Embedding machine learning methods into the data decoding units may enable the extraction of complex information making the tactile sensing systems intelligent. This paper presents and compares the implementations of a convolutional neural network model for tactile data decoding on various hardware platforms. Experimental results show comparable classification accuracy of 90.88% for Model 3, overcoming similar state-of-the-art solutions in terms of time inference. The proposed implementation achieves a time inference of 1.2 ms while consuming around 900 $\mathsf{\mu }$ J. Such an embedded implementation of intelligent tactile data decoding algorithms enables tactile sensing systems in different application domains such as robotics and prosthetic devices.

Recurrent Neural Networks (RNNs) are mainly designed to deal with sequence prediction problems and they show their effectiveness in processing data originally represented as time series. This paper investigates the time series characteristics of RNNs to classify touch modalities represented as spatio temporal 3D tensor data. Different approaches are followed in order to propose efficient RNN models aimed at tactile data classification. The main idea is to capture long-term dependence from data that can be used to deal with long sequences represented by employing Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) architectures. Moreover, a case specific approach to dataset organization of the 3D tensor data is presented. The target is to provide efficient hardware-friendly touch modality classification approaches suitable for embedded applications. To this end, the proposed work achieves effective performance in terms of hardware complexity by reducing the FLOPS by 99.98% and the memory storage by 98.34%, with respect to the state-of-art solutions on the same benchmark dataset. This directly affects the time latency and energy consumption of the embedded hardware. Besides, the implemented models shows a classification accuracy higher than those state-of-art solutions. Results demonstrate that the proposed computing architecture is scalable showing acceptable complexity when the system is scaled up in terms of input matrix size and number of classes to be recognized.

Embedding Machine Learning enables integrating intelligence in recent application domains such as Internet of Things, portable healthcare systems, and wearable devices. This paper presents an assessment of approximate computing methods at algorithmic, architecture, and circuit levels and draws perspectives for further developments and applications. The main goal is to investigate how approximate computing may reduce the complexity and enable the feasibility of embedded Machine Learning (ML) systems. Though ML is a powerful paradigm for applications in the perceptual domain (i.e. vision, touch, hearing, etc.), their computational complexity is very high and consequently real time operation and ultra-low power are still very challenging objectives. On the other hand, approximate computing has emerged as an effective solution to reduce hardware complexity, time latency and to increase energy efficiency.

Machine learning (ML) has increasingly been recently employed to provide solutions for difficult tasks, such as image and speech recognition, and tactile data processing achieving a near human decision accuracy. However, the efficient hardware implementation of ML algorithms in particular for real time applications is still a challenge. This paper presents the hardware architectures and implementation of a real time ML method based on tensorial kernel approach dealing with multidimensional input tensors. Two different hardware architectures are proposed and assessed. Results demonstrate the feasibility of the proposed implementations for real time classification. The proposed parallel architecture achieves a peak performance of 302 G-ops while consuming 1.14 W for the Virtex-7 XC7VX980T FPGA device overcoming state of the art solutions.

In this paper, we propose a novel design approach for the interface electronics of piezoelectric polymer tactile sensing systems. An interface electronics prototype based on commercial off-the-shelf (COTS) components having the same operating principle as the proposed approach has been fabricated. The system has been validated by using an experimental setup where electrical and electromechanical characterization are carried out. The interface electronics measures charge of about 0.15pC for applied forces as low as 12.5mN at the working frequency of 27Hz. The experimental average sensitivity of the system is 0.6pC/kPa in the frequency range from 10Hz to 250Hz. Long term goal is to miniaturize the interface electronics for the development of embedded tactile systems in prosthetic and robotic applications.