Human-Inspired Neurorobotic System for Classifying Surface Textures by Touch

Friedl, Ken E.; Voelker, Aaron R.; Friedl, Ken; Voelker, Aaron; Peer, Angelika; Eliasmith, Chris

Abstract

© 2016 IEEE. Giving robots the ability to classify surface textures requires appropriate sensors and algorithms. Inspired by the biology of human tactile perception, we implement a neurorobotic texture classifier with a recurrent spiking neural network, using a novel semisupervised approach for classifying dynamic stimuli. Input to the network is supplied by accelerometers mounted on a robotic arm. The sensor data are encoded by a heterogeneous population of neurons, modeled to match the spiking activity of mechanoreceptor cells. This activity is convolved by a hidden layer using bandpass filters to extract nonlinear frequency information from the spike trains. The resulting high-dimensional feature representation is then continuously classified using a neurally implemented support vector machine. We demonstrate that our system classifies 18 metal surface textures scanned in two opposite directions at a constant velocity. We also demonstrate that our approach significantly improves upon a baseline model that does not use the described feature extraction. This method can be performed in real-time using neuromorphic hardware, and can be extended to other applications that process dynamic stimuli online.

Citation

Voelker, A. R., Friedl, K. E., Friedl, K., Voelker, A., Peer, A., & Eliasmith, C. (2016). Human-Inspired Neurorobotic System for Classifying Surface Textures by Touch. IEEE Robotics and Automation Letters, 1(1), 516-523. https://doi.org/10.1109/LRA.2016.2517213

• Journal Article Type: Article
• Acceptance Date: Jan 10, 2016
• Online Publication Date: Jan 12, 2016
• Publication Date: Jan 1, 2016
• Publicly Available Date: Jun 7, 2019
• Journal: IEEE Robotics and Automation Letters
• Print ISSN: 2377-3766
• Electronic ISSN: 2377-3766
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 1
• Issue: 1
• Pages: 516-523
• DOI: https://doi.org/10.1109/LRA.2016.2517213
• Keywords: neurorobotics, biologically-inspired robots, force and tactile sensing
• Public URL: https://uwe-repository.worktribe.com/output/914700
• Publisher URL: http://dx.doi.org/10.1109/LRA.2016.2517213