Explanipedia

How Well do Diffusion Policies Learn Kinematic Constraint Manifolds? Open

Lexi Foland, Thomas Cohn, An‐Chi Wei, Nicholas Pfaff, Boyuan Chen , et al. · 2025

Diffusion policies have shown impressive results in robot imitation learning, even for tasks that require satisfaction of kinematic equality constraints. However, task performance alone is not a reliable indicator of the policy's ability t…

Mixed Discrete and Continuous Planning using Shortest Walks in Graphs of Convex Sets Open

Savva Morozov, Tobia Marcucci, Bernhard Paus Graesdal, Alexandre Amice, Pablo A. Parrilo , et al. · 2025

We study the Shortest-Walk Problem (SWP) in a Graph of Convex Sets (GCS). A GCS is a graph where each vertex is paired with a convex program, and each edge couples adjacent programs via additional costs and constraints. A walk in a GCS is …

A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation Open

TRI LBM Team, Jose Barreiros, Andrew Beaulieu, Aditya Bhat, Rick Cory , et al. · 2025

Robot manipulation has seen tremendous progress in recent years, with imitation learning policies enabling successful performance of dexterous and hard-to-model tasks. Concurrently, scaling data and model size has led to the development of…

Physics-Driven Data Generation for Contact-Rich Manipulation via Trajectory Optimization Open

Lujie Yang, H. J. Terry Suh, Тонг Жао, Bernhard Paus Graesdal, Tarik Kelestemur , et al. · 2025

Superfast Configuration-Space Convex Set Computation on GPUs for Online Motion Planning Open

P. Werner, Richard Cheng, Tom Stewart, Russ Tedrake, Daniela Rus · 2025

Steerable Scene Generation with Post Training and Inference-Time Search Open

Nicholas Pfaff, Hongkai Dai, Sergey Zakharov, Shun Iwase, Russ Tedrake · 2025

Training robots in simulation requires diverse 3D scenes that reflect the specific challenges of downstream tasks. However, scenes that satisfy strict task requirements, such as high-clutter environments with plausible spatial arrangement,…

Superfast Configuration-Space Convex Set Computation on GPUs for Online Motion Planning Open

P. Werner, Richard Cheng, Tom Stewart, Russ Tedrake, Daniela Rus · 2025

In this work, we leverage GPUs to construct probabilistically collision-free convex sets in robot configuration space on the fly. This extends the use of modern motion planning algorithms that leverage such representations to changing envi…

Empirical Analysis of Sim-and-Real Cotraining of Diffusion Policies for Planar Pushing from Pixels Open

An‐Chi Wei, Abhinav Agarwal, Boyuan Chen, Rohan Bosworth, Nicholas Pfaff , et al. · 2025

Cotraining with demonstration data generated both in simulation and on real hardware has emerged as a promising recipe for scaling imitation learning in robotics. This work seeks to elucidate basic principles of this sim-and-real cotrainin…

Sampling-Based Motion Planning with Discrete Configuration-Space Symmetries Open

Thomas Cohn, Russ Tedrake · 2025

When planning motions in a configuration space that has underlying symmetries (e.g. when manipulating one or multiple symmetric objects), the ideal planning algorithm should take advantage of those symmetries to produce shorter trajectorie…

Scalable Real2Sim: Physics-Aware Asset Generation Via Robotic Pick-and-Place Setups Open

Nicholas Pfaff, Evelyn Fu, Jeremy Binagia, Phillip Isola, Russ Tedrake · 2025

Simulating object dynamics from real-world perception shows great promise for digital twins and robotic manipulation but often demands labor-intensive measurements and expertise. We present a fully automated Real2Sim pipeline that generate…

Physics-Driven Data Generation for Contact-Rich Manipulation via Trajectory Optimization Open

H. J. Terry Suh, Тонг Жао, Bernhard Paus Graesdal, Jiuguang Wang, Tao Pang , et al. · 2025

We present a low-cost data generation pipeline that integrates physics-based simulation, human demonstrations, and model-based planning to efficiently generate large-scale, high-quality datasets for contact-rich robotic manipulation tasks.…

History-Guided Video Diffusion Open

Ki‐Whan Song, Boyuan Chen, Max Simchowitz, Yilun Du, Russ Tedrake , et al. · 2025

Classifier-free guidance (CFG) is a key technique for improving conditional generation in diffusion models, enabling more accurate control while enhancing sample quality. It is natural to extend this technique to video diffusion, which gen…

Planning Shorter Paths in Graphs of Convex Sets by Undistorting Parametrized Configuration Spaces Open

Shruti Garg, Thomas Cohn, Russ Tedrake · 2024

Optimization based motion planning provides a useful modeling framework through various costs and constraints. Using Graph of Convex Sets (GCS) for trajectory optimization gives guarantees of feasibility and optimality by representing conf…

Faster Algorithms for Growing Collision-Free Convex Polytopes in Robot Configuration Space Open

P. Werner, Thomas Cohn, Rebecca H. Jiang, Tim Seyde, Max Simchowitz , et al. · 2024

We propose two novel algorithms for constructing convex collision-free polytopes in robot configuration space. Finding these polytopes enables the application of stronger motion-planning frameworks such as trajectory optimization with Grap…

Multi-Query Shortest-Path Problem in Graphs of Convex Sets Open

Savva Morozov, Tobia Marcucci, Alexandre Amice, Bernhard Paus Graesdal, Rohan Bosworth , et al. · 2024

The Shortest-Path Problem in Graph of Convex Sets (SPP in GCS) is a recently developed optimization framework that blends discrete and continuous decision making. Many relevant problems in robotics, such as collision-free motion planning, …

