TU Darmstadt / ULB / TUbiblio

MBSlib - An Efficient Multibody Systems Library for Kinematics and Dynamics Simulation, Optimization and Sensitivity Analysis

Wojtusch, Janis and Kunz, Jürgen and Stryk, Oskar von (2016):
MBSlib - An Efficient Multibody Systems Library for Kinematics and Dynamics Simulation, Optimization and Sensitivity Analysis.
In: Robotics and Automation Letters, IEEE, p. 954, 1, (2), DOI: 10.1109/LRA.2016.2527830,
[Article]

Abstract

The dynamic behavior of many technical and biomechanical systems can be modeled, simulated and optimized by using a multibody systems approach. Systems with many degrees of freedom typically result in complex and high-dimensional multibody systems models that necessitate capable modeling approaches and efficient computational algorithms as well as impede the numerical solution of related optimization problems. In this paper, the efficient and modular multibody systems library MBSLIB for kinematics and dynamics simulation, optimization and sensitivity analysis is presented. MBSLIB provides an intuitive modeling interface, modular software architecture, efficient computational algorithms as well as the computation of derivatives with respect to system states, control variables or model parameters. In combination with gradient-based optimization methods, the derivatives can be used to facilitate the numerical solution of optimization problems significantly. The library is open and can be obtained from the MBSLIB website. These features make MBSLIB a powerful, flexible and lightweight modeling, simulation and optimization library suitable for many technical and biomechanical applications.

Item Type: Article
Erschienen: 2016
Creators: Wojtusch, Janis and Kunz, Jürgen and Stryk, Oskar von
Title: MBSlib - An Efficient Multibody Systems Library for Kinematics and Dynamics Simulation, Optimization and Sensitivity Analysis
Language: German
Abstract:

The dynamic behavior of many technical and biomechanical systems can be modeled, simulated and optimized by using a multibody systems approach. Systems with many degrees of freedom typically result in complex and high-dimensional multibody systems models that necessitate capable modeling approaches and efficient computational algorithms as well as impede the numerical solution of related optimization problems. In this paper, the efficient and modular multibody systems library MBSLIB for kinematics and dynamics simulation, optimization and sensitivity analysis is presented. MBSLIB provides an intuitive modeling interface, modular software architecture, efficient computational algorithms as well as the computation of derivatives with respect to system states, control variables or model parameters. In combination with gradient-based optimization methods, the derivatives can be used to facilitate the numerical solution of optimization problems significantly. The library is open and can be obtained from the MBSLIB website. These features make MBSLIB a powerful, flexible and lightweight modeling, simulation and optimization library suitable for many technical and biomechanical applications.

Journal or Publication Title: Robotics and Automation Letters, IEEE
Volume: 1
Number: 2
Uncontrolled Keywords: Biological system modeling;Computational modeling;Heuristic algorithms;Libraries;Mathematical model;Numerical models;Optimization;Calibration and Identification;Direct/Inverse Dynamics Formulation;Multibody Systems;Optimization and Optimal Control;Sensiti
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Simulation, Systems Optimization and Robotics Group
Date Deposited: 20 Jun 2016 23:26
DOI: 10.1109/LRA.2016.2527830
Identification Number: RAL_Wojtusch2016
Related URLs:
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item