Multi-Agent Task Routing Through Dynamic Peer Collaboration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing agent modules can only process user requests within their specific domain of expertise, limiting the ability to handle tasks requiring multiple domain-specific capabilities, and current multi-agent collaboration methodologies are rigid and pre-designed, hindering complete automation of complex tasks.

Innovation Solution

A system and method enabling peer-to-peer natural language-based communication between agents, allowing for automatic selection of optimal agents to collaborate and process user requests that exceed the capabilities of a single agent, through a computing system that determines sub-task distribution and integrates external processing results to generate a final result.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple expert agents are implemented to process specialized tasks, then task processing capability is improved, but system complexity increases

Engineering Contradiction:
Improvetask processing capabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent combines multiple expert agents into a unified collaboration system where agents work together on complex tasks. The task management server integrates agent information, matches tasks to appropriate agents, and coordinates their collaboration, thereby managing complexity while maintaining enhanced processing capabilities across multiple domains.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The task management server serves multiple functions: storing agent information, receiving task requests, matching tasks to agents, managing agent collaboration, and transmitting results. This multi-functional approach consolidates what would otherwise be separate systems into a single coordinating platform, reducing overall system complexity while enabling versatile task processing.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If pre-designed collaboration methodology is used for multi-agent collaboration, then task processing is improved, but flexibility and automation are reduced

Engineering Contradiction:
Improvetask processing efficiencyVSAvoidcollaboration automation
Core Design Contradiction:
ProductivityVSExtent of automation

Solution Approach 1:

The patent implements dynamic agent matching where the task management server selects appropriate agents based on real-time task requirements and agent capabilities. Rather than following fixed pre-designed collaboration patterns, the system dynamically determines which agents to involve and how they should collaborate, enabling flexible automation that adapts to diverse task scenarios.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Agents autonomously execute their assigned sub-tasks and provide results back to the task management server without requiring manual coordination. The system automatically manages the collaboration process, including task distribution, result aggregation, and final answer generation, thereby achieving high automation while maintaining processing efficiency.

Inventive Principle:
Principle #25Self-service

3Device complexity

If a single agent processes all tasks, then system complexity is reduced, but task processing capability is limited

Engineering Contradiction:
Improvesystem complexityVSAvoidtask processing capability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent segments complex tasks into sub-tasks that can be handled by different expert agents based on their specific domain expertise. The task management server divides the overall task, assigns sub-tasks to appropriate agents, and integrates their results to form a complete solution, thereby enabling versatile task processing while keeping each individual agent relatively simple.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4726620A2Method and system for processing task through collaboration between agents
Publication Date: 2026.04.15 SAMSUNG SDS CO LTD
  • EP4726620A2 patent drawingFigure 1
  • EP4726620A2 patent drawingFigure 2
  • EP4726620A2 patent drawingFigure 3

AI summary

The present disclosure relates to a method and system for processing a task through collaboration between agents. According to an aspect of the inventive concept, there is provided a method for processing a task through collaboration between agents, which is performed by a computing system. The method may comprise receiving a user request by a first agent from a user terminal and determining whether the first agent may process the user request by itself, inquiring detailed information on each of a plurality of agents when the first agent determines that it cannot process the user request by itself, and determining a recipient of a first task processing request, which requests to process at least some of a plurality of subtasks constituting a target task for generating a processing result for the user request, based on the detailed information, transmitting the first task processing request to a second agent determined as the recipient of the first task processing request, receiving first external processing result data from the second agent in response to the first task processing request and generating final processing result data for the user request by using the first external processing result data and transmitting the final processing result data to the user terminal.