No abstract

Finger prostheses are devices developed to emulate the functionality of natural human fingers. On top of their aesthetic appearance in terms of shape, size and colour, such biomimetic devices require a high level of dexterity. They must be capable of gripping an object, and even manipulating it in the hand. This paper presents a biomimetic robotic finger actuated by a hybrid mechanism and integrated with a tactile sensor. The hybrid actuation mechanism comprises a DC micromotor and a Shape Memory Alloy (SMA) wire. A customized test rig has been developed to measure the force and stroke produced by the SMA wire. In parallel with the actuator development, experimental investigations have been conducted on Quantum Tunnelling Composite (QTC) and Pressure Conductive Rubber (PCR) towards the development of a tactile sensor for the finger. The viability of using these materials for tactile sensing has been determined. Such a hybrid actuation approach aided with tactile sensing capability enables a finger design as an integral part of a prosthetic hand for applications up to the transradial amputation level.

Brain-Computer Interface (BCI) using Electroencephalography (EEG) enables non-invasive direct control between human brain and machine and opens up new possibilities in providing healthcare solutions for people with severe motor impairment. This paper reviews the recent trends in neuroprostheses and presents a conceptual design for the development of a cost-effective neuroprosthetic hand deploying EEG signals. Towards the development of a brain-computer interface for neuroprostheses, EEG signals are recorded from healthy subjects using the Emotiv Suite Software. The recognition phase and signal analysis are performed using the EEGLab Software. Signal processing is required until clear rhythmic waves are obtained as a command to control a prosthetic hand. A Graphical User Interface (GUI) will be developed using Matlab Software and aided with 3D Animation as a medium of interaction for basic training for the patient before using the prosthetic hand.

No abstract

Shape Memory Alloy (SMA) has been identified as an ideal candidate to actuate the finger joints of hand prosthesis due to its silent operation, small size, high force to weight ratio and operational similarity with human muscles. When current is applied to a SMA wire, the material is heated up. A phase transformation from Martensite to Austenite takes place and the wire contracts. This work evaluates the displacement and force produced from the contraction of a type of SMA wire, i.e. Nitinol (Ni-Ti). The displacement is determined by a linear potentiometer while the force is determined by a load cell. Furthermore the impacts of varying the bias force exerted on the SMA wire and the input current passed through the wire are investigated. Such insight into the characteristics of SMA can be used to realize biomimetic actuation in order to improve the dexterity of hand prosthesis.