GCS*: Forward Heuristic Search on Implicit Graphs of Convex Sets Open

Shao Yuan Chew Chia, Rebecca H. Jiang, Bernhard Paus Graesdal, Leslie Pack Kaelbling, Russ Tedrake · 2024

We consider large-scale, implicit-search-based solutions to Shortest Path Problems on Graphs of Convex Sets (GCS). We propose GCS*, a forward heuristic search algorithm that generalizes A* search to the GCS setting, where a continuous-valu…

Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion Open

Boyuan Chen, Diego Martí Monsó, Yilun Du, Max Simchowitz, Russ Tedrake , et al. · 2024

This paper presents Diffusion Forcing, a new training paradigm where a diffusion model is trained to denoise a set of tokens with independent per-token noise levels. We apply Diffusion Forcing to sequence generative modeling by training a …

OpenVLA: An Open-Source Vision-Language-Action Model Open

Moo Jin Kim, Karl Pertsch, Siddharth Karamcheti, Ted Xiao, Ashwin Balakrishna , et al. · 2024

Large policies pretrained on a combination of Internet-scale vision-language data and diverse robot demonstrations have the potential to change how we teach robots new skills: rather than training new behaviors from scratch, we can fine-tu…

Lyapunov-stable Neural Control for State and Output Feedback: A Novel Formulation Open

Lujie Yang, Hongkai Dai, Zhouxing Shi, Cho‐Jui Hsieh, Russ Tedrake , et al. · 2024

Learning-based neural network (NN) control policies have shown impressive empirical performance in a wide range of tasks in robotics and control. However, formal (Lyapunov) stability guarantees over the region-of-attraction (ROA) for NN co…

Towards Tight Convex Relaxations for Contact-Rich Manipulation Open

Bernhard Paus Graesdal, Shao Yuan Chew Chia, Tobia Marcucci, Savva Morozov, Alexandre Amice , et al. · 2024

We present a novel method for global motion planning of robotic systems that interact with the environment through contacts. Our method directly handles the hybrid nature of such tasks using tools from convex optimization. We formulate the…

Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots Open

Cheng Chi, Zhenjia Xu, Chuer Pan, Eric Cousineau, Benjamin Burchfiel , et al. · 2024

We present Universal Manipulation Interface (UMI) -- a data collection and policy learning framework that allows direct skill transfer from in-the-wild human demonstrations to deployable robot policies. UMI employs hand-held grippers coupl…

PoCo: Policy Composition from and for Heterogeneous Robot Learning Open

Lirui Wang, Jialiang Zhao, Yilun Du, Edward H. Adelson, Russ Tedrake · 2024

Training general robotic policies from heterogeneous data for different tasks is a significant challenge. Existing robotic datasets vary in different modalities such as color, depth, tactile, and proprioceptive information, and collected i…

Shortest Paths in Graphs of Convex Sets Open

Tobia Marcucci, Jack Umenberger, Pablo A. Parrilo, Russ Tedrake · 2024

Given a graph, the shortest-path problem requires finding a sequence of edges\nwith minimum cumulative length that connects a source vertex to a target\nvertex. We consider a variant of this classical problem in which the position\nof each…

Certifying Bimanual RRT Motion Plans in a Second Open

Alexandre Amice, P. Werner, Russ Tedrake · 2023

We present an efficient method for certifying non-collision for piecewise-polynomial motion plans in algebraic reparametrizations of configuration space. Such motion plans include those generated by popular randomized methods including RRT…

Approximating Robot Configuration Spaces with few Convex Sets using Clique Covers of Visibility Graphs Open

P. Werner, Alexandre Amice, Tobia Marcucci, Daniela Rus, Russ Tedrake · 2023

Many computations in robotics can be dramatically accelerated if the robot configuration space is described as a collection of simple sets. For example, recently developed motion planners rely on a convex decomposition of the free space to…

Robot Fleet Learning via Policy Merging Open

Lirui Wang, Kaiqing Zhang, Allan Zhou, Max Simchowitz, Russ Tedrake · 2023

Fleets of robots ingest massive amounts of heterogeneous streaming data silos generated by interacting with their environments, far more than what can be stored or transmitted with ease. At the same time, teams of robots should co-acquire …

Constrained Bimanual Planning with Analytic Inverse Kinematics Open

Thomas Cohn, Seiji Shaw, Max Simchowitz, Russ Tedrake · 2023

In order for a bimanual robot to manipulate an object that is held by both hands, it must construct motion plans such that the transformation between its end effectors remains fixed. This amounts to complicated nonlinear equality constrain…

Non-Euclidean Motion Planning with Graphs of Geodesically-Convex Sets Open

Thomas Cohn, Mark Petersen, Max Simchowitz, Russ Tedrake · 2023

Computing optimal, collision-free trajectories for high-dimensional systems is a challenging problem.Samplingbased planners struggle with the dimensionality, whereas trajectory optimizers may get stuck in local minima due to inherent nonco…

Proximity and Visuotactile Point Cloud Fusion for Contact Patches in Extreme Deformation Open

Jessica Yin, Paarth Shah, Naveen Kuppuswamy, Andrew Beaulieu, Avinash Uttamchandani , et al. · 2023

Visuotactile sensors are a popular tactile sensing strategy due to high-fidelity estimates of local object geometry. However, existing algorithms for processing raw sensor inputs to useful intermediate signals such as contact patches strug…

Fighting Uncertainty with Gradients: Offline Reinforcement Learning via Diffusion Score Matching Open

H. J. Terry Suh, Glen Chou, Hongkai Dai, Lujie Yang, Abhishek Gupta , et al. · 2023

Gradient-based methods enable efficient search capabilities in high dimensions. However, in order to apply them effectively in offline optimization paradigms such as offline Reinforcement Learning (RL) or Imitation Learning (IL), we requir…

Russ Tedrake YOU? Author Swipe