Automatic Differentiation in Automatic Generation of the Linearized Equations of Motion

2021 ◽  
Author(s):  
Bruce Minaker ◽  
Francisco González
Keyword(s):  
Multibody Systems ◽  
Automatic Differentiation ◽  
Equations Of Motion ◽  
Computational Cost ◽  
Order Reduction ◽  
Automatic Generation ◽  
Analysis Model ◽  
Direct Evaluation ◽  
Base Function ◽  
Linearized Equations

Abstract In the ongoing search for mathematically efficient methods of predicting the motion of vehicle and other multibody systems, and presenting the associated results, one of the avenues of continued interest is the linearization of the equations of motion. While linearization can potentially result in reduced fidelity in the model, the benefits in computational speed often make it the pragmatic choice. Linearization techniques are also useful in modal and stability analysis, model order reduction, and state and input estimation. This paper explores the application of automatic differentiation to the generation of the linearized equations of motion. Automatic differentiation allows one to numerically evaluate the derivative of any function, with no prior knowledge of the differential relationship to other functions. It exploits the fact that every computer program must evaluate every function using only elementary arithmetic operations. Using automatic differentiation, derivatives of arbitrary order can be computed, accurately to working precision, with minimal additional computational cost over the evaluation of the base function. There are several freely available software libraries that implement automatic differentiation in modern computing languages. In the paper, several example multibody systems are analyzed, and the computation times of the stiffness matrix are compared using direct evaluation and automatic differentiation. The results show that automatic differentiation can be surprisingly competitive in terms of computational efficiency.

Symbolic Linearized Equations for Nonholonomic Multibody Systems With Closed-Loop Kinematics

2018 ◽  
Author(s):  
Márton Kuslits ◽  
Dieter Bestle
Keyword(s):  
Lagrange Multipliers ◽  
Multibody Systems ◽  
Closed Loop ◽  
Equations Of Motion ◽  
Linear Equations ◽  
Algebraic Equations ◽  
Computationally Efficient ◽  
Linearized Equations ◽  
Nonlinear Equations Of Motion ◽  
Loop Constraints

Multibody systems and associated equations of motion may be distinguished in many ways: holonomic and nonholonomic, linear and nonlinear, tree-structured and closed-loop kinematics, symbolic and numeric equations of motion. The present paper deals with a symbolic derivation of nonlinear equations of motion for nonholonomic multibody systems with closed-loop kinematics, where any generalized coordinates and velocities may be used for describing their kinematics. Loop constraints are taken into account by algebraic equations and Lagrange multipliers. The paper then focuses on the derivation of the corresponding linear equations of motion by eliminating the Lagrange multipliers and applying a computationally efficient symbolic linearization procedure. As demonstration example, a vehicle model with differential steering is used where validity of the approach is shown by comparing the behavior of the linearized equations with their nonlinear counterpart via simulations.

An Ordinary Differential Equation Formulation for Multibody Dynamics: Nonholonomic Constraints

2016 ◽  
Vol 17 (1) ◽  
Author(s):  
Edward J. Haug
Keyword(s):  
Differential Equations ◽  
Ordinary Differential Equations ◽  
Multibody Systems ◽  
Mass Matrix ◽  
Equations Of Motion ◽  
Computational Cost ◽  
Nonholonomic Constraints ◽  
State Variables ◽  
Well Conditioning ◽  
Fifth Order

A method is presented for formulating and numerically integrating ordinary differential equations of motion for nonholonomically constrained multibody systems. Tangent space coordinates are defined in configuration and velocity spaces as independent generalized coordinates that serve as state variables in the formulation, yielding ordinary differential equations of motion. Orthogonal-dependent coordinates and velocities are used to enforce constraints at position, velocity, and acceleration levels. Criteria that assure accuracy of constraint satisfaction and well conditioning of the reduced mass matrix in the equations of motion are used as the basis for updating local coordinates on configuration and velocity constraint manifolds, transparent to the user and at minimal computational cost. The formulation is developed for multibody systems with nonlinear holonomic constraints and nonholonomic constraints that are linear in velocity coordinates and nonlinear in configuration coordinates. A computational algorithm for implementing the approach is presented and used in the solution of three examples: one planar and two spatial. Numerical results using a fifth-order Runge–Kutta–Fehlberg explicit integrator verify that accurate results are obtained, satisfying all the three forms of kinematic constraint, to within error tolerances that are embedded in the formulation.

Sensivity Analysis of Flexible Multibody Systems

2005 ◽  
Author(s):  
M. A. Neto ◽  
R. P. Leal ◽  
J. Ambro´sio
Keyword(s):  
Multibody Systems ◽  
Automatic Differentiation ◽  
Direct Method ◽  
Equations Of Motion ◽  
Flexible Multibody Systems ◽  
Element Analysis ◽  
Design Variables ◽  
Finite Differences Method ◽  
Flexible Multibody ◽  
Analytical Sensitivities

