Robotic Object Placement With Camera-Based Pallet Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems lack the sophistication to efficiently plan and execute the retrieval and placement of objects in complex loading environments, often requiring human intervention for tasks involving larger and more complex operations.
Innovation Solution
A computing system that communicates with a robot equipped with a camera and end effector, capable of detecting object queues, receiving object type identifiers, determining target object poses and placements, and executing motion planning operations for precise retrieval and placement within a loading environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If robotic systems are used to execute physical actions in manufacturing and assembly, then productivity and cost-effectiveness are improved, but the systems lack the sophistication necessary to duplicate human interactions required for executing larger and more complex tasks
Solution Approach 1:
The system segments complex loading environment tasks into distinct operational phases: detection phase (identifying objects and queues), planning phase (determining target poses and placements), and execution phase (motion planning and placement). This segmentation allows the robotic system to handle complex tasks through coordinated specialized modules rather than requiring monolithic sophistication.
Solution Approach 2:
The system performs preliminary detection and planning actions before execution. The camera detects object queues and characteristics in advance, the system receives object type identifiers beforehand, and motion planning operations are computed prior to physical placement. This preliminary action enables the robot to execute complex tasks systematically without requiring human intervention during execution.
2Extent of automation
If robots replicate human actions for dangerous or repetitive tasks, then human involvement is reduced, but the robots lack the sophistication to execute larger and more complex tasks without human intervention
Solution Approach 1:
The robotic system integrates multiple functions into a single automated platform: object detection via camera, identifier recognition, motion planning, and precise placement. This multi-functionality allows one robotic system to handle diverse loading tasks (different object types, queue configurations, and placement requirements) without requiring separate specialized systems or human intervention for each task variant.
Solution Approach 2:
The system introduces an intermediary planning layer between perception and execution. The motion planning operation acts as a mediator that translates detected object characteristics and desired placement locations into executable robot movements. This intermediary layer enables automated handling of complex scenarios by computing appropriate actions without requiring direct human control.
3Manufacturing precision
If the system determines target object pose and placement based on image information, then precision of object placement is improved, but the complexity of processing and planning operations increases
Solution Approach 1:
The system performs preliminary detection of object queues and determination of target poses before execution. By using camera-based detection to identify object characteristics and pre-compute target placement locations, the system establishes precise placement goals in advance. This preliminary action reduces real-time processing complexity during the actual placement operation.
Solution Approach 2:
The system creates a digital representation (copy) of the loading environment through camera image information and object type identifiers. This digital model includes object positions, orientations, and characteristics, allowing the system to plan and simulate placements virtually before executing physical movements. This copying approach enables precise placement planning without requiring complex real-time sensing during execution.
Data Source
AI summary
A computing system including a processing circuit in communication with a robot and a camera having a field of view. The processing circuit obtains image information based on the objects in the field of view and a loading environment, the loading environment which includes loading areas, an object queue, and a buffer zone. The computing system is configured to use the obtained image information in motion planning operations for the retrieval and placement of objects from the object queue into the loading environment. Pallets provided within the loading environment (i.e., within the loading areas) are dedicated to receiving objects having corresponding object type identifiers. The computer system further uses the image information to determine the fill status of pallets existing within the loading environment, and whether new pallets need to be brought into the loading environment and/or swapped out with existing pallets to account for future planning and placement operations.