In the realm of autonomous systems, drones have emerged as the perfect tools to liberate humanity from monotonous, perilous, and grim tasks, known as the "3 Ds" in industry parlance. Among these airborne marvels, low-cost drones, also known as flying robots, have taken the lead, undertaking a myriad of missions ranging from remote sensing to material delivery. The exponential growth in their numbers has led to the deployment of drone swarms, with teams of animators and extensive computer simulations choreographing their synchronized movements.However, a new frontier beckons—an era where drones can dynamically respond to obstacles, predators, vehicles, and even insect swarms. The inspiration for this advancement lies in the remarkable coordination exhibited by birds in flight and fish swimming in schools. Aaron Becker, an associate professor of electrical and computer engineering, is spearheading the development of algorithms that apply these principles to drone swarms. With the support of a generous $1.7 million grant from the Kostas Research Institute at Northeastern University, Becker and his team are poised to revolutionize the fleet-like delivery of drone services.Becker's team, which includes esteemed researchers such as David Jackson, Julien Leclerc, and Daniel Onofrei, seeks to depart from the prevailing swarm research patterns. Instead of relying solely on offline computation or simple rule-based logic, their approach combines the strengths of both computers and human operators. By leveraging fast computation for tactical maneuvers and strategic decision-making skills, they aim to create swarms that behave optimally and adapt fluidly to changing environments.The initial focus will be on two application scenarios. Firstly, in the aerial sensing of forest fires, the drone swarm must track the fire and relay critical information to firefighters. Secondly, for aerial security coverage of commercial facilities and campuses, drones must escort entering and exiting vehicles while managing limited battery life and recharging requirements. Becker's extensive experience in robotics, including controlling massive swarms of robots, equips him with the expertise needed to enhance drone swarms and streamline complex tasks with minimal instructions.The vision is clear, to unleash the potential of drone swarms through coordinated control. By combining computation on drones, clear visualizations on operator computers, and high-level decision-making by human operators, these swarms can navigate and respond optimally to diverse environments. The implications extend far beyond the current applications, paving the way for advancements in fields such as disaster response, environmental monitoring, and infrastructure management.
In the realm of autonomous systems, drones have emerged as the perfect tools to liberate humanity from monotonous, perilous, and grim tasks, known as the "3 Ds" in industry parlance. Among these airborne marvels, low-cost drones, also known as flying robots, have taken the lead, undertaking a myriad of missions ranging from remote sensing to material delivery. The exponential growth in their numbers has led to the deployment of drone swarms, with teams of animators and extensive computer simulations choreographing their synchronized movements.However, a new frontier beckons—an era where drones can dynamically respond to obstacles, predators, vehicles, and even insect swarms. The inspiration for this advancement lies in the remarkable coordination exhibited by birds in flight and fish swimming in schools. Aaron Becker, an associate professor of electrical and computer engineering, is spearheading the development of algorithms that apply these principles to drone swarms. With the support of a generous $1.7 million grant from the Kostas Research Institute at Northeastern University, Becker and his team are poised to revolutionize the fleet-like delivery of drone services.Becker's team, which includes esteemed researchers such as David Jackson, Julien Leclerc, and Daniel Onofrei, seeks to depart from the prevailing swarm research patterns. Instead of relying solely on offline computation or simple rule-based logic, their approach combines the strengths of both computers and human operators. By leveraging fast computation for tactical maneuvers and strategic decision-making skills, they aim to create swarms that behave optimally and adapt fluidly to changing environments.The initial focus will be on two application scenarios. Firstly, in the aerial sensing of forest fires, the drone swarm must track the fire and relay critical information to firefighters. Secondly, for aerial security coverage of commercial facilities and campuses, drones must escort entering and exiting vehicles while managing limited battery life and recharging requirements. Becker's extensive experience in robotics, including controlling massive swarms of robots, equips him with the expertise needed to enhance drone swarms and streamline complex tasks with minimal instructions.The vision is clear, to unleash the potential of drone swarms through coordinated control. By combining computation on drones, clear visualizations on operator computers, and high-level decision-making by human operators, these swarms can navigate and respond optimally to diverse environments. The implications extend far beyond the current applications, paving the way for advancements in fields such as disaster response, environmental monitoring, and infrastructure management.
