Robot dynamics: A recursive algorithm for efficient calculation of Christoffel symbols

•Christoffel symbols of the first kind are calculated recursively.•Calculations based on inertial parameters and transformations matrices.•Execution time and numerical error as performance indicators.•Compares favourably with existing methods.•MATLAB running code is available. Christoffel symbols of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Mechanism and machine theory Jg. 142; S. 103589
Hauptverfasser: Safeea, Mohammad, Neto, Pedro, Bearee, Richard
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.12.2019
Elsevier
Schlagworte:
Christoffel symbols
Dynamicss
Engineering Sciences
Recursive algorithms
Recursive algorithms
Christoffel symbols
Dynamicss
Bioengineering
Mechanics of Materials
ISSN:0094-114X, 1873-3999
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Christoffel symbols of the first kind are calculated recursively.•Calculations based on inertial parameters and transformations matrices.•Execution time and numerical error as performance indicators.•Compares favourably with existing methods.•MATLAB running code is available. Christoffel symbols of the first kind are very important in robot dynamics. They are used for tuning various proposed robot controllers, for determining the bounds on Coriolis/Centrifugal matrix, for mathematical formulation of optimal trajectory calculation, among others. In the literature of robot dynamics, Christoffel symbols of the first kind are calculated from Lagrangian dynamics using an off-line generated symbolic formula. In this study we present a novel and efficient recursive, non-symbolic, method where Christoffel symbols of the first kind are calculated on-the-fly based on the inertial parameters of robot’s links and their transformation matrices. The proposed method was analyzed in terms of execution time, computational complexity and numerical error. Results show that the proposed algorithm compares favorably with existing methods.
ISSN:0094-114X
1873-3999
DOI:10.1016/j.mechmachtheory.2019.103589