A Novel Online Data-Driven Algorithm for Detecting UAV Navigation Sensor Faults

Sun, Rui; Cheng, Qi; Wang, Guanyu; Ochieng, Washington Yotto

doi:10.3390/s17102243

Cited by 69 publications

(31 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…where v i k is the ith row element of the new residual vector v k and it's the difference between∇ and v i k . According to (14), v i k ≈ 0can be established. From the expressions ofṽ k and v k , it can be seen that there is nearly no new measurement information added in the measurement update of the fault adaptation method compared with the fault isolation method.…”

Section: Performance Comparison Of the Two Methodsmentioning

confidence: 99%

“…Besides, fault detection based on artificial intelligence has attracted more attention and been widely studied. Adaptive Neuron Fuzzy Inference System (ANFIS)-based approach was presented for detecting the navigation sensor faults in UAVs [14]. Reference [15] proposed a novel fault detection method based on Gaussian Process Regression (GPR) to improve the detection performance for gradual fault.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Fault Isolation and System Reconfiguration Method for GNSS/INS Integration

et al. 2020

View full text Add to dashboard Cite

Appropriate fault isolation and system reconfiguration scheme is of great significance to ensure the reliability and precision of the tightly coupled global navigation satellite system/inertial navigation system (GNSS/INS). Currently, the commonly used methods include fault isolation method and fault adaptation method. However, the performance of them is not compared and analyzed fully in the existing literatures. In this paper, the principle of them is analyzed and the performance of them under different conditions is compared in theory firstly. On this basis, to improve the effectiveness and adaptability of fault detection and isolation, an adaptive fault isolation and system reconfiguration method is proposed. The radial basis function neural network (RBFNN) is used to predict the pseudo-GNSS measurement for the measurement reconfiguration. Besides, taking the variety of observation conditions into consideration, an adaptive adjustment criterion is introduced to realize the switch of fault isolation and measurement reconfiguration. The performance of the proposed method is verified and compared with the traditional methods by using the field test data. The results show that the fault isolation method can obtain higher filtering precision than the fault adaptation method in theory, compared with these two methods, the proposed method has better adaptability to the complex environment, and can improve the reliability and precision of the navigation system effectively.

show abstract

Section: Performance Comparison Of the Two Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive Fault Isolation and System Reconfiguration Method for GNSS/INS Integration

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The main characteristics of the first two generations of products include large volume, high power consumption, large error, high cost, single function etc. Sun et al [4] pointed out that, at present, laser positioning is mainly developing towards the direction of miniaturization, low power consumption, high precision, low cost, and multifunction.…”

Section: Literature Reviewmentioning

confidence: 99%

Intelligent Navigation Design of Fully Autonomous Unmanned Aerial Vehicle Based on Laser Positioning

2019

IJOMAM

View full text Add to dashboard Cite

In order to study the intelligent navigation design of fully autonomous UAV (Unmanned Aerial Vehicle) for laser positioning, the laser positioning UAV is mainly discussed. An intelligent navigation design system of fully autonomous UAV based on laser positioning is adopted, which can be used not only for material measurement in large mines and wharfs, but also for mine safety and intelligent mining system. Compared with radar technology, acoustic detection and radio scanning, this method has the characteristics of high efficiency and easy realization, and is more suitable for the detection and positioning of low-speed flying small targets in ultra-low altitude visual range. Finally, the software of the system is programmed, including time, inclination, pendulum, positioning and other measurements. The experimental results are illustrated by flow chart. Laser positioning is designed separately according to the actual needs. The UAV and GPS (Global Position System) positioning sensor system are studied and analysed, and the two are combined to get three-dimensional coordinates. The test results show that the system can basically fulfill the requirement of autonomous range of laser positioning, and provide data support for the later positioning system.

show abstract

“…In this method, the Levenberg-Maquardt algorithm was used to improve the traditional BP neural network, which greatly improves the diagnostic level. In reference [14], a novel data-driven adaptive neuron fuzzy inference system (ANFIS)-based approach was proposed to detect on-board navigation sensor faults in UAVs. The main advantages of this algorithm are that it allows the Kalman filter to estimate real-time model-free residual and ANFIS to build a reliable fault detection system.…”

Section: Introductionmentioning

confidence: 99%

Design and Analysis for Early Warning of Rotor UAV Based on Data-Driven DBN

Xiong

et al. 2019

Electronics

View full text Add to dashboard Cite

The unmanned aerial vehicle (UAV), which is a typical multi-sensor closed-loop flight control system, has the properties of multivariable, time-varying, strong coupling, and nonlinearity. Therefore, it is very difficult to obtain an accurate mathematical diagnostic model based on the traditional model-based method; this paper proposes a UAV sensor diagnostic method based on data-driven methods, which greatly improves the reliability of the rotor UAV nonlinear flight control system and achieves early warning. In order to realize the rapid on-line fault detection of the rotor UAV flight system and solve the problems of over-fitting, limited generalization, and long training time in the traditional shallow neural network for sensor fault diagnosis, a comprehensive fault diagnosis method based on deep belief network (DBN) is proposed. Using the DBN to replace the shallow neural network, a large amount of off-line historical sample data obtained from the rotor UAV are trained to obtain the optimal DBN network parameters and complete the on-line intelligent diagnosis to achieve the goal of early warning as possible as quickly. In the end, the two common faults of the UAV sensor, namely the stuck fault and the constant deviation fault, are simulated and compared with the back propagation (BP) neural network model represented by the shallow neural network to verify the effectiveness of the proposed method in the paper.

show abstract

A Novel Online Data-Driven Algorithm for Detecting UAV Navigation Sensor Faults

Cited by 69 publications

References 16 publications

Adaptive Fault Isolation and System Reconfiguration Method for GNSS/INS Integration

Adaptive Fault Isolation and System Reconfiguration Method for GNSS/INS Integration

Intelligent Navigation Design of Fully Autonomous Unmanned Aerial Vehicle Based on Laser Positioning

Design and Analysis for Early Warning of Rotor UAV Based on Data-Driven DBN

Contact Info

Product

Resources

About