In a Model Predictive Control (MPC) setting, the precise simulation of the behavior of the system over a finite time window is essential. This application-oriented benchmark study focuses on a robot arm that exhibits various nonlinear behaviors. For this arm, we have a physics-based model with approximate parameter values and an open benchmark dataset for system identification. However, the long-term simulation of this model quickly diverges from the actual arm’s measurements, indicating its inaccuracy. We compare the accuracy of black-box and purely physics-based approaches with several physics-informed approaches. These involve different combinations of a neural network’s output with information from the physics-based model or feeding the physics-based model’s information into the neural network. One of the physics-informed model structures can improve accuracy over a fully black-box model.

We describe the design and implementation of RSTATION, an object-oriented, modular robot simulator with hierarchical analysis capabilities. Modularity is achieved via the features of design encapsulation and enables grouping a set of interconnected components into a single component and dividing the robot system into several sets of subordinate modules recursively. By careful construction of the data types and classes, RSTATION allows for hierarchical simulation of the kinematics, and the dynamics at three levels: considering only main links (high-level), using simplified models including dynamic properties of transmission elements (intermediate level), and taking into account the detailed kinematics and dynamics of transmission elements (low-level). Submodules can be set to different resolution during a single simulation. The data types and classes also exploit a recent set of coordinate invariant robot analysis algorithms based on modern screw theory. Central to the low-level dynamic analysis capability is an algorithm for systematically extracting the constraint equations for general gearing systems. The various features of RSTATION are illustrated with a detailed case study of a commercial industrial robot.