Wildlife tracker challenge winners use Nordic-powered wireless connectivity
05
Jan
2021
2021
Oslo, Norway
Two of the successful projects in the Hackster.io and Smart Parks backed ‘ElephantEdge’ wildlife tracker challenge employed Nordic’s multiprotocol nRF52840 SoC
Nordic Semiconductor today announces that two of the winning projects from the ‘ElephantEdge’ wildlife tracker challenge are employing Nordic’s nRF52840 Bluetooth® 5.2/Bluetooth Low Energy (Bluetooth LE) advanced multiprotocol System-on-Chip (SoC) to provide the wireless connectivity for the world’s most advanced elephant tracking collar.
The ElephantEdge challenge—which aims to replace traditional and manual methods of elephant monitoring to support elephant conservation efforts in Africa—is a joint initiative of Avnet community, Hackster.io, and pro-conservation organization Smart Parks in combination with leading technology and conservation partners Nordic Semiconductor, Edge Impulse, Microsoft, Taoglas, u-blox, Vulcan EarthRanger, and Western Digital. The objective is to build a wildlife tracker that attaches to elephants via a collar, to better protect this endangered species. Elephants are under threat of extinction in as little as 10 years. The tracker should help park rangers reduce animal loss from illegal ivory poaching, trophy hunting, human conflict, and environmental degradation.
The objective is to build a wildlife tracker that attaches to elephants via a collar, to better protect this endangered species
Machine learning support
The challenge called on the world’s technology community to help build machine learning (ML) models using the Edge Impulse Studio and tracking dashboards using Avnet's IoTConnect (an advanced unified IoT platform featuring sensors and gateways), which will be deployed onto 10 production-grade collars manufactured by engineering partner Institute IRNAS and deployed by Smart Parks. Challenge participants did not need any hardware to build the ML models and could instead use datasets to sample, analyze, and build TinyML models. Participants could also use their smartphones to run simulated data collections and deployment. Final software and hardware is documented and shared freely under the open source license of opencollar.io.
The winning ElephantEdge challenge solutions are designed to help conservation efforts through the monitoring of poaching risk, human conflict, elephant musth (a periodic condition in male elephants characterized by highly aggressive behavior and accompanied by a large rise in reproductive hormones), elephant activity, and elephant communication. The top ten challenge projects were recently rewarded with $500 of prizes including an Apple Watch 3.
Winners from the ‘Top 5 Tracking Dashboards’ category were asked to build an Avnet IoTConnect dashboard to be used for elephant collar deployments, helping park rangers track, monitor, and receive on-demand alerts that are critical to conservation efforts. In this category, Chamal Ayesh received an award for an elephant movement tracking system using Avent IOTConnect and Nordic’s nRF52840 Development Kit (DK) which offers multiprotocol wireless support for Bluetooth LE, Bluetooth mesh, Thread, Zigbee, 802.15.4, and ANT. A simple accelerometer (MPU6050) connects to the nRF52840 DK and when movement occurs the DK reads values from the built-in accelerometer and sends this data to an IoT gateway (an Android smartphone) via Nordic-enabled Bluetooth LE connectivity. The IoT gateway then relays those readings to Avnet IOTConnect via nRF52840 SoC-supported Message Queue Telemetry Transport (MQTT), a lightweight connectivity protocol for machine-to-machine communication.
Tracking wildlife wirelessly
Winners from the ‘Top 5 Machine Learning Models’ category were asked to build ML models with Edge Impulse to be used for collar deployments. These models create a new human to elephant language, powered by TinyML. In this category, Dhruv Sheth received an award for ‘EleTect’, a TinyML and IoT smart wildlife tracker employing the nRF52840 SoC as well as an accelerometer, camera and microphone. Sheth’s different TinyML models included: Camera vision models to monitor the risk of poaching and predators or to monitor elephant movements; accelerometer data models to predict and classify common elephant behaviors; and audio data models to detect and classify elephant musth data and mood swings. These models are ready for deployment in three forms including a C++ library, Arduino library, and OpenMV library. The deployed models are available on GitHub.
The nRF52840 SoC combines a 64MHz, 32-bit Arm® Cortex® M4 processor with floating point unit (FPU) with a 2.4GHz multiprotocol radio (supporting Bluetooth 5.2, ANT™, Thread, Zigbee, IEEE 802.15.4, and proprietary 2.4GHz RF protocol software) with 1MB Flash memory and 256kB RAM. The chip supports all the features of Bluetooth 5 (including 4x the range or 2x the raw data bandwidth (2Mbps)) compared with Bluetooth 4.2.
The SoC is supplied with Nordic’s S140 SoftDevice, a Bluetooth 5-certified software protocol stack for building long range and high data Bluetooth LE applications. The S140 SoftDevice offers concurrent Central, Peripheral, Broadcaster, and Observer Bluetooth LE roles, and supports high throughput and long range modes as well as advertising extensions.