Conservation Edge AI Lab: Design a Real-Time Wildlife & Threat Alert System

To develop an AI-driven, real-time wildlife and threat alert system using edge devices and sensor networks that can automatically detect, classify, and flag potential risks (poaching, habitat encroachment, animal distress) to enable faster, data-driven conservation decisions and on-ground intervention.

• Introduce participants to real-world challenges in wildlife conservation, including poaching, human–wildlife conflict, and bycatch detection.
• Demonstrate how camera traps, sensors, and basic AI/ML models can be combined into a practical, field-ready detection workflow.
• Equip participants to prototype simple, interpretable alert rules that can run on edge devices with limited connectivity and power.
• Build system-thinking skills by mapping actors, data flows, devices, and policies for a conservation area.
• Embed ethics, governance, and community considerations into the design of AI-enabled conservation systems.
• Foster collaboration between ecologists, conservation practitioners, and technologists working on data-driven conservation.

📅 Day 1 – Camera Traps, Sensor Data & Detection Basics

• Context: poaching, human–wildlife conflict, bycatch detection, and why early detection matters.
• Hands-on: load a small set of camera-trap images (classes: human, animal, empty).
• Apply a pre-trained object detection model or simple image classifier to tag images.
• Analyse outputs: count events such as “human at night”, “animal near boundary”, and “empty frame”.
• Discuss detection errors (missed detections, false alarms) and the limits of off-the-shelf models.

👉 Outcome: Basic wildlife detection pipeline + 01_camera_trap_detection_demo.ipynb.

📅 Day 2 – Simple Alert Rules & Edge Constraints

• Concept: edge vs cloud – latency, patchy connectivity, energy limits, and device constraints.
• Hands-on: build a simple rule-based alert layer over the detection outputs.
• Example rule: if person is detected in a sensitive zone within a risky time window → raise a “high-priority alert”.
• Log alerts into a small table with timestamp, location, alert_type, and confidence.
• Prioritisation: distinguish high, medium, and low-priority alerts for rangers and control rooms.

👉 Outcome: Minimal threat alert engine + 02_conservation_alert_logic.ipynb.

📅 Day 3 – System Design Canvas & Governance

• Ethics & governance: privacy of local communities, false positives, ranger fatigue, and safety risks.
• Hands-on: fill a system design canvas for one park/reserve (actors, data flows, devices, policies).
• Define alert routing: who receives which alerts, via which channel (SMS, app, radio), and in what format.
• Design escalation and response: expected response times, hand-offs, and log/feedback loops.
• Group sharing and refinement: compare designs and stress-test them with “what if” scenarios.

👉 Outcome: Deployable concept note for a conservation area + 03_conservation_edge_ai_system_canvas.docx.

• Undergraduate and postgraduate students in Wildlife Biology, Ecology, Environmental Science, Forestry, Geography, GIS/Remote Sensing, Computer Science, ECE, AI/ML, or related fields.
• PhD scholars, postdocs, and early-career researchers working on conservation, human–wildlife conflict, biodiversity monitoring, or environmental data analytics.
• Professionals from conservation NGOs, wildlife trusts, government departments, forest/wildlife agencies, protected area management, and impact-driven startups.
• Data scientists, engineers, and technologists interested in deploying AI/ML solutions in low-resource, edge-computing environments for conservation.
• Basic familiarity with programming and data handling (preferably in Python) is recommended, but no prior deep learning expertise is required.

• Understand how camera traps, sensors, and basic AI models can support wildlife monitoring and anti-poaching operations.
• Build a simple end-to-end wildlife detection pipeline using camera-trap images and a pre-trained model.
• Design and implement rule-based alert logic for prioritising threats (e.g., human intrusion, animals near boundaries).
• Gain practical insight into edge vs cloud deployment constraints: latency, connectivity, energy, and device limits.
• Develop a system design canvas for a real or hypothetical conservation area, including data flows, actors, and policies.
• Integrate ethics and governance into system design, addressing privacy, false positives, ranger workload, and safety.
• Take away reusable artefacts: Jupyter notebooks, alert logic templates, and a draft concept note for a deployable edge-AI conservation system.