Thermodynamics-informed neural networks for physically realistic mixed reality
This work addresses the need for real-time, interactive physics simulation in mixed reality to enhance user experience, representing an incremental improvement in domain-specific applications.
The paper tackles the problem of achieving physically realistic simulations for mixed reality by developing a deep learning method that computes dynamic responses of deformable objects in real-time, ensuring thermodynamic consistency and demonstrating performance through examples.
The imminent impact of immersive technologies in society urges for active research in real-time and interactive physics simulation for virtual worlds to be realistic. In this context, realistic means to be compliant to the laws of physics. In this paper we present a method for computing the dynamic response of (possibly non-linear and dissipative) deformable objects induced by real-time user interactions in mixed reality using deep learning. The graph-based architecture of the method ensures the thermodynamic consistency of the predictions, whereas the visualization pipeline allows a natural and realistic user experience. Two examples of virtual solids interacting with virtual or physical solids in mixed reality scenarios are provided to prove the performance of the method.