Tactile Texture Recognition On Uneven Surfaces Using Self-Attention Based Neural Networks
Bibliographic record
Abstract
Tactile texture recognition is a cornerstone of robotic perception, enabling systems to discern and interact with their environments through tactile feedback. In this work, we introduce a comprehensive methodology for texture classification using time-series data acquired from advanced tactile sensors. Our dataset comprises MARG data captured from a variety of distinct textures on uneven surfaces, which poses significant challenges due to variations in material properties and surface irregularities. To address these challenges, we developed a structured pipeline that begins with preprocessing raw sensor signals—implementing noise reduction, normalization, and segmentation—to enhance subsequent feature extraction. While approaches such as 1D-Convolutional Neural Networks (1D-CNNs), Long Short-Term Memory networks (LSTMs), and hybrid CNN-BiLSTM architectures have been recently explored in this domain, our study proposes an innovative informer-based neural network that leverages transformer and attention mechanisms. This architecture is designed to capture both temporal dependencies and spatial patterns inherent in the tactile data more effectively. A rigorous experimental setup employing cross-validation was used to assess the model’s performance and its ability to generalize to unseen surfaces. Experimental results demonstrate that the proposed informer-based model outperforms the existing approaches, with hybrid architectures yielding the highest accuracy. Overall, our contribution advances the field of tactile perception by providing a robust and scalable framework for texture classification, which is pivotal for enhancing the adaptability and precision of robotic systems. Our best model has reached 93.88% accuracy which is the highest accuracy in the literature on uneven surfaces.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 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".