1,720,991 research outputs found

    Making sense of artificial hands

    No full text
    A review of sensors for artificial hands is presented in terms of their range, specifications and characteristics. There is a growing need for sensors due to the development of prosthetic hands that have multiple degrees of freedom requiring finger coordination into different postures. The sensing of force, position (angle), object-slip and temperature allows for the control of these hands automatically and frees the user from cognitive burden. To make the best possible use of individual sensing elements, future controllers will need to combine data from different types of sensor. They may also have an integral power supply using a small battery or harvest energy from their environment and transmit data wirelessly

    Thick-film force, slip and temperature sensors for a prosthetic hand

    No full text
    Thick-film static and dynamic force sensors have been investigated for their suitability to measure the grip forces exerted upon an object held by a prosthetic hand, and to detect and correspondingly react to the possible slip of a gripped item. The static force sensors exploit the piezoresistive characteristics of commercially available thick-film pastes whilst the dynamic slip sensors utilise the piezoelectric behaviour of proprietary PZT pastes (lead zirconate titanate). The sensors are located upon stainless steel cantilever type structures that will be placed at the fingertips of each digit of the prosthetic hand. Temperature sensors are also included to provide temperature compensation for the force sensors and to prevent accidental thermal damage to the prosthesis. Results have shown that the static force sensor is capable of measuring fingertip forces in excess of 100 N, with an electrical half-bridge configuration sensitivity approaching 10 ?V/V•N (with scope for improvement) and maximum hysteresis below 4% of full scale, depending on the manner by which the cantilever sensor array is attached to the finger. Failure in the bonding mechanism that secures the PZT layer to the stainless steel cantilever meant that the proposed dynamic force sensor could not be evaluated. However, investigations using the same sensor design fabricated on an alumina substrate have shown the potential of the PZT dynamic force sensor to measure vibration and hence potentially operate as a slip sensor

    The functional form of the lognormal distribution as a sum of decaying exponential and sinusoidal terms applied to the isometric pinch force of human fingers

    No full text
    The transient data of the pinch force produced between the human forefinger and thumb have been shown to fit the functional form of the well-known lognormal density function. Isometeric force generation is achieved by the stochastic recruitment of individual motor units, which sum together. Evidence from animal and human experiments demonstrates that the force generation can be modelled by underdamped terms. It is shown that a lognormal time series (distribution) can be fitted to a sum of exponential decaying sinusoidal terms

    Surface EMG pattern analysis of the wrist muscles at different speeds of contraction

    No full text
    The aim of this study is to investigate the characteristics of surface electromyographic signals, particularly in pattern analysis. The data were collected from the wrist muscles (flexor carpi ulnaris and extensor carpi radialis) of 20 healthy participants. The study focuses on the movement of the wrist muscles at different frequencies. Participants were asked to contract their muscles at four different speeds (60, 90 and 120 cycles a minute and maximum speed) during wrist flexion and extension, co-contraction and isometric contraction. In this work, moving approximate entropy, mean absolute value and kurtosis are evaluated from the surface electromyographic signals at the four speeds. Moving approximate entropy and kurtosis analysis show that there are significant differences at three states of contraction; start, middle and end. It is shown that there are more regular data in a surface electromyographic signal at the beginning and end of a muscle contraction with low regularity during the middle part

    Texture sensor for a prosthetic hand

    No full text
    This research has shown for the first time that the physical dimensions of small surface patterns in an object surface are detectable from sensors integrated into the fingertip and mechanical links of a prosthetic hand. A further novel aspect of this work is the use of the standard deviation of data used in the analysis. Charge amplifiers are used to extract the signal from the piezoelectric sensors when an object that has two grating surfaces moves past a fingertip. Similar signals have been observed from all the sensors. An analysis of the data has shown that the repeating pattern from the gratings is detectable from a calculation of the mean standard deviation. An estimate of the grating widths can also be made from this analysis. Approximately 32 and 58 grooves are in contact with fingertip (width 15mm) representing a resolution of 2 grooves and 4 grooves mm-1 respectively
    corecore