Search

Presented was an investigation on the performance of the feedback linearization control (FLC) method for the FES induced STS movement application in paraplegia. Improving the operation condition of the FES induced movement is one of the challenges currently receiving attention by so many scholars in the field. The Newton-Euler and Euler-Lagrange methods are utilised to check its correctness together with the muscle model proposed by Ferrarin and Pedotti. The FLC was evaluated for the system with and without fatigue disturbance. Although the fatigue disturbance was introduced in the system, results revealed that the cumulative current, cumulative rate of change in current the cumulative rate of change in current, RMSE, tracking error, cumulative tracking error and the level of the control signal, were by far better for the FL than that of the PID control technique with wide margins. Therefore, results portrayed that the FL control system the tendency to annul the effects of the effect of fatigue with higher capability over the gain scheduling based PID control method. It also indicates superiority in terms of system stability.

The paper presented an evaluation using the Back Stepping control (BSC) for the regulation of Functional Electrical Stimulation (FES) aided sit-to-stand (STS) motion restoration by investigating the effect of human mass distribution across the world. The study focused on eradicating or reducing to the minimum the tasks of changing the control system parameters when applying the device for different subjects. The comparison was made with the Proportional Integral Derivative (PID) based gain scheduling based control approach. Results do indicate the robustness of the BSC control method as it portrays that despite the alteration of the masses to the extremes the control system was able to keep the tracking error approaching zero or very close to the desired with and without changes. It outperformed the PID based gain scheduling control scheme. It also operates within the permissible limits specified. Hence, it reveals the effectiveness of the BSC scheme for solving the problem of continuous readjustments making the equipment universal. It, therefore, relieve the burden of the necessity of changing the control variables continuously. The study is novel because the BSC has never been evaluated on four segments based induced STS movement in paraplegia model. The universal human mass distribution has also not been considered using the control scheme on the model

The work presented an enhanced trajectory planning for better handling of Functional Electrical Stimulation (FES) actuated revival of sit-to-stand (STS) manoeuvre. Currently there the problem of lack of FES aided devices as results of the failures of the proposed equipment to pass clinical tests. Employing control systems techniques have been yielding fruitful results. Literature indicated that further enhancement of trajectory planning could help in achieving milestones towards better performances. Continuous and coupled with bump-free trajectories are some essential properties for better control systems designs for such systems. The proposed trajectory planning utilises the six-order polynomial. Comparisons are made with the five and seven order polynomials. The fifth order polynomials give an insight of the performance of the system with lower order polynomials while the seventh order gave that of the higher order counterparts. It was employed to enhance the desired movement trajectories. Obtained results portrayed that the trajectory planning achieved are continuous and the tendencies of the appearances of jerks or spikes are eliminated. Others are an improvement on execution time by 11%, reducing the upper and lower bound terminal velocities by 16.9% and 20.9% accordingly. Thus, these indicate the possibilities of realisations of higher performance control systems for the induced movement tasks.

Presented is an evaluation of the sliding mode controller (SMC) for regulation of Functional Electrical Stimulation (FES) induced sit-to-stand (STS) movement considering a global human mass variation. It is aimed at the elimination of the retuning issue associated with control systems for different subjects. Results show that despite the changes in mass the control system maintained the tracking error close to zero; that is close to the desired, and without any disturbance scenario, it was also achieved within the suitable specifications. Thus, it shows that the SMC system can be employed to solve the issue of universality; therefore relieving the necessity of altering the control parameters always.

Artificial neural approach has been executed in various recorded, and a champion amongst the most understood widespread approximators. Neural framework has for quite a while been known for its ability to handle a complex nonlinear system without a logical model and can learn refined nonlinear associations gives. Theoretically, the most surely understood computation to set up the framework is the backpropagation (BP) count which relies on upon the minimization of the mean square error (MSE). This paper exhibits the improvement of quadriceps muscle model by utilizing counterfeit smart procedure named backpropagation neural network nonlinear autoregressive (BPNN-NAR) model in view of utilitarian electrical incitement (FES). A progression of tests utilizing FES was led. The information that is gotten is utilized to build up the quadriceps muscle model. 934 preparing information, 200 testing and 200 approval information set are utilized as a part of the improvement of muscle model. It was found that BPNN-NAR is suitable and efficient to model this type of data. A neural network model is the best approach for modelling non-linear models such as active properties of the quadriceps muscle with one input, namely output namely muscle force.

