一句话：提出耦合物理仿真与照片级可视化的可扩展混合仿真平台，支持城市级多无人机任务与空域合规验证。
论文针对城市规模下自主小型无人机系统（sUAS）验证的仿真需求，指出现有系统普遍并发 sUAS 数量少，且难以对城市环境中的视觉导航、传感器退化以及不断演化的空域法规遵从性进行高保真评估。为此，作者提出 SADE-SIM：一种将物理仿真与照片级真实感可视化耦合的可扩展混合架构，集成了基于 Web 的任务规划、兼容标准协议的实时控制、视频流实时可视化以及飞行数据归档回放。评估上以 32 架异构 sUAS 机队在受限空域中的案例研究展开，无人机动态请求区域进入许可，系统在授予或拒绝前评估关键任务参数。作者认为该平台可支撑自主导航评估、城市峡谷中 GPS 退化导航测试以及 FAA 动态空域管理合规协议的验证流程。
Validating autonomous small Unmanned Aerial Systems (sUAS) at city scale requires simulation frameworks that support large multi-sUAS deployments while maintaining both physical accuracy and geospatial realism. Existing systems typically support few concurrent sUAS and lack fidelity for evaluating vision-based navigation, sensor degradation in urban environments, and compliance with evolving airspace regulations. We introduce SADE-SIM: a scalable hybrid architecture that couples physical simulation with photorealistic visualization. The SADE-SIM integrates web-based mission planning, real-time control compatible with standard protocols, live visualization of video stream, and archival review of flight data. We demonstrate a case study using a 32-sUAS heterogeneous fleet across one restricted airspace zones with diverse operational profiles, where sUAS dynamically request zone entry permissions and the system evaluates mission-critical parameters before granting or denying access. SADE-SIM will enable novel validation workflows such as the evaluation of autonomous navigation systems, GPS-degraded navigation testing in metropolitan canyons, and validation of the FAA compliance protocol for dynamic airspace management. By bridging flight dynamics, photorealistic environments, and regulatory constraints, SADE-SIM advances the research computing infrastructure needed for safe deployment of autonomous aerial systems at scale.
@inproceedings{VadnalaPZMAT26,
title = {SADE-SIM: A Scalable Simulation Platform for Validating City-Scale Multi-sUAS Missions},
author = {Laxminarayana Vadnala and Lucas Parzianello and Bohan Zhang and Michael Murphy and Ankit Agrawal and Douglas Thain},
booktitle = {FSE},
year = {2026},
doi = {10.1145/3803437.3805540}
}