Wen-June Wang

In this study, we proposed a versatile and robust postal code recognition system. The system can be applied to billing envelop sorting machines to help the post In this study, we proposed a versatile and robust postal code recognition system. The system can be applied to billing envelop sorting machines to help the post

In this study, we propose an assistive system for helping visually impaired people walk outdoors. This assistive system contains an embedded system—Jetson AGX Xavier (manufacture by Nvidia in Santa Clara, CA, USA) and a binocular depth camera—ZED 2 (manufacture by Stereolabs in San Francisco, CA, USA). Based on the CNN neural network FAST-SCNN and the depth map obtained by the ZED 2, the image of the environment in front of the visually impaired user is split into seven equal divisions. A walkability confidence value for each division is computed, and a voice prompt is played to guide the user toward the most appropriate direction such that the visually impaired user can navigate a safe path on the sidewalk, avoid any obstacles, or walk on the crosswalk safely. Furthermore, the obstacle in front of the user is identified by the network YOLOv5s proposed by Jocher, G. et al. Finally, we provided the proposed assistive system to a visually impaired person and experimented around an MRT station in Taiwan. The visually impaired person indicated that the proposed system indeed helped him feel safer when walking outdoors. The experiment also verified that the system could effectively guide the visually impaired person walking safely on the sidewalk and crosswalks.

Localization for the indoor aerial robot remains a challenging issue because global positioning system (GPS) signals often cannot reach several buildings. In previous studies, navigation of mobile robots without the GPS required the registration of building maps beforehand. This paper proposes a novel framework for addressing indoor positioning for unmanned aerial vehicles (UAV) in unknown environments using a camera. First, the UAV attitude is estimated to determine whether the robot is moving forward. Then, the camera position is estimated based on optical flow and the Kalman filter. Semantic segmentation using deep learning is carried out to get the position of the wall in front of the robot. The UAV distance is measured using the comparison of the image size ratio based on the corresponding feature points between the current and the reference of the wall images. The UAV is equipped with ultrasonic sensors to measure the distance of the UAV from the surrounded wall. The ground station receives information from the UAV to show the obstacles around the UAV and its current location. The algorithm is verified by capture the images with distance information and compared with the current image and UAV position. The experimental results show that the proposed method achieves an accuracy of 91.7% and a computation time of 8 frames per second (fps).

Detection of moving objects by unmanned aerial vehicles (UAVs) is an important application in the aerial transportation system. However, there are many problems to be handled such as high-frequency jitter from UAVs, small size objects, low-quality images, computation time reduction, and detection correctness. This paper considers the problem of the detection and recognition of moving objects in a sequence of images captured from a UAV. A new and efficient technique is proposed to achieve the above objective in real time and in real environment. First, the feature points between two successive frames are found for estimating the camera movement to stabilize sequence of images. Then, region of interest (ROI) of the objects are detected as the moving object candidate (foreground). Furthermore, static and dynamic objects are classified based on the most motion vectors that occur in the foreground and background. Based on the experiment results, the proposed method achieves a precision rate of 94% and the computation time of 47.08 frames per second (fps). In comparison to other methods, the performance of the proposed method surpasses those of existing methods.

Currently, the anti‐viral therapy has been extensively utilised to reduce the viral burden and switch off certain infectious sources for hepatitis B virus (HBV) infected patients in clinical treatment. Several pieces of existing evidence have demonstrated that large‐scale coverage with anti‐viral therapy has obtained a certain great contribution in hygiene and disease control. In this study, an anti‐HBV mathematical model is considered and its control strategy of the drug treatment is designed. Based on the Lyapunov theory, this study derives three main theorems to propose three different control strategies, respectively, for drug treatments and , such that all states of the anti‐HBV model can finally converge to the infection‐free equilibrium point asymptotically. Especially, the designed drug treatment or is not a fixed value, but it is time‐varying and dependent on states. In Theorem 1, the single drug treatment without is synthesised. Theorem 2 considers the single drug treatment without . In Theorem 3, the combination therapy of and is designed. Finally, there are several simulations to show that the proposed combination therapy is much more effective to cure HBV infected patients than the drug treatment with solely single or single .

