Volcanoes (sometimes dormant) can erupt suddenly without warning. This is why there have been so many deaths in eruptions such as the Nevado del Ruiz stratovolcano in Tolima, Colombia, on November 13, 1985. There had been countless warning signs from as far as a year before in late 1984, and evacuation plans have been made. However, due to a storm on the fateful day itself, communications were down and residents had no means of receiving instructions to evacuate. The storm also forced residents to remain in their homes, not knowing if the eruption has began or ended. In total, the death toll was approximately 23,000.
Our Team realised that the lack of communication was they key cause of the massive death toll. Hence, we set out to design and build a system that is capable of processing data from sensors placed on volcanoes within the device enclosure itself, allowing it to determine if an eruption is imminent. Recent technology breakthroughs in Artificial Intelligence (AI) and Machine Learning (ML) has inspired this development pathway. This decision-making will be developed and refined by training machine learning algorithms in MATLAB.
We plan to use a Raspberry Pi, which provides us with ample processing power for the machine learning required. Sensors connected to the Raspberry Pi are able to collect data and feed them for processing. If the data is determined to be of an imminent eruption, the alarm system will be triggered and this should afford residents that few more minutes of time to evacuate or hide in a designated safe area before they get into grave danger.
It is our hope and vision that our product is widely used around the world, placed near volcanoes and potentially save even more lives than the current available systems.
To find out more about the dangers and currently deployed warning systems, visit Background.