Sky Guardians: Revolutionizing Wildlife Monitoring with Drones and IoT by Dennis Drost

Learn how a Dutch startup revolutionized wildlife monitoring by combining drones, IoT sensors & ML to solve a $100M agricultural problem: deterring geese from farmland.

Key takeaways
  • Company started by using drones to monitor and deter geese from farmlands, causing millions in annual crop damage

  • Traditional geese deterrent methods (hunting, scarecrows, gas cannons) become ineffective within weeks as geese adapt quickly

  • Developed IoT sensors using Raspberry Pis, microphones and machine learning to detect geese presence through audio recognition and automatically alert drone pilots

  • System evolved from basic prototypes to weatherproof units with solar power, capable of 24/7 monitoring across hundreds of hectares

  • Infrastructure runs on AWS using containers, Lambda functions, and automatic scaling to process audio data through ML models

  • Solution proved more effective over time, contrary to other methods - reduced required drone flights from 6-8 to 2-3 times daily

  • Key learnings: Start simple, test in real conditions early, iterate based on actual problems rather than theoretical ones

  • Technical approach: Converting audio to spectrograms for ML analysis, achieved >70% accuracy in geese detection

  • Focus shifted from drone automation to sensor networks as core business after discovering gap in real-time wildlife monitoring market

  • Project demonstrated viability of combining IoT/ML technology with traditional agriculture while maintaining regular employment