Random Target Localization for an Upper Limb Prosthesis

To achieve the purpose of accurately grasping a random target with the upper limb prosthesis, the acquisition of target localization information is especially important. For this reason, a novel type of random target localization algorithm is proposed. Firstly, an initial localization algorithm (ILA) that uses two 3D attitude sensors and a laser range sensor to detect the target attitude and distance is presented. Secondly, an error correction algorithm where a multipopulation genetic algorithm (MPGA) optimizes backpropagation neural network (BPNN) is utilized to improve the accuracy of ILA. Thirdly, a general regression neural network (GRNN) algorithm is proposed to calculate the joint angles, which are used to control the upper limb prosthetic gripper to move to the target position. Finally, the proposed algorithm is applied to the 5-DOF upper limb prosthesis, and the simulations and experiments are proved to demonstrate the validity of the proposed localization algorithm and inverse kinematics (IK) algorithm.

Download Full-text

Smartphone supported upper limb prosthesis

Current Directions in Biomedical Engineering ◽

10.1515/cdbme-2015-0116 ◽

2015 ◽

Vol 1 (1) ◽

pp. 484-487

Author(s):

D. Hepp ◽

J. Kirsch ◽

F. Capanni

Keyword(s):

Neural Network ◽

Upper Limb ◽

Design Development ◽

Test Setup ◽

Software Application ◽

Patient Feedback ◽

Emg Signal ◽

Upper Limb Prosthesis ◽

Upper Limb Prostheses ◽

Limb Prosthesis

AbstractState of the art upper limb prostheses offer up to six active DoFs (degrees of freedom) and are controlled using different grip patterns. This low number of DoFs combined with a machine-human-interface which does not provide control over all DoFs separately result in a lack of usability for the patient. The aim of this novel upper limb prosthesis is both offering simplified control possibilities for changing grip patterns depending on the patients’ priorities and the improvement of grasp capability. Design development followed the design process requirements given by the European Medical Device Directive 93/42 ECC and was structured into the topics mechanics, software and drive technology. First user needs were identified by literature research and by patient feedback. Consequently, concepts were evaluated against technical and usability requirements. A first evaluation prototype with one active DoF per finger was manufactured. In a second step a test setup with two active DoF per finger was designed. The prototype is connected to an Android based smartphone application. Two main grip patterns can be preselected in the software application and afterwards changed and used by the EMG signal. Three different control algorithms can be selected: “all-day”, “fine” and “tired muscle”. Further parameters can be adjusted to customize the prosthesis to the patients’ needs. First patient feedback certified the prosthesis an improved level of handling compared to the existing devices. Using the two DoF test setup, the possibilities of finger control with a neural network are evaluated at the moment. In a first user feedback test, the smartphone based software application increased the device usability, e.g. the change within preselected grip patterns and the “tired muscle” algorithm. Although the overall software application was positively rated, the handling of the prosthesis itself needs to be proven within a patient study to be performed next. The capability of the neural network to control the hand has also to be proven in a next step.

Download Full-text