Agile Flight
We address the state estimation, control, and trajectory tracking problems for agile flight with aerial robots, from developing resilient controllers to autonomous perching on inclined surfaces. Our research focuses on planning and executing dynamically feasible trajectories to navigate to the desired locations with onboard sensing and computation.
Multi-Robot Systems
Multi-robot systems such as swarms of aerial robots are naturally suited to offer additional flexibility, resilience, and robustness in several tasks compared to a single robot by enabling cooperation among the agents. We investigate the problem of coordination of multi-robot systems for transportation of cable-suspended rigid body payloads and collaborative perception.
Learning Model Representations and Actions
We investigate learning representations for control, perception, and planning, including topics such as deep reinforcement learning, supervised learning of system dynamics, learning-based end-to-end control, and continual learning. The goal of our research is to enable our aerial robots to perceive and interact with the environment in full autonomy.