Antiviral therapy widely used in HIV treatment has obtained some achievements in public health. In this study, an HIV infection dynamic mathematical model is considered and three types of control strategy for this model are designed. On the basis of Lyapunov theory, we propose three main theorems in which three types of control strategy for drug treatments u1(t) or/and u2(t) are designed to push all states to the infection-free equilibrium point E1. Theorem 1 or Theorem 2 proposes the monotherapy with u1(t) only or with u2(t) only. Combination therapy with u1(t) and u2(t) together is constructed in Theorem 3. In these three theorems, u1(t) or/and u2(t) can be a function of states or a fixed constant within a certain range. Finally, there are several simulation results to illustrate that the proposed controls are effective to treat the HIV infection. In the simulation section, there is a detailed discussion about the state responses with the proposed three types of control strategy for drug treatments.

In human immunodeficiency virus (HIV) infection, many factors may influence the counts of healthy cells, immune cells and viruses. Drug treatment design for the HIV dynamic system is a valuable subject to be studied. This study considers an HIV dynamic system model with some unknown parameters and unmeasurable CD8 + T cell count and proposes a switching control strategy to force all states of the system to achieve a healthy status. It is a switching form with two different drug therapies and is designed based on the Lyapunov function theory such that the states of the HIV system approach the health equilibrium asymptotically without the influence of unknown parameters and unmeasurable cell counts. The values of all states and drug concentrations are assured to be positive in the control process so that the control strategy can satisfy the actual treatment requirements. Finally, a numerical example is illustrated to show the effectiveness of the proposed control strategy.

This paper proposes a novel approach for designing the observer and controller for Takagi-Sugeno (T-S) fuzzy systems with an Enlarged Class of Disturbances. The class of the disturbances in this study is significantly enlarged because they do not need to satisfy the constraints such as bounded, generated from an exogenous system, or finite derivative is equal to zero. First, the state/disturbance observer is proposed to estimate the un-measurable states and disturbances, simultaneously. Then the controller incorporating with the observer is synthesized to attenuate the disturbances completely and stabilize the T-S fuzzy systems. On the basis of the Lyapunov function and linear matrix inequality (LMI) technique, the sufficient conditions for the synthesis of the observer and controller are derived in the main theorems. Finally, the effectiveness of the proposed method is verified via two examples.

The controller design for a polynomial system with uncertainties and some/all un‐measurable states is a hard issue for researchers. Furthermore, if there is no information for the upper bounds of uncertainties, the design work will be more challenging. In order to overcome these difficulties, this study proposes a novel approach based on the unknown input method and disturbance observer to synthesise an observer‐based controller for the mentioned system such that the states are estimated and stability of the system is achieved asymptotically. First, based on the unknown input method, let the uncertainties of the polynomial system be regarded as the disturbance, and then a new observer is proposed to estimate the disturbance and states simultaneously. Next, the observer‐based controller is synthesised with suitable parameters which are computed from the conditions in the main theorems. Those parameters are obtained by using Matlab tool to solve a set of polynomial linear matrix inequalities which are derived in the main theorems too. With the aid of Lyapunov theory and sum‐of‐square technique, the above two main theorems are derived. It is noted that this study allows the uncertainty bounds are unknown and some/all states are un‐measurable. Finally, two examples are presented to show the effectiveness of the proposed method.

Purpose: Based on laser-range-finder (LRF) sensing, the control design of location and orientation stabilization for the mobile robot is investigated. However, the practical limitation of the LRF sensing is usually ignored in the control design, which leads to incorrect localization and unexpected control results. The purpose of this study is to design the fuzzy controller subject to the practical limitation on the LRF-based localization for a differentially driven wheeled mobile robot. Design/methodology/approach: First, the Takagi–Sugeno (T-S) fuzzy model is derived from the polar kinematic model of a differentially driven mobile robot. Then, the fuzzy controller is designed to the derived T-S fuzzy kinematic model in accordance with the Lyapunov stabilization theorem. The derived Lyapunov stabilization conditions for the fuzzy control design are expressed as the linear matrix inequality (LMI) form and effectively solved by LMI tools. The practical limitation on the LRF-based localization is also expressed as the LMI form and simultaneously solved with the control design. Finding: The location and posture stabilization experiments are carried out on a mobile robot with LRF-based localization to prove the effectiveness of the proposed T-S fuzzy model-based control design. Furthermore, the ground truth experiment evaluates the accuracy of LRF-based localization. Originality/value: The contribution of this study is to develop the fuzzy control law for a differentially driven wheeled mobile robot under the practical limitation on LRF-based localization. The proposed control design can be applied to other robots with practical limitations on the sensors.