ForestGuard

What it does

ForestGuard is an IoT-based sensor system that detects fires in real-time. The system notifies authorities when CO2 and temperature levels rise, allowing them to intervene before the fire spreads. The system can also measure air quality and toxic gas levels.

Your inspiration

As a team of nature lovers, we were on the site of the 2021 forest fires in Turkey. As a drone pilot, one of our members aided S&R efforts from the air, others provided help to the locals. After returning home, we brainstormed about the things that we can improve in the fight against the wildfire. Our team decided to focus on how to detect the fires in the early, more manageable stage. We said why not trees themselves warn us, in case of a fire, so we started to design the ForestGuard, a “wearable technology for trees”.

How it works

The ForestGuard works as a neural network in the forest by using cutting-edge LoRA mesh-network technology. The sensor modules have sensitive wide spectrum sensors which can measure gas levels, temperature, humidity, and air pressure sensors. These sensor modules can be installed on tree trunks via flexible straps to allow any movement and growth of trees. The sensor modules collect environmental data from the surroundings and process it to detect any abnormal situation like sudden changes in such variables. When any emergency situation occurs, sensor modules transmit the collected data to the Internet server through other modules. If an emergency condition is verified on the server, it will inform the authorities and firefighters immediately to take action

Design process

We started out with development boards in August 2021. In a short period of 2 months, we developed our first fully functional modules and deployed them in the forests of Bodrum/Turkey (a severe fire risk area). After a successful proof of concept period, we started developing our own mainboards, carefully selecting each and every component with power efficiency in mind. We successfully manufactured our own boards in February and started testing them non-stop. In May, we deployed our new modules in Istanbul/Turkey, in collaboration with local authorities. Currently, we are testing these modules, in regards to communication, sensor stability, and response time in the event of a fire. Our preliminary tests indicate that a camouflaged appearance will be more suitable for anti-theft and anti-tamper aspects. We are now working on a tree bark-shaped and colored skin for our modules.

How it is different

It is faster, cheaper, easy to install, does not require any electrical or data infrastructure, and most importantly, it detects fire in the ultra-early, smoke stage. Our tests show that ForestGuard modules can detect a controlled burn, from 70m, within the first 5-15 minutes. Sensors have a range of 100m radius and modules can communicate up to 15km. It is cost-effective: with ForestGuard, you can protect 40 hectares of forest, at a cost of forestation of one hectare of forest. Only one module covers a hectare of forested area. Our specially designed mainboard is modular, so the system can be easily modified to work with GSM/NB-iot/Wifi communication protocols. Additional sensors can be added in case of a different use case scenario.

Future plans

At the moment, we are developing our AI algorithm to predict fires more accurately. We are working on implementing solar panels and supercapacitors for a better alternative to Li-Ion batteries. This will help our modules operate for decades, without the usage of rare earth metals and other pollutants in batteries. For the remote areas that are without gsm coverage, we are planning on implementing satellite internet. We are planning on contacting Starlink and Project Kuiper in the following months.

Awards