Teaching method, device, computing device and storage medium for multimodal fusion

Abstract

The invention discloses a multi-modal fusion teaching method, wherein the method comprises the steps: constructing a three-dimensional point cloud of the surface of a workpiece, and obtaining a data point set of the surface of the workpiece according to the three-dimensional point cloud of the surface of the workpiece; determining a starting point and an ending point of a virtual clamp guide trackaccording to the data point set on the surface of the workpiece in combination with a machining strategy, obtaining a discrete data point set by increasing a fixed step length, and interpolating thediscrete data point set to obtain a virtual clamp guide track; judging whether the direction of the virtual clamp guide track is greater than 90 degrees or not, and if so, finishing partial teaching through manual dragging; and if the angle is smaller than 90 degrees, assisting in completing part of teaching through a voice instruction. According to the method, the information of the surface shapeof the workpiece is obtained through three-dimensional reconstruction, so that the tail end of the robot can basically and accurately reach or approach an ideal machining point, an operator only needs to apply certain force to the tail end of the robot for fine adjustment or voice regulation and control when the virtual clamp is inaccurate, and the teaching process is intelligent.

Description

technical field [0001] The invention belongs to the technical field of robot intelligent processing, and more specifically relates to a multi-modal fusion teaching method, device, computing device and storage medium. Background technique [0002] Aeroengines, steam turbines, water turbines and other blade-type fluid machines have the characteristics of large quantity, complex shape, high processing precision and high surface roughness. At present, most factories still use manual grinding for the above-mentioned curved surface parts with complex trajectories. This method has low production efficiency, poor consistency, and is harmful to human health. Therefore, robot technology is widely used in the processing of the above-mentioned complex trajectory parts. At present, the processing of complex trajectory curved surface parts by robots is mainly planned by experienced experts through off-line programming and teaching method of teaching pendant to obtain the processing traje...

AI Technical Summary

Problems solved by technology

[0003] The method for obtaining the machining trajectory of the above-mentioned robot has the following disadvantages. First, the time cost of offline programming is high, the economic cost is high, and the complexity is high; Third, the current teaching methods are mostly based on a single input, such as force input or visual input, with low stability and less interaction and feedback

Images