New PINN model learns hybrid systems with discrete jumps

Researchers have developed Hybrid-time PINNs (HT-PINNs), a novel approach to modeling hybrid dynamical systems that exhibit both continuous flow and discrete jumps. Unlike traditional PINNs that struggle with discontinuities, HT-PINNs utilize separate networks for each flow interval and jointly train them to identify jump times and unknown parameters. This method has demonstrated success in accurately predicting the number of jumps and estimating physical parameters in simulations, such as a bouncing ball and a robotic disassembly task. AI

IMPACT Introduces a new method for modeling complex systems with both continuous and discrete dynamics, potentially improving AI's ability to simulate and control real-world physical processes.

RANK_REASON Research paper introducing a novel method for learning hybrid dynamical systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• HT-PINNs
• Hybrid-time PINNs
• MuJoCo
• physics-informed neural networks
• Shivanshu Tripathi

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →