Abstract—In this paper we provide a method for identifying and temporally localizing tactile force actions from measured force signals. Our key idea is to use the continuous wavelet transform (CWT) with the Complex Morlet wavelet to transform force signals into feature vectors amenable to machine learning algorithms. Our method uses these feature vectors to train a classiﬁer that recognizes different actions. We demonstrate our approach in a system that records human activities with an instrumented set of tongs. Our system successfully identiﬁes a wide range of actions based on a small set of labeled examples.
Subramani, G., Rakita, D., Wang, H., Black, J., Zinn, M., & Gleicher, M. (2017, September). Recognizing actions during tactile manipulations through force sensing. In Intelligent Robots and Systems (IROS), 2017 IEEE/RSJ International Conference on (pp. 4386-4393). IEEE.
“© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.”
While this approach involves using the CWT and complex Morlet wavelet for action recognition using forces signals, it does not limit its application here. We believe we can apply this approach to many different types of signals as long as the phenomenon that needs to be recognized can be characterized by its inherent frequency content.