Robot Control Device Gripping Strategy for Adjacent Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot control systems face issues with accurately gripping target objects without causing deformation when multiple objects are adjacent, as they fail to differentiate between non-overlapping and overlapping regions effectively.
Innovation Solution
A robot control device that utilizes first and second region information to determine the optimal gripping position, avoiding regions that overlap with other target objects, thereby preventing deformation and ensuring accurate object handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the robot grips target objects based on picked-up images using conventional tracking methods, then the robot can handle multiple target objects in a predetermined movable region, but the gripper may come into contact with adjacent target objects causing deformation
Solution Approach 1:
The picked-up image is divided into multiple regions of interest, each associated with a different target object. The control device calculates separate handling times for each region based on position coordinates and directions, enabling selective gripping of specific target objects while avoiding adjacent ones. This segmentation approach allows the robot to handle multiple objects efficiently without causing deformation to non-target objects.
2Loss of time
If the robot calculates moving times from gripper position to target objects to determine handling order, then the robot can optimize handling sequence, but it cannot distinguish between target objects that should be gripped and those that should be avoided when regions overlap
Solution Approach 1:
Different regions of the picked-up image are assigned different properties: regions corresponding to target objects to be gripped are marked with first region information, while regions corresponding to adjacent objects to be avoided are marked with second region information. The control device uses this local quality differentiation to selectively calculate handling times only for valid target objects, ensuring both time optimization and accurate object selection.
3Ease of operation
If the robot uses a single region designation system for target objects, then the system is simple to operate, but it cannot differentiate between regions that should be gripped and regions that should be avoided
Solution Approach 1:
The region information structure is designed to serve multiple functions: it identifies target objects, determines their positions and directions, and distinguishes between grippable and avoidable regions. By encoding both first region information (for target objects) and second region information (for adjacent objects) within the same region information framework, the system maintains operational simplicity while gaining the versatility to handle complex multi-object scenarios.
Data Source
AI summary
On the basis of received first region information indicating a first region, which is a region designated for an acquired picked-up image of a plurality of target objects, and second region information indicating a second region different from the first region, a robot control device causes a robot to grip the target object for which the second region not overlapping the first region of another of the target objects is designated and does not cause the robot to grip the target object, the second region for which overlaps the first region of the other target object.