A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Planning in Learned Latent Action Spaces for Generalizable Legged Locomotion
[article]
2021
arXiv
pre-print
Preserved Fulltext
Other Versions
Once this latent space is learned, we plan over continuous latent actions in a model-predictive control fashion, using a learned high-level dynamics model. ...
Hierarchical learning has been successful at learning generalizable locomotion skills on walking robots in a sample-efficient manner. ...
This motivates the importance of model-based planning in learned latent action spaces for generalization versus learning a model-free policy. ...
arXiv:2008.11867v5
fatcat:4hy5buncavbwrm3ygu43hc2r4y
Learning Actionable Representations with Goal-Conditioned Policies
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Most prior work on representation learning has focused on generative approaches, learning representations that capture all underlying factors of variation in the observation space in a more disentangled ...
In this paper, we instead aim to learn functionally salient representations: representations that are not necessarily complete in terms of capturing all factors of variation in the observation space, but ...
We thank Pim de Haan, Aviv Tamar, Vitchyr Pong, and Ignasi Clavera for helpful insights and discussions. ...
arXiv:1811.07819v2
fatcat:pphhr45mt5dt5mfbkfcb7gwuj4
Real-Time Trajectory Adaptation for Quadrupedal Locomotion using Deep Reinforcement Learning
2021
2021 IEEE International Conference on Robotics and Automation (ICRA)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Additionally, in order to capture terrain information, we include a latent representation of the height maps in the observation space of the RL environment as a form of exteroceptive feedback. ...
We train a policy using deep reinforcement learning (RL) to introduce additive deviations to a reference trajectory in order to generate a feedback-based trajectory tracking system for a quadrupedal robot ...
These corrective controllers have been employed for manipulation tasks [19] and also for legged locomotion [20] . ...
doi:10.1109/icra48506.2021.9561639
fatcat:ohxlvuhgnrc47oy4tsygvq5dna
Learning nonparametric policies by imitation
2008
2008 IEEE/RSJ International Conference on Intelligent Robots and Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
The novel contribution of this work is a method for learning a nonparametric policy which generalizes a fixed action plan to operate over a continuous space of task variation. ...
The robot uses inference in a graphical model to learn sensor-based dynamics and infer a stable plan from a teacher's demonstration of an action. ...
In this example the learned policy maps a 3D state representation of two latent posture dimensions and a foot sensor based dynamics dimension to actions in the 2D latent posture space. ...
doi:10.1109/iros.2008.4650778
dblp:conf/iros/GrimesR08
fatcat:2kplfe3dkvf4je6ldjmnczmqxi
Multi-Agent Manipulation via Locomotion using Hierarchical Sim2Real
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our method hinges on the use of hierarchical sim2real -- a simulated environment is used to learn low-level goal-reaching skills, which are then used as the action space for a high-level RL controller, ...
Manipulation and locomotion are closely related problems that are often studied in isolation. ...
Acknowledgments We thank Chad Richards, Byron David, Matt Neiss, Krista Reymann, Ben Eysenbach, Sergey Levine, and the rest of Robotics at Google for helpful thoughts and discussions. ...
arXiv:1908.05224v2
fatcat:ywya6tymcfbp3ev7kdvvct72di
Learning Vision-Guided Quadrupedal Locomotion End-to-End with Cross-Modal Transformers
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, we introduce LocoTransformer, an end-to-end RL method that leverages both proprioceptive states and visual observations for locomotion control. ...
While learning-based locomotion has made great advances using RL, most methods still rely on domain randomization for training blind agents that generalize to challenging terrains. ...
Acknowledgement: This work was supported, in part, by gifts from Meta, Qualcomm, and TuSimple. ...
arXiv:2107.03996v3
fatcat:l7tjgxb3prgv7p5vju4k5se2im
Language2Pose: Natural Language Grounded Pose Forecasting
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
This joint embedding space is learned end-to-end using a curriculum learning approach which emphasizes shorter and easier sequences first before moving to longer and harder ones. ...
Generating animations from natural language sentences finds its applications in a a number of domains such as movie script visualization, virtual human animation and, robot motion planning. ...
Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of National Science Foundation or Oculus VR. ...
arXiv:1907.01108v2
fatcat:jdrxu46mwfcbnkd47tx3nvj3vm
HyperDynamics: Meta-Learning Object and Agent Dynamics with Hypernetworks
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
It outperforms existing dynamics models in the literature that adapt to environment variations by learning dynamics over high dimensional visual observations, capturing the interactions of the agent in ...
We test HyperDynamics on a set of object pushing and locomotion tasks. ...
MODEL UNROLLING AND ACTION PLANNING Action-conditioned dynamics models can be unrolled forward in time for long-term planning and control tasks. ...
arXiv:2103.09439v1
fatcat:42kmfav7bfedba3phtlgcoh2xq
D2RL: Deep Dense Architectures in Reinforcement Learning
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2020 pre-print arXiv:2010.09163v1 fatcat:d4kkg622dbc4zeddpoianzg54y
While improvements in deep learning architectures have played a crucial role in improving the state of supervised and unsupervised learning in computer vision and natural language processing, neural network ...
architecture choices for reinforcement learning remain relatively under-explored. ...
INTRODUCTION Deep Reinforcement Learning (DRL) is a general purpose framework for training goal-directed agents in high dimensional state and action spaces. ...
arXiv:2010.09163v2
fatcat:vt2x32wbnnekfadpbuvafsgdqe
SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our hybrid simulator combines neural networks and traditional physics simulation to balance expressiveness and generalizability, and alleviates the need for a carefully selected parameter set in System ...
We show that our approach outperforms multiple strong baselines on six robotic locomotion tasks for domain adaptation. ...
In the future, we plan to address these limitations, and test our method for transferring locomotion policies from simulation to a real Laikago robot. ...
arXiv:2101.06005v2
fatcat:3upuanwppffitenrenybagq5qa
Causal Curiosity: RL Agents Discovering Self-supervised Experiments for Causal Representation Learning
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
The learned behavior allows the agents to infer a binary quantized representation for the ground-truth causal factors in every environment. ...
We introduce causal curiosity, a novel intrinsic reward, and show that it allows our agents to learn optimal sequences of actions and discover causal factors in the dynamics of the environment. ...
Many thanks also to Alexander Neitz for sourcing of the CEM planning code. ...
arXiv:2010.03110v4
fatcat:fk4vsjibc5bi7iyhjk7wwe6izu
A Survey on Model-based Reinforcement Learning
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
For non-tabular environments, there is always a generalization error between the learned environment model and the real environment. ...
As such, it is of great importance to analyze the discrepancy between policy training in the environment model and that in the real environment, which in turn guides the algorithm design for better model ...
For high-dimensional state space such as images, representation learning that learns informative latent state or action representation will much benefit the environment model building so as to improve ...
arXiv:2206.09328v1
fatcat:cox3e76nhnewjpbdzmqgmgijnq
2019 Index IEEE Robotics and Automation Letters Vol. 4
2019
IEEE Robotics and Automation Letters
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Guadarrama-Olvera, J.R., +, LRA Oct. 2019 4418-4423 Robot Motion Planning in Learned Latent Spaces. ...
., +, LRA Oct. 2019 4298-4305 Robot Motion Planning in Learned Latent Spaces. Ichter, B., +, LRA July 2019 2407-2414 Safe Navigation With Human Instructions in Complex Scenes. ...
Permanent magnets Adaptive Dynamic Control for Magnetically Actuated Medical Robots. ...
doi:10.1109/lra.2019.2955867
fatcat:ckastwefh5chhamsravandtnx4
Unified Simulation, Perception, and Generation of Human Behavior
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We also discuss the lessons learned and our vision for what is next for human behavior modeling. ...
In this thesis, we take a holistic approach to human behavior modeling and tackle its three essential aspects -- simulation, perception, and generation. ...
For 3D Locomotion, Transform2Act creates a spider-like agent with long legs. ...
arXiv:2204.13678v1
fatcat:ls2kmgifbjbaffgwodgdiilstm
Table of Contents
2022
IEEE Robotics and Automation Letters
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Lueth Dynamics-Aware Metric Embedding: Metric Learning in a Latent Space for Visual Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...
Rimon Configuration Space Decomposition for Scalable Proxy Collision Checking in Robot Planning and Control . . . . . . . . . . . . ...
doi:10.1109/lra.2022.3165102
fatcat:enjzebowe5hn7hsfwklc7nieuy
« Previous
Showing results 1 — 15 out of 133 results