Research On Intelligent Control System Of Rotor-flying Manipulator

The rotor-flying manipulator effectively combines the multi-rotor UAV and the manipulator to form a new type of aerial work robot,which has both the powerful maneuverability of the UAV and the ability of the manipulator to automate and accurately operate.However,the addition of a robotic arm will make the entire system more complex,making it difficult to accurately model and control the system.In view of the above problems,this paper uses independent modeling and co-simulation to establish mathematical models of each subsystem,design an adaptive neural network sliding mode attitude control strategy,break through the intelligent flight control technology under compound disturbance,and combine the model based on BFSJSPCNN.The proposed multi-point task path planning method provides valuable reference and basis for the autonomous operation of the rotor-flying manipulator.The main research contents are as follows:1.Aiming at the problem that the dimension of the whole system increases with the addition of the manipulator,which makes it difficult to accurately model the rotor flying manipulator,the mathematical models of each subsystem are established by independent modeling.Based on the established coordinate system,the multi rotor UAV and the manipulator are regarded as two independent systems.The dynamic model of the multi rotor UAV,the dynamic model of the manipulator and the external simulated wind disturbance model are established by using Newton Euler equation,which provides a theoretical basis for the subsequent design of controller and system stability control.2.Aiming at the problems of model parameter measurement error,unknown external disturbance,manipulator reaction torque and poor effect of traditional sliding mode control in the actual UAV control system,an adaptive neural network sliding mode control scheme is proposed.Firstly,RBF neural network is used to estimate the model parameter measurement error and external disturbance in real time and compensate them to the controller,and then RBF neural network is used to adjust the important parameters in sliding mode control online,so as to reduce the chattering problem of sliding mode control.Finally,a simulation model is built in Simulink to verify the reliability and effectiveness of the algorithm,and the stability of the system is proved by Lyapunov method.3.Aiming at the problem that the traditional heuristic search algorithm takes a long time,has low accuracy and is easy to fall into local optimization in the optimal flight path planning of rotor flying manipulator when performing multi-point tasks,a path planning method based on(bfs-jspcnn)model is proposed.Based on the traditional pulse coupled neural network(PCNN)model,combined with the idea of breadth first search(BFS)and jump point search strategy(JS),a new PCNN neural network(bfs-jspcnn)model is constructed.BFS idea is used to constrain the activation state of neurons in the search process,and the jump point search strategy is used to select the optimal sub path.The simulation results show that the model not only improves the accuracy of path search,but also is much less than the traditional heuristic search algorithm in time.