Robotic Picking Verification Using Multi-Sensor Object Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems for picking goods are not capable of performing complex picking processes error-free and require significant human intervention, as existing sensors are not sophisticated enough to ensure accuracy and efficiency in identifying and handling various objects.
Innovation Solution
A robotic system equipped with a combination of sensors such as ultrasonic sensors, vacuum sensors, RFID readers, cameras, and 3D recognition systems to accurately identify and handle objects, allowing for independent operation with minimal human intervention by verifying the presence and type of objects being held or placed, and using RFID tags to track objects in source and target loading aids.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple sensors are used in the robotic picking system, then the device complexity is reduced, but the measurement precision and reliability of object detection deteriorate
Solution Approach 1:
The patent combines multiple sensor types (ultrasonic sensors, vacuum sensors, RFID readers, cameras, and 3D recognition systems) into an integrated sensor system. This merging of different sensing technologies enables comprehensive object detection with high precision while maintaining manageable system complexity through unified control architecture.
Solution Approach 2:
The sensor system is designed to perform multiple functions simultaneously: ultrasonic sensors detect object presence, vacuum sensors verify gripping force, RFID readers identify object types, cameras capture visual information, and 3D recognition systems determine spatial positioning. This multi-functionality allows a single integrated system to handle complex detection tasks without requiring separate specialized systems for each function.
2Reliability
If multiple sensors are integrated into the robotic system, then the reliability and accuracy of picking processes improve, but the device complexity increases
Solution Approach 1:
The patent implements feedback mechanisms where sensor data is continuously monitored and used to adjust robot actions in real-time. The controller receives input from all sensors, processes the information, and makes dynamic adjustments to gripping force, positioning, and object identification. This feedback loop ensures high reliability by automatically correcting detection or execution errors without human intervention.
Solution Approach 2:
The controller serves as an intermediary that coordinates between multiple sensors and the robot's actuators. It integrates data from diverse sensor sources, reconciles conflicting information, and translates sensor inputs into coordinated robot actions. This intermediary architecture manages system complexity by providing a centralized interface between the complex sensor network and the robot's execution mechanisms.
3Reliability
If comprehensive sensor verification is performed for each object, then the error rate in picking processes decreases, but the productivity and speed of operation reduce
Solution Approach 1:
The patent performs preliminary object identification and verification using RFID readers and cameras before the robot attempts to grasp and place objects. By pre-identifying object types, locations, and required handling parameters, the system reduces on-the-fly decision-making time. This preliminary action ensures accuracy while maintaining speed by preparing verification data in advance of the physical picking operation.
Solution Approach 2:
The sensor verification process is applied selectively rather than uniformly to all objects. The system performs comprehensive verification only when necessary based on object characteristics, while using faster, simpler verification methods for routine objects. This partial application of full verification maintains high accuracy for critical operations while preserving overall productivity by avoiding unnecessary verification steps.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system enables efficient and accurate picking and placement of goods, reducing errors and the need for human intervention by utilizing a multi-sensor approach to ensure that the correct objects are removed and placed, thereby enhancing the reliability and efficiency of the picking process.
Implementation Method 1
The first sensor system can include an ultrasonic sensor
Implementation Method 2
if the robot is equipped with a vacuum gripper, a vacuum sensor
Implementation Method 3
The first sensor system can also have an RFID reader
Data Source
Figure 1~2
Figure 3~4
AI summary
A method for picking objects (2a..2d) is specified, in which at least one object (2a..2d) is removed from a source loading aid (5a, 5b) and is placed in a target loading aid (6a..6c) using a robot (3). After the object (2a..2d) has been removed, a first sensor system (11) of the robot (3) is used to check whether at least one object (2a..2d) is held by the robot (3). A second sensor system (14a, 14b, 17, 18) is used to determine a number and/or a type of the at least one removed object (2a..2d). The operation of placing the at least one object (2a..2d) in the target loading aid (6a..6c) is aborted or modified if an object (2a..2d) is not held by the robot (3) or the number and/or the type of the at least one removed object (2a..2d) do(es) not contribute to completing a picking order which defines a desired number and/or desired type of objects (2a..2d) in the target loading aid (6a..6c). A device (1a..1c) and a computer program product for carrying out the presented method are also specified.