Baike.dev
All toolsOnline toolsAI codingTrendingOpen sourceNewsSubmit
Log in
< Back to tools
I

ipp-marl

> 编程语言
Open source

Multi-UAV Adaptive Path Planning Using Deep Reinforcement Learning

199 stars0 likes1 views
WebsiteGitHub

About

Multi-UAV Adaptive Path Planning Using Deep Reinforcement Learning

Multi-UAV Adaptive Path Planning Using Deep Reinforcement Learning

Jonas Westheider, Julius Rückin, Marija Popovic University of Bonn

This repository contains the implementation of our paper Multi-UAV Adaptive Path Planning Using Deep Reinforcement Learning submitted to IROS2023 (under review). The implementation is currently being cleaned up.

Abstract

Efficient aerial data collection is important in many remote sensing applications. In large-scale monitoring scenarios, deploying a team of unmanned aerial vehicles (UAVs) offers improved spatial coverage and robustness against individual failures. However, a key challenge is cooperative path planning for the UAVs to efficiently achieve a joint mission goal. We propose a novel multi-agent informative path planning approach based on deep reinforcement learning for adaptive terrain monitoring scenarios using UAV teams. We introduce new network feature representations to effectively learn path planning in a 3D workspace. By leveraging a counterfactual baseline, our approach explicitly addresses credit assignment to learn cooperative behaviour. Our experimental evaluation shows improved planning performance, i.e. maps regions of interest more quickly, with respect to non-counterfactual vari- ants. Results on synthetic and real-world data show that our approach has superior performance compared to state-of-the-art non-learning-based methods, while being transferable to varying team sizes and communication constraints.

Overview

At each time step during a mission, each UAV takes a measurement and updates its local map state. The local map is input to an actor network, which outputs a policy from which an action is sampled. During training, a centralised critic network is additionally trained using global map information and outputs Q-values for each action from the current state, i.e. the expected future return.

Maintainer

Jonas Westheider, [email protected], Ph.D. student at PhenoRob - University of Bonn.

Requirements

matplotlib==3.5.1

numpy==1.22.2

opencv-python==4.5.5.62

scipy==1.8.1

torch==1.13.0+cu117

Funding

This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy - EXC 2070 – 390732324. Authors are with the Cluster of Excellence PhenoRob, Institute of Geodesy and Geoinformation, University of Bonn.

Issues· 3 open

View all issuesOpen on GitHub

No open issues yet, or sync has not completed.

> Tags

Python

No comments yet. Be the first to share.

> Details

PublishedAug 1, 2026
UpdatedSep 18, 2026
Category编程语言
PricingOpen source

> Related tools

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言