Semantic Object Marking for Flexible Autonomous Assembly Tasks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automation programming methods are inflexible and machine-specific, limiting the scalability of autonomous systems to handle diverse production tasks and requiring extensive, error-prone sequences of instructions, which are difficult to manage and correct, especially in dynamic manufacturing environments where customized products are increasingly demanded.
Innovation Solution
An autonomous system is developed with a controller and processor that generates a world model with data objects representing physical objects, including workpieces and devices, and uses semantic markers to trigger skills for task performance, allowing for flexible configuration and operation based on high-level objectives and environmental understanding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional machine-centric programming is used with fixed sequences of instructions, then the automation can perform specific assembly tasks, but the system lacks flexibility and scalability to handle diverse production tasks and customized products
Solution Approach 1:
The patent segments the assembly task into discrete, reusable skills (e.g., pick, place, assemble) that can be independently defined and combined. Each skill represents a modular unit of functionality that can be applied to different objects and contexts, replacing the need for extensive fixed instruction sequences and enabling flexible reconfiguration for diverse production tasks.
Solution Approach 2:
The patent creates universal skills that can be applied across multiple contexts and objects. A single skill definition (e.g., an assemble skill) can work with different workpieces, devices, and assembly configurations, eliminating the need for separate programming for each specific task variant and thereby improving adaptability while reducing programming complexity.
2Reliability
If extensive sequences of fixed commands are written for assembly tasks, then the machine can execute specific operations, but the program becomes difficult to manage and error-prone
Solution Approach 1:
The patent performs preliminary action by pre-defining skills with their associated parameters, preconditions, and effects before actual assembly tasks are executed. This allows the system to validate task feasibility, check for errors, and plan execution sequences in advance, improving reliability while making the programming process more systematic and easier to manage through structured skill definitions rather than ad-hoc instruction sequences.
3Adaptability or versatility
If automation programs are directed to assembly of parts with machine-specific context, then the program can execute tasks for a particular device, but it cannot be scaled to allow other machines to participate without modification
Solution Approach 1:
The patent defines skills with universal parameters and interfaces that are not tied to specific machine implementations. Skills can be applied to different devices and workpiece types through parameter instantiation, allowing the same skill set to scale across multiple machines and production configurations without requiring program modification, thereby improving scalability while managing integration complexity through standardized interfaces.
Solution Approach 2:
The patent uses parameter changes to adapt universal skills to specific contexts. Instead of rewriting programs for different machines, the system instantiates skills with appropriate parameters (e.g., device identifiers, workpiece properties, task specifications), allowing the same skill definitions to operate across diverse machines and configurations, thus enabling scalability without increasing system integration complexity.
Data Source
AI summary
An autonomous system used for a production process includes a device configured to manipulate workpieces according to production process tasks. A device controller generates world model of the autonomous system to include data objects representing respective physical objects in the production process, such as workspace, workpieces, and the device. Semantic markers attached to the data objects include information related to a skill to accomplish a task objective. Semantic markers may be activated or deactivated depending on whether the physical object is currently available for a task performance. The device is controlled to perform tasks guided by the semantic markers while relying on an anticipation function with reasoning operations based on types of physical objects, types of skills, and configuration of the data objects.