Model-free fault diagnosis for autonomous underwater vehicles using sequence Convolutional Neural Network

A survey on tracking control of unmanned underwater vehicles: Experiments-based approach

“…Expensive cost of sensors and actuators: The high cost of accurate sensors and actuators is one of the financial constraints when designing a new robot Harris and Whitcomb (2021). This problem becomes more critical for the case of building a low-cost version of the robots to be deployed for special intervention missions Ji et al (2021). In fact, some advanced sensors are not available at any cost Yuh (2000).…”

Fault diagnosis of AUV based on sliding-mode observer

“…According to equation of motion of six degrees of freedom [4], AUV's equation of motion of five degrees of freedom can be acquired when it doesn't take the disturbances into account while assume gravity and buoyancy reach balance:…”

Design of Underwater Thruster Fault Detection Model Based on Vibration Sensor Data: Generative Adversarial Network-based Fault Data Expansion Approach for Data Imbalance

“…(11) For reliable learning performance with the data-training-based deep learning method, high-quality data in large volumes is a prerequisite. (12) In recent data-driven fault detection and diagnosis studies, Ji et al acquired data by implementing and operating the platform's thruster blade breakage for the fault diagnosis of an autonomous underwater vehicle on the basis of deep learning, (13) and Chu et al validated a fault detection system by referring to a vast dataset acquired in the past. (14) However, if the platform is operated in a fault state for a long time to acquire data, performance degradation of the thruster, accidents, or platform loss may occur, or significant time and cost may be required to acquire a vast stored dataset from the past.…”