Design And Analysis Of Tank Climbing Robot

Wall-climbing robot eombines mobile robot technology with adsorption technology.It can replace human beings to accomplish specific tasks in extremely dangerous environment by carrying corresponding equipment.It has high application value in petrochemical industry,shipbuilding industry,nuclear industry,construction industry and other fields.This paper takes the large-scale storage tanks used in the petrochemical industry as the background.Aiming to overcome the shortcomings of small adsorption force of wheeled magnetic adsorption wall climbing robot and poor flexibility of tracked magnetic adsorption wall climbing robot,a quadruped magnetic adsorption wall climbing robot was designed by combining foot-moving and permanent magnet adsorption.It has the advantages of strong adsorption capacity,strong load capacity,good mobility and easy to cross obstacles.Legged wall-climbing robots have a closed-chain structure and a large number of joints.Their kinematics and dynamics are more complex.Therefore,it has certain theoretical depth and research value to establish the correct kinematics and dynamics model for legged wall-elimbing robots.In this paper,the robot is regarded as a parallel robot.The kinematics and dynamics of the robot are analyzed by using the theory of spatial mechanism,which provides a theoretical basis for the follow-up gait planning and control of the robot.This paper mainly completes the following aspects of work.(1)The design technical requirements of the wall-climbing robot were analyzed in combination with the use of the tank in the petrochemical industry.Starting with the moving and adsorbing modes,the overall structure scheme of the quadruped magnetic adsorbing wall climbing robot is determined,and the mechanical structure of the robot was designed.(2)The kinematics mathematical model of the robot was established.The degree of freedom of the robot was analyzed by using the modified Grubler-Kutzbach(G-K)formula.The D-H method was used to establish the link coordinate system of the walking legs,and the kinematics of them was analyzed.The robot was regarded as a parallel mechanism,the forward and inverse kinematics of the carrier platform were analyzed,the analytic solution of the inverse kinematics was given,and a numerical algorithm based on Newton's method was used to solve the redundant equations to obtain the numerical solution of the forward kinematics.The calculation program was compiled in Matlab according to the established mathematical model,and the correctness of the mathematical model is verified by comparing with the simulation results of Adams.By using the method of space search and inverse kinematics,the working space of the robot platform is obtained,which provides the design basis for the gait and path planning of the robot.(3)The Jacobian matrix of the velocity from the platform to the joint was obtained by using the derivative method and the screw theory method respectively.The characteristics of the two methods are analyzed,and the advantages of the screw theory method were illustrated.Combining the Grassman line geometry theory,the singularities of the robot were analyzed by Using Jacobian matrix based on screw theory,and a forward kinematics singularity of a non-redundant drive is verified.The inverse kinematics singularity was used to illustrate a no power consumption configuration.The Jacobian matrix from the velocity of the platform to the velocity of each component was deduced by using the screw theory.The velocity and acceleration of each component were calculated,which paves the way for dynamic modeling.The consistency of velocity and acceleration simulation results in Matlab and Admas indicates the correctness of the calculation and deduction.(4)The static equation of the robot was deduced by using the principle of virtual work,and the dynamic equation of the robot was deduced by combining the d'Alembert principle.When quadruped robots are driven redundantly,the solution of joint driving force is an over-determined input problem,which needs to be allocated by a certain algorithm.Based on the optimization theory,a quadratic programming method was proposed to optimize the distribution of driving force.The joint driving force was calculated using the principle of minimum power,and the rationality of joint driving design is verified.(5)Based on the inverse dynamics of the robot,a feedforward decentralized PID control was designed in Simulink by using the method of co-simulation of Adams and Matlab.The simulation results show that the control scheme has high tracking performance and control accuracy,and verify the correctness of dynamic solution under redundant driving.