US 12,233,996 B1
Underwater submersible robot and control method and control apparatus therefor
Airong Liu, Guangzhou (CN); Jiaqiao Liang, Guangzhou (CN); Jiyang Fu, Guangzhou (CN); Jiajian Liang, Guangzhou (CN); Bingcong Chen, Guangzhou (CN); Hai Lin, Guangzhou (CN); Jialin Wang, Guangzhou (CN); Jiawei He, Guangzhou (CN); Fobao Zhou, Guangzhou (CN); Yixiao Zhang, Guangzhou (CN); and Haoxiang Zhou, Guangzhou (CN)
Assigned to GUANGZHOU UNIVERSITY, Guangzhou (CN)
Filed by Guangzhou University, Guangzhou (CN)
Filed on Mar. 28, 2024, as Appl. No. 18/620,130.
Claims priority of application No. 202311524619.8 (CN), filed on Nov. 16, 2023.
Int. Cl. B63G 8/14 (2006.01); B63G 8/00 (2006.01); B63G 8/04 (2006.01); B63G 8/08 (2006.01); G05D 1/49 (2024.01); G05D 1/606 (2024.01); G05D 109/30 (2024.01)
CPC B63G 8/14 (2013.01) [B63G 8/001 (2013.01); B63G 8/04 (2013.01); B63G 8/08 (2013.01); G05D 1/49 (2024.01); G05D 1/606 (2024.01); G05D 2109/38 (2024.01)] 10 Claims
OG exemplary drawing
 
1. An underwater submersible robot, characterized by comprising:
a fuselage;
a detection apparatus, disposed on the fuselage and configured to detect an underwater target;
a propeller assembly, disposed on the fuselage and configured to drive the underwater submersible robot underwater;
a sensing apparatus, disposed on the fuselage and configured to sense an attitude and depth of the fuselage; and
a control apparatus, disposed on the fuselage, connected to the detection apparatus, the sensing apparatus and the propeller assembly, and configured to:
calculate a first difference between a current actual attitude of the fuselage and a desired attitude and a second difference between an actual depth and a desired depth;
input the first difference and the second difference into a set terminal sliding mode surface to obtain an output value of the terminal sliding mode surface;
use the output value as an input of a preset high-order observer, a radial basis function neural network and a terminal sliding mode control law, respectively, and use an output of the high-order observer and an output of the radial basis function neural network as a compensation input of the terminal sliding mode control law to obtain a virtual force outputted by the terminal sliding mode control law;
perform power distribution for each propeller of the propeller assembly on the basis of the virtual force to obtain a propelling force of each propeller; and
control a propelling operation of the propellers on the basis of the propelling force.