This paper details a strategy of discriminating finger motions using surface electromyography (EMG) signals, which could be applied to teleoperating a dexterous robot hand or controlling the advanced multi-fingered myoelectric prosthesis for hand amputees. Finger motions discrimination is the key problem in this study. Thus the emphasis is put on myoelectric signal processing approaches in this paper. The EMG signal classification system was established based on the surface EMG signals from the subject's forearm. Four pairs of electrodes were attached on the subjects to acquire the signals during six types of finger motions, i.e. thumb extension, thumb flexion, index finger extension, index finger flexion, middle finger extension, and middle finger flexion. In order to distinguish these finger motions. A combination of autoregressive (AR) model and an Artificial Neural Network (ANN) was used in the system. The discrimination procedure consists of two steps. Firstly, the AR model is used to preprocess the surface EMG signals to reduce the scale of the data. These data will be imported into the myoelectric pattern classifier. Secondly the coefficients of AR model are imported into the ANN to identify the finger motions. The experimental results show that the discrimination system works with satisfaction.

A: A low power front-end ASIC, named WASA has been developed for the time projection chamber for the CEPC (Circular Electron Positron Collider) experiment. Power consumption becomes very critical and is addressed by using 65 nm CMOS process and circuit structure with simple analog circuits. Three prototype ASIC chips have been designed and fabricated, including the 5-channel analog front-end (AFE) chip, the SAR ADC chip and the mixed-signal chip with the AFE and the ADC together. Only the design and the test results of the AFE and the SAR ADC chips are reported in this paper. The power consumptions of the AFE and the SAR ADC core were measured to be 2.0 mW and 1.0 mW per channel respectively. The gain and the ENC noise of the AFE were 8.91 mV/fC and 644 electrons @ 10 pF input capacitance, including the contribution of 460 electrons from the external buffer for test purpose. The INL was less than 0.5% for a dynamic range of 145 fC. The maximum crosstalk between two adjacent channels was 0.39%. The SNDR and SFDR of ADC were measured to be 57.2 dB and 79.4 dBc at 50 MS/s for a 2.4 MHz input sine signal, corresponding to an ENOB of 9.2 bit. The total ionizing dose (TID) tests were also done for the AFE and the SAR ADC. No significant performance degradation was observed for the total dose up to 1 Mrad (Si). K: Analogue electronic circuits; Electronic detector readout concepts (gas, liquid); Frontend electronics for detector readout 1Corresponding author.

Bearing intelligent fault diagnosis has been researched comprehensively in recent years. However, the scarcity of labelled training samples and various working conditions seriously hinder the widespread application of deep learning based fault diagnosis methods. To address this problem, we propose a transfer multiscale adaptive convolutional neural network (TMACNN), which significantly enhances the performance of deep learning based methods on few-shot and cross-domain bearing fault diagnosis in terms of network architecture and transfer strategy. On the one hand, we design a novel multiscale adaptive convolutional neural network (MACNN) framework that effectively improves the feature extraction and generalization abilities for bearing fault diagnosis by introducing mega-scale convolutions and continuous stacked multiscale convolutions. On the other hand, we propose an innovative transfer strategy for the MACNN that freezes the six stacked multiscale convolutional feature extraction units and fine-tunes the mega-scale convolution unit and the classifier, which are more suitable for few-shot transfer learning. In experiments on the CWRU dataset and Paderborn dataset, our proposed TMACNN outperforms various advanced baseline models for few-shot and cross-domain bearing fault diagnosis.

No abstract