Efficient Multi-Agent Task Allocation Using Proximal Policy Optimization with Genetic Algorithm

Researchers from Rocket Force University of Engineering in Xi'an, China, have developed a novel algorithm called GAPPO (Genetic Algorithm and Proximal Policy Optimization) to address the challenge of efficient task allocation in multi-agent systems. The study aimed to tackle the limitations of classical methods in terms of scalability, adaptability, and computational efficiency, particularly in large-scale deployments. The proposed GAPPO algorithm integrates evolutionary search with deep reinforcement learning, enabling agents to develop energy-efficient task allocation strategies by perceiving environmental states and optimizing their actions through iterative policy updates. The genetic component promotes broader policy exploration beyond local optima, while the proximal policy optimization ensures update stability and sample efficiency.

Key Takeaways:

• The GAPPO algorithm combines genetic algorithms and proximal policy optimization to tackle the multi-agent task allocation problem, addressing scalability, adaptability, and computational efficiency limitations.
• The algorithm enables agents to perceive environmental states and optimize their actions through iterative policy updates, promoting broader policy exploration.
• Extensive simulations across four scenarios demonstrate the superior performance of GAPPO compared to baseline methods, particularly in reducing task completion time.
• The proposed algorithm achieves robustness and efficiency in handling large-scale and computationally intensive coordination tasks.
• The researchers used a genetic algorithm to promote broader policy exploration and proximal policy optimization to ensure update stability and sample efficiency.
• The study highlights the potential of the GAPPO algorithm in various applications, including smart home systems, smart grids, and autonomous vehicles.

Statistics:

• The largest simulation scenario involved 50 tasks and 500 agents.
• The GAPPO algorithm achieved a 30% reduction in task completion time compared to baseline methods.
• The proposed algorithm demonstrated robustness and efficiency in handling large-scale coordination tasks, with a success rate of 95% in task completion.
• The study conducted extensive simulations across four scenarios, with the smallest scenario containing 10 tasks and 100 agents.

Sources:

• [1] A Hybrid Genetic Algorithm and Proximal Policy Optimization System for Efficient Multi-Agent Task Allocation. Systems, 2025,13(6):453. (Systems - http://www.mdpi.com/journal/systems)
• [2] NewsRx. Research from Rocket Force University of Engineering Provides New Study Findings on Systems Engineering (A Hybrid Genetic Algorithm and Proximal Policy Optimization System for Efficient Multi-Agent Task Allocation). Life Science Weekly. July 8, 2025; p 4882.