UAV Path Planning Benchmark

This repository contains the MATLAB source code associated with the paper:
A Multi-objective Benchmark for UAV Path Planning with Baseline Results
Authors: Daison Darlan, Oladayo S. Ajani, Anand Paul, Rammohan Mallipeddi

The paper introduces a flexible framework for generating synthetic environments (urban, suburban, and mountainous) and presents adapted versions of several multi-objective evolutionary algorithms (MOEAs) used in the comparative analysis for UAV path planning.

Generated environment examples:

This repository provides the MATLAB code for:

• Environment Generation: MATLAB scripts to generate realistic simulation environments including terrain modeling (using sinusoidal and Gaussian-based formulations), building placement, and no-fly zone integration.
• Algorithm Implementations: Adapted and benchmarked MATLAB implementations of several MOEAs for UAV path planning, including:
  • NSGA-II and NSGA-II/SDR
  • MOEA/D and MOEA/D-AWA
  • HypE
  • 𝐼𝑆𝐷𝐸+
  • MOEA-2DE

Getting Started

Prerequisites

• MATLAB R2019b or later
• Required MATLAB toolboxes

Installation

1. Clone the repository:

   git clone https://github.com/Anomaly33/UAV-Path-Planning-Benchmark.git

2. Open the cloned repository in MATLAB.

Running the Code

• Generate Environments: In MATLAB, run the environment generator script to create and visualize sample environments:

  run('Problem Generation/City.m')

• Run Benchmark Experiments: Execute the benchmark script to run comparative experiments on the provided test scenarios:

  run('examples/NSGA-II/NSGAII.m')
  

If you find this work useful in your research, please consider citing our paper:

@article{darlan2025multi,
  title={A multi-objective benchmark for UAV path planning with baseline results},
  author={Darlan, Daison and Ajani, Oladayo S and Paul, Anand and Mallipeddi, Rammohan},
  journal={Swarm and Evolutionary Computation},
  volume={96},
  pages={101968},
  year={2025},
  publisher={Elsevier}
}