Fmb: a functional manipulation benchmark for generalizable robotic learning
In this paper, we propose a real-world benchmark for studying robotic learning in the context
of functional manipulation: a robot needs to accomplish complex long-horizon behaviors by …
of functional manipulation: a robot needs to accomplish complex long-horizon behaviors by …
Empower dexterous robotic hand for human-centric smart manufacturing: A perception and skill learning perspective
Recent rapid developments of dexterous robotic hands have greatly enhanced the
manipulative capabilities of robots, enabling them to perform industrial tasks in human-like …
manipulative capabilities of robots, enabling them to perform industrial tasks in human-like …
Continuous control with coarse-to-fine reinforcement learning
Despite recent advances in improving the sample-efficiency of reinforcement learning (RL)
algorithms, designing an RL algorithm that can be practically deployed in real-world …
algorithms, designing an RL algorithm that can be practically deployed in real-world …
Serl: A software suite for sample-efficient robotic reinforcement learning
In recent years, significant progress has been made in the field of robotic reinforcement
learning (RL), enabling methods that handle complex image observations, train in the real …
learning (RL), enabling methods that handle complex image observations, train in the real …
Learning prehensile dexterity by imitating and emulating state-only observations
When human acquire physical skills (eg, tool use) from experts, we tend to first learn from
merely observing the expert. But this is often insufficient. We then engage in practice, where …
merely observing the expert. But this is often insufficient. We then engage in practice, where …
Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning
Reinforcement learning (RL) holds great promise for enabling autonomous acquisition of
complex robotic manipulation skills, but realizing this potential in real-world settings has …
complex robotic manipulation skills, but realizing this potential in real-world settings has …
Diff-Transfer: Model-based Robotic Manipulation Skill Transfer via Differentiable Physics Simulation
The capability to transfer mastered skills to accomplish a range of similar yet novel tasks is
crucial for intelligent robots. In this work, we introduce $\textit {Diff-Transfer} $, a novel …
crucial for intelligent robots. In this work, we introduce $\textit {Diff-Transfer} $, a novel …
Hybrid Reinforcement Learning from Offline Observation Alone
We consider the hybrid reinforcement learning setting where the agent has access to both
offline data and online interactive access. While Reinforcement Learning (RL) research …
offline data and online interactive access. While Reinforcement Learning (RL) research …
Continuously Improving Mobile Manipulation with Autonomous Real-World RL
R Mendonca, E Panov, B Bucher, J Wang… - arXiv preprint arXiv …, 2024 - arxiv.org
We present a fully autonomous real-world RL framework for mobile manipulation that can
learn policies without extensive instrumentation or human supervision. This is enabled by 1) …
learn policies without extensive instrumentation or human supervision. This is enabled by 1) …
Predicting blind-use test (BUT) results from sensory testing using Bayesian bootstrapping
A Ping - Frontiers in Analytical Science, 2024 - frontiersin.org
Cosmetic researchers recruit consumers to evaluate new formulas as part of the product
development process. This screens out poorly performing formulas in favor of better ones for …
development process. This screens out poorly performing formulas in favor of better ones for …