In this work a general formulation for the computation of the first order analytical sensitivities based on the direct method is presented. The direct method for sensitivity calculation is obtained by differentiating the equations that define the response of the flexible system with respect to the design variables. The design variables used here are the ply orientations of the laminated. The analytical sensitivities are compared with the numerical results obtained by using the finite differences method. For the beam composite material elements, the section properties and their sensitivities are found using an asymptotic procedure that involves a two-dimensional finite element analysis of their cross-section. The equations of the sensitivities are obtained by automatic differentiation and integrated in time simultaneously with the equations of motion of the multibody systems. The equations of motion and sensitivities of the flexible multibody system are solved and the accelerations and velocities and sensitivities of accelerations and velocities are integrated. Through the application of the methodology to a single flexible multibody systems the difficulties and benefices of the procedure are discussed.

scholarly journals About the Gibbs-Appel equations for multibody systems

2006 ◽  
Vol 28 (4) ◽  
pp. 225-229
Author(s):  
Nguyen Van Khang
Keyword(s):  
Multibody Dynamics ◽  
Multibody Systems ◽  
Equations Of Motion ◽  
Automatic Generation ◽  
Matrix Form ◽  
Dynamical Equations

In this paper a matrix form of Gibbs-Appel function is recommended for multibody dynamics formulations. The form proposed in this paper seems to be more clear and suitable for automatic generation of dynamical equations of motion. The advantages followed from the formulation proposed are illustrated through an example.

Some Applications of Automatic Differentiation to Rigid, Flexible, and Constrained Multibody Dynamics

2005 ◽  
Author(s):  
D. Todd Griffith ◽  
James D. Turner ◽  
John L. Junkins
Keyword(s):  
Dynamical Systems ◽  
Multibody Dynamics ◽  
Multibody Systems ◽  
Automatic Differentiation ◽  
Equations Of Motion ◽  
Numerical Results ◽  
Flexible Multibody Systems ◽  
New Approach ◽  
Constrained Dynamical Systems ◽  
Constrained Multibody Dynamics

In this paper, we discuss several applications of automatic differentiation to multibody dynamics. The scope of this paper covers the rigid, flexible, and constrained dynamical systems. Particular emphasis is placed on the development of methods for automating the generation of equations of motion and the simulation of response using automatic differentiation. We also present a new approach for generating exact dynamical representations of flexible multibody systems in a numerical sense using automatic differentiation. Numerical results will be presented to detail the efficiency of the proposed methods.

Two Lie Group Formulations for Dynamic Multibody Systems With Large Rotations

2011 ◽  
Author(s):  
Olivier Bru¨ls ◽  
Martin Arnold ◽  
Alberto Cardona
Keyword(s):  
Numerical Solution ◽  
Lie Group ◽  
Multibody Systems ◽  
Equations Of Motion ◽  
Time Integration ◽  
Computational Cost ◽  
Differential Algebraic Equations ◽  
Rigid Body Dynamics ◽  
Time Step ◽  
Iteration Matrix

This paper studies the formulation of the dynamics of multibody systems with large rotation variables and kinematic constraints as differential-algebraic equations on a matrix Lie group. Those equations can then be solved using a Lie group time integration method proposed in a previous work. The general structure of the equations of motion are derived from Hamilton principle in a general and unifying framework. Then, in the case of rigid body dynamics, two particular formulations are developed and compared from the viewpoint of the structure of the equations of motion, of the accuracy of the numerical solution obtained by time integration, and of the computational cost of the iteration matrix involved in the Newton iterations at each time step. In the first formulation, the equations of motion are described on a Lie group defined as the Cartesian product of the group of translations R3 (the Euclidean space) and the group of rotations SO(3) (the special group of 3 by 3 proper orthogonal transformations). In the second formulation, the equations of motion are described on the group of Euclidean transformations SE(3) (the group of 4 by 4 homogeneous transformations). Both formulations lead to a second-order accurate numerical solution. For an academic example, we show that the formulation on SE(3) offers the advantage of an almost constant iteration matrix.

EQUATIONS OF MOTION OF COMPLEX MULTIBODY SYSTEMS USING KINEMATICAL DIFFERENTIALS

1989 ◽  
Vol 13 (4) ◽  
pp. 113-121 ◽  
Author(s):  
M. HILLER ◽  
A. KECSKEMETHY
Keyword(s):  
Multibody Systems ◽  
Degrees Of Freedom ◽  
Multibody System ◽  
Equations Of Motion ◽  
Linear Equations ◽  
Nonholonomic Systems ◽  
Automatic Generation ◽  
Minimal Order ◽  
Closed Form Solutions ◽  
Constraint Equations

In complex multibody systems the motion of the bodies may depend on only a few degrees of freedom. For these systems, the equations of motion of minimal order, although more difficult to obtain, give a very efficient formulation. The present paper describes an approach for the automatic generation of these equations, which avoids the use of LAGRANGE-multipliers. By a particular concept, designated “kinematical differentials”, the problem of determining the partial derivatives required to state the equations of motion is reduced to a simple re-evaluation of the kinematics. These cover the solution of the global position, velocity and acceleration problems, i.e. the motion of all bodies is determined for given generalized (independent) coordinates. For their formulation and solution, the multibody system is mapped to a network of nonlinear transformation elements which are connected by linear equations. Each transformation element, designated “kinematical transformer”, corresponds to an independent multibody loop. This mapping of the constraint equations makes it possible to find closed-form solutions to the kinematics for a wide variety of technical applications, and (via kinematical differentials) leads also to an efficient formulation of the dynamics. The equations are derived for holonomic, scleronomic systems, but can also be extended to general nonholonomic systems.

Sign in / Sign up

Export Citation Format

Share Document