Research on controlling autonomous walking of humanoid robot based on Linux and ROS environment

In response to the requirement of high stability when walking for humanoid robots, ZMP theory is used to analyze and ensure the smoothness of the robot's gait, but efficient, natural and stable humanoid motion control is still a major difficulty. The combination of Linux system and ROS framework is used to take advantage of their advantages in robot development, because the modular design of ROS is conducive to the fusion of multi-sensors, and a variety of sensors are added to the humanoid robot, which is conducive to maintaining the smoothness of the robot body and feeding back real-time data such as the robot's position, speed, attitude, etc. The experimental robot is able to successfully meet the task requirements, and the feasibility of this set of control strategies is verified. The feasibility of this set of control strategies.