No description available.

This thesis considers the optimisation of vehicle suspension systems via a reinforcement learning technique The aim is to assess the potential of learning automata to learn 'optimum' control of suspension systems, which contain some active element under electronic control, without recourse to system models. Control optimisation tasks on full-active and senu-active suspension systems are used for the feasibility assessment and subsequent development of the learning automata technique. The quarter-vehicle simulation model, with ideal full-active suspension actuation, provides a well-known environment for initial studies applying classical discrete learning automata to learn the controller gains of a linear state-feedback controller. Learning automata are shown to be capable of acquiring near optimal controllers without any explicit knowledge of the suspension environment. However, the methodology has to be developed to allow safe on-line application. A moderator is introduced to prevent excessive suspension deviations as a result of possible unstable control actions applied during learning. A hardware trial is successfully implemented on a test vehicle fitted with semi-active suspension, excited by a hydraulic road simulation rig. During these initial studies some inherent weaknesses of the discrete automata are noted A discrete action set provides insufficient coverage of a continuous controller space so optima may be overlooked. Subsequent methods to increase the resolution of search lead to a forced convergence and hence an increased likelihood of local optima location. Th1s motivates the development of a new formulation of learning automaton, the CARLA, which exhibits a continuous action space and a reinforcement generalisation. The new method is compared w1th discrete automata on vanous stochastic function optimisatwn case stui1es, demonstrating that the new functionality of CARLA overcomes many of the identified shortcomings of discrete automata. Furthermore, CARLA shows a potential capability to learn in non-stationary environments. Repeatmg the earlier suspension tasks with CARLA applied, including an on-line hardware study, further demonstrates a performance gain over discrete automata Finally, a complex multi-goal learning task is considered A dynamic roll-control strategy is formulated based on the senu-active suspension hardware of the test vehicle. The CARLA is applied to the free parameters of this strategy and is seen to successfully synthesise improved roll-control over passive suspension.

Thesis (M.Sc.), Faculty of Engineering & the Built Environment, University of the Witwatersrand, 2003

University of Macau / Faculty of Science and Technology. / Department of Electromechanical Engineering

University of Macau / Faculty of Science and Technology. / Department of Electromechanical Engineering

Photocopy of a carbon copy of the original / xi, 331 leaves : ill. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (Ph.D.)--University of Adelaide, Dept. of Mechanical Engineering, 1966

Zugl.: Kaiserslautern, Techn. Universiẗat, Diss.

No description available.

All-terrain vehicle(ATV) is very popular in Europe and North America because of its excellent riding performance. The Design of suspension and steering mechanism is one of the most important parts in ATV design, because it is much related to the riding performance of ATVs. Therefore, this research focuses on investigating the design of suspension and steering mechanism for ATVs. There are two main objectives to be achieved in this research. One is to build the analysis theory, and the other is to develop a systematic design process of suspension and steering mechanism for ATVs. Firstly, we survey papers about suspension and steering mechanism to find its basic functions and properties. Secondly, we use engineering design method to classify the design requirements and frame the design process. Thirdly, we list new suspension and steering mechanisms by applying the creative mechanism design procedure, then separate the new mechanisms separate into two types. One type is suitable only for back-wheel drive vehicles, and the other type is suitable for back-wheel drive and four-wheel drive vehicles. Finally, we complete the kinematic design of the type1.1(double-wishbone) and type8.3(new mechanism) suspension and steering mechanisms based on the Genetic algorithm (GA), and we can obtain the better mechanism dimensions than Yamaha 660 and 450 ATV.

The design, fabrication, and testing of a quarter-car facility coupled with various control algorithms are presented in this thesis. An experimental linear tubular motor, capable of producing a 52-N force, provides control actuation to the model. Controllers consisting of two designs were implemented: a classical controller employing lead and lag networks and a state-space feedback design. Each design was extensively simulated to screen for receptiveness to actuation force limitations and robustness regarding the inexact tire modeling. The goal of each controller was to minimize the acceleration of the sprung mass in the presence of simulated road disturbances, modeled by both sinusoidal and step input excitation wheels. Different reference velocity inputs were applied to the control scheme. Responses to a zero reference were juxtaposed to those that resulted from tracking a reference built off a model that incorporated inertial-frame damping attached to the sprung mass. The outcome of this comparison was that low-frequency disturbances were attenuated better when tracking a zero reference, but the reference relaxation introduced by the inertialframe damping model allowed for better-attenuated high frequency signals. Employing an inertial-frame damping value of 250 N-s/m, the rejected frequency component of the system response synchronous with the disturbance input excitation of 40 rad/s bettered by 33% and 28% when feeding control force from the classical controller and state-space controller, respectively. The experimental analysis conducted on the classical and state-space controllers produced sinusoidal disturbance rejection of at worst 50% within their respective bandwidths. At 25 rad/s, the classical controller was able to remove 80% of the base component synchronous with the disturbance excitation frequency, while the state-space controller filtered out nearly 60%. Analysis on the system's ability to reject step disturbances was greatly confounded with the destructive lateral loading transferred during the excitation process. As a result, subjection to excitation could only occur up to 25 rad/s. At the 20 rad/s response synchronous to the disturbance excitation, the classical and state-space controllers removed 85% and 70% of the disturbance, respectively. Sharp spikes in timebased amplitude were present due to the binding that ensued during testing.