A Hierarchical Framework and Marginal Return Optimization for Dynamic Task Allocation in Heterogeneous UAV Networks

Area of Science:

• Robotics and Autonomous Systems
• Multi-Agent Systems
• Operations Research

Background:

• Coordinating heterogeneous Unmanned Aerial Vehicles (UAVs) for complex, multi-stage tasks is challenging.
• Traditional linear models struggle with emergent synergistic effects and dynamic multi-agent collaboration.
• Existing approaches lack robust methods for efficient dynamic task allocation in complex scenarios.

Purpose of the Study:

• To propose a novel hierarchical framework for coordinating heterogeneous UAVs.
• To introduce a theoretical structure for modeling multi-agent collaboration, including Mission Chains (MCs), Execution Paths (EPs), Task Networks (TNs), and Solution Spaces (SSs).
• To develop an efficient dynamic task allocation algorithm for complex missions.

Main Methods:

• Defined a hierarchical framework based on the Mission Chain (MC) concept.
• Modeled key elements: Mission Chains (MCs), Execution Paths (EPs), Task Networks (TNs), and Solution Spaces (SSs).
• Formulated the problem as a Sensor-Effector-Target Assignment challenge and proposed the Marginal Return-Based Heuristic Algorithm (MRBHA).

Main Results:

• The MRBHA significantly outperformed standard greedy and random assignment strategies.
• Achieved a 14% higher total expected mission value compared to greedy assignment.
• Achieved a 77% higher total expected mission value compared to random assignment, demonstrating effective capitalization on synergistic opportunities.

Conclusions:

• The proposed hierarchical framework and MRBHA provide a robust and scalable solution for complex UAV coordination.
• The approach effectively manages dynamic task allocation in complex environments.
• Potential applications include search-and-rescue, environmental monitoring, and intelligent logistics.