WMSDsNet: A Deep Learning Framework for Real-Time Ergonomic Risk Prediction in Human-Robot Collaboration in Disassembly
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Disassembly tasks are increasingly vital for sustainable manufacturing and the circular economy, as they facilitate component recovery and waste reduction. While humanrobot collaboration (HRC) is often promoted for reducing physical ergonomic challenges compared to tasks performed entirely by humans, studies have largely overlooked the unique ergonomic issues inherent to HRC. These environments can still present challenges that, if neglected, can contribute to work-related musculoskeletal disorders (WMSDs). This study introduces WMSDsNet, a dual-head deep-learning framework that automates ergonomic risk assessment by simultaneously classifying subtasks and predicting ergonomic risks, offering realtime, cumulative risk evaluation using wearable sensor data. Unlike traditional methods, which rely on subjective and timeintensive manual observations, or previous works that primarily focus on posture-based risk assessments to recognize awkward postures for immediate alerts or feedback, WMSDsNet detects changes in posture over a specific period of time. Based on this information, the frequency and duration of awkward postures can be analyzed to understand their cumulative effects on ergonomic risks. We analyzed the task of disassembling a Programmable Logic Controller (PLC) and selected specific subtasks to be performed by human operators in collaboration with the robot, including unscrewing components, detaching cables, sorting components, and changing the cobot's tools. Data was collected in numerical form using wearable sensors, enabling the framework to evaluate risk levels and predict ergonomic risks with over 90% accuracy in task classification and risk assessment. By providing real-time ergonomic assessments, this framework supports proactive interventions, offering a significant advancement in ergonomic evaluation for industrial HRC environments.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it