Researchers at Purdue University are advancing a new generation of intelligent robotic systems that can navigate complex environments, perceive, learn, and collaborate autonomously. Backed by a five-year, $1.5 million cooperative agreement from the U.S. Army Combat Capabilities Development Command Army Research Laboratory (DEVCOM ARL), Purdue’s Intelligent Design for Exploration and Augmented Systems (IDEAS) Lab is developing AI-enabled air–ground robotic teams capable of sensing, mapping, and maneuvering through hostile terrain without reliance on GPS.
Led by Aniket Bera, associate professor of computer science and director of the Purdue IDEAS Lab, the research focuses on coordinated teams of aerial and ground robots that can operate collaboratively in GPS-denied and high-risk environments. These systems are intended to support and protect military operations by autonomously navigating uncertain terrain where rapid decision-making is critical and human deployment may be unsafe.

This project focuses on advanced terrain mapping and classification to support autonomous navigation, enabling dynamic adaptation to complex environments and potential threats. (Graphic courtesy of Aniket Bera)
The project aims to create trusted robotic teammates that integrate the aerial perspective of uncrewed aerial vehicles with the ground-level situational awareness of unmanned ground systems. By combining these complementary viewpoints, the research team is developing AI-enabled systems that can operate effectively in complex, unpredictable conditions while sharing information, coordinating actions, and adapting in real time.
The AI principles developed through this program are expected to support a range of defense applications, including autonomous scouting, threat reconnaissance, logistics route planning, and tactical surveillance. The integration of aerial and ground intelligence enables faster, safer, and more accurate situational awareness for military operations.
The research is conducted within Purdue’s IDEAS Lab, which brings together postdoctoral researchers, Ph.D. students, and undergraduate assistants working at the intersection of artificial intelligence, computer vision, and robotics. The lab specializes in embodied AI systems that combine physical sensing, learning, and reasoning to support intelligent decision-making in real-world environments.
“Our lab focuses on building intelligent robotic systems that integrate machine learning, motion planning, and human–robot collaboration,” said Bera. “Research will take place in Purdue’s Hicks Robotics and Autonomy Testbed, a 13,000+ square-foot space equipped with humanoid robots, quadrupeds, aerial drones, and advanced sensor platforms. This environment allows seamless integration between simulation and real-world deployment, a crucial step for developing AI-enabled aerial and ground collaboration systems that must function autonomously under uncertainty.”
The IDEAS Lab collaborates with several major Purdue research initiatives that strengthen the university’s robotics and AI ecosystem, including the Institute for Physical AI (IPAI), the Institute for Control, Optimization, and Networks (ICON), and the Purdue Computes AI Initiative. Purdue Computes emphasizes four pillars of Purdue’s computational environment: computing departments, physical AI, quantum science, and semiconductor innovation.
“Purdue offers a unique ecosystem for developing embodied AI, blending deep technical expertise, cutting-edge infrastructure, and strong partnerships with federal and industrial stakeholders,” said Bera. “The university provides humanoid and quadruped robots, UAV fleets, sensor arrays, and high-performance GPU clusters. Cross-campus collaborations through IPAI, ICON, and Purdue Computes foster connections between AI theorists, engineers, and applied roboticists. This ecosystem enables Purdue to prototype, deploy, and operationalize AI-driven robotic systems rapidly, translating academic innovation into military and industrial impact.”
This award builds on prior DEVCOM ARL–supported research led by the IDEAS Lab on terrain intelligence and adaptive navigation for autonomous systems in dynamic and hostile environments. Earlier work focused on single-agent autonomy and navigation in uncertain terrain, whereas the current project extends this to multi-agent systems that integrate aerial and ground intelligence. Planned next steps include integrating adaptive learning, conducting field evaluations, and scaling to larger, distributed robotic networks.
“This project is exciting because it bridges AI research and defense technology in a meaningful way,” said Bera. “It allows us to bring the latest advances in large-scale learning, multi-agent reasoning, and intelligent planning into physical platforms that operate autonomously in the field.”
The research also supports Purdue’s broader leadership in intelligent robotics and embodied AI, integrating sensing, reasoning, and action to support future Army missions.
“Working with DEVCOM ARL, we are contributing to the Army’s mission of operationalizing science by transitioning academic breakthroughs in AI and robotics into field-ready capabilities,” said Bera. “Our work advances the Army’s modernization goals by developing AI systems that think, adapt, and collaborate, enabling soldiers and intelligent machines to operate together safely and effectively. We are building robots that don’t just move, they think, learn, and serve.”
“I often describe our systems as a team of intelligent scouts and navigators,” said Bera. “The drones serve as the scouts, surveying the environment and providing global awareness, while the ground robots act as navigators, interpreting the terrain in detail and executing the mission plan.
The connection between them, the AI framework we design, acts like a shared nervous system. It fuses perception, communication, and decision-making, allowing these systems to operate as a cohesive unit. This aerial–ground collaboration transforms machines from isolated tools into intelligent teammates.”
The funding will also support expansion of the IDEAS Lab team, including additional postdoctoral researchers, doctoral candidates, and undergraduate researchers focused on AI-enabled autonomy.
“These researchers will work on perception, motion planning, data fusion, and control algorithms that enable safe and intelligent behavior in multi-robot teams,” said Bera. “Beyond scientific advancement, this funding also helps train the next generation of scientists and engineers who will drive AI and robotics innovation for national defense, which blends computer science, machine learning, and systems engineering to create practical, trustworthy autonomy.”
The funded project, titled Advanced Air–Ground Teaming for Autonomous Terrain Mapping and Threat-Aware Navigation, is supported through a five-year, $1.5 million cooperative agreement awarded by DEVCOM ARL, the U.S. Army’s corporate research laboratory.
About the Purdue IDEAS Lab
The Intelligent Design for Exploration and Augmented Systems (IDEAS) Lab is a research laboratory within the Department of Computer Science at Purdue University. Led by Aniket Bera, the lab focuses on embodied artificial intelligence, integrating machine learning, computer vision, robotics, and human–robot interaction. The IDEAS Lab conducts research on autonomous systems operating in complex, uncertain environments, with applications spanning defense, robotics, and real-world AI deployment. Its work emphasizes multi-agent coordination, perception-driven decision-making, and AI-enabled autonomy for aerial and ground robotic systems. For more information, please click here.
Source: Purdue University IDEAS Lab
(Editor’s Note: All trademarks mentioned in this article, including company names, product names, and logos, are the property of their respective owners. Use of these trademarks is for informational purposes only and does not imply any endorsement.)
- Featured Expert: Comau - February 8, 2026
- Featured Expert: Universal Robots - February 7, 2026
- Featured Expert: Stäubli Robotics - February 5, 2026
