Answers are grounded in this paper and cite their sources.
Rigid-body motion requires formulations that account for both rotational and translational dynamics. Two Lie groups, the special orthogonal group [Formula: see text] and the space of quaternions [Formula: see text], are commonly used to represent attitude. When considering rigid-body pose, i.e., position and attitude, the special Euclidean group [Formula: see text] and the space of dual quaternions [Formula: see text] are frequently utilized. These groups are also Riemannian manifolds. Riemannian optimization is an attractive approach to tackling optimal control and trajectory optimization problems on these spaces because it does not require the imposition of additional manifold-preserving constraints on the solver since they are accounted for in the optimization algorithm. Consequently, implementing these methods as the nonlinear program solvers in optimal control problems through direct transcription offers a reduction in the number of constraints imposed on the problem. This paper conducts the requisite work to apply a Riemannian augmented Lagrangian method to solve constrained trajectory optimization problems on matrix Lie groups. The unified optimization framework is applied to the rigid-body configuration spaces [Formula: see text] and [Formula: see text]. By maintaining an identical solver architecture across both formulations, the investigation isolates and permits a comparison of the influence of configuration space geometry on convergence behavior, constraint satisfaction, and control structure.
No takes yet. Share an insight, caveat, or question.
McCann et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: