Quantization-aware distributed deep reinforcement learning for dynamic multi-robot scheduling

doi:10.1016/j.eswa.2025.129027

Quantization-aware distributed deep reinforcement learning for dynamic multi-robot scheduling

2025-07-16

0

PRE

AI

DOI：10.1016/j.eswa.2025.129027

Original

Original request for help

Save

Abstract

En 中文

• Distributed RL framework for real-time cargo scheduling in dynamic multi-port multi-robot scenarios. • MaxNextQ with ε-greedy balances exploration and exploitation of promising decisions in high-dimensional optimization. • Fine-tuned Quantization-Aware Training to accelerate model convergence and enhance deployment efficiency. • Achieving 5.75% higher scores and 22.95% faster completion vs standard benchmark datasets.

Keywords:

Distributed Reinforcement Learning

Multi-port Multi-robot Scheduling

MaxNextQ

Quantization-Aware Training

Real-time Optimization

Quantization-aware distributed deep reinforcement learning for dynamic multi-robot scheduling

Abstract

Journal

Organization