Method and robot for detecting pallet status
By graying and mean filtering the pallet image, combined with image difference judgment and reference image verification, the instability problem of robot pallet detection is solved, fast and accurate pallet status detection is achieved, and the robot service level is improved.
CN114663409BActive Publication Date: 2025-09-26KEENON ROBOTICS CO LTD
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202210324488.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-29
- Publication Date
- 2025-09-26
- Estimated Expiration
- 2042-03-29
AI Technical Summary
Technical Problem
In the existing technology, image processing for robot pallet item detection is difficult, and detection instability leads to low detection success rate, which affects the scheduling logic of the robot to perform return tasks.
Method used
By obtaining the reference image and current image of the pallet, graying and mean filtering are performed, the image differences are compared, and the pallet status is judged using the preset ratio and threshold. The current image status is verified or corrected in combination with the reference image.
Benefits of technology
It achieves fast and accurate pallet status detection, improves the efficiency of the robot in performing return tasks, and enhances customer experience.
✦ Generated by Eureka AI based on patent content.
Abstract
The present invention provides a method for detecting the status of a pallet, wherein the pallet is located on a robot. The method comprises: obtaining a reference image at the pallet and a status corresponding to the reference image; obtaining a current image; comparing the current image with a previous image to determine the difference between the current image and the previous image; determining the status corresponding to the current image based on the difference between the current image and the previous image; comparing the current image with the reference image to determine the difference between the current image and the reference image; and verifying or correcting the status of the current image based on the difference between the current image and the reference image. The technical solution of the present invention eliminates the need to process image depth; only the relationship between the change in the current image and a threshold is needed to determine whether the pallet is occupied. This method provides fast detection speed and high accuracy, and the robot can execute the scheduling logic for the return task without having to wait for an extended period of time. This improves the robot's service level and thus enhances the customer experience.
Need to check novelty before this filing date? Find Prior Art