Container Packing Placement Using 2D Frames From 3D Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for automatically placing objects into containers using 3D point cloud data are computationally intensive due to the large amount of data processing required, which can be time-consuming with limited hardware capabilities.
Innovation Solution
A method and system that utilize a processing unit to obtain object and space dimensions, generate simplified 2D virtual frames from 3D point clouds, and determine unoccupied areas to efficiently place objects into containers using a holding unit, reducing computational load by simplifying data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If 3D point cloud data is used for real-time object recognition and packing, then automation and recognition accuracy are improved, but computational time and processing resources increase significantly
Solution Approach 1:
The patent segments the complex 3D point cloud processing into multiple stages: initial object detection, dimension extraction, packing position determination, and placement execution. By dividing the computational task into smaller manageable segments, the system achieves real-time processing without sacrificing automation capability.
Solution Approach 2:
The patent extracts only the essential dimension data from the complete 3D point cloud information. Instead of processing all point cloud data, the system extracts key dimensional parameters (length, width, height) needed for packing decisions, significantly reducing computational load while maintaining automation.
2Measurement precision
If complete 3D point cloud processing is performed, then object recognition accuracy is improved, but processing speed decreases due to large data volume
Solution Approach 1:
The system extracts only the necessary dimensional information from the 3D point cloud rather than processing the complete point cloud dataset. This extraction approach maintains accurate object recognition by capturing essential geometric parameters while dramatically improving processing speed by reducing data volume.
Solution Approach 2:
The patent transitions from processing full 3D point cloud data to working with simplified dimensional parameters (length, width, height). This dimensional reduction allows the system to maintain measurement precision for packing decisions while achieving faster processing speeds through reduced computational complexity.
Data Source
AI summary
A method for automatically placing a to-be-packed object into a container that defines an accommodation space is provided. A processing unit obtains an object dimension data piece that indicates dimensions of the to-be-packed object, and obtains, through a camera unit that captures images of the to-be-packed object and the accommodation space, an unoccupied area related to the accommodation space. Based on the object dimension data piece and the unoccupied area, the processing unit determines whether the container is capable of accommodating the to-be-packed object.