Engineering

Prize-winning drone system can assist first responders in indoor searches

Penn State’s robotics competition club develops a two-drone system with 3D mapping technology

Rachel Axten and Venkatarishnan Iyer are two members of the Penn State robotics team that recently developed a low-cost autonomous drone system. Credit: Provided by Rachel Axten. All Rights Reserved.

UNIVERSITY PARK, Pa. — Indoor search and rescue operations are some of the most dangerous tasks that law enforcement and first responders must face, but drone technology has revolutionized how they approach these intense situations, according to graduate students in Penn State’s Autonomous Robotics Competition Club (ARCC). Drones can be used to locate people in claustrophobic, dark and GPS-limited environments like collapsed buildings, but developing these drones is difficult and expensive.  

To answer this challenge, the students recently developed a low-cost, unmanned aircraft system (UAS). The two-drone system includes 2D and 3D mapping technology that can assist law enforcement and first responders in indoor searches. They published a paper detailing their system in Aerospace Research Central, a partner site to the American Institute of Aeronautics and Astronautics.  

The paper follows the robotics club’s participation in the 2023 First Responder UAS 3D Mapping Challenge, where they earned second place and $20,000 in prize money. Teams were tasked with creating a cost-effective UAS solution with one or more autonomous or semi-autonomous vehicles to provide enhanced situational awareness for first responders entering a building.  

“We developed a two-drone system: Our first small and easy-to-fly drone had essentially no onboard computers other than what was necessary for flight control,” said Rachel Axten, a doctoral candidate in aerospace engineering, president of the ARCC and co-author of the paper. “The other vehicle had a sensor to help us build the 3D map of the indoor environment, and a relatively powerful computer that allowed for the map to be generated entirely onboard.” 

The two-drone fleet can operate in both a single or multi-vehicle configuration, which allows the team to distribute tasks between two vehicles: a smaller, nimbler vehicle equipped with thermal and night vision capabilities alongside a live camera feed, and a bulkier vehicle that can generate a 3D map of its environment, providing first responders with valuable information. This solution provides greater flexibility depending on the needs of first responders, allowing teams to customize and reconfigure the drones based on the situation. 

By using off-the-shelf pieces in addition to customized software, algorithms and hardware components, ARCC developed their drone system to cost just a few thousand dollars — much less than other U.S. developed drone systems, which can cost tens of thousands of dollars to manufacture. The team applied lessons they learned from participating in past competitions and connecting with first responders, helping them simplify their design and adapt to what search and rescue teams need out of a reconnaissance drone. 

“We strived to achieve a balance between functionality, weight, size and price,” said Venkatakrishnan Iyer, a doctoral candidate in aerospace engineering, treasurer of the ARCC and co-author of the paper. “For example, the thermal camera used allows only direct line of sight detection, meaning that heat signatures behind a wall may not be detected. Such features require expensive sensors leading to an overall increase in cost, so we had to determine what feature trade-offs we had to make in the interest of keeping the system affordable.” 

The team developed the drone over just a few weeks, but spent much of that time testing it at the State College Airport, according to Vítor Valente, assistant research professor of aerospace engineering and co-author of the paper. During the system’s development, Valente was the vice president of ARCC and obtaining his doctorate in aerospace engineering. 

There is always something that needs to be adjusted or improved after you run tests,Valente said. “Our tests showed us that the camera system responsible for mapping needed to be adjusted. Finding the right solution was time consuming, but the process not only fixed the camera, it made the whole system better.” 

ARCC has no intention of slowing down, Axten said. The team recently won stage one of the GoAERO Challenge sponsored by Boeing, where teams are tasked with developing an autonomous aerial vehicle to help first responders with medical evacuation efforts. This public safety project not only aims to save lives with the proposed vehicles but will also provide the winning team with over a million dollars in prize money. Stage two of the competition, the building stage, is open now. Winners of this phase will be announced on Nov. 18. 

“About 50% of helicopter accidents today are on the site of a medical evacuation,” Axten said. “Developing this technology can save the lives of first responders and those in need. It’s incredible work that can help a lot of people.”  

Additionally, the team recently qualified for the XPRIZE Wildfire Challenge, a four-year competition that tasks teams with developing and implementing UAS technology to fight wildfires. The Penn State team is in Track B of the competition, which includes a total prize of $5 million for the team who can develop the most effective autonomous firefighting system. 

In addition to Axten, Iyer and Valente, Eric Johnson, professor of aerospace engineering and faculty adviser to ARCC, is a co-author on the paper.  

The team used winnings from previous competitions alongside support from the College of Engineering and the Department of Aerospace Engineering to develop their drone platform. 

Last Updated February 25, 2025

Contact