Adaptive Bin-Picking ROI for Grasping Objects Near Bin Walls
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current approaches to determining regions of interest (ROI) for robotic bin picking are inefficient and fail to accurately estimate grasp points on objects proximate to bin walls, leading to performance issues in robotic grasping and manipulation in dynamic environments.
Innovation Solution
An autonomous system that uses a depth camera to capture images of bins and objects, generating an adaptive ROI by expanding from a default ROI based on the bin bottom, excluding bin walls, and adjusting the ROI based on camera perspective and object positions to include objects near or against walls, thereby improving grasp point determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed default ROI based on bin bottom is used, then the system is simple to implement, but it fails to include objects proximate to bin walls in the analysis
Solution Approach 1:
The patent implements a dynamic ROI that automatically adjusts its boundaries based on detected object positions. The system transitions from a static fixed ROI to a dynamic adaptive ROI that expands to include objects near bin walls while excluding the walls themselves, resolving the contradiction between adaptability and complexity through automated dynamic adjustment
Solution Approach 2:
The system uses feedback from depth image analysis and object detection results to continuously adjust the ROI boundaries. By monitoring object positions and automatically modifying the ROI to include relevant objects while excluding bin walls, the system achieves high adaptability without manual intervention, balancing versatility and complexity
2Measurement precision
If the ROI is expanded to include objects near bin walls, then grasp point accuracy improves, but bin wall interference increases
Solution Approach 1:
The patent extracts and excludes bin wall regions from the ROI while retaining objects near the walls. By specifically removing the harmful bin wall elements while preserving the useful object regions, the system achieves accurate grasp point estimation without wall interference, directly resolving the contradiction between precision and harmful factors
Solution Approach 2:
The system applies different treatment to different regions within the bin: objects near walls are included with high precision for grasp point estimation, while bin wall regions are excluded to prevent interference. This localized quality approach allows the ROI to have different inclusion criteria for different spatial zones, simultaneously achieving accuracy and eliminating harmful effects
3Reliability
If a comprehensive ROI analysis is performed to improve grasp computation, then picking performance improves, but processing time increases
Solution Approach 1:
The patent performs partial ROI analysis by focusing computational resources only on relevant regions: objects within the adaptive ROI are analyzed in detail for grasp computation, while excluded regions (bin walls) are ignored. This partial action approach maintains high picking performance by concentrating analysis on necessary areas, reducing overall processing time compared to comprehensive full-bin analysis
Data Source
AI summary
An autonomous system can include a depth camera configured to capture a depth image of a bin that contains a plurality of objects from a first direction, so as to define a captured image. Based on the bottom end of the bin and the captured image, the system can generate a cropped region that defines a plane along a second direction and a third direction that are both substantially perpendicular to the first direction. Based on the captured image, the system can make a determination as to whether at least one object of the plurality of objects lies outside the cropped region. Based on the determination, the system can select a final region of interest for determining grasp points on the plurality of objects.