21 papers
Globalized Constrained Stein Variational Inference for Diverse Feasible Robot Motion Planning
Jiayun Li, Georgia Chalvatzaki
The paper introduces SteinSQP, a constrained Stein variational inference algorithm that generates diverse, feasible robot motion plans by embedding constraints into a kernel-space…
IMPACT: An Implicit Active-Set Augmented Lagrangian for Fast Contact-Implicit Trajectory Optimization
Jiayun Li, Dejian Gong, Georgia Chalvatzaki
Contact-implicit trajectory optimization (CITO) has attracted growing attention as a unified framework for planning and control in contact-rich robotic tasks. Recent approaches hav…
Bimanual Robot Manipulation via Multi-Agent In-Context Learning
Alessio Palma, Indro Spinelli, Vignesh Prasad +4
Language Models (LLMs) have emerged as powerful reasoning engines for embodied control. In particular, In-Context Learning (ICL) enables off-the-shelf, text-only LLMs to predict ro…
Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning
Rickmer Krohn, Vignesh Prasad, Gabriele Tiboni +1
Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception. However, Reinforcement Learning agents struggle to learn in such…
Robot-DIFT: Correspondence-Sensitive Diffusion Features for Contact-Rich Robot Manipulation
Yu Deng, Yufeng Jin, Xiaogang Jia +3
Robot manipulation often fails in the final millimeters: a policy may recognize the right object yet miss the pose offsets, boundaries, or pre-contact alignments needed for action.…
HARBOR: A Harness Framework for Agentic Robot Reinforcement Learning
Zechu Li, Yufeng Jin, Xiaoyang Liu +4
Reinforcement learning (RL) has become a powerful paradigm for robot learning, particularly in sim-to-real settings, but its broader adoption remains limited by the engineering pip…