AI-powered robot teams pass firefighting trial, navigating obstacles and extinguishing fires with 99.67% success.
Firefighting robots powered by artificial intelligence have completed a successful early trial in Australia, showing they can navigate obstacles and work together to extinguish fires without placing human crews in danger.
The project was led by Cyborg Dynamics Engineering in collaboration with Griffith University and funded by the Queensland Defence Science Alliance. The team demonstrated the system in simulated and hybrid simulation-physical tests using an unmanned ground vehicle, or UGV, alongside up to four virtual robot teammates.In the trial, the physical UGV moved around obstacles and coordinated with its simulated counterparts to find and extinguish multiple simulated fires. The system achieved a 99.67 percent success rate in navigating and putting out two fires.Dr Zhe Hou, project Lead Chief Investigator from Griffith University’s School of Information and Communication Technology, said the results point to real-world potential.“We demonstrated that multiple real and simulated UGVs, trained through a structured three-stage AI learning curriculum, could learn to perform both low-level navigation and high-level collaborative tasks,” Dr Hou said.“This confirms the operational potential of our approach for practical case studies such as autonomous navigation and firefighting.”Robots learn to collaborateThe team built the system using a technique known as multi-agent reinforcement learning, or MARL. In simple terms, multiple AI-driven agents learn by trial and error while interacting with each other and their environment.The researchers designed a three-stage training curriculum. It began with a single robot learning basic navigation. The next stage involved multiple robots moving around obstacles. The final stage required several robots to coordinate and extinguish multiple fires in a complex setting with barriers.The robots were able to self-organize and divide tasks on their own, including splitting into teams to tackle separate fire outbreaks. This reduces the cognitive load on human operators and increases safety in high-risk situations.Cyborg Dynamics Engineering General Manager Ryan Marple said the company already deploys remotely controlled firefighting UGVs on mine sites across Australia.“We have developed the control systems for firefighting UGVs that are currently deployed on mine sites across Australia,” Marple said.“These units are remotely controlled by a human, a bit like an RC car.”“They have been an extremely effective measure in removing human firefighters from dangerous situations and enabling high-value assets to be saved from fires.”Autonomous swarms, faster responseMarple said the next step is automation of low-level control and coordinated swarming behavior.“The future of these kinds of vehicles – and the focus of this research – is the automation of low-level control and swarming behaviour across multiple agents. Such autonomous swarms can respond to complex situations in a way that just isn’t possible with direct manual control.”He added that sensor-driven decision-making allows faster responses than a human operator monitoring a screen.“By ingesting data from a wide variety of sensors, these systems can make decisions quickly, which just isn’t possible by the very limited situational awareness of a human looking at a screen.”The team plans to refine its neural networks and improve sim-to-real transfer methods so that skills learned in simulation can be reliably applied in the field.Researchers are also exploring the use of the same AI framework in other autonomous systems, including underwater and aerial vehicles, or mixed teams of different robotic platforms.The study was published in Engineering Applications for Artificial Intelligence.
Autonomous Swarms Emergency Response Firefighting Robots MARL Reinforcement Learning Robotics Unmanned Ground Vehicle
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Video: How the ‘world’s fastest developed’ humanoid robot was built in record timeUK robotics startup Humanoid has released a 19-minute documentary showcasing the backend scenario of building humanoids under time pressure.
Read more »
Denver breaks heat record Sunday; temperatures 22 degrees above normalTemperatures at DIA reached 68 degrees on Sunday, breaking the previous Feb. 15 heat record of 67 degrees set in 2017, according to the National Weather Service.
Read more »
China: World’s first full-stack humanoid robot open-sourced, draws global developersBeijing startup RoboParty open-sources Roboto Origin, a full-stack humanoid prototype, attracting 1,000+ GitHub Stars globally.
Read more »
Israel’s ‘mechanical octopus’ robot cuts tank cleaning time from 2 days to 2 hoursThe Israeli company Ruby AI has reportedly developed a specialized robotic system for the Israel Defence Forces (IDF).
Read more »
Kung fu ‘master’: China’s new humanoid robot nails tough aerial moves in demoAgibot’s Expedition A3 humanoid performs high-flying maneuvers, showcasing advanced balance, coordination, and real-world robotic agility.
Read more »
Our 67 Best Weeknight Pasta RecipesCheck out this roundup of weeknight pasta recipes, including carbonara recipes, pesto pasta recipes, macaroni and cheese, and more.
Read more »