Taiwan Association of Engineering and Technology Innovation: E-Journals
Not a member yet
887 research outputs found
Sort by
Robust LPV Control for Attitude Stabilization of a Quadrotor Helicopter under Input Saturations
Get PDFThis article investigates the robust stabilization of the rotational subsystem of a quadrotor against external inputs (disturbances, noises, and parametric uncertainties) by the LFT-based LPV technique. By establishing the LPV attitude model, the LPV robust controller is designed for the system. The weighting functions are computed by Cuckoo Search, a meta-heuristic optimization algorithm. Besides, the input saturations are also taken into account through the Anti-Windup compensation technique. Simulation results show the robustness of the closed-loop system against disturbances, measurement noises, and the parametric uncertainties
Synthesis of Formation Control Systems for Multi-Agent Systems under Control Gain Perturbations
Get PDFThis paper proposed a linear matrix inequality (LMI)-based design method of non-fragile guaranteed cost controllers for multi-agent systems (MASs) with leader-follower structures. In the guaranteed cost control approach, the resultant controller guarantees an upper bound on the given cost function together with asymptotical stability for the closed-loop system. The proposed non-fragile guaranteed cost control system can achieve consensus for MASs despite control gain perturbations. The goal is to develop an LMI-based sufficient condition for the existence of the proposed non-fragile guaranteed cost controller. Moreover, a design problem of an optimal non-fragile guaranteed cost controller showe that minimizing an upper bound on the given quadratic cost function can be reduced to constrain a convex optimization problem. Finally, numerical examples were given to illustrate the effectiveness of the proposed non-fragile controller for MASs
Evaluation of Pan-Sharpening Techniques Using Lagrange Optimization
Get PDFEarth’s observation satellites, such as IKONOS, provide simultaneously multispectral and panchromatic images. A multispectral image comes with a lower spatial and higher spectral resolution in contrast to a panchromatic image which usually has a high spatial and a low spectral resolution. Pan-sharpening represents a fusion of these two complementary images to provide an output image that has both spatial and spectral high resolutions. The objective of this paper is to propose a new method of pan-sharpening based on pixel-level image manipulation and to compare it with several state-of-art pansharpening methods using different evaluation criteria. The paper presents an image fusion method based on pixel-level optimization using the Lagrange multiplier. Two cases are discussed: (a) the maximization of spectral consistency and (b) the minimization of the variance difference between the original data and the computed data. The paper compares the results of the proposed method with several state-of-the-art pan-sharpening methods. The performance of the pan-sharpening methods is evaluated qualitatively and quantitatively using evaluation criteria, such as the Chi-square test, RMSE, SNR, SD, ERGAS, and RASE. Overall, the proposed method is shown to outperform all the existing methods
Application of Recent Developments in Deep Learning to ANN-based Automatic Berthing Systems
Get PDFPrevious studies on Artificial Neural Network (ANN)-based automatic berthing showed considerable increases in performance by training ANNs with a set of berthing datasets. However, the berthing performance deteriorated when an extrapolated initial position was given. To overcome the extrapolation problem and improve the training performance, recent developments in Deep Learning (DL) are adopted in this paper. Recent activation functions, weight initialization methods, input data-scaling methods, a higher number of hidden layers, and Batch Normalization (BN) are considered, and their effectiveness has been analyzed based on loss functions, berthing performance histories, and berthing trajectories. Finally, it is shown that the use of recent activation and weight initialization method results in faster training convergence and a higher number of hidden layers. This leads to a better berthing performance over the training dataset. It is found that application of the BN can overcome the extrapolated initial position problem
The Autonomous Shopping-Guide Robot in Cashier-Less Convenience Stores
Get PDFIn recent years, several cashier-less convenience stores have appeared, including Amazon’s. A store without cashiers will become a trend in the near future. To substitute the employee in traditional stores, this research proposes and designs an autonomous shopping cart robot to guide customers’ purchase according to the requested shopping list to enhance their shopping experience. The core techniques of the system are the autonomous driving robot, the traffic control center for robots and the dynamic route planning algorithm. The robot is a self-propelled vehicle developed by ROS (Robot Operating System), in which we achieve the automatic driving via image recognition, April tag identification and driving direction guidance from the path planning and traffic control services. This enables the robot to lead customers to find their commodities following the preplanned route. In conjunction with the vocal service, the robot can notify the customer when arriving at each commodity, he or she plans to buy. We also design a light APP for customers to easily set up and manage their shopping list, call for the robotic shopping cart’s help, and interact with the shopping cart robot. To enhance the shopping experience of customers, we design the dynamic route planning genetic algorithm to dynamically plan the shopping route according to the customer’s request and the traffic condition. Experiments show that our genetic algorithm can provide the most stable performance and always get efficient shopping route planning in a limited time compared to other methods
Estimation of the PCB Production Process Using a Neural Network
Get PDFPrinted Circuit Boards (PCB) are an integral part of all electronic products, and the production process for printed circuit boards is quite complex. As the life cycle of electronic products becomes shorter and shorter, and the precision and signal bandwidth of electronic products become higher and higher, the manufacturing process of printed circuit boards is further complicated. Therefore, how to pre-evaluate the production difficulty before starting the production will effectively increase the efficiency and quality of printed circuit board production.
