Hinge model identification method and robot operation method

Abstract

The invention discloses a hinge model identification method and a robot operation method. The hinge model identification method comprises the steps: (1) enabling an end effector to grab and pull a handle of a hinged object, and recording the pose of the end effector at each moment; (2) obtaining a pose relation between two rigid parts at each moment through calculating to serve as a training sample; (3) setting a hinging model of the hinged object, and obtaining the probability of occurrence of the pose relation between the two rigid parts of a certain hinging model at a certain moment according to the transformation degree of freedom between the two rigid parts; and (4) obtaining a likelihood function of the pose relation between the two rigid parts of a certain hinge model through calculating, and selecting the hinge model capable of enabling the likelihood function to obtain a maximum value as a final hinge model. According to the method, a motion model of the hinged object is automatically generated through active operation of a robot on the hinged object, the parameter identification process is completed, operation of the robot on the object of which the hinge model is unknown is greatly simplified, and the application range of the robot in a complex environment is widened.

Description

technical field [0001] The invention relates to the field of robots, in particular to an identification method of an articulated model and a robot operation method. Background technique [0002] The home environment is a key application area for service robots. To accomplish a given task, robots often need to manipulate various types of objects. But many of them are not rigid bodies, such as refrigerators, doors, drawers, etc. These objects all have movable parts. This object, which can be regarded as multiple rigid bodies connected together in a certain way and can move relative to each other, is called an articulated object. How to operate these objects is a huge challenge for robots. [0003] One of the most common existing methods is to move the robot chassis to a specified position, and collect the trajectory of the robot when operating the articulated object by dragging and teaching. At this time, if the robot needs to perform corresponding operations, it only needs...

AI Technical Summary

Problems solved by technology

The first way to achieve object manipulation through teaching and reproduction, due to the existence of various errors, the contact stress between the robot and the articulated object may be very large, which will greatly reduce the service life of the robot or the object or even cause damage

[0007] The second improvement scheme improves the stress that occurs during contact through impedance control, but both this scheme and the first scheme have the problem of poor applicability, and corresponding teaching work is required for each articulated object that needs to be operated

[0008] The third solution solves the problems of the first two solutions to a certain extent, but the autonomy of the robot is still poor, and each new articulated object requires the user to manually input its model parameters, which is unacceptable for end users of

Images