Two-foot robot track generating and modulating method based on certified program generator (CPG) mechanism

Abstract

The invention relates to a two-foot robot track generating and modulating method based on the certified program generator (CPG) mechanism. The method comprises the following steps: building up a gravity center track generator of a two-foot robot by utilization of a CPG unit; designing a mapping function, and mapping output signals of a CPG into a three-dimensional gravity center track of the robot; modulating a working space track of the robot by utilization of the on-line generated three-dimensional gravity center track, and further obtaining robot tail end track signals with adaptability due to the fact that the CPG has rich dynamic characteristics and can produce gravity center track signals with adaptability; and designing a motion engine to map dynamic characteristics of the tail end of the robot to a joint space of the robot so as to realize adaptable walking of the robot. Compared with the prior art, the method starts from the gravity center and the tail end track of the two-foot robot, explores to improve a bionic method of walking capacity of the two-foot robot, improves walking adaptable capacity of the two-foot robot to a certain extent, and enables the robot to have certain robustness.

Description

technical field [0001] The invention relates to the field of robot motion control, in particular to a method for generating and modulating a trajectory of a biped robot based on a CPG mechanism. Background technique [0002] Recently, in the field of international robotics, humanoid robots, which can fully reflect the characteristics of intelligent robots, have received more and more attention. The basis of the motion control problem of humanoid robot is biped walking control. Compared with multi-legged robots, the alternation of single-foot support and double-foot support makes the stability problem the most fundamental motion control problem of biped robots. With people's in-depth understanding of the walking nature of humanoid robots and the continuous development of neuroscience, as an intelligent control algorithm based on neuroscience and bionics, the humanoid robot based on central pattern generator (CPG) Robot motion control algorithm has become a research hotspot....

AI Technical Summary

Problems solved by technology

But for complex legged robots, the joint control signals are not simple sinusoidal or sinusoidal

It is very difficult to generate arbitrary waveform signals with CPG

Therefore, it is necessary to use an evolutionary algorithm to obtain the desired control signal, but this also increases the complexity of the project

In addition, the CPG is directly mapped to the joint space of the biped robot, and the external feedback is difficult to design

Due to the complex dynamic characteristics of the CPG network, the impact on the network after the feedback information is connected is relatively complicated

Not conducive to engineering applications of robots

Images