Evaluation of a Neural Network-Based Control Strategy for a Cost-Effective Externally-Powered Prosthesis

Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, USA.
Assistive technology: the official journal of RESNA (Impact Factor: 0.51). 09/2012; 24(3):196-208. DOI: 10.1080/10400435.2012.659796
Source: PubMed


This paper presents a control strategy that compensates for the nonlinearity in the inexpensive sensors and hardware of a cost effective prosthetic hand. The control strategy uses neural network-based force control and sensory feedback to detect disturbance induced by slippage. The neural network approach is chosen over other nonlinear models because it is easy to implement and it offered the additional advantage of having its parameters easily adjusted over the life span of the device. The proposed strategy was evaluated on a functional multi-digit underactuated prosthetic hand. The initial and incremental forces exerted from each finger were adjusted to balance the amount of disturbance and the deformation of the objects. Experiments were conducted to test the performance of the protocol in situations encountered in activities of daily living. The displacement of each object under three grasping configurations was measured as a performance criterion while the object's mass was changed. The results showed that with the adjusted parameters for each grasping configuration, the control strategy was able to detect the dynamic changes in mass of the object and was also able to successfully adjust the grasping force before the object drops from the hand.

1 Follower
31 Reads
  • Fontana · Chiu ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Myoelectric pattern recognition systems can translate muscle contractions into prosthesis commands; however, the lack of long-term robustness of such systems has resulted in low acceptability. Specifically, socket misalignment may cause disturbances related to electrodes shifting from their original recording location, which affects the myoelectric signals (MES) and produce degradation of the classification performance. In this work, the impact of such disturbances on wavelet features extracted from MES was evaluated in terms of classification accuracy. Additionally, two principal component analysis frameworks were studied to reduce the wavelet feature set. MES from seven able-body subjects and one subject with congenital transradial limb loss were studied. The electrode shifts were artificially introduced by recording signals during six sessions for each subject. A small drop in classification accuracy from 93.8% (no disturbances) to 88.3% (with disturbances) indicated that wavelet features were able to adapt to the variability introduced by electrode shift disturbances. The classification performance of the reduced feature set was significantly lower than the performance of the full wavelet feature set. The results observed in this study suggest that the effect of electrode shift disturbances on the MES can potentially be mitigated by using wavelet features embedded in a pattern recognition system.
    Assistive technology: the official journal of RESNA 05/2014; 26(2). DOI:10.1080/10400435.2013.827138 · 0.51 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Internal models allow unimpaired individuals to appropriately scale grip force when grasping and lifting familiar objects. In prosthesis users, the internal model must adapt to the characteristics of the prosthetic devices and reduced sensory feedback. We studied the internal models of 11 amputees and eight unimpaired controls when grasping and lifting a fragile object. When the object was modified from a rigid to fragile state, both subject groups adapted appropriately by significantly reducing grasp force on the first trial with the fragile object compared to the rigid object (p < 0.020). There was a wide range of performance skill illustrated by amputee subjects when lifting the fragile object in 10 repeated trials. One subject, using a voluntary close device, never broke the object, four subjects broke the fragile device on every attempt and seven others failed on their initial attempts, but improved over the repeated trials. Amputees decreased their grip forces 51 ± 7 % from the first to the last trial (p < 0.001), indicating a practice effect. However, amputees used much higher levels of force than controls throughout the testing (p < 0.015). Amputees with better performance on the Box and Blocks test used lower grip force levels (p = 0.006) and had more successful lifts of the fragile object (p = 0.002). In summary, amputees do employ internal models when picking up objects; however, the accuracy of these models is poor and grip force modulation is significantly impaired. Further studies could examine the alternative sensory modalities and training parameters that best promote internal model formation.
    Experimental Brain Research 08/2014; 232(12). DOI:10.1007/s00221-014-4071-1 · 2.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This chapter describes the implementation of a neural network-based predictive control system for driving a prosthetic hand. Nonlinearities associated with the electromechanical aspects of prosthetic devices present great challenges for precise control of this type of device. Model-based controllers may overcome this issue. Moreover, given the complexity of these kinds of electromechanical systems, neural network-based modeling arises as a good fit for modeling the fingers' dynamics. The results of simulations mimicking potential situations encountered during activities of daily living demonstrate the feasibility of this technique.
    Methods in molecular biology (Clifton, N.J.) 01/2015; 1260:179-94. DOI:10.1007/978-1-4939-2239-0_11 · 1.29 Impact Factor

Similar Publications