Edge-Based Object Refinement for Accurate Robotic De-Palletizing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems face challenges in accurately identifying the physical characteristics of packages on pallets, leading to potential mishandling and increased risk of injury to human workers or damage to equipment.
Innovation Solution
The implementation of a robotic system with a post-detection refinement mechanism that uses pairwise edge analysis to identify offsets and alignment errors between initial object estimations and detected features, thereby updating and verifying the object detection results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If initial object estimation is used without refinement, then processing speed is maintained, but measurement precision deteriorates
Solution Approach 1:
The system performs preliminary edge detection and corner identification on the captured image before final object estimation. By pre-processing the image to extract edge information and potential corner locations, the refinement stage can quickly verify and adjust initial estimations without performing exhaustive analysis, thus improving precision while limiting time loss.
Solution Approach 2:
The refinement mechanism applies partial action by only adjusting object estimations when edge-based detection indicates significant deviations. Instead of re-processing all objects uniformly, the system selectively refines detections that fall outside acceptable tolerance thresholds, reducing overall processing time while maintaining precision for critical cases.
2Measurement precision
If edge-based refinement is applied to all objects, then measurement precision improves, but device complexity increases
Solution Approach 1:
The system applies local quality by using different detection strategies for different objects based on their characteristics. Objects with high precision requirements or ambiguous initial detections receive full edge-based refinement, while clearly detectable objects use simpler methods. This selective approach improves overall precision without uniformly increasing system complexity.
Solution Approach 2:
The detection process is segmented into multiple independent stages: initial object estimation, edge detection, corner identification, and refinement. Each stage operates independently and can be selectively applied. This segmentation allows the system to implement complex refinement only where necessary, managing overall system complexity while achieving high precision when needed.
3Reliability
If manual registration is required due to detection failure, then measurement precision can be corrected, but productivity decreases
Solution Approach 1:
The system implements feedback by continuously monitoring detection confidence levels and edge alignment quality. When initial detections fall outside acceptable thresholds, the refinement mechanism automatically activates to correct errors before finalization. This closed-loop feedback ensures high reliability by catching and correcting detection failures automatically, preventing the need for manual intervention and maintaining productivity.
Data Source
AI summary
The present disclosure relates to verifying an initial object estimation of an object. A two-dimensional (2D) image representative of an environment including one or more objects may be obtained. The 2D image may be inspected to detect edges of an object. The edges may be processed to verify or update an initial object estimation to increase the accuracy of an object detection result.