Research On Decentralized Force/Position Control For Reconfigurable Manipulator With Environment Constraint

With the rapid development of modern science and technology,reconfigurable manipulator has penetrated into many areas in modern life.Due to the structure limitation of the traditional manipulator,it can only carry out a specific task,which brings a lot of inconvenience in practical application.The emergence of reconfigurable manipulators will greatly change the previous dilemma.According to the different external environment,reconfigurable manipulator with different size and a group of the information exchange module,can assemble into different configurations to meet the different needs of the mission.However,the end actuator of reconfigurable manipulator will inevitably contact with the external environment.In the process of contacting with environment,the control problem of force and position will have to be considered under environmental constraints.We know that there are only a few scholars to study this problem by consulting previous documents.Therefore,based on previous research of reconfigurable manipulator position control,this paper further researches force/position control for constrained reconfigurable manipulator,which ensures accuracy of force and position tracking and makes the control torque smooth.In order to solve the reconfigurable manipulator dynamic model including force and position,a sliding mode force / position control method based on extended state observer,a force / position control method based on non-singular terminal sliding mode and a decentralized terminal sliding mode force / position control method based on neural network are proposed respectively.The specific contents of this research include the following aspects:1.First of all,the background and significance of this topic are introduced in detail,the research status at home and abroad and the application of the related algorithms are also introduced at the same tome.2.For the force / position control problem of the reduced order constrained reconfigurable manipulator,the extended state observer is used to approximate of reconfigurable manipulator model to compensate model uncertainty,then the force / position controller is designed combined with the sliding mode.Finally,the stability of the designed controller is proved through Lyapunov theory and simulations verify its effectiveness.3.According to the model in the third chapter,a force / position control method for reconfigurable manipulator based on non-singular terminal sliding mode is proposed.The dynamic model were solved by the state of separation and neural network is used to estimate separately the uncertain and coupled items.The force/ position controller is designed based on non-singular terminal sliding mode technique,which not only guarantees the precision of force /position control but also effectively suppress the chattering effect compared with the linear sliding mode.The simulation results verify the superiority and effectiveness of the proposed force / position control with different configurations.4.Based on neural network,this chapter presents a decentralized terminal sliding mode force/position control method for constrained reconfigurable manipulators.Through the establishment of the error dynamics model,the system correlation and nonlinear terms are separated,and the neural network is used to compensate effectively.The effectiveness of the method is verified by simulations.Finally,the research contents of this paper are summarized,and the prospect of further research is pointed out.