Transfer learning for vision-based tactile sensing

Due to the complexity of modeling the elastic properties of materials, the use of machine learning algorithms is continuously increasing for tactile sensing applications. Recent advances in deep neural networks applied to computer vision make vision-based tactile sensors very appealing for their high-resolution and low cost. A soft optical tactile sensor that is scalable to large surfaces with arbitrary shape is discussed in this paper. A supervised learning algorithm trains a model that is able to reconstruct the normal force distribution on the sensor’s surface, purely from the images recorded by an internal camera. In order to reduce the training times and the need for large datasets, a calibration procedure is proposed to transfer the acquired knowledge across multiple sensors while maintaining satisfactory performance.

Authors

• Carmelo Sferrazza
• Raffaello D'Andrea

Keywords

• Training
• Surface reconstruction
• Computational modeling
• Force
• Transfer learning
• Supervised learning
• Tactile sensors
• Computer architecture
• Skin
• Calibration
• Tactile Sensor
• High-resolution
• Normal Distribution
• Neural Network
• Material Properties
• Learning Algorithms
• Deep Neural Network
• Training Time
• Calibration Procedure
• Normal Force
• Force Distribution
• Arbitrary Shape
• Elastic Material Properties
• Training Data
• Level Characteristics
• Feed-forward Network
• Viewing Angle
• Optical Flow
• Image Position
• Automatic Procedure
• Use Of Cameras
• Distribution Of Markers
• Types Of Forces
• Fisheye Lens
• Different Types Of Objects
• Silicone Gel
• Label Vector
• Changes In Light Intensity
• Sensor Design