This work presents a shooting algorithm to compute the periodic responses of geometrically nonlinear structures modelled under the special Euclidean (SE) Lie group formulation. The formulation is combined with a pseudo-arclength continuation method, while special adaptations are made to ensure compatibility with the SE framework. Nonlinear normal modes (NNMs) of various two-dimensional structures including a doubly clamped beam, a shallow arch, and a cantilever beam are computed. Results are compared with a reference displacement-based FE model with von Kármán strains. Significant difference is observed in the dynamic response of the two models in test cases involving large degrees of beam displacements and rotation. Differences in the contribution of higher-order modes substantially affect the frequency-energy dependence and the nonlinear modal interactions observed between the models. It is shown that the SE model, owing to its exact representation of the beam kinematics, is better suited at adequately capturing complex nonlinear dynamics compared to the von Kármán model.
A two-field version of the floating frame-of-reference approach is developed to describe the dynamics of a flexible body in the case of motion in a uniformly rotating frame. Full decoupling of the kinetic energy results from splitting of the motion into rigid-body modes of the floating center of mass and a velocity field orthogonal to them. The motion equations are developed from a canonical variational expression in which displacement and velocity fields play independent roles. They turn out to keep the same global structure as in the case of motion in an inertial frame. The discretization of all inertia terms is discussed in depth. It is also shown that, in the case of discretization of the continuum with 3D solid elements, all gyroscopic terms can be derived from a mass kernel and associated $\mathbf{S} _{i}$ matrices. Model reduction is proposed using the Herting–Martinez method in order to automatically satisfy orthogonality of the velocity field to rigid-body modes. Two application examples of high-speed rotating systems are developed at length to assess the efficiency of the proposed methodology.
An independent velocity field is introduced via Legendre transformation of the kinetic energy of a geometrically-exact beam, leading to a first-order system of twice as many governing equations as a one-field formulation. Nevertheless, the new field does not have to be assembled across elements and can be eliminated at the element level, so that the assembled system has the same size as a one-field formulation. Furthermore, because the new field does not have to satisfy the compatibility equations that the original velocity field is subjected to, its finite-element discretization is simpler and leads to simplified inertial forces.
This paper presents a finite element implementation of plates and shells for the analysis of flexible multibody systems. The developments are set within the framework of the motion formalism that (1) uses configuration and motion to describe the kinematics of flexible multibody systems, (2) couples their displacement and rotation components by recognizing that configuration and motion are members of the Special Euclidean group, and (3) resolves all tensors components in local frames. The formulation based on the motion formalism (1) provides a theoretical framework that streamlines the formulation of shell elements, (2) leads to governing equations of motion that are objective, intrinsic, and present a reduced order of nonlinearity, (3) improves the efficiency of the solution process, (4) circumvents the shear locking phenomenon that plagues shell formulations based on classical kinematic descriptions, and (5) prevents the occurrence of singularities in the treatment of finite rotation. Numerical examples are presented to illustrate the advantageous features of the proposed formulation.
26 September 2024 in Journal of Intelligent & Robotic Systems
The aim of coordinated planning is to avoid robot-to-robot collisions in a multi-robot system, and there are two standard solution approaches: centralized planning and decoupled planning. Our first contribution is a decoupled planning approach that ensures $$\varvec{\mathcal {C}^2}$$ -continuous control commands with zero velocities and zero accelerations at the start and goal. We benchmark our decoupled approach with a centralized approach. Contrary to literature, we show that for a standard motion planning pipeline, such as the one used by MoveIt!, centralized planning is superior to decoupled planning in dual-arm manipulation: It has a lower computation time and a higher robustness. Our second contribution is an optimization that minimizes the rotational motion of an end-effector while considering obstacle avoidance. We derive the analytic gradients of this optimization problem, making the algorithm suitable for online motion planning. Our optimization extends an existing path quality improvement method. Integrating it into our decoupled approach overcomes its shortcomings and provides a motion planning pipeline that is robust at up to 99.9 % with a planning time of less than 1 s and that computes high-quality paths.
7 January 2022 in Journal of Computational and Nonlinear Dynamics
This paper describes a finite element approach to the analysis of flexible multibody systems. It is based on the motion formalism that (1) uses configuration and motion to describe the kinematics of flexible multibody systems, (2) recognizes that these are members of the special Euclidean group thereby coupling their displacement and rotation components, and (3) resolves all tensors components in local frames. The goal of this review paper is not to provide an in-depth derivation of all the elements found in typical multibody codes but rather to demonstrate how the motion formalism (1) provides a theoretical framework that unifies the formulation of all structural elements, (2) leads to governing equations of motion that are objective, intrinsic, and present a reduced order of nonlinearity, (3) improves the efficiency of the solution process, and (4) prevents the occurrence of singularities.