Work Cell Deployment Planning With Action Chunks and Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for work cell operation in unstructured assembly environments require inefficient replanning for each task and are prone to failure due to environmental changes during computation.
Innovation Solution
An operation environment deployment method and system that utilizes a cell model, pre-stored action chunks, and semantic scene graphs to generate an action chunk sequence, build a configurable control graph, and perform object detection for adaptable task execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a new execution sequence is replanned for each task in an unstructured environment, then the work cell can adapt to different tasks, but the planning efficiency deteriorates
Solution Approach 1:
The system pre-plans and stores multiple execution sequences for different task types before actual operation. When a task arrives, the system directly selects from pre-planned sequences based on task classification, eliminating the need for real-time replanning and significantly improving planning efficiency while maintaining adaptability
Solution Approach 2:
The planning system is segmented into task classification modules and execution sequence selection modules. Each module handles specific functions independently, allowing the system to efficiently match tasks with appropriate pre-planned sequences without requiring complete replanning, thus resolving the contradiction between adaptability and efficiency
2Reliability
If the work cell starts operation only after task plan verification, then task execution reliability is improved, but the response time to environmental changes deteriorates
Solution Approach 1:
The system performs partial verification of task plans by checking only critical parameters and constraints before execution, rather than complete verification. This allows the work cell to start operation quickly while still ensuring reliability for key aspects, reducing the time loss without compromising essential reliability
Solution Approach 2:
The system implements real-time feedback monitoring during task execution, continuously checking actual execution status against planned parameters. If deviations occur, the system can immediately adjust or halt operations, maintaining reliability through ongoing monitoring rather than exhaustive pre-verification, thus reducing response time to environmental changes
Data Source
AI summary
An operation environment deployment method, adapted to an operation environment including a work cell, wherein the work cell includes multiple equipment, and the method includes: obtaining a first current deployment status and a target deployment status of the operation environment from a cell model and performing problem planning, using pre-stored action chunks to perform domain planning, generating an action chunk sequence according to a pre-stored semantic scene graph, dismantling the action chunk sequence into target action features, obtaining a second current deployment status of the operation environment from the cell model, generating a configurable control graph according to the target action features and the second current deployment status, outputting control commands to the equipment according to the configurable control graph and obtaining corresponding feedback status, and building a three-dimensional geometric scene and a local semantic scene according to sensory data corresponding to the equipment.