Improving Haptic Adjective Recognition with Unsupervised Feature Learning

Richardson, Benjamin A.; Kuchenbecker, Katherine J.

doi:10.1109/icra.2019.8793544

Cited by 18 publications

(13 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another type of feature-extracting algorithm, unsupervised dictionary learning, has been successfully used to extract features from raw tactile data for multiple haptic classification tasks (Madry et al, 2014). We additionally demonstrated the viability of these methods in our previous work (Richardson and Kuchenbecker, 2019), in which the learned features greatly outperformed hand-crafted features in the binary adjective classification tasks presented by Chu et al (2015). We use the same unsupervised dictionary feature-extraction algorithms in this work.…”

Section: Introductionmentioning

confidence: 91%

“…Although our primary contribution pertains to the method mapping the learned features to the labels, this section describes the process used to extract the features from the raw data, shown in the first two columns of Figure 5. Specifically, we used unsupervised dictionary learning, which has proven far more effective than using hand-crafted features (Richardson and Kuchenbecker, 2019).…”

Section: Unsupervised Feature Learningmentioning

confidence: 99%

“…The specific parameter values for K-SVD and ST-HMP were selected because they previously showed good performance in a binary adjective classification task using the same raw tactile data (Richardson and Kuchenbecker, 2019). Additionally, because unsupervised feature learning is not the primary focus of this work, full parameter optimization was not a priority.…”

Section: Feature Extraction Proceduresmentioning

confidence: 99%

“…our previous work (Richardson and Kuchenbecker, 2019), we measure the contribution of different exploratory actions and haptic sensor modalities to the learning and prediction of the adjectives. The secondary goal is to analyze the labels gathered by Chu et al (2015) and provide insight into the antonymous relationships between common haptic adjectives.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning to Predict Perceptual Distributions of Haptic Adjectives

Richardson

Kuchenbecker

2020

Front. Neurorobot.

Self Cite

View full text Add to dashboard Cite

When humans touch an object with their fingertips, they can immediately describe its tactile properties using haptic adjectives, such as hardness and roughness; however, human perception is subjective and noisy, with significant variation across individuals and interactions. Recent research has worked to provide robots with similar haptic intelligence but was focused on identifying binary haptic adjectives, ignoring both attribute intensity and perceptual variability. Combining ordinal haptic adjective labels gathered from human subjects for a set of 60 objects with features automatically extracted from raw multi-modal tactile data collected by a robot repeatedly touching the same objects, we designed a machine-learning method that incorporates partial knowledge of the distribution of object labels into training; then, from a single interaction, it predicts a probability distribution over the set of ordinal labels. In addition to analyzing the collected labels (10 basic haptic adjectives) and demonstrating the quality of our method's predictions, we hold out specific features to determine the influence of individual sensor modalities on the predictive performance for each adjective. Our results demonstrate the feasibility of modeling both the intensity and the variation of haptic perception, two crucial yet previously neglected components of human haptic perception.

show abstract

Section: Introductionmentioning

confidence: 91%

Section: Unsupervised Feature Learningmentioning

confidence: 99%

Section: Feature Extraction Proceduresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning to Predict Perceptual Distributions of Haptic Adjectives

Richardson

Kuchenbecker

2020

Front. Neurorobot.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Perhaps most similar to our application, [27] used tactile sensors to classify interactions between a robot hand and its environment. There are similar works in this area (e.g., [28,29,30,31,32]) that propose the use of ML to relate complex tactile data to object and material classes. In this paper, we aim to use a similar approach in that we propose the novel application of ML and proprioceptive sensing to classify the excavation media in industrial excavation activities.…”

Section: Machine Learning For Classificationmentioning

confidence: 99%

What lies beneath: Material classification for autonomous excavators using proprioceptive force sensing and machine learning

Fernando

Marshall

2020

Automation in Construction

View full text Add to dashboard Cite

The ability of robotic excavators to acquire meaningful knowledge about materials during digging can augment their autonomous functionality, as well as optimize downstream operations in construction and mining. Some material properties, such as rock sizes, can be determined visually, but these methods cannot see what lies beneath. In this work, a classification methodology that utilizes only proprioceptive force data acquired from an autonomous digging system and machine learning algorithms is proposed for excavation material identification. The consistent performance synonymous with autonomous digging systems allows for the use of basic features extracted from the force data for classification. A proof of concept of this novel approach to excavation material classification is demonstrated through a binary classification of rock and gravel materials. Force data were obtained from full-scale autonomous loading trials with a 14-tonne capacity load-haul-dump machine at a mining and construction test facility. Preliminary results achieved a classification accuracy of 90 %.

show abstract