Skip to content
RobotWorld
← Back to Videos

Sample-Efficient Model-Based RL for Spot Loco-Manipulation

Loading video
Loading video

BNN dynamics with a kinematic prior enable sample-efficient model-based RL for dynamic end-effector tracking on Spot with arm.

Boston Dynamics SpotBNNModel-Based-RLLoco-manipulationBoston DynamicsMobile ManipulationModel-Driven Reinforcement LearningQuadruped robotsPaper
Category: research
Author: @breadli428
Date: 2025-10-31T00:00:00
Duration: 41.726s
Source: https://x.com/i/status/1984300011056717893
Reference: https://sites.google.com/view/learning-more-with-less

Related Videos

MessyMem: Persistent Spatial Memory for Robots
0:48

MessyMem: Persistent Spatial Memory for Robots

@anuvabanwasi · 854 views · 2026-09-16
Persistent MemoryMobile Manipulation3D scene graphs
Arkshel MX01 transformable robot
0:46

Arkshel MX01 transformable robot

@techniahqrobot · 442 views · 2026-08-13
Transformable robotsArkshelMX01
VideoJev: A System One Model for Video Understanding and Interaction
1:31

VideoJev: A System One Model for Video Understanding and Interaction

@_akhaliq · 38 views · 2026-10-14
VideoJevVideo understandingTemporal Grounding
Veda Sparse Attention Speeds Up Video Character Replacement in ComfyUI
0:27

Veda Sparse Attention Speeds Up Video Character Replacement in ComfyUI

@ai_hakase_ · 128 views · 2026-10-13
Video GenerationSparse AttentionComfyUI
Draft: Parametric Robot Design Exploration Without Rebuilding CAD
0:26

Draft: Parametric Robot Design Exploration Without Rebuilding CAD

@spaceandtech_ · 211 views · 2026-10-11
Robot DesignMJCFMuJoCo
RoboJEPA: The First Scaling Law for Multi-Embodiment World Models
1:06

RoboJEPA: The First Scaling Law for Multi-Embodiment World Models

@artemZholus · 209 views · 2026-10-10
World ModelsJEPAScaling Law

As an Amazon Associate, we earn from qualifying purchases.