To improve the efficacy of robotic exoskeleton-based rehabilitation training, active joint torque of subjects should be detected. This paper presents a practical and adaptive method to estimate active joint torque using electromyography (EMG) signals for a custom lower limb robotic exoskeleton with two degrees of freedom (DOFs). This estimator, constructed of radial basis function neural networks (RBFNNs), was used to form an extended Slotine-Li controller. This extended controller eliminated the need for the calibration for EMG-torque model. The adaptive control of exoskeleton and adaptive estimation of active joint torque were performed within the same framework. By introducing a two-step learning strategy into the controller, the estimator can continuously adapt to changes in the EMG-torque model, and overcome the problems due to the time-varying property of EMG signals. Simulation and experimental results show that the presented estimator can predict the active joint torque of subjects in a practical and adaptive manner. Additionally, the accurate movement control of exoskeleton is also guaranteed. At present, the experiments are conducted only for the swing phase due to the lack of the force plate sensors.

Objective. Recent studies have reported that the classification performance of electromyographic (EMG) signals degrades over time without proper classification retraining. This problem is relevant for the applications of EMG pattern recognition in the control of active prostheses. Approach. In this study we investigated the changes in EMG classification performance over 11 consecutive days in eight able-bodied subjects and two amputees. Main results . It was observed that, when the classifier was trained on data from one day and tested on data from the following day, the classification error decreased exponentially but plateaued after four days for able-bodied subjects and six to nine days for amputees. The between-day performance became gradually closer to the corresponding within-day performance. Significance. These results indicate that the relative changes in EMG signal features over time become progressively smaller when the number of days during which the subjects perform the pre-defined motions are increased. The performance of the motor tasks is thus more consistent over time, resulting in more repeatable EMG patterns, even if the subjects do not have any external feedback on their performance. The learning curves for both able-bodied subjects and subjects with limb deficiencies could be modeled as an exponential function. These results provide important insights into the user adaptation characteristics during practical long-term myoelectric control applications, with implications for the design of an adaptive pattern recognition system.

Restoring the interaction between disabled people and the 3-D physical world via a brain-computer interface (BCI) is an exciting topic. To this end, we designed a wearable BCI system based on the steady-state visual evoked potential (SSVEP), which enables 3-D navigation of quadcopter flight with immersive first-person visual feedback using a head-mounted device. In addition, to alleviate the user's operational burden, this paper provides asynchronous switch control for the users. The transitional state due to head movement in an asynchronous BCI was isolated online and translated into hover to eliminate its influence. The experimental results in the physical environment showed that the subjects could accomplish the 3-D flight tasks accurately and smoothly using our system. In particular, in this paper, we proposed an information transfer rate metric that is suitable for the asynchronous task. We demonstrated the feasibility of using the head-mounted device and a proper control strategy to facilitate the portability and practicability of the SSVEP-based BCI system for its navigation utility.

Robotic exoskeletons for physical rehabilitation have been utilized for retraining patients suffering from paraplegia and enhancing motor recovery in recent years. However, users are not voluntarily involved in most systems. This paper aims to develop a locomotion trainer with multiple gait patterns, which can be controlled by the active motion intention of users. A multimodal human-robot interaction (HRI) system is established to enhance subject's active participation during gait rehabilitation, which includes cognitive HRI (cHRI) and physical HRI (pHRI). The cHRI adopts brain-computer interface based on steadystate visual evoked potential. The pHRI is realized via admittance control based on electromyography. A central pattern generator is utilized to produce rhythmic and continuous lower joint trajectories, and its state variables are regulated by cHRI and pHRI. A custom-made leg exoskeleton prototype with the proposed multimodal HRI is tested on healthy subjects and stroke patients. The results show that voluntary and active participation can be effectively involved to achieve various assistive gait patterns.

Background: The nonstationary property of electromyography (EMG) signals usually makes the pattern recognition (PR) based methods ineffective after some time in practical application for multinational prosthesis. The conventional EMG PR, which is accomplished in two separate steps: training and testing, ignores the mismatch between training and testing conditions and often discards the useful information in testing dataset. Method: This paper presents a novel self-enhancing approach to improve the classification performance of the electromyography (EMG) pattern recognition (PR). The proposed self-enhancing method incorporates the knowledge beyond the training condition to the classifiers from the testing data. The widely-used linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) are extended to self-enhancing LDA (SELDA) and self-enhancing QDA (SEQDA) by continuously updating their model parameters such as the class mean vectors, the class covariances and the pooled covariance. Autoregressive (AR) and Fourier-derived cepstral (FC) features are adopted. Experimental data in two different protocols are used to evaluate performance of the proposed methods in short-term and long-term application respectively. Results: In protocol of short-term EMG, based on AR and FC, the recognition accuracy of SEQDA and SELDA is 2.2% and 1.6% higher than conventional that of QDA and LDA respectively. The mean results of SEQDA(C) and SEQDA (M) are improved by 2.2% and 0.75% for AR, and 1.99% and 1.1% for FC respectively when compared to QDA. The mean results of SELDA(C) and SELDA (M) are improved by 0.48% and 1.55% for AR, and 0.67% and 1.22% for FC when compared to LDA. In protocol of long-term EMG, the mean result of SEQDA is 3.15% better than that of QDA. Conclusion: The experimental results show that the self-enhancing classifiers significantly outperform the original versions using both AR and FC coefficient feature sets. The performance of SEQDA is superior to SELDA. In addition, preliminary study on long-term EMG data is conducted to verify the performance of SEQDA.