This robot can walk on a slackline, ride a skateboard, and even fly. LEONARDO is a multimodal locomotion robot that can both fly and walk using thrusters and legs. Developed by a team at the Caltech Center for Autonomous Systems and Technologies (CAST), it offers a new solution for future robots designed to handle more complex tasks and challenges. | GIGadgets

Discrete-Time Hybrid Automata Learning: Legged Locomotion Meets Skateboarding Discrete-time Hybrid Automata Learning (DHAL) is a framework that uses on-policy Reinforcement Learning to identify and execute mode-switching without needing trajectory segmentation or event function learning. Hybrid dynamical systems, which combine continuous flow and discrete mode switching, are useful for modeling tasks like legged robot locomotion. Traditional model-based methods rely on predefined gaits, while mo