Research On Control Of Software Operation Remote Control System Based On Underwater Robot

With the continuous deepening of human exploration of the ocean,the cable remote operated underwater robot(Remotely Operated Vehicle,ROV),which can replace human beings in small,dangerous and unknown environments,is constantly developing as a new technology.This thesis takes the self-developed remote operating system of ROV and software operation hand as the research object,and studies the new method of improving the tracking control accuracy of the end position of the software operation hand and the new method of the overall underwater movement attitude control of ROV and software operation hand.The main research contents are as follows :First of all,this thesis focuses on the new method of improving the tracking control of the end position of the soft working hand,and establishes a kinematics model and a dynamic model for the index finger actuator of the soft working hand.Predictive Control Algorithms.The algorithm first makes a reasonable prediction of the end position of the index finger of the software main hand at the current moment,and then through feedback correction and rolling optimization,the tracking control of the soft hand to the end position of the software main hand is realized,which is verified by simulation,the algorithm effectively improves the tracking effect of the software operator’s end position on the software main hand.Secondly,a new method of underwater movement attitude control of ROV and software operator is studied.When ROV and software operator perform grasping tasks underwater,because their own posture is easily disturbed by ocean currents,a method based on difference is proposed.Evolution combined with simulated annealing algorithm for RBF(DE-SA-RBF)neural network S-plane attitude control algorithm.Aiming at the problem that the input of the RBF neural network is difficult to determine,the differential evolution combined with simulated annealing algorithm(DE-SA)is used to find the optimal input of the RBF neural network;at the same time,the differential evolution(DE)algorithm is prone to fall into local Optimum,the simulated annealing(SA)algorithm is used to optimize the differential evolution algorithm.Therefore,the DE-SA algorithm has a strong ability to escape from local extremum values;it does not depend on the advantages of initial values and has global optimization capabilities.The LM(Levenberg-Marquardt)algorithm is used to train and analyze the RBF neural network,and the optimal output of the RBF neural network is obtained to control the parameters of the S surface to achieve the purpose of controlling the motion posture.The simulation experiment verified the effectiveness of the S-plane attitude control algorithm based on DE-SA-RBF neural network for ROV and software operator’s overall underwater attitude control.Then,a remote operating system for software operations based on underwater robots was developed.The main components include the development of the ROV control system and the remote operation control system for software operations.The ROV control system is divided into a surface control system and an underwater control system.Design and software design;and completed the hardware and software design of the software operating hand remote operation control system.Finally,the experiment and result analysis of the software remote operating system based on the underwater robot were carried out,and the land grasping experiment of the software remote operating system was completed;the ROV motion control experiment and the software operating system based on the underwater robot were completed.Teleoperation experiment,the experimental results show the stability and effectiveness of the software operator system grasping.