EMG pattern recognition is essential for the control of a multifunction myoelectric hand. The main goal of this study was to develop an efficient feature projection method for EMG pattern recognition. To this end, a linear supervised feature projection is proposed that utilizes a linear discriminant analysis (LDA). First, a wavelet packet transform (WPT) is performed to extract a feature vector from four channel EMG signals. To dimensionally reduce and cluster the WPT features, an LDA then incorporates class information into the learning procedure, and identifies a linear matrix to maximize the class separability for the projected features. Finally, a multilayer perceptron classifies the LDA-reduced features into nine hand motions. To evaluate the performance of the LDA for WPT features, the LDA is compared with three other feature projection methods. From a visualization and quantitative comparison, it is shown that the LDA produces a better performance for the class separability, plus the LDA-projected features improve the classification accuracy with a short processing time. A real-time pattern recognition system is then implemented for a multifunction myoelectric hand. Experiments show that the proposed method achieves a 97.4 % recognition accuracy, and all processes, including the generation of control commands for the myoelectric hand, are completed within 97 msec. Consequently, these results confirm that the proposed method is applicable to real-time EMG pattern recognition for multifunction myoelectric hand control.
This paper proposes a new learning method for a Gaussian mixture model (GMM). First, a traditional maximum a posterior (MAP) parameter estimate is used to achieve regularization based on conjugate priors. Next, a model order selection criterion is derived from Bayesian-Laplace approaches such that the conjugate prior distribution can be used to measure the uncertainty in the estimated parameters. As a result, the proposed learning method avoids the possibility of convergence toward local minima in the parameter space, and is also capable of selecting the optimal order for a GMM using an additional complexity penalty for the prior distribution. The proposed method is applied to electromyogram (EMG) pattern recognition for controlling a multifunction myoelectric hand, and experiments conducted to recognize nine kinds of hand motion from EMG signals for ten subjects. In conclusion, the proposed learning method effectively estimated the change of feature vectors according to the subject and the GMM classifier
demonstrated a high recognition accuracy.
A multifunction myoelectric hand is designed with underactuated mechanisms. The finger design allows an adaptive grasp, including adaptation between fingers and phalanges with respect to the shape of an object. In addition, a self-lock is embedded in the metacarpophalangeal joint to prevent back driving when external forces act on the fingers. The thumb design also provides adaptation between phalanges and adds an intermittent rotary motion to the carpometacarpal joint. As a result, the hand can perform versatile grasping motions using only two motors, and is capable of natural and stable grasping without complex sensor and servo systems. Moreover, the adaptive grasping capabilities reduce the requirements of electromyogram pattern recognition, as analogous motions, such as cylindrical and tip grasps, can be classified as one motion.
The infrared (IR) light emitting diode (LED)-based tactile sensor on the each finger that can independently measure the normal and tangential force between the hand and an object. The developed IR LED-based sensor has several advantages over other technologies, including a low price, small size, and good sensitivity. The design of the first prototype is described and some experiments are conducted to show output characteristics of the sensor. Furthermore, the effectiveness of the proposed sensor is demonstrated through anti-slip control in a multifunction myoelectric hand, called the KNU Hand, which includes several novel mechanisms for improved grasping capabilities. The experimental video clips show that slippage was avoided by simple force control using feedback on the normal and tangential force from the proposed sensor. Thus, grasping force control was achieved without any slippage or damage to the object.
Orthosis system driven by a new antagonistic SMA actuator with Bowden-cable housing has been developed. Many orthosis actuators have already been developed to overcome the limitation of electric motor-driven orthosis systems, which are usually bulky and heavy. In contrast to conventional approaches, SMA actuators are lightweight with a high power density and silent actuation, and can have several mechanical configurations: winding-type, weaving-type, straight-line-type, coil-type, and Bowden-cable-type. Among these alternatives, the winding-type is superior as regards force generation, the amount of displacement within a restricted volume. Therefore, A novel mechanical design, including a new winding method for the SMA wire and new control system is investigated. The developed SMA actuator is applied to an elbow orthosis system and its performance verified through experiments.