As the first part, this paper presents an overview on the existing works on fault detection and diagnosis (FDD) and fault-tolerant control (FTC) for unmanned rotorcraft systems. Considered faults include actuator and sensor faults for single and multi-rotor systems. As the second part, several FDD and FTC techniques developed recently at the Networked Autonomous Vehicles Lab of Concordia University are detailed along with experimental application to a unique and newly developed quadrotor helicopter testbed.

Development of advanced FDD and FTC techniques with application to an unmanned quadrotor helicopter testbed

Plant diseases and pests are important factors determining the yield and quality of plants. Plant diseases and pests identification can be carried out by means of digital image processing. In recent years, deep learning has made breakthroughs in the field of digital image processing, far superior to traditional methods. How to use deep learning technology to study plant diseases and pests identification has become a research issue of great concern to researchers. This review provides a definition of plant diseases and pests detection problem, puts forward a comparison with traditional plant diseases and pests detection methods. According to the difference of network structure, this study outlines the research on plant diseases and pests detection based on deep learning in recent years from three aspects of classification network, detection network and segmentation network, and the advantages and disadvantages of each method are summarized. Common datasets are introduced, and the performance of existing studies is compared. On this basis, this study discusses possible challenges in practical applications of plant diseases and pests detection based on deep learning. In addition, possible solutions and research ideas are proposed for the challenges, and several suggestions are given. Finally, this study gives the analysis and prospect of the future trend of plant diseases and pests detection based on deep learning.

/pdf/plant-diseases-and-pests-detection-based-on-deep-learning-a-518qywcd1i.pdf

Plant diseases and pests detection based on deep learning: a review.

Faults or failures are inevitable to occur and their prompt detection and isolation are essential for the dependability of various systems and for avoiding damages to the system itself, persons and the environment. Therefore, the safety of helicopter platforms have attracted the attention of many researchers in the past two decades. In order to deal with these problems, this paper presents an overview of the recent development and current researches in the field of fault diagnosis, including analytical/model-based, signal processing-based and knowledge-based techniques, and also passive/active fault- tolerant control approaches. Among various helicopters, single-rotor aerial vehicles, i.e. manned helicopters, unmanned helicopters, two and three degree-of-freedom unmanned helicopter experimental platforms, are considered for providing an overall picture of the fault diagnosis and fault-tolerant control approaches based on the review of journal articles in last two decades, conference articles in last several years and some books.

A Review on Fault Diagnosis and Fault Tolerant Control Methods for Single-rotor Aerial Vehicles

In the past two decades, the reliability and safety of helicopters have attracted the attention of many researchers. Especially with the development of Unmanned Helicopters, Fault Diagnosis approaches were used for manned or unmanned helicopter platforms. In this paper, a brief state-of-the-art on Fault Diagnosis approaches is presented for helicopters, including analytical model-based, signal processing-based and knowledge-based Fault Diagnosis approaches. A short description of fault types is presented before the main part of the review.

A literature review on Fault Diagnosis methods for manned and unmanned helicopters

With the development of unmanned helicopters, the dependability of helicopters have attracted more and more attention of many researchers. In order to deal with these problems, fault diagnosis and fault tolerant control methods were used for manned or unmanned helicopter platforms. This paper presents an overview of the existing works on fault diagnosis, including analytical/model-based, signal processing-based and knowledge-based techniques, and passive/active fault tolerant control approaches for helicopters mainly with single rotor. Before the main part of the review, a short description of fault classification is presented. Compared with the former work, this review contains some in-depth discussion of various fault diagnosis techniques and a survey for fault tolerant control methods.

Fault diagnosis and fault tolerant control methods for manned and unmanned helicopters: A literature review

Early and accurate detection of plant diseases is necessary to maximize crop yield. The artificial intelligence based deep learning method plays a vital role in the detection of the diseases using a huge volume of plant leaves images. However, to detect disease with small datasets is a challenging task using deep learning methods. Transfer learning is one of the popular deep learning methods used to accurately detect plant disease with minimal plant image data. In this paper, the transfer learning-based deep convolution neural network model to identify tomato leaf disease has proposed. The model performs detection of disease using real-time images and stored tomato plant images. Furthermore, the performance of the proposed model is evaluated using adaptive moment estimation (Adam), stochastic gradient descent (SGD), and RMSprop optimizers. The experimental result demonstrates that the proposed model using the transfer learning approach is effective in automated tomato leaf disease classification. The Adam optimizer achieves better accuracy compared with SGD and RMSprop optimizers.

Automated tomato leaf disease classification using transfer learning-based deep convolution neural network

Accurate and fast tomato plant disease identification is significant to enhance its sustainable agricultural productivity. In the conventional technique, human experts in the field of agriculture have been accommodated to find out the anomalies in tomato plants caused by pests, diseases, climatic conditions, and nutritional deficiencies. Automatic tomato leaf disease identification is initially solved through conventional image processing and machine learning approaches which result in less accuracy. In order to produce greater prediction accuracy, deep learning-based classification is introduced. This paper provides an overall review of recent work performed in the field of tomato leaf disease identification using image processing, machine learning, and deep learning approaches. And also discuss both public and private datasets available to detect tomato leaf disease, methods employed, and adopted deep learning frameworks. Consequently, suggestions are provided to figure out the appropriate techniques in order to obtain the better prediction accuracy. Finally, the challenges encountered in implementing the machine learning and deep learning models are discussed.

