Deep Lagrangian Networks for end-to-end learning of energy-based control for under-actuated systems

Understand

Applying Deep Learning to control has a lot of potential for enabling the intelligent design of robot control laws.

• Unfortunately common deep learning approaches to control, such as deep reinforcement learning, require an unrealistic amount of interaction with the real system, do not yield any performance guarantees, and do not make good use of extensive insights from model-based control.
• In particular, common black-box approaches -- that abandon all insight from control -- are not suitable for complex robot systems.
• We propose a deep control approach as a bridge between the solid theoretical foundations of energy-based control and the flexibility of deep learning.