Action control method of under-actuated biped robot

Abstract

The invention discloses an action control method of an under-actuated biped robot with high real-time performance and engineering practicability, comprising the following steps of (1) establishing an action control mathematical model of the under-actuated biped robot; (2) selecting a synchronous variable of the under-actuated biped robot and marking the synchronous variable as theta (q); (3) carrying out computation to obtain the virtual constraint output of the under-actuated biped robot and recording the virtual constraint output as hd (theta (q)), actual output as h0 (q); (4) obtaining an error, marking the error as y which is determined by the expression of y=ho (q)-hd (theta (q)), carrying out feedback correction to modify the synchronous variable in the step (2) and making y incline to zero; (5) calculating y by a multi-model controller; (6) selecting one control variable from the control variables obtained in the step (5) according to the working condition of the under-actuated biped robot and recording the control variable as vi; (7) substituting the control variable vi obtained in the step (6) into the following expression to output an actual control variable after computation; and (8) repeating the steps from (2) to (7).

Description

technical field [0001] The invention relates to the technical field of robot control, in particular to a motion control method for an underactuated biped robot. Background technique [0002] Time-invariant gait is the main method of motion planning for underactuated biped robots. The current motion control methods for underactuated biped robots are mainly specific technical solutions based on the theory of time-invariant gait. Time-invariant gait is It means that the gait of the robot is related to the state, but not related to time. The motion control method of the underactuated biped robot generally uses an attitude variable related to the passive degree of freedom as a synchronization variable, and by controlling the active joints, the synchronization of the active joints and the passive joints is realized to achieve the purpose of controllable attitude. The motion trajectories of joints are called virtual constraints. The core problem of the motion control method of th...

AI Technical Summary

Problems solved by technology

The core problem of the motion control method of the underactuated biped robot is how to better realize the control of the single-leg support period, so that the underactuated biped robot can be quickly stabilized in one step period, while the current underactuated biped robot The motion control method is to output the control quantity through a single controller, which leads to the speed of stabilization of the underactuated biped robot in one cycle is not fast enough. In other words, the current motion control method of the underactuated biped robot is not real-time. Not practical

Images