Combining Physics and Deep Learning to learn ContinuousTime Dynamics Models
Abstract
Deep learning has been widely used within learning algorithms for robotics. One disadvantage of deep networks is that these networks are blackbox representations. Therefore, the learned approximations ignore the existing knowledge of physics or robotics. Especially for learning dynamics models, these blackbox models are not desirable as the underlying principles are well understood and the standard deep networks can learn dynamics that violate these principles. To learn dynamics models with deep networks that guarantee physically plausible dynamics, we introduce physicsinspired deep networks that combine first principles from physics with deep learning. We incorporate Lagrangian mechanics within the model learning such that all approximated models adhere to the laws of physics and conserve energy. Deep Lagrangian Networks (DeLaN) parametrize the system energy using two networks. The parameters are obtained by minimizing the squared residual of the EulerLagrange differential equation. Therefore, the resulting model does not require specific knowledge of the individual system, is interpretable, and can be used as a forward, inverse, and energy model. Previously these properties were only obtained when using system identification techniques that require knowledge of the kinematic structure. We apply DeLaN to learning dynamics models and apply these models to control simulated and physical rigid body systems. The results show that the proposed approach obtains dynamics models that can be applied to physical systems for realtime control. Compared to standard deep networks, the physicsinspired models learn better models and capture the underlying structure of the dynamics.
 Publication:

arXiv eprints
 Pub Date:
 October 2021
 arXiv:
 arXiv:2110.01894
 Bibcode:
 2021arXiv211001894L
 Keywords:

 Computer Science  Machine Learning;
 Computer Science  Robotics