Research On MPC-Based Trajectory Tracking Control Of Submarine Cable Inspection Robot

With the increase of human energy demand and the development of ocean technology,the total amount and scale of submarine cable laying have rapidly expanded.In order to ensure the safety and reliability of submarine cables,regular maintenance is necessary to detect and solve potential faults and risks in a timely manner.The use of a cable inspection robot is an effective technical solution for the maintenance of submarine cables.As the motion control system of the cable inspection robot is a complex nonlinear system with multiple inputs and outputs,in order to ensure the accuracy of trajectory tracking and the stability of the control,this paper adopts a model predictive control algorithm(MPC)for trajectory tracking control.The specific research contents of the paper are as follows:(1)In order to obtain an accurate motion model of the cable inspection robot,this paper starts from force analysis and applies rigid body dynamics,hydrodynamics theory,and factors such as restoring force and propeller thrust to derive specific kinematic and dynamic models.On this basis,the motion control model of the robot is subdivided into one to three dimensions to provide a basic theoretical model for subsequent algorithms.(2)For trajectory tracking control of the cable inspection robot,this paper introduces model predictive control,sliding mode control,backstepping control,and PID control algorithms,and introduces the theoretical principles and corresponding controller design for each algorithm.By comparing and analyzing the control performance indicators such as rising time,overshoot,adjustment time,and cumulative error of different algorithms in step response experiments,the unique advantages of the MPC algorithm in convergence speed and stability are verified.(3)This paper provides two different control strategies,linear model predictive control(LMPC)and nonlinear model predictive control(NMPC),for nonlinear motion control models of the cable inspection robot.Based on the two control strategies,the paper carries out model discretization processing,penalty function design,system constraint setting,and optimization solution respectively.Simulation results show that LMPC and NMPC have good control accuracy and stability in trajectory tracking,but LMPC has certain deficiencies in anti-interference.When encountering large disturbances,LMPC can cause the system output to oscillate.Therefore,NMPC is finally selected as the core algorithm of this paper.(4)For the control performance differences of the cable inspection robot under different speeds,this paper designs an MPC control method based on an adaptive prediction time domain strategy.Firstly,multiple sets of simulation experiments with different prediction time domains are carried out under the same speed scenario,and the prediction time domain corresponding to the optimal control effect at each speed is selected.Secondly,a univariate quaternary polynomial is used to fit the relationship between the prediction time domain and speed,and this is used as the basis for selecting the prediction time domain.Finally,the experimental results show that the adaptive prediction time domain MPC has good control accuracy and stability under different speeds,which verifies that the algorithm has strong adaptability and robustness.