Aerial Manipulation
Aerial manipulation integrates the mobility of drones with the dexterity of robotic arms to perform tasks in challenging environments. Current research emphasizes robust onboard perception (using vision and tactile sensing), advanced control algorithms (including model-predictive control, reinforcement learning, and hybrid motion-force controllers), and cooperative strategies for multi-robot systems, often employing model-based deep reinforcement learning for tasks like grasping and pushing. This field is significant for its potential to automate complex tasks in agriculture, infrastructure inspection, and disaster response, while also advancing robotics research in areas such as dynamic control and multi-sensor fusion.
Papers
Modular Adaptive Aerial Manipulation under Unknown Dynamic Coupling Forces
Rishabh Dev Yadav, Swati Dantu, Wei Pan, Sihao Sun, Spandan Roy, Simone Baldi
The Power of Input: Benchmarking Zero-Shot Sim-To-Real Transfer of Reinforcement Learning Control Policies for Quadrotor Control
Alberto Dionigi, Gabriele Costante, Giuseppe Loianno