Food Portioning Gripper Control for Uneven-Density Weight Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing food portioning systems struggle to accurately grip a target weight of food with uneven density or voids, leading to inaccuracies in portioning.
Innovation Solution
A food portioning device equipped with a gripper, imaging unit, weighing unit, and decision unit that adjusts gripping positions and weights based on captured images and learned models to ensure precise portioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a gripper is used to grip food from a container, then food portioning can be automated, but accurate gripping of target weight is difficult when food density is uneven or voids are present
Solution Approach 1:
The imaging unit captures images of the food container before the gripper grips the food. The decision unit processes these images to identify the picking position and determine the target weight, performing preliminary analysis to guide the subsequent gripping operation and ensure accurate portioning despite uneven food density.
Solution Approach 2:
The system uses imaging feedback to continuously monitor the food container state and adjusts the picking position and target weight decisions based on the captured images. This feedback loop enables the decision unit to adapt to variations in food density and voids, maintaining accurate target weight gripping.
2Quantity of substance
If the gripper inserts deeper into the container to grip more food, then the portioned weight increases, but the error in target weight accuracy increases
Solution Approach 1:
The decision unit determines a specific picking position within the food container based on image analysis, rather than using a uniform gripping approach. This local optimization allows the gripper to target areas with more consistent density, reducing weight error even when gripping larger quantities of food.
Solution Approach 2:
The system dynamically adjusts the target weight parameter based on the analyzed picking position and food characteristics observed in the images. By changing the target weight parameter to match the actual conditions at the selected picking position, the system minimizes weight error regardless of the quantity being gripped.
3Measurement precision
If multiple gripping operations are performed to achieve precise target weight, then portioning accuracy improves, but processing time increases
Solution Approach 1:
The decision unit performs preliminary calculations to determine the optimal number of gripping operations and the target weight for each operation before execution. This advance planning reduces the need for iterative adjustments during actual portioning, minimizing processing time while maintaining accuracy.
Solution Approach 2:
The system may perform a slightly excessive first gripping operation followed by a smaller corrective grip, rather than attempting to achieve the exact target weight in a single precision grip. This approach can reduce total processing time by avoiding multiple small adjustments, while still achieving the required portioning accuracy.
Data Source
AI summary
A food portioning device 1 includes a gripper 11 configured to grip a portion of food 21 having an indefinite shape, decision units ((313, 314)) configured to decide a first target value of a weight of the food to be portioned by one gripping operation, which is calculated based on a final target value to be portioned, and to decide a first picking position corresponding to the first target value based on a captured image, and a control unit configured to execute portioning of the food 21 by moving the gripper 11 to the first picking position. When a weight of the food portioned by the one gripping operation by the gripper 11 satisfies the first target value, the decision unit decides a second target value of a weight of the food to be portioned by a next gripping operation and decides a second picking position corresponding to the second target value.