Underactuated biped robot excitation planning and control method

Abstract

The invention discloses an underactuated biped robot excitation planning and control method, which has the advantages of high real-time property and strong engineering practicability. The method comprises the following steps: (1) establishing a mathematic model and converting the mathematic model into a system state equation; (2) selecting an excitation signal; (3) inputting the excitation signal; (4) taking joint angle positions, joint angle speeds and joint angle acceleration obtained in the step (3) as given values, recording an angle excitation plan for a passive joint to be qpd and recording an actual output angle of an underactuated joint to be qp; (5) performing sampling according to a sampling period and calculating an actual underactuated joint angle qp, an underactuated joint angle speed and underactuated joint angle acceleration by using a sensor; (6) comparing qpd with qp in each sampling period to obtain an error, determining the error with an expression and performing sliding mode variable structure control correction so as to make e tend to zero; (7) outputting an active joint control variable ua after the operation of the sliding mode variable structure control correction; and (8) repeating the steps (3) to (7) to obtain the active joint control variables of each sampling period.

Description

Technical field [0001] The invention relates to the technical field of robot control, in particular to an excitation planning and control method for an under-actuated biped robot. Background technique [0002] The underactuated biped robot is an inherently nonlinear system, and the dynamic constraints are not integrable. It is a second-order non-holonomic system, and its trajectory planning and control are a difficult problem. Due to the lack of tracking capabilities for arbitrary state space trajectories, in under-actuated systems, the generation of motion trajectories is much more difficult than general conventional systems, and the achievability of the trajectory completely depends on the dynamic relationship. The under-driving system is a system that is not completely controllable, and the general smooth state feedback control method cannot stabilize it. [0003] At present, many researchers are studying the situation where passive joints (underactuated joints) contain brakes....

AI Technical Summary

Problems solved by technology

The underactuated system is an incompletely controllable system, and the general smooth state feedback control method cannot stabilize it

[0003] At present, many researchers are studying the situation where the passive joints (underactuated joints) contain brakes, but for the situation where the passive joints do not have brakes, there is still no incentive planning for underactuated biped robots with high real-time performance and strong engineering practicability. and control methods appear

Images