Mechanical Modeling And Trajectory Planning Of Mobile System For Luanr Quadruped Robots

With the development of outer space exploration,the motion ability of exploration equipment restricts the further exploration mission.Planetary rovers are mainly wheeled vehicles,which have a limited mobility in the face of unstructured terrain.In contrast,legged robots have a potential.At present,some research institutions of the US,Germany and other countries have developed legged robots for Planetary exploration,but most of them are static walker,but have a limited mobility,as well as its detection range.If motion of those robots can be well planned and the dynamic walking gait can be adopted,the detection range and the task efficiency can be greatly improved.Legged robot trajectory planning problem is about how to generate legged robot motion under the condition of physical constraints,which is a hot topic in the field of legged robot.Researchers have proposed their own trajectory planning method,most considered the foot-ground interaction as a contact process between two rigid bodies.So,those methods are not suitable for planetary exploration robot.Therefore,this paper takes the lunar quadruped robot as the research object,and applies a new method to generate motion trajectory of the lunar quadruped robot.This paper also provides support for achieving the dynamic motion of the lunar quadruped robot.The kinematics model of lunar quadruped robot is established according to the characteristics of the soft and deformable lunar surface.The position and the velocity of robot body is described,and the position and speed of robot leg are analyzed.The relationship among foot-ground relative motion,joint motion and body motion of robot is analyzed,and the Jacobian matrix of robot is solved.The multi-body dynamics model,the foot-ground interaction model and the joint mechanical model of the lunar quadruped robot are established.The simulation system of the lunar quadruped robot is built.Given the initial state of the robot and the time history of the control variable,the state trajectory of the lunar quadruped robot is solved,which provides a foundation for the verification of the trajectory planning method.The dynamic motion process of lunar quadruped robot is described and analyzed.Then,the multi-interval Gauss pseudo-spectral method is used to transform it into a nonlinear optimization problem,and the sequential quadratic programming method is adopted to solve it.Motion process of the lunar quadruped robot is simulated to verify the trajectory planning results,and the robot's motion under different forward velocity and different contact stiffness is analyzed.