/MouseTrap

By instrumenting trap’s with IoT sensors, cataloging the specific positions, and capturing trap & clear events in the cloud, we can apply machine learning algorithms to understand patterns and provide predictive analytics to optimally schedule trap visits and ideal location placement for trap & field technician efficiency.

Primary LanguageHTMLMIT LicenseMIT

Purpose:

To digitally transform pest control through operational efficiency driven by intelligent information.

Hypothesis:

By instrumenting trap’s with IoT sensors, cataloging the specific positions, and capturing trap & clear events in the cloud, we can apply machine learning algorithms to understand patterns and provide predictive analytics to optimally schedule trap visits and ideal location placement for trap & field technician efficiency.

Technology Overview:

PaaS Services used for easy Scale-out , Reliable “Always On” SLA, and low cost of maintenance. Services loosely coupled together so the architecture is flexible to future enhancements & maintenance.

Trap Sensor: Windows IoT Core

  • A closed circuit on GPIO is applied to a traditional trap, which is connected to an IP cloud-connected gateway device to collect and transmit data.
  • Raspberry Pi 2 running Windows IOT Core
  • UWP App monitors circuit and sends events to Azure IoT Hub

In production scenarios, it would be more cost efficient to use multiple sensors connected to a single gateway device, ideally through a wireless low power device (i.e. zigbee, zwave or Bluetooth)