Development Of A Unity-Based Drone Simulation Framework For Precision Agriculture

Document Type

Conference Proceeding

Publication Date

1-1-2026

School

Computing Sciences and Computer Engineering

Abstract

To bridge the gap between development and real-world deployment, we present a Unity-based drone simulation framework for precision agriculture that uniquely combines the Unity game engine with Microsoft AirSim’s physics library. This integration leverages Unity’s versatile rendering and environment design with AirSim’s high-fidelity flight dynamics, a novel approach given AirSim’s only recently available experimental Unity port. The result is a realistic simulation of drone flight, sensor outputs, and farm environments, achieving photorealistic visuals and accurate physics. Key objectives of our framework include testing autonomous flight algorithms (e.g., coverage path planning and basic obstacle avoidance) in a risk-free, virtual farm setting and evaluating their performance on agricultural tasks. The methodology shows the simulation environment design, including high-fidelity crop modeling, aerial dynamics, camera imaging, and data logging strategies. Results from our experiments demonstrate that the simulated drone can precisely follow planned waypoints and collect high-resolution aerial images of crops, indicating that the Unity+AirSim approach yields credible flight behavior and data for analysis. The realism of the simulation (e.g., natural drone accelerations, camera sensor noise, and detailed crop models) provides confidence that algorithms refined in this virtual setting will transfer well to physical drones. We also outline forward-looking uses of the framework, including incorporating multi-drone operations and crop disease models. In ongoing and future work, we plan to validate the simulation’s effectiveness with real-world field trials, using the insights gained to further improve fidelity. This Unity-based simulation framework thus offers a novel, practical tool for developing and de-risking autonomous drone strategies in precision agriculture before deployment on actual farms. The paper concludes with a discussion of the simulation’s contributions, current limitations, and future directions for enhancing realism and supporting multi-drone operations in precision farming.

Publication Title

Communications in Computer and Information Science

Volume

2720 CCIS

First Page

3

Last Page

17

Link to Full Text

Share

COinS