System Identification and Parameter Self-Tuning Controller on Deep-Sea Mining Vehicle
Qi-wang Weng,Jian-min Yang,Qiong-wen Liang,Jing-hang Mao,Xiao-xian Guo
DOI: https://doi.org/10.1007/s13344-023-0005-7
IF: 1.253
2023-03-28
China Ocean Engineering
Abstract:System identification is a quintessential measure for real-time analysis on kinematic characteristics for deep-sea mining vehicle, and thus to enhance the control performance and testing efficiency. In this study, the system identification algorithm, recursive least square method with instrumental variables (IV-RLS), is tailored to model 'Pioneer I', a deep-sea mining vehicle which recently completed a 1305-meter-deep sea trial in the Xisha area of the South China Sea in August, 2021. The algorithm operates on the sensor data collected from the trial to obtain the vehicle's kinematic model and accordingly design the parameter self-tuning controller. The performances demonstrate the accuracy of the model, and prove its generalization capability. With this model, the optimal controller has been designed, the control parameters have been self-tuned, and the response time and robustness of the system have been optimized, which validates the high efficiency on digital modelling for precision control of deep-sea mining vehicles.
engineering, mechanical, civil, ocean,water resources