We explore the science of flight across all drone types, fixed-wing aircraft, quadcopters, and helicopters, to build smarter, safer, and more resilient autonomous systems. 🚁 Our lab focuses on optimal control, trajectory planning, and decision-making under uncertainty, developing algorithms that allow drones to think and react in real time. We’re particularly interested in interception and avoidance strategies for intruder drones, combining physics-based modeling with intelligent guidance and sensing. By merging theory with hands-on experimentation, from high-fidelity simulation to flight testing, we aim to create aerial systems that can cooperate, defend, and adapt safely in dynamic environments. 🌍 As a team, we’re pushing the boundaries of autonomy to make the skies safer, smarter, and more connected. ☁️📡

We use artificial intelligence to figure out what kind of drone we’re looking at: quadcopter, fixed-wing, or helicopter. By studying how each drone moves, our system learns to recognize its unique flight pattern. This helps us detect unusual or unsafe behavior and decide how to respond quickly and safely. Our AI models can tell us what type of drone it is. Our algorithm showed an accuracy for identifying the correct drone type of 96%.

We study how uncertainty affects the performance of machines and structures from cranes and drones to precision robots. ⚙️📊 Our research focuses on global sensitivity analysis (GSA), which helps us find out which parameters (like mass, stiffness, or damping) have the biggest impact on how a system vibrates or behaves. 🎯 

To make this process faster, we use a method called Polynomial Chaos Expansion (PCE), a powerful alternative to traditional Monte Carlo (MC) simulations. Instead of running thousands of random tests, PCE builds a smart mathematical model that captures the same information with only a fraction of the effort. For instance: In one of our projects, we applied GSA to vibratory structures and reduced computational time from over 10 hours to less than 1 minute using PCE, while keeping 99.4% accuracy. ⏱️📈

We also use Shapley values from game theory 🎮 to understand how different uncertainties “cooperate” or “compete” to influence system behavior. By combining Shapley-based GSA with PCE, we create fast, explainable, and reliable tools for designing robust engineering systems, where uncertainty becomes quantifiable.📊