Physics-informed Neural Networks to Model and Control Robots: a Theoretical and Experimental Investigation

Physics-inspired neural networks are proven to be an effective modeling method by giving more physically plausible results with less data dependency. However, their application in robotics is limited due to the non-conservative nature of robot dynamics and the difficulty in friction modeling. Moreover, these physics-inspired neural networks do not account for complex input matrices, such as those found in underactuated soft robots. This paper solves these problems by extending Lagrangian and Hamiltonian neural networks by including dissipation and a simplified input matrix. Additionally, the loss function is processed using the Runge-Kutta algorithm, circumventing the inaccuracies and environmental susceptibility inherent in direct acceleration measurements. First, the effectiveness of the proposed method is validated via simulations of soft and rigid robots. Then, the proposed approach is validated experimentally in a tendon-driven soft robot and a Panda robot. The simulations and experimental results show that the modified neural networks can model different robots while the learned model enables decent anticipatory control.

Trajectory Tracking for Robotic Arms with Input Saturation and Only Position Measurements

In this work, we propose a passivity-based control approach that addresses the trajectory tracking problem for a class of mechanical systems that comprises a broad range of robotic arms. The resulting controllers can be naturally saturated and do not require velocity measurements. Moreover, the proposed methodology does not require the implementation of observers, and the structure of the closed-loop system permits the identification of a Lyapunov function, which eases the convergence analysis. To corroborate the effectiveness of the methodology, we perform experiments with the Philips Experimental Robot Arm.