PINN-Based Predictive Control Combined With Unknown Payload Identification for Robots With Prismatic Quasi-Direct-Drives

Abstract

This study introduces a unified control framework that addresses the challenge of precise robots with Quasi-Direct-Drives under unknown payloads, named as online payload identification-based physics-informed neural network predictive control (OPI-PINNPC). By integrating online payload identification with physics-informed neural networks (PINNs), our approach embeds identified payload parameters directly into the neural network's loss function, ensuring physical consistency while adapting to changing load conditions. The physics-constrained neural representation serves as an efficient surrogate model within our nonlinear model predictive controller, enabling real-time optimization despite the complex dynamics of robots with Quasi-Direct-Drives. Experimental validation on our robot platform demonstrates 35% improvement in position and orientation tracking accuracy across diverse payload conditions, with substantially faster convergence compared to previous adaptive control methods. Our framework provides an adaptive solution for maintaining tracking performance under variable payload conditions without sacrificing computational efficiency.