Can On-device Machine Learning Prevent Forest Fires?

Author: Luka Banović


We are proud to present a short video that our partners at EdgeImpulse have put together. In the video, you can get a glimpse of RAM-1, an electrical grid monitoring system with applied machine learning which we have been working on together with Izoelektro. What excites us about RAM-1 is that it very clearly puts on display the technical finesse of everyone that has been involved with it.

We love working on technological solutions that have a high social or environmental impact. RAM-1 is a very visible demonstration of added value we can create with adding on-device machine learning to an advanced technological system. With RAM-1, we can directly improve the environmental and public security of regions where it is deployed.

What have we learned from RAM-1

The challenge we were confronted with taking on development of RAM-1 aligns with practically every aspect of product development that we are uniquely specialized in. As you may have read in our previous posts, one of our expertise is design of low-power solutions, reaching battery lives of 20+ years in most extreme environmental conditions. The components that we chose to include in the design, are perfectly capable to withstand the requirements. Having said that, it is important to apply them correctly. The only way we could do this is through partnerships with technology providers EdgeImpulse and Arm. Having engineers from all partners readily available to support us in the process and share their expertise is absolutely crucial in the process of bringing a breakthrough product like RAM-1 to the market, which integrates all latest available technology in the field of industrial IoT.

Exponential rise of technological advancements on component-level always develops in parallel with the application-level. For that to be efficient, we have to rely on the massive amount of knowledge that every player in the field has acquired through their years of existence; be it semiconductor manufacturers, connectivity infrastructure experts, power consumption optimization tools, device management capabilities, or anything else.

How we want to continue our mission

Blending all this knowledge together into advanced integrated systems has become somewhat of a mission for us. It is a crucial channel for us that enables us to perform well in the work we do. We make sure to be early adopters of every novelty, often even before it is made available to the general public, so we can start accumulating that knowledge as early as possible.

We have been purposefully designing our team and workflow to be able to follow this path. As our CEO, Luka Mustafa, has recently explained, one of our key priorities is to maintain partnerships with leading partners in the field. We want to grow a network of people and companies we believe in. As a company, we wish to mature together with the technology, which will allow us to deploy more of advanced systems similar to RAM-1 that directly improve people’s well-being.

About the author

Luka Banović

Luka Banović

Is fascinated by the comprehensive experience that product development offers. He makes sure to maximize the value of our work every step of the way. He ensures complete functionality and usability of our solutions and always works hard to find a quick and efficient way of delivering our solutions to the customers.

WEBINAR: Hands-on experience with LR1110 LoRaWAN transceiver with GNSS and WiFi functionality

Semtech LR1110 LoRaWAN transceiver with GNSS and WiFi scanning evaluation has shown promising results. To best share our hands-on experience at the first instance we are organizing this session to explore the capabilities of the solution and share our initial experience. 

Share this post

Share on facebook
Share on linkedin
Share on twitter
Share on print
Share on email

Stay Updated!

Subscribe to receive News on interesting projects we’re involved in and open hardware projects we create.