An Agentic AI Framework with Large Language Models and Chain-of-Thought for UAV-Assisted Logistics Scheduling with Mobile Edge Computing

2026-03-05 · Source: cs.AI updates on arXiv.org · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Operations & Process Management · Depth: Expert, long

Summary

A new agentic AI framework addresses the complex hybrid scheduling problem of UAV-assisted logistics and mobile edge computing (MEC) in cloud manufacturing. This problem involves UAVs collecting products and simultaneously providing MEC services, where routing decisions impact both logistics and computational task scheduling, including energy budgets and task deadlines. The framework features two main components: an agentic AI that uses large language models (LLMs), retrieval-augmented generation (RAG), and chain-of-thought (CoT) reasoning to translate user input into an interpretable mathematical formulation. The second component is a hierarchical deep reinforcement learning (DRL) approach based on proximal policy optimization (PPO), with an upper layer for UAV routing and a lower layer for per-slot task execution and resource allocation. Simulations demonstrate the framework's ability to generate consistent formulations, with the hierarchical PPO achieving 99.6% full product collection and 100% deadline satisfaction.

Key takeaway

For research scientists developing intelligent manufacturing systems, you should consider integrating agentic AI with hierarchical DRL to tackle complex, coupled scheduling problems. This approach can help you efficiently derive interpretable mathematical models and achieve robust, high-performance solutions for joint logistics and computational task management, ensuring both operational efficiency and task deadline adherence in dynamic environments.

Key insights

An agentic AI framework optimizes UAV logistics and mobile edge computing via LLMs and hierarchical DRL.

Principles

• Decompose complex problems into sequential decision layers.
• Combine LLM reasoning with RAG and CoT for model generation.

Method

The method involves an agentic AI for mathematical formulation using LLMs, RAG, and CoT, followed by a hierarchical DRL approach with PPO for solving, separating UAV routing from task execution and resource allocation.

In practice

• Use agentic AI for complex combinatorial problem formulation.
• Apply hierarchical DRL for coupled logistics-computation scheduling.

Topics

• Agentic AI Framework
• Large Language Models
• UAV-Assisted Logistics
• Mobile Edge Computing
• Hybrid Scheduling

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, Robotics Engineer

Related on AIssential

• An Agentic AI Framework with Large Language Models and Chain-of-Thought for UAV-Assisted Logistics Scheduling with Mobile Edge Computing — An agentic AI framework optimizes UAV logistics and MEC task scheduling using LLMs and hierarchical DRL.
• Agentic AI for Trip Planning Optimization Application — Orchestrated agentic AI significantly improves trip planning optimization and robustness over single-agent or uncoordinated multi-agent systems.
• Article: Orchestrating Agentic and Multimodal AI Pipelines with Apache Camel — Orchestrate agentic and multimodal AI with integration frameworks for production reliability and clear separation of concerns.
• From data overload to actionable insights: How Verizon Connect scaled agentic AI to 100,000 users — Agentic AI transforms vast fleet data into actionable insights by dynamically investigating patterns and adapting analysis.
• Agentic MPC for Semantic Control System Resynthesis — An agentic MPC framework integrates LLM-based agents to enable context-aware, semantically adaptive control synthesis.
• AI agents need context everywhere they run, even where the cloud can't follow — Enterprise AI competitive advantage shifts to platforms providing real-time, context-aware agent memory across diverse environments.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.AI updates on arXiv.org.