Intelligent cleaning device, cleaning mode selection method and computer storage medium

Abstract

The invention discloses a cleaning mode selection method based on a convolutional neural network. The method comprises the following steps of S1, collecting a current working scene image of an intelligent cleaning device; S2, inputting the collected current working scene image to a working scene convolutional neural network classification model of the intelligent cleaning device to determine the type of a current working scene, wherein the working scene convolutional neural network classification model is established according to a training sample set, and pre-collected working scene images inthe training sample set are marked with corresponding working scene type labels; S3, adopting a corresponding cleaning mode according to the type of the current working scene. By collecting the current working scene image and inputting the current working scene image into the trained working scene convolutional neural network classification model, the type of the current working scene of the intelligent cleaning device can be obtained, and the corresponding cleaning mode is adopted accordingly. By means of the method, selection of the cleaning mode is more efficient, intelligent and diversified, and the demands of consumers are met.

Description

technical field [0001] The invention relates to the technical field of intelligent cleaning, in particular to a method for selecting a cleaning mode based on a convolutional neural network, an intelligent cleaning device, and a computer storage medium. Background technique [0002] Existing smart cleaning equipment usually provides position information and motion status information to the control system through its own sensing device, and then realizes path planning, obstacle avoidance, etc. Intelligent cleaning equipment using this method cannot "see" objects. Objects in its working environment can only be overcome or moved away as obstacles. Moreover, the existing smart cleaning devices cannot directly distinguish the room they are in, and the user can only use the mobile terminal to divide the map to inform its specific location and what cleaning mode should be adopted. The above methods are not intelligent enough and require too much manual intervention, resulting in lo...

AI Technical Summary

Problems solved by technology

[0002] Existing smart cleaning equipment usually provides position information and motion status information to the control system through its own sensing device, and then realizes path planning, obstacle avoidance, etc. Intelligent cleaning equipment using this method cannot "see" objects. Objects in its working environment can only be overcome or moved away as obstacles

Moreover, the existing smart cleaning devices cannot directly distinguish the room they are in, and the user can only divide the map through the mobile terminal to inform the specific location and what cleaning mode should be adopted

The above method is not smart enough and requires too much manual intervention, resulting in low cleaning efficiency

Images