweight RAG architecture using sentence-transformers for agricultural knowledge retrieval, grounded in NC Cooperative Extension publications and peer-reviewed soil science.
- Knowledge base construction from extension publications (20+ documents)
- Embedding strategy: sentence-transformers for agricultural text
- Context injection: real-time sensor data as RAG context
- Evaluation: answer relevance, source attribution, hallucination rate
- Comparison: RAG vs. fine-tuned LLM vs. rule-based expert system
- Farmer usability study: trust, comprehension, actionability
10–12 pages
Deployed & testable
Opportunity 6 — HCI / Visualization
Floating Tooltip Interfaces for Proactive AI Insights in Agricultural Dashboards: A Design Study
Int'l J. Human-Computer Studies (IF: 5.4)
ACM CHI (~25% acceptance)
Computers in Human Behavior (IF: 9.0)
The UX design of proactive AI insight delivery — floating tooltips anchored to sensor markers, priority color-coding, dismiss patterns, and alert fatigue reduction.
- Design space: notifications vs. tooltips vs. ambient displays
- Implementation: map-anchored floating tooltips with priority hierarchy
- User study: farmer comprehension and response time
- Alert fatigue analysis: dismiss patterns, insight relevance over time
- Accessibility: color-blind safe palettes, screen reader compatibility
- Design guidelines for proactive AI interfaces in agriculture
8–10 pages
Implementation complete
Opportunity 7 — Digital Twin
A Digital Twin Framework for Agricultural IoT Testbed Design: Virtual Sensor Placement and Irrigation Strategy Optimization
Digital Twin (Springer) (IF: 4.2)
ASABE Annual Meeting
Simulation Modelling Practice & Theory (IF: 4.2)
The simulator functions as a digital twin — virtual sensor placement, irrigation strategy testing, failure mode analysis before hardware deployment. Validate twin accuracy against real field data.
- Digital twin architecture: physics-based soil model + sensor simulation
- Virtual sensor placement optimization (coverage vs. cost)
- What-if scenarios: drought, equipment failure, sensor drift
- Irrigation strategy comparison: fixed schedule vs. AI-driven
- Twin calibration with real-world sensor data from pilot deployment
10–12 pages
Simulation framework complete