Robotic Gripper and Vision Timing for Fast Chaotic Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic gripper systems face challenges in precisely identifying and handling products that are touching or overlapping on a conveyor belt, leading to slower picks, incorrect picks, or potential damage, and struggle with variability in product size, shape, and weight, requiring adaptable gripping mechanisms.
Innovation Solution
A method for coordinating actions between a robotic arm and a vision system by transmitting control signals and updating pick target locations, allowing the vision system to perform resource-intensive analyses only when the robotic arm is outside the sensor's field of view, and using machine learning constructs for object discrimination and tracking, with real-time grasp quality detection and analytics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the vision system continuously analyzes product locations in real-time, then picking accuracy improves, but system speed and throughput deteriorate due to resource-intensive processing
Solution Approach 1:
The vision system performs resource-intensive analysis only when the robotic arm is outside the field of view, preparing pick target locations in advance. This preliminary action allows the system to maintain high picking accuracy while avoiding processing bottlenecks during the actual picking operation, thereby preserving system throughput.
Solution Approach 2:
Instead of continuous real-time analysis, the vision system operates periodically - analyzing product locations only at specific moments when the robotic arm is positioned appropriately (outside the field of view). This periodic operation reduces computational load while maintaining sufficient picking accuracy.
2Productivity
If the robotic arm moves quickly to maintain high throughput, then productivity improves, but the timing coordination with vision system deteriorates, causing downtime
Solution Approach 1:
The system performs vision analysis in advance, before the robotic arm completes its movement. By preparing pick target locations while the arm is still in transit or positioned outside the field of view, the system eliminates waiting time and maintains continuous operation, thereby preserving both throughput and minimizing downtime.
Solution Approach 2:
The robotic arm maintains continuous motion without idle waiting periods. The vision system's coordinated periodic operation ensures that analysis results are ready exactly when needed, allowing the arm to continuously perform useful picking actions without interruption or downtime.
3Measurement precision
If the vision system processes images at high resolution for accurate product identification, then measurement precision improves, but processing time increases, reducing system speed
Solution Approach 1:
High-resolution image processing is performed in advance when the robotic arm is outside the field of view. This preliminary processing allows the system to maintain high product identification accuracy while completing intensive computations before they are needed, avoiding speed bottlenecks during the actual picking operation.
Solution Approach 2:
The vision system processes images at high resolution periodically rather than continuously, only when timing conditions are favorable (arm outside field of view). This periodic high-resolution processing maintains identification accuracy while reducing overall processing time and preserving system speed.
Data Source
AI summary
Exemplary embodiments relate to a machine-learning based approach to detecting individual items in a chaotic moving pick-and-place environment. In such an environment, objects may move relative to a robotic arm. As the objects move through the environment, their locations may change. A relatively more-processing-intensive procedure is employed once on an upstream side of the pick and place station in order to identify or initially segment objects in the environment. Identified items are then tracked using less intensive methods as the object moves through the environment. In order to provide rapid picks, the robot's vision system coordinates with the robot's gripper to image the target when the gripper is out of the image area. A pick location is transmitted back to the gripper within a few hundred milliseconds of the last pick.