Posted by Mumbai: A Lebanese software firm Eurisko Mobility has built a system using IoT and AI/ML to help protect forests against fires. Called as ‘Smart Forest’ – the system uses the combined power of IoT and AI/Machine Learning.
It can be very effective in monitoring forest ecosystem round the clock and taking proactive measures against forest fires. Governments, forest authorities and local municipalities can deploy this system in forest regions.
Power of IoT and AI/Machine Learning
Probably, it’s the first time a powerful mix of IoT and AI/ Machine Learning has been leveraged in a smart system that mainly is focused on environment and sustainability.
“Absolutely. Being weather-proof, cost and energy efficient, environment-friendly, IoT is the perfect technology for monitoring real-time metrics,” said Eurisko Mobility‘s Co-founder and CTO, Ziad Tawk told TechHerald.in
“Combined with the power of AI & Machine Learning, such systems can protect our different natural resources by providing prevention and real-time alerting,” added Tawk.
He, along with his brother Edgard Tawk founded the software development firm Eurisko Mobility in 2009.
Last month, the software firm announced to offer its Smart Forest system to seven forests across Lebanon.
The system’s overall development, deployment and maintenance will cost around $700,000. The nationwide implementation, covering forests areas is expected to be completed in the second quarter, 2020.
However, prior to the deployment, the company currently has undertaken a study to quantify the sensors, its testing under real-life fire scenarios and overall functioning.
“Eurisko Mobility is studying, how many sensors are needed, its gathering of historical data in order to train the Machine Learning models, deploying sensors & testing the system in real-life by emulating bush fires in each forest,” informed Tawk.
The Middle East nation has suffered extremely due to numerous wildfire incidents across its forests in recent years. These fires alarmingly reduced the overall forest cover in Lebanon to below 12% endangering the nation’s ecosystem with bigger risks.
Since Lebanon lacks a national level environment related policy, Tawk reckoned that Eurisko Mobility’s Smart Forest system can play a crucial role in protecting and sustaining remaining forests in his country.
Besides, the IoT and AI/ML combination, the Smart Forest system relies on powerful sensors that captures data across parameters, including air and soil temperature, relative humidity, air pressure, CO (carbon monoxide) and CO2 (carbon dioxide) levels, wind speed and precipitation level.
Deep Machine Learning/ Artificial Intelligence engine
At the heart of this smart system is a Deep Machine Learning/ Artificial Intelligence engine and its ability to predict Risk index. This engine is trained using artificial neural networks, which collect data sets from different forests in Lebanon.
With each forest or geographical location being different, the engine needs a separate training based on specific factors like soil type, climate and altitude, etc. The engine training is seamless and fully automated.
That’s how the engine is able to compute multi-forest fire risk indexes and able to send out real-time notifications over email, SMS and automated phone calls This notifications are based on the recipients’ preferences and sending rules.
The system comes with a real-time rich dashboard that shows Risk Index in different geographical locations, historical data (day, week, month) of different parameters in comparison with the previous year’s data.
It provides a secure access to all the data through a responsive web portal available anywhere. This gives a good understanding of Smart Forest’s technical and functioning aspects.
Unsupervised Machine Learning
What makes this system so unique in terms of dealing with nature related issues, according to Tawk, is the power of “Unsupervised Machine Learning.”
“Unsupervised Machine Learning can allow learning the specific natural characteristics of each forest even with the dynamicity of climate change,” claimed Tawk.
For large forests in the US, Australia and other parts of the world, Tawk stated that AI can help to compute the accurate Fire Risk Index, IoT can measure CO/CO2 emissions in real-time in order to send alerts as soon as fires start.
Though AI can promptly send alerts and notifications, Tawk cautioned about the physical challenge involved in dealing with fire.
“The big challenge remains dispatching units as fast as possible before fires become uncontrollable. Real-time integration with fire extinguishing drones can be a promising add-on for relatively large forests,” commented Tawk.
Although, Eurisko Mobility’s Smart Forest system is being deployed in Lebanon forests, Tawk concluded that it can definitely be rolled in any country in the world.
(Image source – Shutterstock)