Self-adaptive collision detection method and device and storage medium

Abstract

The invention discloses a self-adaptive collision detection method and device and a storage medium. The self-adaptive collision detection method comprises the steps of: recognizing the friction torque of each shaft in a robot to obtain the friction torque corresponding to the current speed of each shaft, and correcting the friction torque to obtain the corrected friction torque; calculating a first theoretical torque according to the speed and the position of each shaft and the corrected friction torque; calculating an external force moment according to the feedback moment and the first theoretical moment; judging whether the external force moment is greater than a preset moment threshold; if the external force moment is greater than or equal to a preset moment threshold value, judging that collision is detected; and if the external force torque is smaller than the preset torque threshold value, judging that collision is not detected. In this way, the self-adaptive collision detection method and device and the storage medium have the advantages that the collision detection precision can be improved, and the probability of damage caused by collision is reduced.

Description

technical field [0001] The present application relates to the field of robot technology, in particular to an adaptive collision detection method, device and storage medium. Background technique [0002] Collision detection refers to judging whether the robot has collided by sensing the force of the external environment on the robot during the movement of the industrial robot, thereby protecting the safety of personnel and equipment. [0003] Commonly used collision detection methods in the prior art include: first, by installing a torque sensor on the base to obtain the torque value of the external environment and the robot, and by setting the torque threshold to determine whether a collision occurs, but the price of the torque sensor is high. Adding a torque sensor will increase the cost of the product; the second method is to predict the external torque by detecting the change of the motor current in the robot, and determine whether a collision occurs by setting the torque...

AI Technical Summary

Problems solved by technology

[0003] Commonly used collision detection methods in the prior art include: first, by installing a torque sensor on the base to obtain the torque value of the external environment and the robot, and by setting the torque threshold to determine whether a collision occurs, but the price of the torque sensor is high. Adding a torque sensor will increase the cost of the product; the second method is to predict the external torque by detecting the change of the motor current in the robot, and determine whether a collision occurs by setting the torque threshold. This method is limited by the friction torque identification accuracy. It is difficult to accurately estimate the magnitude of the external force, resulting in insensitive collision detection or false alarms

Images