HT-PINNs：面向混合动力系统学习的间断感知物理信息神经网络
Hybrid-time PINNs: Discontinuity-Aware Physics-Informed Neural Networks for Learning Hybrid Dynamical Systems
AI 导读
HT-PINNs 为混合时间域的每个流动区间分配一个网络，并与跳跃时间和未知参数联合训练，解决单一 PINNs 网络因时间平滑而无法建模跳跃的问题。该方法能恢复弹跳球的跳跃次数和恢复系数；在 MuJoCo 机器人拆卸仿真中，摩擦系数估计误差仅 0.1% 和 1.7%。
正文 · 原文
Abstract:Hybrid dynamical systems flow continuously but jump at discrete events. Physics-informed neural networks (PINNs) fit data while enforcing known dynamics, but a single network is smooth in time and cannot model these jumps. We show that the more closely a PINN follows a jump, the larger its physics loss must be. We propose hybrid-time PINNs (HT-PINNs), which assign one network to each flow interval of the hybrid time domain and train the networks jointly with the jump times and the unknown parameters. The number of jumps and the sequence of modes are found by enumeration. HT-PINNs recover the number of jumps and the restitution coefficient of a bouncing ball. In a MuJoCo simulation of robotic disassembly, where a robot pulls a press-fitted pin out of a housing through a spring, HT-PINNs find the number of jumps and the sequence of modes, and estimate the static and kinetic friction coefficients with small errors of 0.1% and 1.7%.
| Subjects: | Systems and Control (eess.SY); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33562 [eess.SY] |
| (or arXiv:2610.33562v1 [eess.SY] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33562 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shivanshu Tripathi [view email]
[v1]
Sat, 24 Oct 2026 09:29:58 UTC (533 KB)
来源：arXiv cs.AI · arxiv.org