A5088964798- Robotic Path Planning Algorithms
- Adaptive Control of Nonlinear Systems
- Aerospace and Aviation Technology
- Control Systems and Identification
- Robotics and Sensor-Based Localization
- Advanced Algorithms and Applications
- Robotic Locomotion and Control
- Advanced Control Systems Optimization
- Neural Networks and Applications
- Control and Dynamics of Mobile Robots
- Real-time simulation and control systems
- Fuzzy Logic and Control Systems
- Vehicle Dynamics and Control Systems
- Video Surveillance and Tracking Methods
- Simulation and Modeling Applications
- Fault Detection and Control Systems
- Advanced Control Systems Design
- Robot Manipulation and Learning
- Autonomous Vehicle Technology and Safety
- Water Quality Monitoring Technologies
- Industrial Technology and Control Systems
- IoT-based Smart Home Systems
- Smart Parking Systems Research
- Target Tracking and Data Fusion in Sensor Networks
- Adaptive Dynamic Programming Control
National Taiwan Ocean University
2016-2025
Weatherford College
2021
University of Missouri
1998-2002
This correspondence presents a new approach that utilizes evolutionary computation and proportional-integral differential (PID) control to multi-input multioutput (MIMO) nonlinear system. is demonstrated through laboratory helicopter called the twin rotor MIMO system (TRMS). The goals of are stabilize TRMS in significant cross-couplings, reach desired position, track specified trajectory efficiently. proposed scheme includes four PID controllers with independent input. In order reduce total...
This study proposes using unmanned aerial vehicles (UAVs) to carry out tasks involving path planning and obstacle avoidance, explore how improve work efficiency ensure the flight safety of drones. One applications under consideration is aquaculture cage detection; net-cages used in sea-farming are usually numerous scattered widely over sea. It necessary save energy consumption so that drones can complete all detections return their base on land. In recent years, application reinforcement...
This study proposes a marine trash detection system based on unmanned aerial vehicles (UAVs) and aims to replace manpower with UAVs detect efficiently provide information government agencies regarding real-time pollution. Internet technology computer–machine interaction were applied in this study, which involves the deployment of drone’s onboard computer for calculations. Images provided train modified YOLO model (You Look Only Once networks). The UAV was shown be able fly along predefined...
This study applies path planning and obstacle avoidance to drone control for conducting riverbank inspections. Given that the river’s surrounding environments are often windy filled with overgrown branches unknown obstacles, this improves enable drones complete inspection tasks using planned optimal route. Multiple used larger-scale areas reduce time consumption increase efficiency. Regarding planning, A* algorithm is improved a grid-based approach. For avoidance, depth cameras installed on...
This paper presents an adaptive neural network, designed to improve the performance of conventional automatic landing systems (ALS). Real-time learning was applied train network using gradient-descent error function adaptively update weights. Adaptive rates were obtained through analysis Lyapunov stability guarantee convergence learning. In addition, we a DSP controller VisSim/TI C2000 Rapid Prototyper develop embedded control system and establish on-line real-time control. Simulations show...
Abstract Purpose This paper presents our team’s results to establish an AIoT smart cage culture management system. Methods According the built system, farmed field information is transmitted data platform of Ocean Cloud, and all collected analysis can be applied after bigdata analysis. Results system successfully integrates AI IoT technologies in culture. Using underwater biological images feeding as examples, this explains how into a feasible framework that constantly acquire about health...
Applications of drones in the military and daily life have increased recent years. However, it is necessary to obstacle avoidance capability. Path planning also needed for automated tasks. In this study, path based on a reinforcement learning algorithm are implemented an unmanned aerial vehicle (UAV). For global planning, Q-learning utilized find between waypoints first. Then local system applied train UAV by Deep Microsoft AirSim unreal environment.
Neural network applications to aircraft automatic landing control are presented. Conventional systems (ALSs) can provide a smooth landing, which is essential the comfort of passengers. However, these work only within specified operational safety envelope. When conditions beyond envelope, such as turbulence or wind shear, they often cannot be used. The objective this paper investigate use neural networks in ALSs and make more intelligent. Current flight law adopted intelligent controller...
This paper presents an application of optical word recognition and fuzzy control to a smartphone automatic test system. The system consists robot arm two webcams. After the words from panel that represent commands are recognized by system, performs corresponding actions smartphone. One webcams is utilized capture on screen panel, other recognize tested method image processing based Red-Green-Blue (RGB) Hue-Saturation-Luminance (HSL) color spaces reduce influence light. Fuzzy theory used in...
