A human in-hand motion (HIM) recognition system based on multi-modal perception information fusion is proposed in this paper, which can observe the state information between the object and the hand by using customized ten kinds of HIM manipulation in order to recognize the complex HIMs. First, combined with the characteristics of HIM capture, ten kinds of HIM sets are designed, and finger trajectory, contact force and electromyographic signal data are acquired synchronously through the multi-modal data acquisition platform; second, motion segmentation is realized through the threshold segmentation method, the multi-modal signal preprocessing is realized by Empirical Mode Decomposition (EMD), and multi-modal signal feature extraction is realized by Maximum Lyapunov Exponent (MLE); then, a detailed non-linear data analysis is carried out. A detailed analysis and discussion are presented from the results of the Random Forest (RF) recognizing HIMs, the comparison results of motion recognition rates of different subjects, the comparison results of motion recognition rates of different perceptrons, and the comparison results of the motion recognition rates of different machine learning methods. The experimental results show that the multi-modal perception information based HIM recognition system proposed in this paper can effectively recognize ten different HIMs, with an accuracy rate of 93.72%.

The bipedal robot should be able to maintain standing balance even in the presence of disturbing forces. The control schemes of bipedal robot are conventionally developed based on system models or fixed torque-ankle states, which often lack robustness. In this paper, a variable impedance control based on artificial muscle activation is investigated for bipedal robotic standing balance to address this limitation. The robustness was improved by applying the artificial muscle activation model to adjust the impedance parameters. In particular, an ankle variable impedance model was used to obtain the antidisturbance torque which combined with the ankle dynamic torque to estimate the desired ankle torque for robotic standing balance. The simulation and prototype experimentation results demonstrate that the control method improves the robustness of bipedal robotic standing balance control.

Surface electromyography (SEMG) signals have emerged as a natural and intuitive control method, offering significant advantages for the operation of bionic hands. This technology has been widely applied in various fields, including medical rehabilitation and human-computer interaction. This paper proposes an innovative human in-hand motion (HIM) recognition system that leverages SEMG signals to identify 10 self-defined in-hand operations.Firstly, acknowledging that conventional gestures and postures are insufficient to capture the complexity of human hand operations, we designed 10 HIM sets,including classical motions such as transfer, translation, and rotation. The original SEMG signals were collected using high-precision sensors.Subsequently, the collected SEMG signals were denoised using the ensemble empirical mode decomposition (EEMD) algorithm, effectively mitigating noise interference and enhancing signal quality. Feature extraction was then performed using the multivariate autoregressive (MVAR) method, which provides detailed nonlinear data analysis and captures the intricate temporal dynamics of the SEMG signals. Finally, the gradient boosting decision tree (GBDT) algorithm was employed to analyze the extracted features comprehensively. The results were discussed from three perspectives: the accuracy of human in-hand motion recognition, the inter-subject variability in motion recognition rates, and the comparative performance of different machine learning methods.The experimental results demonstrate that the proposed SEMG-based HIM recognition system achieves a high accuracy of 92.78% in identifying the 10 different HIMs. This study highlights the potential of integrating advanced signal processing and machine learning techniques to enhance the functionality and precision of bionic hand control systems.

The multi-fingered dexterous robotic hand is increasingly used to achieve more complex and sophisticated human-like manipulation tasks on various occasions. This paper proposes a hybrid Surface Electromyography (SEMG) and Kinect-based human in-hand motion (HIM) capture system architecture for recognizing complex motions of the humans by observing the state information between an object and the human hand, then transferring the manipulation skills into bionic multi-fingered robotic hand realizing dexterous in-hand manipulation. First, an Adaptive Directed Acyclic Graph (ADAG) algorithm for recognizing HIMs is proposed and optimized based on the comparison of multi-class support vector machines; second, ten representative complex in-hand motions are demonstrated by ten subjects, and SEMG and Kinect signals are obtained based on a multi-modal data acquisition platform; then, combined with the proposed algorithm framework, a series of data preprocessing algorithms are realized. There is statistical symmetry in similar types of SEMG signals and images, and asymmetry in different types of SEMG signals and images. A detailed analysis and an in-depth discussion are given from the results of the ADAG recognizing HIMs, motion recognition rates of different perceptrons, motion recognition rates of different subjects, motion recognition rates of different multi-class SVM methods, and motion recognition rates of different machine learning methods. The results of this experiment confirm the feasibility of the proposed method, with a recognition rate of 95.10%.

With the continuous development of wireless technology, automatic modulation recognition plays an increasingly prominent role in military and civilian fields. However, the complex communication environment and diversified strategy bring challenges to modulation signal recognition. Therefore, an automatic modulation signal recognition technique combining truncated migration processing and convolutional neural network is proposed. Then multi-task learning is used to optimize and distinguish easily confused modulated signals. Three datasets, RadioML2016.10A, RadioML2016.10B and RadioML2016.04C are used for experiments. The results show that the modulation signal automatic recognition method proposed in this paper has good recognition accuracy. When the signal-to-noise ratio was 14dB, it reached a maximum of 95.46%. In addition, the Floating Point Operations of the proposed method were 1.71G, the number of parameters was 25636712, and the operation time was 228s, which showed that the method was light in weight and high in calculation efficiency. The optimized recognition technology can effectively distinguish three groups of easily confused signals, and the recognition rate is as high as 100%, and the minimum is not less than 90%. The proposed modulated signal automatic recognition technology has achieved remarkable results in improving the recognition accuracy and processing efficiency, which provides a strong technical support for the development of wireless communication technology, and also provides a new tool for researchers and engineers in related fields.