ABSTRACT
This chapter presents a survey of the existing algorithms and tasks applied for tactile data processing. Data processing algorithms presented in the literature could be divided into two categories: preprocessing and classification/regression. Preprocessing algorithms involve feature extraction and dimensionality reduction, while classification and regression algorithms are grouped into machine- and deep learning algorithms. Machine learning algorithms are an efficient solution for processing tactile data in various applications. Deep learning is a kind of artificial neural network where the network has more hidden layers inside it. The chapter provides guidelines for selecting appropriate hardware platforms for the algorithm’s implementation. The algorithms are compared in terms of computational complexity and hardware implementation requirements. The chapter introduces the experimental setup used for touch modality classification in terms of the used dataset, preprocessing techniques and the performance of the algorithms in terms of classification accuracy.
