Yu Hongliu

Random Re-Connection Leaning Algorithm of CMAC Model in Prosthetic Knee Control

A new learning algorithm of random re-connection (RRC) originating from the problem of mapping precision of CMAC controller used for prosthetic knee, which connects input layer with neural cell layer, is put forward from the view point of structure optimization in this paper. After the learning process of RRC, the in-degree distribution of neural cell becomes to follow power-law, which indicates that the effect of every cell in pattern identification is different. The simulation result of applying the RRC algorithm to prosthesis control shows that the algorithm of random re-connection can significantly improve the mapping precision of CMAC network model.

Analysis and comparison of intelligent control methods for computer-controlled artificial leg

The computer-controlled artificial legs, most of which are above-knee prostheses(AKP) can better adapt to the human gait and walking modes, automatically distinguishing terrain and coordinate the symmetry of gait. Due to the complexity and non-linearity of AKP control, it is required to design a kind of controller being intelligent enough for it. Research and product development of computer-controlled AKP is comprehensively discussed in the paper. The expert controller based on finite-state machine method and BP neural network controller based on PD supervision are especially analyzed. The major intelligent control methods applied for existing AKP products are also compared with each other here. The development of intelligent control technology in the future is pointed out.

Study on the usability evaluation of prosthetic leg products based on ergonomics

The prosthetic leg (PL) is a typical human-machine system in which the dynamic interaction between the human body and PL (machine) determines a high requirement of ergonomics design for PL and consequently needs to consider an indicator of usability indicating the performance of gait biomechanics. How to evaluate the indexes of usability for PL products is critical to the design technology of PL based on ergonomics. The gait symmetry of PL products, which is a core usability index, was experimentally analyzed by using a specially designed testing device in this paper. The test results show that the swing speed symmetry for intelligent prosthetic leg (IPL) is high up to 96.5%, which indicates a high performance for gait tracking. Then, a comprehensive evaluation was made for the usability of PL products with the analysis method of Grey Correlation Degree. Three types of PL product, including ordinary PL, IPL and Gait-following IPL (GL-IPL), were used in the evaluation. The evaluation results show that the GL-IPL product is the best in usability while both of IPL and GF-IPL products based on ergonomics are obviously better than the ordinary PL. The method of usability evaluation studied here is expected to help directing the design of prosthetic leg products.

A Real-Time Force-Displacement Detection and Evaluation System for Wearable Hand Exoskeleton Based on Multi-sensors

The real-time accurate judgment of force and angular displacement is the basis of controlling for wearable hand rehabilitation robot. This paper puts forward to and set up a kind of real-time detection system of force and the angular displacement, which includes an elastic pressure glove and three thin film pressure sensors and three angular displacement sensors. And the system can accurately detect the pressure on the knuckles of the thumb and forefinger and middle finger, at the meantime three angular displacements of IP of the thumb and the PIP of the index finger and middle finger during hand rehabilitation training. Finally, the motion of the hand can be assessed and the curative effect of rehabilitation treatment can be judged objectively. So this study could provide the basis for the selection of appropriate treatment technology and the hand display technology and other ways to reduce the limited range of motion (ROM).

Compound control for intelligent artificial Leg based on Fuzzy-CMAC

The features of complex system model, such as nonlinearity and parameter uncertainty, in the Intelligent Artificial Leg (IAL) system determines the need of intelligent control methods. On the other hand, the traditional mathematic model is not suitable for the actual control because the IAL knee torque is indirectly caused by the nonlinear damping at the knee joint. A dynamic model of a self-made hydraulic IAL with the nonlinear damper control parameters and hip torque was established, and an inverse dynamic compound controller of PD/Fuzzy-CMAC for tracking the knee swing was designed in the paper. A case simulation shows that an arbitrary trajectory like a desired walking pattern can be tracked in less than 5 seconds, which proves that the designed controller has high performance of real-time and precision.