Robot Picking Control with Precalculated Placement Candidates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing picking systems face challenges in reducing the time required for the picking task while minimizing impacts on objects during transfer, as accurate three-dimensional shape measurement and calculation of optimal placement positions are computation-intensive and time-consuming.
Innovation Solution
A picking system that employs multiple measuring instruments to acquire shape information from different directions, allowing for initial calculation of position candidates based on initial measurements and subsequent calculation of the robot hand's placement position using subsequent measurements, enabling efficient transfer without stopping the robot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If accurate three-dimensional shape measurement and calculation of optimal placement positions are performed, then manufacturing precision is improved, but loss of time increases
Solution Approach 1:
The system performs preliminary action by calculating multiple position candidates based on initial shape measurements before the robot hand contacts the object. This allows the placement position to be predetermined in advance, so that when the object is picked up and shape changes occur, the system only needs to select from pre-calculated candidates rather than performing full optimization calculations, significantly reducing real-time computation time while maintaining placement accuracy.
Solution Approach 2:
The system applies partial action by performing full precision calculations only for the initial placement position candidates, then using simpler selection criteria to choose the final position from these candidates. This partial calculation approach achieves sufficient precision for the task while avoiding the time cost of exhaustive optimization, resolving the contradiction between accuracy and speed.
2Manufacturing precision
If the robot stops to perform calculations during the picking task, then manufacturing precision is improved, but productivity decreases
Solution Approach 1:
The system performs the computationally intensive work of calculating placement positions as preliminary actions before the robot picks up the object. By preparing multiple position candidates in advance based on initial measurements, the system eliminates the need to stop during the picking operation for calculations, maintaining both high precision and continuous productivity.
Solution Approach 2:
The system maintains continuity of useful action by ensuring the robot never stops for calculations. The measurement and calculation processes are overlapped with robot movements, and position candidates are pre-calculated so that the robot can continuously pick and place objects without interruption, maximizing productivity while achieving precise placement through the pre-prepared candidate positions.
3Measurement precision
If multiple measuring instruments are used to acquire shape information from different directions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system applies universality by using the robot hand itself as a multi-functional element that both manipulates the object and serves as a measurement reference. The shape measurement function is integrated into the existing robot system rather than adding completely separate measurement devices, reducing overall system complexity while maintaining the ability to acquire three-dimensional shape information from multiple directions through coordinated robot movements and measurements.
Data Source
AI summary
According to one embodiment, a picking system includes a picking robot and a control device. The picking robot transfers an object from a first space to a second space by using a robot hand. The control device controls the picking robot. When a first measurement result related to a shape of the object in the first space when viewed along a first direction is acquired, the control device performs a first calculation of calculating a position candidate for placing the object in the second space based on the first measurement result. When a second measurement result related to a shape of the object when viewed along a second direction is acquired, the control device performs a second calculation of calculating a position of the robot hand when placing the object in the second space based on the second measurement result and the position candidate.