No video

Versatile Multi-Contact Planning and Control for Legged Loco-Manipulation

  Рет қаралды 10,338

Robotic Systems Lab: Legged Robotics at ETH Zürich

Robotic Systems Lab: Legged Robotics at ETH Zürich

Күн бұрын

Abstract:
Loco-manipulation planning skills are pivotal for expanding the utility of robots in everyday environments. These skills can be assessed on the basis of a system’s ability to coordinate complex holistic movements and multiple contact interactions when solving different tasks. However, existing approaches have been merely able to shape such behaviors with hand-crafted state machines, densely engineered rewards, or prerecorded expert demonstrations. Here, we propose a minimally guided framework that automatically discovers whole-body trajectories jointly with contact schedules for solving general loco-manipulation tasks in premodeled environments. The key insight is that multimodal problems of this nature can be formulated and treated within the context of integrated task and motion planning (TAMP). An effective bilevel search strategy is achieved by incorporating domain-specific rules and adequately combining the strengths of different planning techniques: trajectory optimization and informed graph search coupled with sampling-based planning. We showcase emergent behaviors for a quadrupedal mobile manipulator exploiting both prehensile and nonprehensile interactions to perform real-world tasks such as opening/closing heavy dishwashers and traversing spring-loaded doors. These behaviors are also deployed on the real system using a two-layer whole-body tracking controller.
Paper: www.science.or...
arxiv.org/abs/2...
Authors: Jean-Pierre Sleiman, Farbod Farshidian, Marco Hutter

Пікірлер: 5
@stevenjensjorgensen
@stevenjensjorgensen Жыл бұрын
Excellent work! The sampling-based bi-level optimization is the key novelty and seems very efficient.
@therobotstudio
@therobotstudio 11 ай бұрын
Fantastic work, as always!
@SHAINON117
@SHAINON117 Жыл бұрын
So awesome great work ❤
@NeuroScientician
@NeuroScientician 7 ай бұрын
Is this an open source projectt?
@gf2e
@gf2e Жыл бұрын
Great work, and a great explanation. I noticed at 2:06 it seemed to indicate that a foot contacting the door was forbidden. But at 4:10 I see a foot holding the door so it doesn’t spring shut. Is there a difference between the two scenarios?
DeepGait: Planning and Control of Quadrupedal Gaits using Deep Reinforcement Learning (Presentation)
8:18
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 20 М.
Combining Learning-Based Locomotion with Model-Based Manipulation for Legged Mobile Manipulators
3:41
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 6 М.
OMG what happened??😳 filaretiki family✨ #social
01:00
Filaretiki
Рет қаралды 13 МЛН
ISSEI & yellow girl 💛
00:33
ISSEI / いっせい
Рет қаралды 22 МЛН
КТО ЛЮБИТ ГРИБЫ?? #shorts
00:24
Паша Осадчий
Рет қаралды 1,4 МЛН
DTC: Deep Tracking Control
6:46
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 15 М.
Internet is going wild over this problem
9:12
MindYourDecisions
Рет қаралды 135 М.
Unlocking Agile and Object-Aware Loco-Manipulation (ICRA'24 ATOMPC Workshop)
31:05
Robot Motor Intelligence (RoMI) Lab
Рет қаралды 450
Tutorial: Gait and Trajectory Optimization for Legged Robots
28:41
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 37 М.
80 Year Olds Share Advice for Younger Self
12:22
Sprouht
Рет қаралды 2 МЛН
TAMOLS: Terrain-Aware Motion Optimization for Legged Systems
5:25
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 7 М.
Cassie: Dynamic Planning on Stairs
3:37
Agility Robotics
Рет қаралды 1,5 МЛН
Hybrid Locomotion for Wheeled-Legged Robots (Presentation)
8:58
Robotic Systems Lab: Legged Robotics at ETH Zürich
Рет қаралды 22 М.
OMG what happened??😳 filaretiki family✨ #social
01:00
Filaretiki
Рет қаралды 13 МЛН