Dual-Arm Peg-in-Hole Assembly Using DNN with Double Force/Torque Sensor
Bibliographic record
Abstract
Assembly tasks executed by a robot have been studied broadly. Robot assembly applications in industry are achievable by a well-structured environment, where the parts to be assembled are located in the working space by fixtures. Recent changes in manufacturing requirements, due to unpredictable demanded products, push the factories to seek new smart solutions that can autonomously recover from failure conditions. In this way, new dual arm robot systems have been studied to design and explore applications based on its dexterity. It promises the possibility to get rid of fixtures in assembly tasks, but using less fixtures increases the uncertainty on the location of the components in the working space. It also increases the possibility of collisions during the assembly sequence. Under these considerations, adding perception such as force/torque sensors have been done to produce useful data to perform control actions. Unfortunately, the interaction forces between mating parts produced non-linear behavior. Consequently, machine learning algorithms have been considered an alternative tool to avoid the non-linearity. In this work we introduce an assembly strategy for an industrial dual arm robot based on the combination of a discrete event controller and Deep Neural Networks (DNN) to solve the peg-in-hole assembly. Our results show the difference between the use of DNN with one and with two force/torque sensors during the assembly task and demonstrate a 30% increase in the assembly success ratio when using a double force/torque sensor.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".