The GummiArm Project: A Replicable and Variable-Stiffness Robot Arm for Experiments on Embodied AI
Bibliographic record
Abstract
Robots used in research on Embodied AI often need to physically explore the world, to fail in the process, and to develop from such experiences. Most research robots are unfortunately too stiff to safely absorb impacts, too expensive to repair if broken repeatedly, and are never operated without the red kill-switch prominently displayed. The GummiArm Project was intended to be an open-source "soft" robot arm with human-inspired tendon actuation, sufficient dexterity for simple manipulation tasks, and with an eye on enabling easy replication of robotics experiments. The arm offers variable-stiffness and damped actuation, which lowers the potential for damage, and which enables new research opportunities in Embodied AI. The arm structure is printable on hobby-grade 3D printers for ease of manufacture, exploits stretchable composite tendons for robustness to impacts, and has a repair-cycle of minutes when something does break. The material cost of the arm is less than $6000, while the full set of structural parts, the ones most likely to break, can be printed with less than $20 worth of plastic filament. All this promotes a concurrent approach to the design of "brain" and "body," and can help increase productivity and reproducibility in Embodied AI research. In this work we describe the motivation for, and the development and application of, this 6 year project.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".