AutoDex: An Automated Real-World System for Dexterous Grasping Data Collection
AutoDex: An Automated Real-World System for Dexterous Grasping Data Collection
This paper introduces AutoDex, a system that automates the collection of real-world dexterous grasping data. By combining generated grasp candidates with multi-view object tracking and residual-torque collision monitoring, AutoDex plans, executes, labels, and resets grasps without continuous human supervision. The authors report collecting over 3,500 trials across 100 objects using Allegro and Inspire hands, reducing human effort from an estimated 9.3 hours to 0.2 hours for 500 trajectories compared to teleoperation. Retrieval from the collected library achieves a 76% success rate in new configurations, outperforming simulation-only screening at 34%. Code and data are available publicly.
The system automates data collection for dexterous grasping, reducing human effort significantly while validating physical execution against simulated candidates.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Data origin
- Robot trajectoriesOpen source S4
Reported numbers
grasp trials
500 attempts
Reported trials
View original evidence
500 grasp trials across 100 objects with Allegro and Inspire hands, recording both successes and failures together with synchronized visual observations and robot states. AutoDex reduces human effort by more than an order of magnitude, requiring an estimated 0.2 human-hours to collect 500 trajectories compared with 9.3 for teleoperation. As AutoDex collects trials, it builds a library of physically validated grasps f
Open source S4objects
100
View original evidence
500 grasp trials across 100 objects with Allegro and Inspire hands, recording both successes and failures together with synchronized visual observations and robot states. AutoDex reduces human effort by more than an order of magnitude, requiring an estimated 0.2 human-hours to collect 500 trajectories compared with 9.3 for teleoperation. As AutoDex collects trials, it builds a library of physically validated grasps f
Open source S4
- code: Available Artifact link Open source S5
- dataset: not_reported
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
nuous human operation. Given generated grasp candidates, AutoDex plans and executes grasps, labels their outcomes, and resets the object to continue collection across stable poses. Multi-view object tracking supports labeling under hand-object occlusion, while residual-torque collision monitoring and pose-directed reset enable the loop to operate without continuous human supervision. Using AutoDex, we collect over 3,
Open source S3
500 grasp trials across 100 objects with Allegro and Inspire hands, recording both successes and failures together with synchronized visual observations and robot states. AutoDex reduces human effort by more than an order of magnitude, requiring an estimated 0.2 human-hours to collect 500 trajectories compared with 9.3 for teleoperation. As AutoDex collects trials, it builds a library of physically validated grasps f
Open source S4
or subsequent execution. In new object placements and obstacle configurations, retrieval from this library achieves 76% success, compared with 34% for candidates screened only in simulation. Code and data are available at https://snuvclab.github.io/AutoDex/.
Open source S5
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Match the buyer’s robot embodiment, action space, sensors and timestamp requirements before planning collection.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org