Optimizing Multi-Robot Task Allocation with Dynamic Crisis Response: A Genetic Algorithm Approach with Task Resumption and Island Model Enhancement
Open Access
Abstract
This paper presents an optimization framework for Multi-Robot Task Allocation (MRTA) for a heterogeneous robot fleet operating in dynamic, failure-prone environments. In contrast to traditional MRTA approaches that handle only the initial allocation, our system extends functionality by integrating real-time crisis response and intelligent task recovery from failure points. The framework combines island model genetic algorithm-based initial optimization with an event-driven architecture for handling robot failures during mission execution. Our key contribution is the integration of crisis-aware capabilities with the island model paradigm, enabling task resumption from failure points and dynamic reoptimization, while preserving the diversity benefits of multi-population evolution. When a robot fails, the system intelligently substitutes replacement robots and resumes interrupted tasks from their exact failure point, rather than restarting from the beginning. This significantly improves mission efficiency and resilience. We introduce a temporal scheduling mechanism that tracks actual task execution states and calculates remaining work upon failure, enabling true task continuation. Experimental validation across 57 diverse scenarios with 2340 independent runs demonstrates that the island model achieves higher fitness scores, maintains greater population diversity, exhibits more consistent performance, and recovers faster from crisis events compared to the standard single-population genetic algorithm.Keywords
Funding Information
- National Plan for Science, Technology, and Innovation (MAARIFAH), King Abdulaziz City for Science and Technology, Kingdom of Saudi Arabia (13-SPA1130-02)
This publication has 28 references indexed in Scilit:
- MRTASim: An agent-based multi-robot task allocation simulationSoftware Impacts, 2025
- SA-reCBS: Multi-robot task assignment with integrated reactive path generationIFAC-PapersOnLine, 2023
- Solving the Multi-robot task allocation with functional tasks based on a hyper-heuristic algorithmApplied Soft Computing, 2023
- Market Approaches to the Multi-Robot Task Allocation Problem: a SurveyJournal of Intelligent & Robotic Systems, 2023
- Resilient Robot Teams: a Review Integrating Decentralised Control, Change-Detection, and LearningCurrent Robotics Reports, 2022
- Multi-robot task allocation in disaster response: Addressing dynamic tasks with deadlines and robots with range and payload constraintsRobotics and Autonomous Systems, 2022
- Dynamic multi-robot task allocation under uncertainty and temporal constraintsAutonomous Robots, 2021
- Coupled task scheduling for heterogeneous multi-robot system of two robot types performing complex-schedule order fulfillment tasksRobotics and Autonomous Systems, 2020
- Scheduling of a Robot’s Tasks With the TaskER FrameworkIEEE Access, 2020
- A comparative review of approaches to prevent premature convergence in GAApplied Soft Computing, 2014