Unconstrained text recognition is an important computer vision task, featuring a wide variety of different sub-tasks, each with its own set of challenges. One of the biggest promises of deep neural networks has been the convergence and automation of feature extractors from input raw signals, allowing for the highest possible performance with minimum required domain knowledge. To this end, we propose a data-efficient, end-to-end neural network model for generic, unconstrained text recognition. In our proposed architecture we strive for simplicity and efficiency without sacrificing recognition accuracy. Our proposed architecture is a fully convolutional network without any recurrent connections trained with the CTC loss function. Thus it operates on arbitrary input sizes and produces strings of arbitrary length in a very efficient and parallelizable manner. We show the generality and superiority of our proposed text recognition architecture by achieving state of the art results on seven public benchmark datasets, covering a wide spectrum of text recognition tasks, namely: Handwriting Recognition, CAPTCHA recognition, OCR, License Plate Recognition, and Scene Text Recognition. Our proposed architecture has won the ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.

Unconstrained text recognition is an important computer vision task, featuring a wide variety of different sub-tasks, each with its own set of challenges. One of the biggest promises of deep neural networks has been the convergence and automation of feature extractors from input raw signals, allowing for the highest possible performance with minimum required domain knowledge. To this end, we propose a data-efficient, end-to-end neural network model for generic, unconstrained text recognition. In our proposed architecture we strive for simplicity and efficiency without sacrificing recognition accuracy. Our proposed architecture is a fully convolutional network without any recurrent connections trained with the CTC loss function. Thus it operates on arbitrary input sizes and produces strings of arbitrary length in a very efficient and parallelizable manner. We show the generality and superiority of our proposed text recognition architecture by achieving state of the art results on seven public benchmark datasets, covering a wide spectrum of text recognition tasks, namely: Handwriting Recognition, CAPTCHA recognition, OCR, License Plate Recognition, and Scene Text Recognition. Our proposed architecture has won the ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.

Unconstrained text recognition is an important computer vision task, featuring a wide variety of different sub-tasks, each with its own set of challenges. One of the biggest promises of deep neural networks has been the convergence and automation of feature extractors from input raw signals, allowing for the highest possible performance with minimum required domain knowledge. To this end, we propose a data-efficient, end-to-end neural network model for generic, unconstrained text recognition. In our proposed architecture we strive for simplicity and efficiency without sacrificing recognition accuracy. Our proposed architecture is a fully convolutional network without any recurrent connections trained with the CTC loss function. Thus it operates on arbitrary input sizes and produces strings of arbitrary length in a very efficient and parallelizable manner. We show the generality and superiority of our proposed text recognition architecture by achieving state of the art results on seven public benchmark datasets, covering a wide spectrum of text recognition tasks, namely: Handwriting Recognition, CAPTCHA recognition, OCR, License Plate Recognition, and Scene Text Recognition. Our proposed architecture has won the ICFHR2018 Competition on Automated Text Recognition on a READ Dataset.

Active magnetic bearing (AMB) systems have attracted much attention in the high speed rotating machinery industry. This paper presents an application of discrete-time model predictive control (MPC) subject to input/states constraints to control an AMB system based on linear timeinvariant (LTI) model. The main control objectives are to levitate the rotor shaft of the AMB system while tracking a reference trajectory and to reject possible disturbances without violating the input and state constraints. A nonlinear (NL) model of the AMB system is considered; at each sampling instant, a finite horizon MPC problem is solved to compute the optimal control input. The performance and the efficiency of the proposed MPC is validated via simulation and comparison with another classical PID controller.

Active magnetic bearing (AMB) systems have attracted much attention in the high speed rotating machinery industry. This paper presents an application of discrete-time model predictive control (MPC) subject to input/states constraints to control an AMB system based on linear timeinvariant (LTI) model. The main control objectives are to levitate the rotor shaft of the AMB system while tracking a reference trajectory and to reject possible disturbances without violating the input and state constraints. A nonlinear (NL) model of the AMB system is considered; at each sampling instant, a finite horizon MPC problem is solved to compute the optimal control input. The performance and the efficiency of the proposed MPC is validated via simulation and comparison with another classical PID controller.

Active magnetic bearing (AMB) systems have attracted much attention in the high speed rotating machinery industry. This paper presents an application of discrete-time model predictive control (MPC) subject to input/states constraints to control an AMB system based on linear timeinvariant (LTI) model. The main control objectives are to levitate the rotor shaft of the AMB system while tracking a reference trajectory and to reject possible disturbances without violating the input and state constraints. A nonlinear (NL) model of the AMB system is considered; at each sampling instant, a finite horizon MPC problem is solved to compute the optimal control input. The performance and the efficiency of the proposed MPC is validated via simulation and comparison with another classical PID controller.

Active magnetic bearing (AMB) systems have attracted much attention in the high speed rotating machinery industry. This paper presents an application of discrete-time model predictive control (MPC) subject to input/states constraints to control an AMB system based on linear timeinvariant (LTI) model. The main control objectives are to levitate the rotor shaft of the AMB system while tracking a reference trajectory and to reject possible disturbances without violating the input and state constraints. A nonlinear (NL) model of the AMB system is considered; at each sampling instant, a finite horizon MPC problem is solved to compute the optimal control input. The performance and the efficiency of the proposed MPC is validated via simulation and comparison with another classical PID controller.

This paper presents as application of discrete-time model predictive control (MPC) to control a utility scale wind turbine based on linearized models. The main objective of the controller is to allow the wind turbine to extract from the wind a prespecified desired amount of power according to the wind speed during the whole range of operation. A nonlinear model of a 225 kW wind turbine is considered; at each sampling instant, a linearized model of the corresponding operating point is computed and used to obtain the optimal control input by solving an infinite horizon MPC problem. The procedure is repeated in the subsequent samples to control the nonlinear model. This MPC scheme can guarantee the stability of the closed-loop system at the operating point and its neaighborhood and it demonstrates high control performance.

This paper presents as application of discrete-time model predictive control (MPC) to control a utility scale wind turbine based on linearized models. The main objective of the controller is to allow the wind turbine to extract from the wind a prespecified desired amount of power according to the wind speed during the whole range of operation. A nonlinear model of a 225 kW wind turbine is considered; at each sampling instant, a linearized model of the corresponding operating point is computed and used to obtain the optimal control input by solving an infinite horizon MPC problem. The procedure is repeated in the subsequent samples to control the nonlinear model. This MPC scheme can guarantee the stability of the closed-loop system at the operating point and its neaighborhood and it demonstrates high control performance.

This paper presents as application of discrete-time model predictive control (MPC) to control a utility scale wind turbine based on linearized models. The main objective of the controller is to allow the wind turbine to extract from the wind a prespecified desired amount of power according to the wind speed during the whole range of operation. A nonlinear model of a 225 kW wind turbine is considered; at each sampling instant, a linearized model of the corresponding operating point is computed and used to obtain the optimal control input by solving an infinite horizon MPC problem. The procedure is repeated in the subsequent samples to control the nonlinear model. This MPC scheme can guarantee the stability of the closed-loop system at the operating point and its neaighborhood and it demonstrates high control performance.