Low-cost imprecise robots achieve 80-90% success on six fine bimanual manipulation tasks using imitation learning with a new Action Chunking with Transformers algorithm trained on only 10 minutes of demonstrations.
Holo-dex: Teaching dexterity with immersive mixed reality
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.RO 5roles
baseline 1polarities
baseline 1representative citing papers
DexTwist detects tripod pinches, estimates the intended screw axis and twist magnitude, then applies real-time joint refinement to track turning progress while stabilizing the robot's tripod geometry.
Play2Perfect uses task-agnostic RL play pretraining on diverse objects to build reusable manipulation priors, then fine-tunes for assembly, yielding 33x sample efficiency gains and 60% success on 0.5mm-clearance insertions in sim-to-real transfer.
A robot generalizes one demonstration of placing a single brick to build walls of arbitrary length and layout by approximating motions as screw sequences and using ScLERP/RMRC for planning.
VR-DAgger is a VR-centered human-in-the-loop framework that applies MC dropout uncertainty to select and correct failure segments in diffusion policy rollouts, yielding up to 23 percentage point gains over behavioral cloning and 40% lower per-sample collection time on three dexterous tasks.
citing papers explorer
-
Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
Low-cost imprecise robots achieve 80-90% success on six fine bimanual manipulation tasks using imitation learning with a new Action Chunking with Transformers algorithm trained on only 10 minutes of demonstrations.
-
DexTwist: Dexterous Hand Retargeting for Twist Motion via Mixed Reality-based Teleoperation
DexTwist detects tripod pinches, estimates the intended screw axis and twist magnitude, then applies real-time joint refinement to track turning progress while stabilizing the robot's tripod geometry.
-
Play2Perfect: What Matters in Dexterous Play Pretraining for Precise Assembly?
Play2Perfect uses task-agnostic RL play pretraining on diverse objects to build reusable manipulation priors, then fine-tunes for assembly, yielding 33x sample efficiency gains and 60% success on 0.5mm-clearance insertions in sim-to-real transfer.
-
Manipulation Planning for Construction Activities with Repetitive Tasks
A robot generalizes one demonstration of placing a single brick to build walls of arbitrary length and layout by approximating motions as screw sequences and using ScLERP/RMRC for planning.
-
VR-DAgger: Immersive VR for Dexterous Data Collection and Uncertainty-Guided On-Policy Correction
VR-DAgger is a VR-centered human-in-the-loop framework that applies MC dropout uncertainty to select and correct failure segments in diffusion policy rollouts, yielding up to 23 percentage point gains over behavioral cloning and 40% lower per-sample collection time on three dexterous tasks.