simulation-based pretraining, real-world learning, and iterative adaptation. The first phase will consist of
training locomotion... of the policy [FineTune25,WM25]. Alternatively, the project will explore direct
training on the robot using off-policy... -
Voir cette offre d'emploi