No description available.

Abstract In this dissertation the generation and tuning of fuzzy membership function parameters are considered as a part of the fuzzy model development process. The automatic generation and tuning of fuzzy membership function parameters are needed for the fast adaptation and tuning of fuzzy models of various nonlinear dynamical systems. The developed methods are especially useful in automatic fuzzy membership function generation and tuning when dynamic of application area is fast enough to exclude manual tuning. The fuzzy model development process and development methods, modelling environment and nature of application area as well as algorithm development parameters are extensively discussed, because each of them sets their own restrictions on the design parts and parameters used in the modelling. The developed methods have been applied in different kinds of applications (in forecasting the demand of signal transmission products, power control and code tracking of cellular phone system, fuzzy reasoning in radio resource functions of cellular phone systems), where other approaches are either very difficult or too time consuming to implement. The professional areas of the thesis are fuzzy modelling and control in telecommunications.

This thesis presents the development and experimental testing of a lower-limb exoskeleton system. The device supplies assistive torque at the knee joint to alleviate the loading at the knee, and thus reduce the muscular effort required to perform activities of daily living. The hypothesis is that the added torque would facilitate the execution of these movements by people who previously had limited mobility. Only four specific movements were studied: level-waking, gradient-walking, sit-to-stand-to-sit and ascending stairs. All three major components of the exoskeleton system, i.e. the exoskeleton actuators and actuator control system, the user intention estimation algorithm, and the mechanical construction of the exoskeleton, were investigated in this work. A leg brace was fabricated in accordance with the biomechanics of the human lower-limb. A single rotational degree of freedom at the knee and ankle joints was placed to ensure that the exoskeleton had a high kinematic compliance with the human leg. The position of the pneumatic actuators and sensors were also determined after significant deliberation. The construction of the device allowed the real-world testing of the actuator control algorithm and the user intention estimation algorithms. Pneumatic artificial muscle actuators, that have high power to weight ratio, were utilized on the exoskeleton. An adaptive fuzzy control algorithm was developed to compensate for the inherent nonlinearities in the pneumatic actuators. Experimental results confirmed the effectiveness of the adaptive controller. The user intention estimation algorithm is responsible for interpreting the user's intended movements by estimating the magnitude of the torque exerted at the knee joint. To accomplish this, the algorithm utilizes biological signals that emanate from the knee extensor and flexor muscles when they are activated. These signals combined with the knee angle data are used as inputs to the estimation algorithm. The output is the magnitude and direction of the estimated torque. This value is then scaled by an assistance ratio, which determines the intensity of the assistive torque provided to the user. The experiments conducted verify the robustness and predictability of the proposed algorithms. Finally, experimental results from the four activities of daily living, affirm that the desired movements could be performed successfully in cooperation with the exoskeleton. Furthermore, muscle activity recorded during the movements show a reduction in effort when assisted by the exoskeleton.

No description available.

Robot soccer provides a fertile environment for the development of artificial intelligence techniques. Robot controls require high speed lower level reactive layers as well as higher level deliberative functions. This thesis focuses on a number of aspects in the robot soccer arena. Topics covered include boundary avoidance strategies, vision detection and the application of evolutionary learning to find fuzzy controllers for the control of mobile robot. A three input, two output controller using two angles and a distance as the input and producing two wheel velocity outputs, was developed using evolutionary learning. Current wheel velocities were excluded from the input. The controller produced was a coarse control permitting only either forward or reverse facing impact with the ball. A five input controller was developed which expanded upon the three input model by including the current wheel velocities as inputs. The controller allowed both forward and reverse facing impacts with the ball. A five input hierarchical three layer model was developed to reduce the number of rules to be learnt by an evolutionary algorithm. Its performance was the same as the five input model. Fuzzy clustering of evolved paths was limited by the information available from the paths. The information was sparse in many areas and did not produce a controller that could be used to control the robots. Research was also conducted on the derivation of simple obstacle avoidance strategies for robot soccer. A new decision region method for colour detection in the UV colour map to enable better detection of the robots using an overhead vision system. Experimental observations are given.

This thesis proposes and designs an autonomous transportation robot. It can provide elders and disabled people to cross the crosswalk safely with sensor information fusion for real time decision-making and control by fuzzy inference and the information of image, radar, and encoder etc. In this study, the vision system makes feedback of position offset, and declination angle through the fuzzy controller, the robot could modify the attitude error to cross the crosswalk completely. This thesis considers obstacles in the forward path. Using the vision feedback information to design the fuzzy controller, it can supply obstacle avoidance to make the transportation smoothly. Besides, we also address the robot guide at a crosswalk interface point. The detection of special purpose road for the blinds on the vision system can guide the robot to make a right turn or left turn to the next intersection. In cooperate with the intersection-agent system for collision avoidance, the robot could pass through the next crosswalk safely in order to finish the whole safely intersection system.