Artificial neural network has been implemented in many filed, and one of the most famous estimators. Neural network has long been known for its ability to handle a complex nonlinear system without a mathematical model and has the ability to learn sophisticated nonlinear relationships provides. Theoretically, the most common algorithm to train the network is the backpropagation (BP) algorithm which is based on the minimization of the mean square error (MSE). Subsequently, this paper displays the change of quadriceps muscle model by using fake savvy strategy named backpropagation neural system nonlinear autoregressive (BPNN-NAR) model in perspective of utilitarian electrical affectation (FES). A movement of tests using FES was driven. The data that is gotten is used to develop the quadriceps muscle model. 934 planning data, 200 testing and 200 endorsement data set are used as a part of the change of muscle model. It was found that BPNN-NARMA is suitable and efficient to model this type of data. A neural network model is the best approach for modelling nonlinear models such as active properties of the quadriceps muscle with one input, namely output namely muscle force.

FES induced movement control is a significantly challenging area due to complexity and non-linearity of musculo-skeletal system. The goal of this study is to design a cycle-to-cycle control of FES-induced swinging motion. In this approach only the quadriceps muscle is stimulated by controlling the amount of stimulation pulsewidth. This time dependent behaviour is successfully compensated for using a cycle-to cycle fuzzy controller, which computes the amount of knee extension stimulation on the basis of the achieved flexion angle in previous cycles. The capability of fuzzy control in automatic generation of stimulation burst duration is assessed in computer simulations using a musculo-skeletal model. This paper presents the development of a fuzzy logic control scheme based on discrete-time cycle to cycle control strategies without predefined trajectory. The results show the effectiveness of the approach in controlling FES-induced swinging motion

The use of electrical signals to restore the function of paralyzed muscles is called functional electrical stimulation (FES). FES is a promising method to restore mobility to individuals paralyzed due to spinal cord injury. FES induced movement control is a significantly challenging area, mainly emanating from various characteristics of the underlying physiological/biomechanical system. An approach of fuzzy trajectory tracking control of swinging motion optimized with genetic algorithm is presented. The results show the effectiveness of the approach in controlling FES-induced swinging motion. In this approach only the quadriceps muscle is stimulated to perform the swinging motion by controlling the amount of stimulation pulsewidth

In this paper, we propose a series of techniques to enhance the computational performance of existing Belief Propagation (BP) based stereo matching that relies on automatic estimation of the Markov random field (MRF) parameters. First, we show how convergence in matching can be achieved faster than with the existing message comparison technique by skipping comparisons in early inferences. Second, assuming that a stereo pair is captured with identical cameras, we apply a hypothesis called noise equivalence to pre-estimate the likelihood parameters and thus, avoid costly nested inferences to reduce the computational time. The likelihood parameters and intensity information are used for accelerated message propagation in image regions lacking gradients. Third, the prior model parameters are estimated with a combination of maximum likelihood (ML) estimation and disparity gradient constraint to further reduce the computational time. Supporting experiments for the proposed algorithms show encouraging results on ground truth test images.

Micro-electro-mechanical systems (MEMS) are found in area applications such as the automotive industry, the aviation industry, the semiconductor industry, the medical field, and various other fields where miniaturization is taking over. The accurate measurement of features on the surface of MEMS is an important tool for the assessment and monitoring of product quality. Presented here are the algorithms and results of 3D model reconstructions of MEMS devices using a variety of microscopic sensors. These sensors include an atomic force microscope, a scanning electron microscope, and a laser scanning confocal microscope. MEMS devices with micron-size features were first scanned with these microscopes. 3D models were then built and visualized using methods specific to each microscope. This allows for the models use in applications such as inspection, study of wear and tear, behavior, and reaction of such systems to pressure, heat, and friction.

This paper summarizes the various components of face recognition research conducted at the IRIS Lab. First, fusion of visual and thermal infrared (IR) images for robust face recognition is discussed. Two techniques are implemented: data fusion and decision fusion. With the knowledge that eyeglasses block the emission of thermal energy, an algorithm is designed to detect and replace eyeglasses with an eye template in thermal images. A commercial face recognition software (FaceIt/spl reg/) is used in the evaluation of the various fusion algorithms. Comparison results show that fusion-based face recognition outperforms individual visual or thermal face recognizers under illumination variations and facial expressions. Efforts in the 3D arena are also described. Results of high resolution stereo-based 3D reconstruction of faces are shown and analyzed, in a first approach, then in a second approach, a warping technique is applied to overlay color and thermal textures on 3D mannequin head models, obtained using a laser range scanner.

In the design of a dedicated neural network, the number of precision levels used in the hardware circuitry to store weight values is an important consideration as it will impact the functionality and hence the performance of the neural network. One measure of the functionality is the number of training set examples required to achieve perfect generalization. In this paper, we experimentally determine the training set size required for the threshold Adaline (adaptive linear element) with various levels of weight precision to achieve perfect generalization. In all cases, it was found that the training set size required for the perfect generalization was proportional to the number of weights; for the binary, ternary, and 5-ary Adalines, the constants of the proportionality were found to be 1.36, 2.5, and 4.85 respectively.