The Māui dolphin is the world’s rarest dolphin, with an estimated 63 (or less) in existence. Preventing the extinction of these precious taonga first requires knowing where they are and where they travel - the problem is that there is no robust data on these mammals, nor has there been an efficient way to collect it, until now.
MĀUI63 is a non-profit dedicated to counting, tracking and hopefully saving the Māui dolphins from extinction by improving on the existing methods which are incredibly expensive and inaccurate. The teams combined skills of marine biology, drone enthusiasm and artificial intelligence brought an opportunity to RUSH - how can we display the AI collected data from the surveying process so that regular location data of critically endangered marine animals can finally be utilised?
With the integration of the AI technology with the UAV, there is now an AI-powered tracking drone which can autonomously find, follow, and identify Māui dolphins from other dolphins. It can also switch to ‘follow’ mode when it finds one, telling the drone to circle around the area to keep the dolphin in view.
As data builds up over time MAUI63 hopes to be able to identify individual animals from unique identifying markings, and with continuous monitoring starting in June 2021 there will be new data every 2-4 weeks as opposed to annually. .
The R/VISION platform will also be able to show data produced by AI Driven predictive models, showing areas where Māui dolphins are likely to be.
The automated compilation of collected location data will be presented in geo spatial models and made freely available for scientists, government agencies, fishing companies, conservationists and the general public. This means that up-to-date data can actively inform stakeholders of where these animals are, and by knowing their location we can keep them from harm or look for areas of water that require testing for toxoplasmosis threats.
As the continuous monitoring plan unfolds and data builds up, predictive models start working, the required flight time will eventually reduce, allowing cost reduction and scale to technical development of future features including unique identification, thermal identification, and UAV enhancements such as 4k.