Unknown Slope Oriented Research On Trot Gait For A Quadruped Robot

Compared with the traditional wheeled robots and crawler robots,the quadruped robot has a good application prospect in the fields of military and civilian with better environmental adaptability and movement flexibility.The quadruped robot in Trot gait walking has a faster movement speed than the Walk gait,and has better motion stability compared to the Bound gait.As a typical geomorphic feature in the natural environment,the dynamic stability of the quadruped robot walking on the slope is the prerequisite and guarantee for the practical application.Therefore,this paper studies the problem of movement perception and motion planning of quadruped robot walking on the slope in Trot gait.Mainly studied the following four aspects:(1)Based on the kinematics and dynamics modeling and analysis of quadruped robot,this paper adopts the overall control strategy of inner and outer ring stratification.Firstly,the kinematics model of quadruped robot is established by D-H method.Secondly,the Lagrange dynamics model is established on the basis of the kinematic model.Because of the high dimensional and strong nonlinearity of the quadruped robot dynamics model,the overall control strategy based on the inner and outer ring stratification has better achievability than the traditional control method based on the precise dynamic model.(2)Based on slope information estimation,this paper adjusts the expected attitude of the body.First,the position of the foot end in the operating space is obtained by homogenization method based on the joint position and the body attitude angle.Secondly,this paper uses several sets of foot end point operation space position to calculate the slope of the unknown slope and the current uphill direction of the quadruped robot by solving the linear system.Finally,this paper uses the Kalman filter algorithm to deny the preliminary estimation results of the slope.Then the filtered slope information is used for posture control.(3)In this paper,we plan the trajectory of the mass center and the trajectory of the swing leg.First,according to the ZMP stability criterion,the stability condition of the quadruped robot walking on the slope is deduced.Secondly,using the LQR optimization method,the stability condition of the quadruped robot walking on slope is taken as the constraint,and the body mass center acceleration is used as the optimization index to optimize the expected trajectory of the body center.Finally,this paper uses the swing leg starting point position,the swing leg starting point speed,the swing leg end point position,the swing leg end point speed as the boundary condition,to plan the swing leg foot movement trajectory.(4)This paper designs simulation experiments to verify the feasibility of the algorithm.According to the modeling method and the uphill control strategy proposed in this paper,the model is set up in the Recurdyn dynamic simulation software,and the control circuit is set up in Matlab Simulink,and the simulation experiment is carried out.This paper realizes the stable walking of the quadruped robot in the terrain of the slope.The data shows that the feasibility of the slope information estimation method and the body position attitude control strategy proposed in this paper.