This Homemade AI Drone Software Finds People When Search and Rescue Teams Can’t
British Mountain Rescue workers have developed an automated drone system that can scour a landscape far quicker and more thoroughly than human eyes.
You are viewing a single comment's thread from:
I'd be happy to provide a more detailed account of the story of Charlie Kelly's tragic fate and the development of MR Maps.
Charlie Kelly's Tragic Fate
Charlie Kelly, a 56-year-old hillwalker and forensic psychologist, had been planning to tackle Creise, a 1,100-meter-high peak in the Scottish Highlands, on September 6, 2023. He had been an avid hillwalker for many years and had already completed several Munros, Scotland's highest peaks, which are defined as mountains over 3,000 feet in height.
Kelly's plan was to climb Creise and then attempt to bag a second Munro, which he had been wanting to do for some time. He had been preparing for the climb for weeks and had packed extra food, water, and warm clothing, as well as a first aid kit and a map.
However, Kelly never made it off the hill, and his body was not discovered until six weeks later. The investigation that followed was one of the most extensive in recent Scottish history, involving sniffer dogs, quad bikes, multiple helicopters, and drones equipped with infrared and conventional camera equipment.
Despite the efforts of the Glencoe Mountain Rescue team, which was comprised of professionals from the Coastguard, police Scotland, and the Royal Air Force, as well as dozens of highly trained volunteers from 10 different Mountain Rescue (MR) teams, Kelly's body was not found until October 24.
The search for Kelly was a major operation, with many resources being dedicated to finding him. The search area was vast, covering over 100 square miles of rugged terrain, and the weather conditions were often poor, with strong winds and heavy rain.
The Development of MR Maps
Meanwhile, two Mountain Rescue team members, Dan Roach and David Binks, from the Lake District, had been working on a new piece of piloting and image analysis software, designed to help drones find missing people more effectively.
Roach, a keen climber and enthusiast of drones, had been developing the idea of using drones for search and rescue purposes since 2012. However, he faced resistance from his peers, who were skeptical about the effectiveness of drones in such situations.
Undeterred, Roach continued to work on the project, eventually partnering with Binks, a retired software developer who had experience in writing software for MR teams.
Binks had a background in writing software for MR teams and had previously worked on a program called MR Maps, which allowed teams to track their members in near real-time, based on regular pings from their radios.
Roach and Binks realized that MR Maps could be used to analyze images taken by drones, rather than simply inputting specific colors to search for. By counting the number of pixels of a particular color, and then highlighting clusters of unusual colors, the software could effectively locate missing persons in dense or rugged terrain.
The MR Maps software was tested in the wild, behind the Glencoe Ski Center, where Kelly's body was eventually found. The area had been extensively searched by Glencoe Mountain Rescue, but Kelly's body had fallen into a gully, obscuring the line of sight of rescuers on the ground.
The muted colors of Kelly's clothing, meanwhile, had made him aLL but invisible from the air. However, the MR Maps software was able to pick up on the subtle differences in color and highlight Kelly's location.
The Breakthrough
The breakthrough in Kelly's case came when Binks and Roach realized that the software could be used to analyze images taken by drones, rather than simply inputting specific colors to search for. By counting the number of pixels of a particular color, and then highlighting clusters of unusual colors, the software could effectively locate missing persons in dense or rugged terrain.
The software was tested in the wild, behind the Glencoe Ski Center, where Kelly's body was eventually found. The area had been extensively searched by Glencoe Mountain Rescue, but Kelly's body had fallen into a gully, obscuring the line of sight of rescuers on the ground.
The muted colors of Kelly's clothing, meanwhile, had made him all but invisible from the air. However, the MR Maps software was able to pick up on the subtle differences in color and highlight Kelly's location.
The Impact
The successful deployment of MR Maps in Kelly's case marked a turning point in the search and rescue industry, highlighting the potential of drones to speed up the process and improve the chances of finding missing persons.
Since then, the software has been used in multiple other searches, mostly by helping eliminate areas of interest. The MR Maps software has also had a significant impact on the Mountain Rescue teams in the UK, which are facing mounting pressure due to an increasing number of callouts.
The number of callouts has been increasing for years, but the COVID-19 pandemic has led to a massive leap in the number of incidents. The rise of social media has also helped publicize particular spots, leading to a surge in visitor numbers to the UK's national parks.
The Future
The MR Maps software is free for anyone involved in search and rescue, and has already been adopted by several police forces in the UK. However, the creators of the software, Roach, Binks, and Parsons, have zero interest in commercializing it, and are content to continue working on the project as a hobby.
The development of MR Maps is a testament to the ingenuity and determination of the Mountain Rescue teams in the UK, who continue to work tirelessly to save lives and bring closure to families.
In addition, the MR Maps software has the potential to be used in a wide range of search and rescue applications, including missing person cases, natural disasters, and humanitarian crises. The software's ability to analyze images taken by drones and highlight areas of interest makes it an attractive tool for search and rescue teams around the world.
Article