The motor imagery brain-computer interface (MI-BCI) based on electroencephalography (EEG) is a widely used human-machine interface paradigm. However, due to the non-stationarity and individual differences among subjects in EEG signals, the decoding accuracy is limited, affecting the application of the MI-BCI. In this paper, we propose the EISATC-Fusion model for MI EEG decoding, consisting of inception block, multi-head self-attention (MSA), temporal convolutional network (TCN), and layer fusion. Specifically, we design a DS Inception block to extract multi-scale frequency band information. And design a new cnnCosMSA module based on CNN and cos attention to solve the attention collapse and improve the interpretability of the model. The TCN module is improved by the depthwise separable convolution to reduces the parameters of the model. The layer fusion consists of feature fusion and decision fusion, fully utilizing the features output by the model and enhances the robustness of the model. We improve the two-stage training strategy for model training. Early stopping is used to prevent model overfitting, and the accuracy and loss of the validation set are used as indicators for early stopping. The proposed model achieves within-subject classification accuracies of 84.57% and 87.58% on BCI Competition IV Datasets 2a and 2b, respectively. And the model achieves cross-subject classification accuracies of 67.42% and 71.23% (by transfer learning) when training the model with two sessions and one session of Dataset 2a, respectively. The interpretability of the model is demonstrated through weight visualization method.

To improve the recognition rate of lower limb actions based on surface electromyography (sEMG), an effective weighted feature method is proposed, and an improved genetic algorithm support vector machine (IGA-SVM) is designed in this paper. First, for the problem of high feature redundancy and low discrimination in the surface electromyography feature extraction process, the weighted feature method is proposed based on the correlation between muscles and actions. Second, to solve the problem of the genetic algorithm selection operator easily falling into a local optimum solution, the improved genetic algorithm-support vector machine is designed by championship with sorting method. Finally, the proposed method is used to recognize six types of lower limb actions designed, and the average recognition rate reaches 94.75%. Experimental results indicate that the proposed method has definite potentiality in lower limb action recognition.

The inability of new users to adapt quickly to the surface electromyography (sEMG) interface has greatly hindered the development of sEMG in the field of rehabilitation. This is due mainly to the large differences in sEMG signals produced by muscles when different people perform the same motion. To address this issue, a multi-user sEMG framework is proposed, using discriminative canonical correlation analysis and adaptive dimensionality reduction (ADR). The interface projects the feature sets for training users and new users into a low-dimensional uniform style space, overcoming the problem of individual differences in sEMG. The ADR method removes the redundant information in sEMG features and improves the accuracy of system motion recognition. The presented framework was validated on eight subjects with intact limbs, with an average recognition accuracy of 92.23% in 12 categories of upper-limb movements. In rehabilitation laboratory experiments, the average recognition rate reached 90.52%. The experimental results suggest that the framework offers a good solution to enable new rehabilitation users to adapt quickly to the sEMG interface.

Joint angle estimation during continuous motion based on surface electromyography (sEMG) plays a crucial role in the exoskeleton robot applications. To estimate joint angles using sEMG signals, a Hill-based muscle model (HMM) can be incorporated into forward dynamics algorithms. Nevertheless, obtaining muscle parameters remains challenging, and the strongly nonlinear ’open-loop’ system can lead to significant cumulative errors. To address these issues, this article applies a novel ’closed-loop’ sEMG-driven state space model to estimate the elbow joint angle in the human upper limb during continuous motion. The forward kinematics of the human joints are integrated with the HMM, and a torque-based musculoskeletal parameter estimation (TBMPE) is introduced to derive more accurate state transition function by utilizing angular velocity as a cost function. The observation function are constructed by utilizing sEMG features to design the ‘closed-loop’ sEMG-driven state space model. Subsequently, the particle filter (PF) algorithm is introduced to stably estimate the joint angles within the strongly nonlinear model. Experiments were conducted based on the elbow joints of six healthy subjects. The experimental results indicate that the proposed new model has an average root mean square error of 0.09 for cases with and without loads. The estimation results are better than those of similar algorithms, proving the validity of the proposed approach.

In the presented work, the practical trajectory tracking controller is designed for a kind of nonholonomic system under full state constraints. Firstly, a finite-time tracking controller is proposed to drive the action of relay switching which divides the entire control process into two stages. Then, using state transformations, the remaining error original system is converted into an unconstrained system for which the tracking controller is designed based on dynamic surface control. Meanwhile, another tracking control law is also addressed to ensure the boundedness of tracking errors before the action of switching. In stability analysis, it has been shown that the developed tracking control strategy can achieve the desired tracking control objectives and the predefined full state constraints. At the same time, all signals keep bounded in the whole control process. Simulation results are provided finally to demonstrate the effectiveness of the tracking control algorithm.

In this study, a new cable-driven rehabilitation robot is designed, the overall design of the robot is given, and the kinematic equation of the lower limbs in the supine state of the human body is addressed. Considering that cable winders move along the rail brackets, the closed vector method is applied to establish the kinematic model of the robot, and the relationship between the human joint angle and the cable length change was deduced. Considering joint compliance, a fifth-order polynomial trajectory planning method based on an S-shaped curve is proposed by introducing an S-shaped velocity curve, and the changes in cable length displacement, velocity, and acceleration are simulated and analyzed. Three planning methods are compared based on two indices, and experimental verification is carried out on the rehabilitation experiment platform. The simulation and experimental results show that the trajectory planning method presents low energy consumption and strong flexibility, and can achieve better rehabilitation effect, which builds a good basis for the subsequent study of dynamics and control strategy.