Physics-informed neural networks for the prediction of robot dynamics considering motor and external force couplings
Regular Papers|Updated:2026-01-12
|
Physics-informed neural networks for the prediction of robot dynamics considering motor and external force couplings
Enhanced Publication
考虑电机与外力耦合的机器人动力学预测物理信息神经网络
“Physics-informed neural networks (PINNs) have made significant progress in modeling conservative systems of rigid-body dynamics. This study proposes two enhanced PINNs that integrate motor dynamics and external force modeling, providing solutions to solve high-precision torque estimation problems in complex external force scenarios.”
Frontiers of Information Technology & Electronic EngineeringVol. 26, Issue 12, Pages: 2604-2622(2025)
Affiliations:
College of Automation, Beijing Information Science and Technology University, Beijing 100192, China
Fengyu SUN, Shuangshuang WU, Zhiming LI, et al. Physics-informed neural networks for the prediction of robot dynamics considering motor and external force couplings[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(12): 2604-2622.
DOI:
Fengyu SUN, Shuangshuang WU, Zhiming LI, et al. Physics-informed neural networks for the prediction of robot dynamics considering motor and external force couplings[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(12): 2604-2622. DOI: 10.1631/FITEE.2500254.
Physics-informed neural networks for the prediction of robot dynamics considering motor and external force couplingsEnhanced Publication