Abstract Aircraft landing control based on fuzzy modelling networks is presented. The proposed scheme uses a controller combined with linearized inverse aircraft model. A multi-layered neural network used as the controller, providing signals at each stage of aircraft-landing phase. algorithm to train Backpropagation Through Time. model provides error that will be back-propagate through stage. objective this study improve performance conventional automatic systems. simulation results are...
This paper presents an intelligent control scheme which utilizes a single neuron PID controller to experimental propeller setup called the twin rotor multi-input multi-output system (TRMS). The objective is make TRMS move quickly and accurately desired attitudes. pitch angle azimuth in conditions of decoupled cross-coupled between vertical horizontal axes are considered. It difficult design suitable because influence two nonlinear movement. For easy demonstrate movement on plane separately,...
Many rivers are polluted by trash and garbage that can affect the environment. Riverbank inspection usually relies on workers of environmental protection office, but sometimes places unreachable. This study applies unmanned aerial vehicles (UAVs) to perform task, which significantly relieve labor work. Two UAVs used cover a wide area riverside capture riverbank images. The images from different stitched using scale-invariant feature transform (SIFT) algorithm. Static dynamic image stitching...
A neural network architecture is developed for the gait synthesis of a five-link biped walking robot. The learning scheme uses multilayered feedforward combined with linearized inverse model. It can generate by giving reference trajectory which defines desired in several stages. algorithm used to train known as back-propagation time-delay or so-called backpropagation through time. three-layered controller, it provides control signals each stage gait. model calculates error will be back...
This paper presents a control problem involving an experimental propeller setup that is called the twin rotor multi-input multi-output system (TRMS). The objective to make beam of TRMS move quickly and accurately desired attitudes, both pitch angle azimuth in condition decoupling between two axes. It difficult design suitable controller because influence axes nonlinear movement. For easy demonstration vertical horizontal separately, decoupled by main tail rotor. An intelligent scheme which...
The purpose of this study is to integrate image processing techniques, fuzzy theory, wireless communications, and smartphone a wheeled mobile robot (WMR) for real-time object recognition, tracking indoor surveillance. Image control algorithms are coded by C# language. Zigbee utilized transmit command signals between webcam, WMR, computer. WMR uses webcam capture its surroundings. Classification color features based on Hue-Saturation-Value (HSV) space. calculates the relative position target...
In this study, we apply laser and infrared sensors to a wheeled mobile robot (WMR) for simultaneous localization mapping (SLAM).The utilizes measurement sensor detect obstacles identify unknown environments.Fuzzy theory the iterative closest point (ICP) algorithm are applied control design.The proposed scheme can WMR movement along walls avoid obstacles.In addition, k-dimensional (k-D) tree is used reduce computation time achieve real-time positioning.By calculating rotation translation...
Neural network applications to aircraft automatic landing control based on linearized inverse model are presented. Conventional systems can provide a smooth which is essential the comfort of passengers. However, these work only within specified operational safety envelope. When conditions beyond envelope, such as turbulence or wind shear, they often cannot be used. The objective this study investigate use neural networks with in and make more intelligent. Current flight law adopted...
This paper presents an intelligent automatic landing system that uses a time delay neural network controller and linearized inverse aircraft model to improve the performance of conventional systems. The airplane is enabled only under limited conditions. If severe wind disturbances are encountered, pilot must handle due limits system. In this study, learning-through-time process used in training. Simulation results show can act as experienced guide safe disturbance environments without using...
To determine the relationships between exposure to environmental contaminants in water and chronic kidney disease (CKD), we investigated associations of 61 attributes with prevalence CKD End-Stage Renal Disease (ESRD) using data from 2005 2011 all 22 counties cities main island Taiwan. We acquired patient information Taiwan Longitudinal Health Insurance Database calculate age-standardized ESRD rates linked patients’ residences quality monitoring data, which were sampled periodically for a...
In this study, we have crafted an innovative methodology that represents a groundbreaking synthesis of deep learning techniques with cooperative game theory. use the accuracy data prediction by different LSTM models as measurement index and assign corresponding weights through Shapley value calculation method to construct more accurate predictive analysis model. We improved regulation model calibrate long short-term memory (LSTM) neural network using historical freight predict China...