Abstract
Tactile sensing has recently been used in robotics for object identication, grasping,
and material identication. Although human tactile sensing is multimodal,
existing material recognition approaches use vibration information only. Moreover,
material identication through tactile sensing can be solved as an continuous
process, yet state of the art approaches use a batch approach where readings
are taken for at least one second. This work proposes a recursive multimodal
(vibration and thermal) tactile material identication approach. Using the frequency
response of the vibration induced by the material and a set of thermal
features, we show that it is possible to accurately identify materials in less than
half a second. We conducted an exhaustive comparison of our approach with
commonly used vibration descriptors and machine learning algorithms for material
identication such as k-Nearest Neighbour, Articial Neural Network and
Support Vector Machines. Experimental results show that our approach identi
es materials faster than existing techniques and increase the classication
accuracy when multiple sensor modalities are used.
and material identication. Although human tactile sensing is multimodal,
existing material recognition approaches use vibration information only. Moreover,
material identication through tactile sensing can be solved as an continuous
process, yet state of the art approaches use a batch approach where readings
are taken for at least one second. This work proposes a recursive multimodal
(vibration and thermal) tactile material identication approach. Using the frequency
response of the vibration induced by the material and a set of thermal
features, we show that it is possible to accurately identify materials in less than
half a second. We conducted an exhaustive comparison of our approach with
commonly used vibration descriptors and machine learning algorithms for material
identication such as k-Nearest Neighbour, Articial Neural Network and
Support Vector Machines. Experimental results show that our approach identi
es materials faster than existing techniques and increase the classication
accuracy when multiple sensor modalities are used.
Original language | English |
---|---|
Pages (from-to) | 130-139 |
Number of pages | 10 |
Journal | Robotics and Autonomous Systems |
Volume | 106 |
Early online date | 14 May 2018 |
DOIs | |
Publication status | Published (in print/issue) - 31 Aug 2018 |
Keywords
- Material classification
- tactile sensing
- supervised learning