Se (3)-diffusionfields: Learning smooth cost functions for joint grasp and motion optimization through diffusion

J Urain, N Funk, J Peters… - 2023 IEEE International …, 2023 - ieeexplore.ieee.org
Multi-objective optimization problems are ubiquitous in robotics, eg, the optimization of a
robot manipulation task requires a joint consideration of grasp pose configurations …

Motion planning diffusion: Learning and planning of robot motions with diffusion models

J Carvalho, AT Le, M Baierl, D Koert… - 2023 IEEE/RSJ …, 2023 - ieeexplore.ieee.org
Learning priors on trajectory distributions can help accelerate robot motion planning
optimization. Given previously successful plans, learning trajectory generative models as …

Constrained stein variational trajectory optimization

T Power, D Berenson - IEEE Transactions on Robotics, 2024 - ieeexplore.ieee.org
In this article, we present constrained Stein variational trajectory optimization (CSVTO), an
algorithm for performing trajectory optimization with constraints on a set of trajectories in …

Accelerating motion planning via optimal transport

AT Le, G Chalvatzaki, A Biess… - Advances in Neural …, 2024 - proceedings.neurips.cc
Motion planning is still an open problem for many disciplines, eg, robotics, autonomous
driving, due to their need for high computational resources that hinder real-time, efficient …

Hierarchical policy blending as inference for reactive robot control

K Hansel, J Urain, J Peters… - 2023 IEEE International …, 2023 - ieeexplore.ieee.org
Motion generation in cluttered, dense, and dynamic environments is a central topic in
robotics, rendered as a multi-objective decision-making problem. Current approaches trade …

Motion planning diffusion: Learning and adapting robot motion planning with diffusion models

J Carvalho, A Le, P Kicki, D Koert, J Peters - arxiv preprint arxiv …, 2024 - arxiv.org
The performance of optimization-based robot motion planning algorithms is highly
dependent on the initial solutions, commonly obtained by running a sampling-based planner …

Revisiting energy based models as policies: Ranking noise contrastive estimation and interpolating energy models

S Singh, S Tu, V Sindhwani - arxiv preprint arxiv:2309.05803, 2023 - arxiv.org
A crucial design decision for any robot learning pipeline is the choice of policy
representation: what type of model should be used to generate the next set of robot actions …

Sampling constrained trajectories using composable diffusion models

T Power, R Soltani-Zarrin, S Iba… - IROS 2023 Workshop on …, 2023 - openreview.net
Trajectory optimization and optimal control are powerful tools for synthesizing complex robot
behavior using appropriate cost functions and constraints. However, methods for solving the …

Preconditioners for the Stochastic Training of Implicit Neural Representations

SF Chng, H Saratchandran, S Lucey - arxiv preprint arxiv:2402.08784, 2024 - arxiv.org
Implicit neural representations have emerged as a powerful technique for encoding complex
continuous multidimensional signals as neural networks, enabling a wide range of …

Learning stable vector fields on lie groups

J Urain, D Tateo, J Peters - IEEE Robotics and Automation …, 2022 - ieeexplore.ieee.org
Learning robot motions from demonstration requires models able to specify vector fields for
the full robot pose when the task is defined in operational space. Recent advances in …