Robot Hand Regrasp Control for Bulk Workpiece Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot systems face challenges in accurately detecting and retrieving workpieces loaded in bulk, as irregular intervals and orientations within containers can lead to incomplete removal and load collapse, due to issues with three-dimensional sensor imaging and collision detection.
Innovation Solution
A robot system equipped with a hand featuring claw parts and a three-dimensional sensor, including a stereo camera, which uses a controller to acquire and process three-dimensional information, select target workpieces, detect collisions, and determine whether to grip or press workpieces using grip regions or tip parts, allowing for movement and reorientation within the container.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a three-dimensional sensor is used to detect workpiece position and orientation, then the workpiece can be accurately detected and gripped, but the detection may be incomplete due to dead angles caused by irregular workpiece arrangements
Solution Approach 1:
The robot hand performs dynamic exploration by moving to multiple positions and attempting to grip the workpiece from different angles. When the initial grip attempt fails due to dead angles blocking the sensor view, the system dynamically adjusts the hand position and tries alternative grip points, transforming a static detection problem into a dynamic exploration process that eventually succeeds.
2Productivity
If the robot attempts to grip the target workpiece directly, then the operation is efficient, but collision with surrounding workpieces may occur due to inaccurate position detection
Solution Approach 1:
The system performs preliminary exploration actions by moving the hand to the detected workpiece position and attempting grip before final retrieval. This preliminary action serves as a verification step that checks whether the detected position is accurate and whether the path is clear, preventing collisions by identifying positioning errors before they cause damage.
Solution Approach 2:
The system uses feedback from grip attempt results to adjust subsequent operations. When a collision or failed grip is detected, the system receives feedback about the actual workpiece position and orientation, then recalculates the grip plan and tries again, creating a closed-loop control system that improves reliability through continuous correction.
3Adaptability or versatility
If the robot suspends operation to allow load collapse, then workpieces that were previously inaccessible may become reachable, but the process time increases and some workpieces may still remain inaccessible
Solution Approach 1:
Instead of waiting for complete load collapse, the system performs partial exploration by attempting grip operations at multiple positions and angles. This partial action approach allows the robot to retrieve workpieces that become accessible during the gradual collapse process without requiring full suspension of operations, reducing time loss while maintaining adaptability.
Data Source
AI summary
A robot system is provided with a robot, a hand, and a range sensor that acquires information about a distance to a workpiece. Claw parts of the hand include grip regions having a shape for gripping the workpiece. A controller stops the robot when the hand collides with the workpiece. The controller determines whether or not a target workpiece can be gripped by a part other than the grip regions of the hand. When it is determined that the target workpiece can be gripped by the part other than the grip regions, the controller performs control for gripping the target workpiece by using the hand, and for moving the target workpiece inside a container.