| In the modern era of rapid development of science and technology,robotics research is one of the most important areas of research in science and technology in various countries.Among them,biped robots that are close to the shape of humans are a hot topic in the field of robotics research in various countries and represent the current state of high-tech development in a country.Compared with other forms of robots,such as crawler and wheeled robots,biped robots can help people in high-risk work and daily life,respectively.Because of its complex system,the biped robot has multiple technical indicators,among which the gait stability is one of its important indicators.The improved gait stability of the biped robot can ensure that it can be quickly used in practical applications.The transformation of the attitude angle of each joint during the robot's walking has a certain influence on its stability,so this paper studies the attitude angle solution method.Because the posture of the robot changes continuously with walking,in order to facilitate the acquisition of posture information by the posture sensor,it is necessary to make regular gait planning for the walk of the robot in advance.thesis mainly studies from the following aspects:1.Select and explain the hardware of the main control system.The core system board adopts the circuit module with STM32F103C8T6 as the core computing CPU.The package of this module is the smallest in F103.It has the advantages of low power consumption and fast computing.The control module of the steering gear module uses serial communication digital steering gear.The selected model is DX-H25,which has stronger fixed power.The attitude reading module uses the MPU-6050 attitude sensor.The MPU-6050 is an inertial measurement element that includes an accelerometer and a gyroscope.Compared with the combined element,the difference between the accelerometer and the gyroscope time axis is reduced Reduced packaging space.2.Establish the joint link models of each part of the biped robot,and explain the definition of the posture angle in the model.Afterwards,based on the link model,analysis of human kinematics involved,including forward kinematics and inverse kinematics.3.For the stability criterion involved in stability,the ZMP criterion is used to explain,on this basis,a gait planning method that can meet the dynamic walking of biped robots is proposed,that is,the five-point planning method is used to carry out Gait planning,in this planning process,it focuses on planning the hip and ankle joints that mainly affect posture information,and calculates the hip joint trajectory equation and ankle joint trajectory equation in the walking state.4.Research the attitude solution method based on the planned gait and the obtained joint trajectory equation.In the attitude solution of the biped robot,the attitude sensor plays a key role.In the attitude sensor work,Karl is mainly used.Manman filtering algorithm is used for processing.The Kalman filter algorithm is not suitable for processing the signals generated in the nonlinear system,so it has a large error in the attitude calculation process.The integration error generated results in a larger output error ratio.A fusion filter solution method of two complementary filters and a traceless Kalman filter is proposed to solve the attitude.After the experimental analysis of the fusion algorithm,the results show that the algorithm can improve the accuracy and reduce the error to a certain extent in the calculation of the attitude angle. |
PDF Full Text Request