A foot force and moment following control method for a biped robot

Abstract

The invention discloses a foot force and moment following control method of a biped robot. In the method, a double spring damping model is designed, and a force following controller is designed with an LQR optimization method to realize the following of the force and moment of the foot of a biped robot. ; and the expected force on the foot and the expected moment of the foot are calculated by the ZMP distribution method of planning, which finally makes the ZMP of the biped robot follow better and can adapt to certain uneven ground. The present invention abandons the traditional ZMP following control method to realize the stable walking of the biped robot and adapts to uneven ground, directly calculates the expected force and torque of the foot to enable the robot to walk stably, and directly controls the force and torque of the foot. The following of the torque is used for stable control in a more essential and easier way, with faster control response and stronger ability to adapt to uneven ground, and the ZMP follow-up effect is very ideal.

Description

technical field [0001] The invention belongs to the technical field of robots, and in particular relates to a foot force and moment following control method of a biped robot. Background technique [0002] Zero-moment point ZMP is a very important concept in the research of biped robots, and it plays a decisive role in the walking planning and stability of biped robots. An important means to evaluate whether the biped robot walks stably is whether the actual ZMP can follow the planned ZMP well. The force and moment received by the foot can reflect the ZMP of the robot. Generally, the actual ZMP of the robot when walking on flat ground is calculated through the force and moment received by the foot. Therefore, by controlling the actual foot force and moment of the robot, Make it able to follow the planned force and moment, so as to realize the follow of ZMP and make the robot walk more stable. [0003] From a physical point of view, this method can enable the foot to always ...

AI Technical Summary

Problems solved by technology

However, using ZMP to follow to achieve all this has two disadvantages. One is that calculating ZMP on uneven ground is theoretically inconvenient and inaccurate, so it cannot reflect the walking state of the robot well. The effect of using the ZMP following control method alone on the flat ground will not be very good, and we cannot blindly pursue the calculated ZMP of the robot to follow the planning value.

On the other hand, due to the impact and error of the force sensor, the ZMP calculated from the force and moment of the foot is also very inaccurate. In addition to the theoretical error, the real ZMP of the robot calculated in this way is not true at all. It will inevitably affect the control effect

Images