This paper develops a new solution framework for adaptive output feedback fuzzy control systems aimed at effectively dealing with nonlinear systems with multiple input-multiple output (MIMO) delays and in the presence of dynamics and actuator failure uncertainties. Takagi-Sugeno (T-S) fuzzy systems are employed to represent nonlinear systems, which have desired capacity for dynamic system approximation with parametric and structural properties suitable for using rigorous feedback control techniques. A multiple-delay fuzzy system prediction model is derived and its system properties are clarified. Such a prediction model enables the use of a model-based approach for fuzzy control. The design and analysis are presented for an adaptive control scheme for multiple-delay T-S fuzzy systems, as well as for an adaptive actuator failure compensation scheme for systems with redundant actuators subject to uncertain failures, for which new system parametrizations and controller structures are developed. Simulation results are presented to demonstrate the studied new concepts and to verify the desired performance of the new types of adaptive fuzzy control systems.
In order to deal with the uncertainties caused by different operation conditions and unknown actuator failures of high-speed trains, an adaptive failures compensation control scheme is designed based on the piecewise model. A piecewise constant model is introduced to describe the variable system parameters caused by the variable operation environments, and a multiple-particle piecewise model of high-speed trains, with unknown actuator failures, is then established. An adaptive failure compensation controller is developed for the multiple-particle piecewise constant model, by using a direct model refering to the adaptive control method. Such an adaptive controller can not only compensate the uncertainties from unknown actuator failures, but also effectively deal with the uncertainties caused by different operating conditions. Finally, a CRH380A high-speed train model is taken as the controlled object for the simulation study. The simulation results show that the proposed controller ensures the desired system performance in the presence of unknown actuator failures and uncertain operation conditions.
License plate recognition plays a critical role in many practical applications, but license plates of large vehicles are difficult to be recognized due to the factors of low resolution, contamination, low illumination, and occlusion, to name a few. To overcome the above factors, the transportation management department generally introduces the enlarged license plate behind the rear of a vehicle. However, enlarged license plates have high diversity as they are non-standard in position, size, and style. Furthermore, the background regions contain a variety of noisy information which greatly disturbs the recognition of license plate characters. Existing works have not studied this challenging problem. In this work, we first address the enlarged license plate recognition problem and contribute a dataset containing 9342 images, which cover most of the challenges of real scenes. However, the created data are still insufficient to train deep methods of enlarged license plate recognition, and building large-scale training data is very time-consuming and high labor cost. To handle this problem, we propose a novel task-level disentanglement generation framework based on the Disentangled Generation Network (DGNet), which disentangles the generation into the text generation and background generation in an end-to-end manner to effectively ensure diversity and integrity, for robust enlarged license plate recognition. Extensive experiments on the created dataset are conducted, and we demonstrate the effectiveness of the proposed approach in three representative text recognition frameworks.