Research On Adaptive Flexible Control Of Snake Robot

Snake robot is a kind of robot with high redundancy,which is composed of connecting rod in series.But its complex coupling structure makes it difficult to study.In this paper,the corresponding gait algorithm of the snake robot is set according to its structural characteristics,and how to realize its flexible adaptive control is studied.Firstly,this paper introduces the research status,shape and internal structure of snake-shaped robots at home and abroad.The structure of the snake robot studied in this paper is described.Secondly,the research and development status of deep learning recurrent neural network(RNN)and long and short term memory network(LSTM)are introduced.Then,the paper introduces the whole structure of the snake robot,which is divided into mechanical structure,control system and communication system.The following is the introduction and analysis of the snake robot gait control function,focusing on the influence of the parameters of the control function on the movement form of the snake robot.Gazebo simulation system is used to verify the motion simulation of the snake robot based on solidwork modeling.Then the kinematics of the snake robot is analyzed based on spinor theory and the kinematics model and velocity model are established.This lays the theoretical control foundation for the movement of the snake robot.Next,in order to realize the flexible adaptive motion of the snake robot,the LSTM algorithm is mainly analyzed and the combination of the gait control function defined by the snake robot and the LSTM algorithm function is proposed to define a new closed-loop feedback control function with adaptive effect.The defined experimental data are used to verify the accuracy and extensibility of the LSTM structure.Finally,based on gazebo simulation system and actual motion test,the proposed flexible adaptive algorithm is verified.The results of the actual movement of the snake robot are analyzed and the function is modified to realize the decoupling effect of the snake robot to some extent.The correctness of the control algorithm and the flexibility and accuracy of the snake robot's flexible adaptive motion are verified.