În afara temei aiqbench
Această știre a fost preluată de la o sursă pe care o urmărim, dar nu ține de domeniul AI sau cuantic, așa că nu există o perspectivă aiqbench pentru ea.
LEAP: Gradient Flow Augmented Lagrangian Trajectory Optimization for Contact Constrained Robots
arXiv:2610.20936v1 Announce Type: new Abstract: Generating dynamically feasible motions for legged and manipulating robots requires solving a large, nonlinear, non-convex optimal control problem (OCP) in which contact wrenches must satisfy friction-cone constraints throughout the motion, not only at isolated instants. Existing solvers are sensitive to the initial guess and scale poorly with the number of joints. This paper introduces LEAP, which transcribes the OCP with pseudospectral collocation and, instead of a Newton or SQP solve, integrates a primal--dual gradient flow of a Powell--Hestenes--Rockafellar augmented Lagrangian to compute a solution. Because the flow is stiff, this paper designs a linearly implicit integrator with a Gauss--Newton metric, momentum, and a merit-based step controller that keeps the augmented Lagrangian monotone. Across locomotion benchmarks on five robot platforms, from planar hoppers to 32-joint humanoids, LEAP is the only method to solve every tested problem, achieves substantial speedups over Crocoddyl and Aligator on several benchmarks, and reaches trackable references at smaller compute budgets. The resulting trajectories are validated by tracking them in a high-fidelity simulator and on hardware. We will make the code public upon final acceptance of this paper.
Authors:César E. Ramos Chuquiure, Ryan Tsai, Yutong Zhou, Zihang Huang, Vansh Thakur, Akshunn Trivedi, Zachary Lin, Pearl Lin, Ruikai Yang, Challen Enninful Adu, Ram Vasudevan
Abstract:Generating dynamically feasible motions for legged and manipulating robots requires solving a large, nonlinear, non-convex optimal control problem (OCP) in which contact wrenches must satisfy friction-cone constraints throughout the motion, not only at isolated instants. Existing solvers are sensitive to the initial guess and scale poorly with the number of joints. This paper introduces LEAP, which transcribes the OCP with pseudospectral collocation and, instead of a Newton or SQP solve, integrates a primal--dual gradient flow of a Powell--Hestenes--Rockafellar augmented Lagrangian to compute a solution. Because the flow is stiff, this paper designs a linearly implicit integrator with a Gauss--Newton metric, momentum, and a merit-based step controller that keeps the augmented Lagrangian monotone. Across locomotion benchmarks on five robot platforms, from planar hoppers to 32-joint humanoids, LEAP is the only method to solve every tested problem, achieves substantial speedups over Crocoddyl and Aligator on several benchmarks, and reaches trackable references at smaller compute budgets. The resulting trajectories are validated by tracking them in a high-fidelity simulator and on hardware. We will make the code public upon final acceptance of this paper.
Robotics (cs.RO)
arXiv:2610.20936 [cs.RO]
(or arXiv:2610.20936v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.20936 arXiv-issued DOI via DataCite (pending registration)
From: Cesar E. Ramos-Chuquiure [view email] [v1] Wed, 14 Oct 2026 21:04:31 UTC (4,171 KB)
Text integral sub licența deschisă a publicației · 237 cuvinte
Subiecte conexe
MAGMA: Multiplier-Augmented Geometric Motion Algorithm
arXiv · 16 oct. 2026
CQDC-Warp: GPU-Accelerated Differentiable Quasi-Dynamic Simulation for Contact-Rich Manipulation
arXiv · 16 oct. 2026
Exploring the Design Space of Reward Backpropagation for Flow Matching
arXiv · 10 iun. 2026
Learning Stiffness Dependent Fluid Structure Dynamics from Coarse Flow Representations
arXiv · 24 sept. 2026