Robotic Grasp Planning for Feature-Aware Item Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic picking devices struggle to successfully grasp items with features that obstruct scanning or require specific grasping orientations, leading to inefficiencies and potential human intervention.
Innovation Solution
A method and system that utilize imagery analysis, including machine learning procedures, to identify features of interest in items to be picked. This analysis generates a grasping plan that accommodates these features, ensuring the picking device can successfully grasp and, if necessary, position items for scanning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing pixel-level feature detection techniques are used to identify item features, then the picking device can grasp items with visible features, but the technique fails when features are occluded, oriented away from camera, or have high variation in type, shape, and lighting
Solution Approach 1:
The system performs preliminary actions by gathering multiple images of the item from different angles and positions before generating the grasping plan. This allows the imagery analysis procedure to identify features of interest even when they are occluded or oriented away from the camera in any single image, thereby resolving the contradiction between reliability and adaptability.
Solution Approach 2:
The patent introduces an intermediary imagery analysis procedure that processes multiple images and identifies features of interest without requiring direct visibility in all images. This intermediary analysis layer bridges the gap between limited camera views and comprehensive feature detection, enabling reliable picking operations across diverse item configurations.
2Productivity
If the picking device grasps items without considering features of interest, then the grasping operation is simple and fast, but the barcode may be obscured or the item may be grasped incorrectly
Solution Approach 1:
The system performs preliminary imagery analysis to identify features of interest before generating the grasping plan. This preliminary step ensures that the barcode scanner can access scannable indicia and that structural components are properly positioned, thereby improving grasping accuracy without significantly impacting picking operation speed.
Solution Approach 2:
The picking device autonomously performs imagery analysis and generates grasping plans that accommodate features of interest without requiring human intervention. The system self-adjusts the grasping plan based on identified features, maintaining high productivity while improving grasping accuracy through automated feature-aware planning.
3Reliability
If the system performs comprehensive imagery analysis to identify all features of interest, then the grasping plan can accommodate all features, but the processing time and computational resources increase
Solution Approach 1:
The imagery analysis procedure extracts only the essential features of interest that affect the picking operation, such as scannable indicia and critical structural components, rather than analyzing all possible features. This selective extraction reduces processing time and computational resources while maintaining reliable feature accommodation in the grasping plan.
Solution Approach 2:
The system applies different levels of analysis to different regions of the item based on their importance. Critical areas like barcodes and structural components receive detailed analysis, while less important regions receive minimal or no analysis. This localized quality approach ensures reliable feature accommodation while minimizing overall processing time.
Data Source
AI summary
Robotic picking devices and methods for performing a picking operation. The disclosed methods may involve executing an imagery analysis procedure to identify a feature of interest of an item, wherein the feature of interest may affect a picking device's ability to perform a picking operation, and generating a grasping plan executable by the picking device, wherein the grasping plan accommodates the feature of interest so that the picking device is able to perform the picking operation.