Gerber file is the most commonly used data format for the printed circuit board industry. This file contains most of the parameters required for the manufacture of printed circuit boards. Therefore, this study uses a neural network to evaluate new PCB products before they are produced through the production parameters that are more influential in the PCB manufacturing process. This makes it possible to evaluate the difficulty and the required production process before the new PCB product is produced. This will be very beneficial for the PCB production schedule, quality control, and cost
A Multichannel MAC Protocol for IoT-enabled Cognitive Radio Ad Hoc Networks
Get PDFCognitive radios have the ability to dynamically sense and access the wireless spectrum, and this ability is a key factor in successfully building Internet-of-Things (IoT)-enabled mobile ad hoc networks. This paper proposes a contention-free token-based multichannel MAC protocol for IoT-enabled Cognitive Radio Ad Hoc Networks (CRAHNs). In this, secondary users of CRAHNs detect activity on the wireless spectrum and then access idle channels licensed by primary users. CRAHNs are divided into clusters, and the channel to use for transmission is determined dynamically from the probability of finding idle primary-user channels. The token-based MAC window size is adaptive, with adjustment according to actual traffic, which reduces both end-to-end MAC contention delay and energy consumption. High throughput and spatial reuse of channels can also be achieved using a dynamic control channel and dynamic schemes for contention windows. We performed extensive simulations to verify that the proposed method can achieve better performance in mobile CRAHNs than other MAC schemes can
Efficient RSU Selection Approaches for Load Balancing in Vehicular Ad Hoc Networks
Get PDFDue to advances in wireless communication technologies, wireless transmissions gradually replace traditional wired data transmissions. In recent years, vehicles on the move can also enjoy the convenience of wireless communication technologies by assisting each other in message exchange and form an interconnecting network, namely Vehicular Ad Hoc Networks (VANETs). In a VANET, each vehicle is capable of communicating with nearby vehicles and accessing information provided by the network. There are two basic communication models in VANETs, V2V and V2I. Vehicles equipped with wireless transceiver can communicate with other vehicles (V2V) or roadside units (RSUs) (V2I). RSUs acting as gateways are entry points to the Internet for vehicles. Naturally, vehicles tend to choose nearby RSUs as serving gateways. However, due to uneven density distribution and high mobility nature of vehicles, load imbalance of RSUs can happen. In this paper, we study the RSU load-balancing problem and propose two solutions. In the first solution, the whole network is divided into sub-regions based on RSUs’ locations. A RSU provides Internet access for vehicles in its sub-region and the boundaries between sub-regions change dynamically to adopt to load migration. In the second solution, vehicles choose their serving RSUs distributedly by taking their future trajectories and RSUs’ loading information into considerations. From simulation results, the proposed methods can improve packet delivery ratio, packet delay, and load balance among RSUs
The Design of Mold Clamping Mechanisms with Two Dead-Position Configurations
Get PDFBlow-molding machines are divided into two types: (1) rotary type blow-molding machine, and (2) linear type blow-molding machine. Good kinematic characteristics of mold clamping mechanism is the goal pursued by manufacturers of blow-molding machine. Traditionally, the mold clamping mechanism has only one dead-position configuration for the mold in the closed position. When the mold is in the open position, the mold clamping mechanism does not have dead-position configuration. This study focuses on the design of mold clamping mechanisms with two dead-position configurations when the mold is in open and closed positions. Firstly, with reference to the existing patents, based on the Yan's creative design methodology, four design concepts of mold clamping mechanisms are synthesized. Then, according to the studies on kinematics of mechanisms, three mold clamping mechanisms with two dead-position configurations are also synthesized. The results of this paper will enhance the research and development (R&D) capability of the mold clamping mechanisms to improve industrial competitiveness
Shear Behavior of RC Deep Beam Strengthened by V-Shaped External Rods
Get PDFThis research investigated the shear strengthening technique of Reinforced Concrete (RC) deep beams using a V-shaped external rod system. Shear behavior, the stress in an external rod, and the shear capacity at the diagonal shear failure of a strengthened beam were focused mainly. Experimental tests of control and two strengthened beams were carried out to observe the effect of the external rod on shear behavior of RC deep beam. A theoretical approach to compute the stress in the external rod and the nominal strength of the strengthened beam in the diagonal shear failure were examined based on the experimental test results and verified using Finite Element Method (FEM) in ABAQUS. The computed nominal shear strength of the strengthened beam was 10% higher than the experimental test. The strengthening technique shifted the brittle shear failure to ductile shear failure and improved the performance of RC deep beam