Artificial intelligence in tomato leaf disease detection: a comprehensive review and discussion

In this paper additive fault detection and isolation method coupled with fault tolerant control architecture are developed in order to deal with component faults for a rotorcraft based unmanned aerial vehicle (RUAV). The failure considered is malfunction with internal components of the helicopter which occurs during the maneuvers: rotor angular rate variations, etc. These faults lead from trivial to catastrophic damage of the system. The proposed fault detection and reconfiguration control is based on a parameter estimation approach which drives a reconfigurable control system (RCS) build with the Pseudo-inverse method. The complete setup is implemented under Hardware-in-the-loop-simulation (HILS). The PC104 board with QNX RTOS platform is used for simulation. Simulation results illustrate the efficiency and effectiveness of the proposed approach.

Fault tolerant controller design for component faults of a small scale unmanned aerial vehicle

Over the past decades, substantial research has been undertaken in the design of intelligent architecture for the rotorcraft-based unmanned aerial vehicles (RUAV). Designing intelligent architecture is a challenging problem because future RUAVs are utterly autonomous and their performance is comparable with that of manned vehicles. This paper deals with the design and development of a layered architectural framework that addresses the issue arising in autonomous intelligent control systems. The architecture consists of two layers. The high-level layer is occupied by planning routines. In this level, the waypoints and mission tasks from the command center are executed. The function of the low-level layer is to stabilize the flight and follow the commanded trajectory from the upper layer. These layers integrate the following functionalities: (1) waypoint navigation and control, which includes auto-landing; (2) obstacle detection and avoidance; (3) fault detection and identification; and (4) system reconfiguration in two levels (high-level and low-level controllers). The resulting layered architecture is discussed in detail. Moreover, the novel fault detection and identification method is developed to address multiplicative and additive faults. A testing environment for RUAV is developed to validate this architecture. Complete setup is carried out using an embedded board run under a real-time operating system. The algorithms are tested and evaluated using hardware-in-the-loop simulation (HILS). The simulation result proves that the proposed architecture demonstrates the desired efficiency and reliability.

Reconfigurable Intelligent Control Architecture of a Small-Scale Unmanned Helicopter

The aggressive waves of ongoing world-wide virus pandemics urge us to conduct further studies on the performability of local computing infrastructures at hospitals/medical centers to provide a high level of assurance and trustworthiness of medical services and treatment to patients, and to help diminish the burden and chaos of medical management and operations. Previous studies contributed tremendous progress on the dependability quantification of existing computing paradigms (e.g., cloud, grid computing) at remote data centers, while a few works investigated the performance of provided medical services under the constraints of operational availability of devices and systems at local medical centers. Therefore, it is critical to rapidly develop appropriate models to quantify the operational metrics of medical services provided and sustained by medical information systems (MIS) even before practical implementation. In this paper, we propose a comprehensive performability SRN model of an edge/fog based MIS for the performability quantification of medical data transaction and services in local hospitals or medical centers. The model elaborates different failure modes of fog nodes and their VMs under the implementation of fail-over mechanisms. Sophisticated behaviors and dependencies between the performance and availability of data transactions are elaborated in a comprehensive manner when adopting three main load-balancing techniques including: (i) probability-based, (ii) random-based and (iii) shortest queue-based approaches for medical data distribution from edge to fog layers along with/without fail-over mechanisms in the cases of component failures at two levels of fog nodes and fog virtual machines (VMs). Different performability metrics of interest are analyzed including (i) recover token rate, (ii) mean response time, (iii) drop probability, (iv) throughput, (v) queue utilization of network devices and fog nodes to assimilate the impact of load-balancing techniques and fail-over mechanisms. Discrete-event simulation results highlight the effectiveness of the combination of these for enhancing the performability of medical services provided by an MIS. Particularly, performability metrics of medical service continuity and quality are improved with fail-over mechanisms in the MIS while load balancing techniques help to enhance system performance metrics. The implementation of both load balancing techniques along with fail-over mechanisms provide better performability metrics compared to the separate cases. The harmony of the integrated strategies eventually provides the trustworthiness of medical services at a high level of performability. This study can help improve the design of MIS systems integrated with different load-balancing techniques and fail-over mechanisms to maintain continuous performance under the availability constraints of medical services with heavy computing workloads in local hospitals/medical centers, to combat with new waves of virus pandemics.

Performability Evaluation of Load Balancing and Fail-over Strategies for Medical Information Systems with Edge/Fog Computing Using Stochastic Reward Nets

Vishnu Kumar Kaliappan

Papers

Automated tomato leaf disease classification using transfer learning-based deep convolution neural network

Artificial intelligence in tomato leaf disease detection: a comprehensive review and discussion

Fault tolerant controller design for component faults of a small scale unmanned aerial vehicle

Reconfigurable Intelligent Control Architecture of a Small-Scale Unmanned Helicopter

Performability Evaluation of Load Balancing and Fail-over Strategies for Medical Information Systems with Edge/Fog Computing Using Stochastic Reward Nets