Autonomous Machine Knowledge Transfer Across Task Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently transferring knowledge among autonomous machines, especially when robots are onboarded or migrated across different domains, leading to issues with false positives and the need for constant feedback from process experts.
Innovation Solution
The implementation of Collaborative Brachiation Transfer Learning (CBTL) for autonomous machines, which involves determining group migration, generating knowledge transfer messages, and updating task performing models based on received knowledge, allowing for seamless knowledge transfer across different groups and domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional knowledge transfer methods are used among autonomous machines, then knowledge can be shared between groups, but false positives occur and constant feedback from process experts is required
Solution Approach 1:
The patent applies preliminary action by pre-processing knowledge before transfer through filtering mechanisms that remove false positives. The source group's knowledge is filtered through a task performance model that predicts task outcomes, preventing unreliable knowledge from being transferred in the first place, thereby eliminating the need for constant expert feedback.
Solution Approach 2:
The patent introduces an intermediary filtering mechanism between knowledge source and destination. The task performance model acts as a mediator that evaluates and filters knowledge transfers, blocking false positives while allowing valid knowledge to pass through, thus reducing system complexity by automating what previously required expert intervention.
2Adaptability or versatility
If autonomous machines operate in different domains, then versatility is improved, but knowledge transfer efficiency decreases due to domain differences
Solution Approach 1:
The patent applies local quality by customizing the task performance model according to the target domain's specific characteristics. When transferring knowledge to a different domain, the model is adapted to reflect that domain's task dynamics and conditions, allowing efficient knowledge transfer while maintaining domain-specific accuracy and relevance.
Solution Approach 2:
The patent utilizes parameter changes by adjusting the task performance model parameters based on the target domain. The model's parameters are modified to account for domain-specific variations in task performance, enabling effective knowledge transfer across diverse domains while maintaining high efficiency through automated parameter adaptation.
3Productivity
If more autonomous machines are deployed to perform complex tasks, then task coverage is improved, but the effort required for employment increases
Solution Approach 1:
The patent implements self-service by enabling autonomous machines to automatically filter and transfer knowledge without human intervention. The task performance model autonomously evaluates knowledge transfers between machines, allowing the system to scale to more machines and complex tasks while reducing employment effort through automated knowledge management.
Solution Approach 2:
The patent applies feedback mechanisms where the task performance model continuously learns from actual task outcomes and uses this feedback to improve future knowledge transfers. This automated feedback loop enables the system to handle more machines and tasks efficiently, reducing the need for human effort in knowledge management while improving task coverage.
Data Source
AI summary
A controller for an automated machine may include including: one or more processors configured to: determine that a group affiliation of the automated machine switched from a first group of automated machines to a second group of automated machines, the first group of automated machines being assigned to one or more first tasks, the second group of automated machines being assigned to one or more second tasks; generate a message for one or more network devices of the second group of automated machines in accordance with a communication protocol, the message including information about a task performing model of the automated machine, the task performing model being based on a result of performing at least one task of the one or more first tasks by the automated machine.