aiqbench analysis
This research introduces "Sumo," a sim-to-real framework that allows legged robots to dynamically manipulate heavy objects using test-time steering of pre-trained policies without additional training. The approach demonstrates significant generalization across diverse tasks and object types, bridging the gap between simulation and real-world deployment on platforms like the Spot quadruped. By enabling whole-body dexterity for complex physical interactions, this work advances the practical capabilities of autonomous robotics in unstructured environments.
Sumo: Dynamic and Generalizable Whole-Body Loco-Manipulation
arXiv:2604.08508v3 Announce Type: replace Abstract: This paper presents a sim-to-real approach that enables legged robots to dynamically manipulate large and heavy objects with whole-body dexterity. Our key insight is that by performing test-time steering of a pre-trained whole-body control policy with a sample-based planner, we can enable these robots to solve a variety of dynamic loco-manipulation tasks. Interestingly, we find our method generalizes to a diverse set of objects and tasks with no additional tuning or training, and can be further enhanced by flexibly adjusting the cost function at test time. We demonstrate the capabilities of our approach through a variety of challenging loco-manipulation tasks on a Spot quadruped robot in the real world, including uprighting a tire heavier than the robot's nominal lifting capacity and dragging a crowd-control barrier larger and taller than the robot itself. Additionally, we show that the same approach can be generalized to humanoid loco-manipulation tasks, such as opening a door and pushing a table, in simulation. Project code and videos are available at https://sumo.rai-inst.com/.
Authors:John Z. Zhang, Maks Sorokin, Jan Brüdigam, Brandon Hung, Stephen Phillips, Dmitry Yershov, Farzad Niroui, Tong Zhao, Leonor Fermoselle, Xinghao Zhu, Chao Cao, Duy Ta, Tao Pang, Jiuguang Wang, Preston Culbertson, Zachary Manchester, Simon Le Cléac'h
Abstract:This paper presents a sim-to-real approach that enables legged robots to dynamically manipulate large and heavy objects with whole-body dexterity. Our key insight is that by performing test-time steering of a pre-trained whole-body control policy with a sample-based planner, we can enable these robots to solve a variety of dynamic loco-manipulation tasks. Interestingly, we find our method generalizes to a diverse set of objects and tasks with no additional tuning or training, and can be further enhanced by flexibly adjusting the cost function at test time. We demonstrate the capabilities of our approach through a variety of challenging loco-manipulation tasks on a Spot quadruped robot in the real world, including uprighting a tire heavier than the robot's nominal lifting capacity and dragging a crowd-control barrier larger and taller than the robot itself. Additionally, we show that the same approach can be generalized to humanoid loco-manipulation tasks, such as opening a door and pushing a table, in simulation. Project code and videos are available at this https URL.
Robotics (cs.RO)
arXiv:2604.08508 [cs.RO]
(or arXiv:2604.08508v3 [cs.RO] for this version)
Full text under the publisher's open licence · 213 words
Related coverage
WholeBodyWAM: Generalizing Pre-trained World-Action Priors to Humanoid Loco-Manipulation via WBC-Grounded Coordination
arXiv · Sep 16, 2026
$\lambda_0$: Scaling Open-World Egocentric Human Data for Humanoid Loco-Manipulation
arXiv · Oct 15, 2026
MotionWAM: Towards Foundation World Action Models for Real-Time Humanoid Loco-Manipulation
arXiv · Jun 9, 2026
Being-M0.7: A Latent World-Action Model for Humanoid Robots
arXiv · Oct 9, 2026