Zobrazeno 1 - 10
of 18
pro vyhledávání: '"Vecerik, Mel"'
Many robot manipulation tasks can be framed as geometric reasoning tasks, where an agent must be able to precisely manipulate an object into a position that satisfies the task from a set of initial conditions. Often, task success is defined based on
Externí odkaz:
http://arxiv.org/abs/2404.13478
Autor:
Vecerik, Mel, Doersch, Carl, Yang, Yi, Davchev, Todor, Aytar, Yusuf, Zhou, Guangyao, Hadsell, Raia, Agapito, Lourdes, Scholz, Jon
For robots to be useful outside labs and specialized factories we need a way to teach them new useful behaviors quickly. Current approaches lack either the generality to onboard new tasks without task-specific engineering, or else lack the data-effic
Externí odkaz:
http://arxiv.org/abs/2308.15975
Autor:
Doersch, Carl, Yang, Yi, Vecerik, Mel, Gokay, Dilara, Gupta, Ankush, Aytar, Yusuf, Carreira, Joao, Zisserman, Andrew
We present a novel model for Tracking Any Point (TAP) that effectively tracks any queried point on any physical surface throughout a video sequence. Our approach employs two stages: (1) a matching stage, which independently locates a suitable candida
Externí odkaz:
http://arxiv.org/abs/2306.08637
Dense object tracking, the ability to localize specific object points with pixel-level accuracy, is an important computer vision task with numerous downstream applications in robotics. Existing approaches either compute dense keypoint embeddings in a
Externí odkaz:
http://arxiv.org/abs/2112.04910
Autor:
Luo, Jianlan, Sushkov, Oleg, Pevceviciute, Rugile, Lian, Wenzhao, Su, Chang, Vecerik, Mel, Ye, Ning, Schaal, Stefan, Scholz, Jon
Over the past several years there has been a considerable research investment into learning-based approaches to industrial assembly, but despite significant progress these techniques have yet to be adopted by industry. We argue that it is the prohibi
Externí odkaz:
http://arxiv.org/abs/2103.11512
Autor:
Vecerik, Mel, Regli, Jean-Baptiste, Sushkov, Oleg, Barker, David, Pevceviciute, Rugile, Rothörl, Thomas, Schuster, Christopher, Hadsell, Raia, Agapito, Lourdes, Scholz, Jonathan
A robot's ability to act is fundamentally constrained by what it can perceive. Many existing approaches to visual representation learning utilize general-purpose training criteria, e.g. image reconstruction, smoothness in latent space, or usefulness
Externí odkaz:
http://arxiv.org/abs/2009.14711
Model-free reinforcement learning algorithms such as Deep Deterministic Policy Gradient (DDPG) often require additional exploration strategies, especially if the actor is of deterministic nature. This work evaluates the use of model-based trajectory
Externí odkaz:
http://arxiv.org/abs/1911.06833
Autor:
Cabi, Serkan, Colmenarejo, Sergio Gómez, Novikov, Alexander, Konyushkova, Ksenia, Reed, Scott, Jeong, Rae, Zolna, Konrad, Aytar, Yusuf, Budden, David, Vecerik, Mel, Sushkov, Oleg, Barker, David, Scholz, Jonathan, Denil, Misha, de Freitas, Nando, Wang, Ziyu
Publikováno v:
Robotics: Science and Systems Conference 2020
We present a framework for data-driven robotics that makes use of a large dataset of recorded robot experience and scales to several tasks using learned reward functions. We show how to apply this framework to accomplish three different object manipu
Externí odkaz:
http://arxiv.org/abs/1909.12200
We propose Generative Predecessor Models for Imitation Learning (GPRIL), a novel imitation learning algorithm that matches the state-action distribution to the distribution observed in expert demonstrations, using generative models to reason probabil
Externí odkaz:
http://arxiv.org/abs/1904.01139
Insertion is a challenging haptic and visual control problem with significant practical value for manufacturing. Existing approaches in the model-based robotics community can be highly effective when task geometry is known, but are complex and cumber
Externí odkaz:
http://arxiv.org/abs/1810.01531