AGV Depth-Vision Route Calibration for Precise Pallet Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated guided vehicles (AGVs) face positioning accuracy issues in large work areas, leading to errors in aligning with objects such as pallets, which affects their ability to fetch objects accurately.
Innovation Solution
The AGV is equipped with an image capturing device, a processor, and a navigation coordinate system that allows it to capture depth images, perform image recognition, and convert reference pixel information into navigation coordinates to determine an accurate object-fetching route.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If SLAM navigation technique is used for positioning in large work areas, then the AGV can operate autonomously, but positioning accuracy deteriorates with errors of about 1m
Solution Approach 1:
The patent introduces an image capturing device as an intermediary between the AGV and the environment. This device captures images of the target object, and through image recognition and coordinate transformation, provides accurate position and orientation information that compensates for SLAM positioning errors, enabling precise alignment despite autonomous navigation limitations
Solution Approach 2:
The patent replaces reliance on mechanical SLAM-based positioning with an optical vision-based positioning system. By using image capturing devices to detect target objects and calculate their positions through coordinate transformations, the system substitutes the mechanical navigation approach with an optical measurement approach that achieves higher precision
2Ease of operation
If map-based navigation is used for route following, then the AGV can reach target positions, but alignment accuracy with objects deteriorates due to positioning errors
Solution Approach 1:
The patent implements a feedback mechanism where the image capturing device continuously monitors the target object's position and orientation. The system captures images, performs recognition, transforms coordinates, and uses this feedback information to correct alignment errors, ensuring the AGV can accurately align with objects despite initial positioning inaccuracies from map-based navigation
Solution Approach 2:
The patent performs preliminary image capture and recognition actions before the AGV completes its navigation to the target position. By capturing images of the target object in advance and calculating its precise location through coordinate transformation, the system prepares accurate alignment data that guides the final positioning and alignment operations
Data Source
AI summary
An automated guided vehicle (AGV) includes an image capturing device, a storage device, a fetching device, a driving device and a processor. The processor is configured to execute: controlling the AGV to move from a starting position to a target position according to a navigation coordinate system, the target position corresponding to the object to be fetched; capturing a depth image from the object to be fetched through the image capturing device; performing image recognition on the depth image to obtain reference pixel information; converting the reference pixel information to the navigation coordinate system according to a coordinate mapping algorithm to obtain a calibrated position; and determining an object-fetching route according to the target position and the calibrated position, and controlling the AGV to move according to the object-fetching route to fetch the object to be fetched.