No description available.

No description available.

Adequate control of the position of the plasma column within the STOR-M tokamak is a chief requirement in order for experimental quality discharges to be obtained. Optimal control over tokamak discharge parameters, including the plasma position, is very difficult to achieve. This is due in large part to the difficulty in modelling the tokamak discharge parameters, as they are highly nonlinear and time varying in nature. The difficulty of modelling the tokamak discharge parameters suggests that a control system, such as a fuzzy logic based controller, which does not require a system model may be well suited to the control of fusion plasma. In order to improve the quality of control over the plasma position within the STOR-M tokamak, the existing analog PID controller was modified. These modifications facilitate the application of a digital controller by a personal computer via the Advantech PCL-711B data acquisition card. The performance of the modified plasma position controller and an Arbitrary Signal Generator developed by the author was evaluated. This modified plasma position controller was applied successfully to the STOR-M tokamak during both normal mode and A.C. mode operation. In both cases, the modified controller provided adequate control over the position of the plasma column within the discharge chamber. Furthermore, the modified controller was more convenient to optimize than the original, existing analog PID controller. By taking advantage of the modifications that were made to the plasma position controller, a fuzzy logic controller was developed by the author. The fuzzy logic based plasma position controller was also successfully applied to the STOR-M tokamak during both normal mode and A.C. operation. The fuzzy controller was demonstrated to reliably provide a higher degree of control over the position of the plasma column within the STOR-M tokamak than the modified PID controller.

Industrial applications increasingly require actuators that offer a combination of high force output, large stroke and high accuracy. The ElectroHydraulic Actuator (EHA) was designed by Drs. Habibi and Goldenberg originally as a high-performance actuator for use in robotics. However, it was determined that the EHA had the potential to achieve high positional accuracy. Little research has been performed in the area of high-accuracy hydraulic positioning systems. Therefore, the objective of this study to achieve nano-scale positional accuracy with the EHA while maintaining large stroke and high force output. It was planned to achieve this objective through modification of the prototype EHA and the use of fuzzy control. During this research project, both hardware and control system modifications to the EHA were performed. A high-precision optical encoder position sensor with a 50 nm resolution was mounted on the inertial load to directly measure the position of the load. A number of device drivers were written to interface the MATLAB real-time control environment with the optical encoder and servo motor amplifier. A Sugeno-inference fuzzy controller was designed and implemented in MATLAB. For comparison purposes, a switched-gain controller and a proportional controller were also implemented in the control environment. The performance of the fuzzy controller was compared to the switched-gain controller and the proportional controller in a number of tests. First, the regulatory and tracking performance of the EHA with an inertial load of 20 kg was examined. It was determined in the regulatory tests that the positional accuracy of the EHA with the fuzzy controller was excellent, achieving a steady state error of 50 ± 25 nm or less for step inputs in the range 5 cm to 200 nm. The positional accuracy during the tracking tests was found to be reduced compared to the regulatory tests since the actuator did not have sufficient time to settle to final accuracy due to the timevarying input signals. In all cases, it was found that the positional accuracy of the EHA with the fuzzy controller was significantly greater than with the switched-gain and proportional controllers for both regulatory and tracking signals. Testing with the inertial load eliminated or changed was not performed because the position sensor was mounted to the load, making it unfeasible to alter the load during the time frame of this study. The regulatory and tracking performance of the EHA with an inertial load of 20 kg plus external resistive loads of 90 to 280 N were investigated. It was found that the positional accuracy of the EHA decreased with the application of an external load to 3.10 ± 0.835 µm for a 1 cm step input (90 N load) and 8.45 ± 0.400 µm for a 3 cm step input (280 N load). Again, the positional accuracy of the EHA decreased during the tracking tests relative to the regulatory tests, for the reason stated above. This implies that the positional accuracy of the EHA with a resistive load is in the microscale, rather than the nano-scale as was put forth as the objective of this study. Nevertheless, the positional accuracy of the EHA with the fuzzy controller was found to be significantly greater than with the switched-gain and proportional controllers. It is postulated that the increase in positional error observed during the external load tests was due to an increase in cross-port leakage, relative to the inertial load tests, caused by the pressure differential induced across the actuator by the external load. Methods of reducing the increase in positional error caused by external loads on the EHA remains an area for future study.