Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning
Premiered Mar 22, 2022
Nvidia presented parts of this work at GTC 2022, revealing our humanoid-quadruped transformer!
Title: Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning
Authors: Eric Vollenweider, Marko Bjelonic, Victor Klemm, Nikita Rudin, Joonho Lee and Marco Hutter
Paper submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems in Kyoto.
Abstract: In recent years, reinforcement learning (RL) has shown outstanding performance for locomotion control of highly articulated robotic systems. Such approaches typically involve tedious reward function tuning to achieve the desired motion style. Imitation learning approaches such as adversarial motion priors aim to reduce this problem by encouraging a pre-defined motion style. In this work, we present an approach to augment the concept of adversarial motion prior-based RL to allow for multiple, discretely switchable styles. We show that multiple styles and skills can be learned simultaneously without notable performance differences, even in combination with motion data-free skills. Our approach is validated in several real-world experiments with a wheeled-legged quadruped robot showing skills learned from existing RL controllers and trajectory optimization, such as ducking and walking, and novel skills such as switching between a quadrupedal and humanoid configuration. For the latter skill, the robot is required to stand up, navigate on two wheels, and sit down. Instead of tuning the sit-down motion, we verify that a reverse playback of the stand-up movement helps the robot discover feasible sit-down behaviors and avoids tedious reward function tuning.
Note: The following parts of the video are sped up:
- Door opening and closing when the robot stands inside the elevator between 00:20 and 00:21 (+200%)
- In-between the standing up and sitting down sequence between 00:49 and 01:10 (+200% only the navigation on two legs)
- Reaction with the crowd between 1:41 and 1:50 (+150%)
- Last drone footage after 2:01 (+200%)
Learn more at Swiss-Mile Robotics AG
Last edited by Airicist2; 20th January 2024 at 10:12.
Rocky VI: robot training montage
May 24, 2022
Finally, after the first Rocky movie in 1976, the Robotic Systems Lab presents a continuation of the iconic series. Our transformer robot visited Philly in 2022 as part of the International Conference on Robotics and Automation.
Credits: Marko Bjelonic, Hendrik Kolvenbach, Takahiro Miki, Nikita Rudin, Vassilios Tsounis, Maria Vittoria Minniti, Julian Nubert, Lorenz Wellhausen, Jonhoo Lee, Dimitrios Kanoulas, Marco Hutter
Robot loco-manipulation for industrial applications
Jun 27, 2022
This video shows uninterrupted operation of ANYmal Bull. The robot performs the following tasks: turning a wheel, pulling a lever, pushing a gate, and pulling a rope to lift a bucket. The robot walks around and up the ramp teleoperated via a remote controller. However, the robot solves each of the manipulation tasks autonomously.
Additional notes: Perception is done with a Vicon motion capture system and with the readings of the robot joint states. The bucket weighs approximately 1.1 kg.
Timestamps:
00:00 Turning the wheel
00:20 Pulling the lever
00:38 Walking around the scaffolding
00:55 Walking up the first ramp
01:12 Opening the gate
01:30 Walking to the top level
01:57 Lifting the bucket
Perceptive Locomotion through Nonlinear Model Predictive Control
Aug 18, 2022
This work is currently under review, a preprint is available:
researchgate.net/publication/362759412_Perceptive_Locomotion_through_Nonlinear_Model_Predictive_Control
Title:
Perceptive Locomotion through Nonlinear Model Predictive Control
Authors:
Ruben Grandia, Fabian Jenelten, Shaohui Yang, Farbod Farshidian, and Marco Hutter
Abstract:
Dynamic locomotion in rough terrain requires accurate foot placement, collision avoidance, and planning of the underactuated dynamics of the system. Reliably optimizing for such motions and interactions in the presence of imperfect and often incomplete perceptive information is challenging. We present a complete perception, planning, and control pipeline, that can optimize motions for all degrees of freedom of the robot in real-time. To mitigate the numerical challenges posed by the terrain a sequence of convex inequality constraints is extracted as local approximations of foothold feasibility and embedded into an online model predictive controller. Steppability classification, plane segmentation, and a signed distance field are precomputed per elevation map to minimize the computational effort during the optimization. A combination of multiple-shooting, real-time iteration, and a filter-based line-search are used to solve the formulated problem reliably and at high rate. We validate the proposed method in scenarios with gaps, slopes, and stepping stones in simulation and experimentally on the ANYmal quadruped platform, resulting in state-of-the-art dynamic climbing.
Advanced skills by learning locomotion and local navigation end-to-end
Sep 27, 2022
Local navigation and locomotion of legged robots are commonly split into separate modules.
In this work, we propose to combine them by training an end-to-end policy with deep reinforcement learning.
Training a policy in this way opens up a larger set of possible solutions, which allows the robot to learn more complex behaviors.
IROS 2022
work by Nikita Rudin, David Hoeller, Marko Bjelonic, and Marco Hutter
paper: https://arxiv.org/abs/2209.12827
project website: sites.google.com/leggedrobotics.com/end-to-end-loco-navigation/home
Slippery surfaces are no match for ANYmal
Mar 14, 2023
ANYmal plowing through belly-deep snow
Mar 15, 2023
Towards legged locomotion on steep planetary terrain
Aug 2, 2023
We present a novel locomotion strategy for the legged robotic platform ANYmal. The paper will be presented at the IROS 2023 conference in Detroit, USA.
Paper: research-collection.ethz.ch/handle/20.500.11850/625001
Authors: Cedric Weibel, Giorgio Valsecchi, Hendrik Kolvenbach, Marco Hutter
Credits: ETH Zurich, Beyond Gravity, ANYmal Research, ESA
Music: Cinematic Atmosphere Score 2, Resistance - GvidonAudio
ANYmal mastering the uneven wooden obstacle at SPRINT Robotics World Conference 2023
Oct 18, 2023
ANYmal, the autonomous legged robot designed for industrial challenging environments, provides the mobility, autonomy, and inspection intelligence to enable safe and efficient inspection operations. In this virtual showcase, discover how ANYmal climbs stairs, recovers from a fall, performs an autonomous mission and avoids obstacles, docks to charge by itself, digitizes analog sensors, and monitors the environment.
The unrivaled performance of ANYmal: 24/7 in all weather
Dec 7, 2023
ANYmal, the autonomous legged robot designed for industrial challenging environments, provides the mobility, autonomy, and inspection intelligence to enable safe and efficient inspection operations. In this virtual showcase, discover how ANYmal climbs stairs, recovers from a fall, performs an autonomous mission and avoids obstacles, docks to charge by itself, digitizes analog sensors, and monitors the environment.
Posting Permissions
Reply With Quote
Социальные закладки