This conference paper was submitted for presentation at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

Many different types of physics can be considered in the simulation of robot performance. A robot is a mechatronics device since it is a mechanical system that can move a load from point to point, guided by a sophisticated controller to meet specifications. The accurate prediction of the dynamic performance of a robot requires a mechanical model that correctly represents the flexibility of the components and joints, as well as an accurate representation of the controller system, including the actual controller firmware code where possible. It is also necessary to represent the realistic dynamic performance of the actuators. Robotics in manufacturing, production, and packaging is of interest to a variety of industries, including transportation and mobility, industrial equipment, aerospace and defense, consumer packaged goods and retail, energy, and materials. This presentation will provide background information and an assortment of case studies that show how motion-focused, multiphysics simulation can accurately predict advanced robotic behavior prior to a hardware prototype being built, reducing cost and risk. Discussion topics include: 1) Role and application of rigid body dynamics, 2) Integration with nonlinear controllers, 3) Application of sensors for closed-loop adaptability, 4) Addition of linear and nonlinear flexible bodies, 5) Consideration of gripper flexibility and contact, 6) System-level validation of dynamic behavior, 7) Impact on complexity when considering a mobile robot, and 8) Integration of particle-based CFD for robot that control a fluid jet with significant momentum. Two case studies will be reviewed in detail. The first considers robot placement accuracy and frequency response of the system considering joints and arms with various stiffness. The second case study adds the complication of heavy fluid flow at the end effector on the positioning accuracy during a sweep of the arm. Namely, a substantial flow of liquid exits a tube during the motion of the robot arm. Insights will be gained regarding the effect of the fluid flow on system vibrations and the ability of the arm to traverse the target curve accurately. Robot designs have been simulated for decades, but not with multiple types of physics at the same time. In many cases the simulation results were accurate and useful. However, in some cases, the interactions between the different types of physics were missed, and the robot exhibited problems that were not indicated in the simulation results. The use of multiphysics simulation techniques reduces the probability of missing key design limitations. Using multiphysics (multibody dynamics, nonlinear FEA, controls, and CFD) for realistic simulation. In order to take full advantage of multiphysics simulation, best practices should be followed in creating the models and checking the results.