Vehicle and unmanned aerial vehicle cooperative scheduling method based on hierarchical reinforcement learning

Through a method based on layered reinforcement learning, a vehicle and drone collaborative scheduling model is built, which solves the complexity and robustness problems in vehicle and drone collaborative scheduling, and realizes efficient, real-time, resource optimization and cost control of urban rapid logistics transportation.

CN119671420AActive Publication Date: 2025-03-21BEIHANG UNIV
View PDF 4 Cites 2 Cited by

Patent Information

Application Number
CN202411696960.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-26
Publication Date
2025-03-21
Estimated Expiration
2044-11-26

AI Technical Summary

Technical Problem

At present, the coordinated scheduling of vehicles and drones faces challenges such as complex scheduling algorithms, poor real-time robustness, imperfect resource optimization and allocation, and difficult transportation cost control, making it difficult to meet the efficient real-time needs of urban rapid logistics and transportation.

Method used

Using a method based on hierarchical reinforcement learning, an interactive simulation environment for rapid urban logistics transportation is built, a dynamic scheduling model of vehicle drones is built using the r-MDP model, a layered reinforcement learning algorithm framework is designed, and demand allocation, fleet scheduling and route planning are realized through the upper-level agent, the lower-level agent and the path-heuristic algorithm layer.

Benefits of technology

It realizes dynamic allocation of transportation needs, real-time scheduling of vehicles and drones and route planning, optimizes resource allocation, controls transportation costs, reduces environmental pollution, and improves urban logistics and transportation efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119671420A_ABST
    Figure CN119671420A_ABST
Patent Text Reader

Abstract

The invention relates to a vehicle and unmanned aerial vehicle cooperative scheduling method based on hierarchical reinforcement learning, belongs to the technical field of traffic transportation and logistics, solves the problems of high cost, low efficiency and unreasonable resource allocation of'ground-air integration 'joint transportation in the prior art, and comprises the steps of S1, constructing an interactive simulation environment of urban rapid logistics transportation; s2, constructing a dynamic scheduling model of the vehicle unmanned aerial vehicle based on the co-scheduling process of the vehicle unmanned aerial vehicle by using r-MDP; s3, designing a hierarchical reinforcement learning algorithm framework, and solving the vehicle unmanned aerial vehicle dynamic scheduling model through the hierarchical reinforcement learning algorithm framework to obtain a cooperative scheduling strategy; and S4, substituting the obtained collaborative scheduling strategy into the constructed interactive simulation environment for verification to obtain a verification result, if the verification result meets a preset requirement, outputting the collaborative scheduling strategy for collaborative scheduling of the vehicle and the unmanned aerial vehicle, otherwise, returning to the step S3 to modify the parameters and reconstructing the hierarchical reinforcement learning algorithm framework.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Instant delivery order distribution system for rider-unmanned vehicle cooperative delivery

    CN116415882A

  • System for providing transportation management system based logistics service

    KR102299328B1

  • System for autonomous operation of multiple hybrid unmanned aerial vehicles supported by recharging stations to perform services

    US20180188747A1

  • Deep reinforcement learning-based taxi dispatching method and system

    WO2021248607A1

Cited By

  • Concrete dynamic scheduling method based on multistage reinforcement learning and intelligent agent simulation

    CN120409194A

  • Environment simulator and enhanced learning method and device based on environment simulator

    CN121480765A