A Modified Multiple Ant Colony Optimization Algorithm with Adaptive Evaporation and Obstacle-Aware Heuristics for UAV Path Planning in Complex 2D and 3D Environments
Description
This dataset accompanies the research article entitled "Modified Ant Colony Optimization Algorithm for UAV Path Planning in 2D and 3D Environments". It contains the complete source code, experimental results, statistical analyses, raw experimental data, and supporting documentation required to reproduce and validate the findings presented in the manuscript. The source code is implemented in Python and includes the proposed Modified Ant Colony Optimization (MACO) algorithm together with four benchmark optimization algorithms, namely Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and Genetic Algorithm (GA). The implementation covers both two-dimensional (2D) and three-dimensional (3D) UAV path planning scenarios, with three benchmark environments provided for each dimensionality. Each environment contains dedicated scripts for algorithm execution, experiment management, environment generation, cost function evaluation, and result visualization. The dataset also includes all experimental outputs generated during the study. These comprise optimized UAV trajectories, convergence curves, algorithm pipeline illustrations, adaptive evaporation rate analysis, statistical summary tables, and graphical representations of the comparative performance of the evaluated algorithms. To ensure the statistical reliability of the reported results, the raw data from 75 independent experimental trials for each benchmark environment are included in Microsoft Excel format. In addition, Python scripts used for statistical analysis, including Wilcoxon signed-rank significance testing and statistical summary generation, are provided together with the corresponding output files. The dataset is organized into separate directories containing: Python source code for all optimization algorithms and experimental environments. Experimental results and generated figures. Raw data from 75 independent trials for each 2D and 3D benchmark case. Statistical analysis scripts and outputs. A comprehensive README file describing the directory structure, software requirements, execution procedure, and interpretation of the results. This dataset enables complete reproducibility of the experiments presented in the associated publication and provides researchers with the necessary resources to validate, compare, extend, or benchmark optimization algorithms for UAV path planning applications.
Files
Steps to reproduce
1) Configure the Python environment by installing the required dependencies listed in the README file. 2) Select the desired experimental scenario (2D or 3D environment) and the corresponding benchmark case (Case 1, Case 2, or Case 3). 3) Execute the Master.py script to initialize the environment, configure algorithm parameters, and run the selected optimization algorithm. 4) Run the experiment scripts (MACO Experiment.py, ACO Experiment.py, PSO Experiment.py, GWO Experiment.py, or GA Experiment.py) to perform UAV path planning simulations for the selected algorithm. 5) Repeat each experiment for 75 independent trials to generate statistically reliable performance data. 6) Save the generated outputs, including optimized paths, convergence curves, and performance metrics. 7) Execute the statistical analysis scripts to compute descriptive statistics and perform Wilcoxon signed-rank significance tests. 8) Export the statistical summaries, graphical results, and raw experimental data to the corresponding results folders. 9) Organize all source code, raw data, statistical analyses, figures, and supporting documentation into the provided directory structure to reproduce the complete dataset associated with the publication.
Institutions
- Air UniversityIslamabad, Islamabad