Problem

Wildfires cause billions of dollars of damage every year and countless lives lost due to slow response times. The U.S experiences on average 8000 wildfires annually. These wildfires destroy hundreds of thousands of acres of land and put human lives at risk. Monitoring and fighting these fires is a dangerous task for those in the front lines and people in its path. In 2020 alone up until September 6,447 fires have been ravaging over 2,246,483 acres and counting resulting in a tremendous death toll directly from these fires as well as an indirect death toll of aproximatly 1,200 from the smoke produced. Those on the front lines fighting these fires do not currently have a way to safely monitor and share information in real time resulting in delayed action, planning and evacuations.


Solution

Through the use of drones and processing aerial images FireScout is committed to understanding these problems giving first responders the tools to make decisions quickly and accurately to save lives. Together with Dr. Fatemeh Afghah and Ph.D candidate Alireza Shamsoshoara, we have proposed a solution to help them achieve this goal. User 1 will pilot the drone towards the fires location and User 2 will start the AI algorithms and other software through the GUI that is sent throught the SDR.With the Jetson Nano attached to the drone, AI algorithims can be ran to detect fires in greater detail and in real time. Once a fire is detected with an image classification model, data and the image containing the fire will be sent to the ground station through the use of an SDR designed by the Elecetrical Engineering Team. Once images and data requested by the user have been received, the GUI will display the results





Poster