Navigation Performance Analysis of LIDAR-Based KISS-ICP Enhanced with Dynamically Updated Global Maps and Integrity Monitoring for UAV Operations in GNSS-Denied Environments

Anne-Sophie Polz, Maarten Uijt de Haag

Peer Reviewed

Abstract:	The increasing use of Uncrewed Aerial Vehicles (UAVs) in low-altitude urban environments demands navigation systems that are robust, accurate, and allow integrity monitoring. However, navigation based on Global Navigation Satellite Systems (GNSS) often has degrading performance in urban airspace due to multipath, shadowing, and non-line-of-sight effects. LiDARbased navigation methods offer a promising alternative for GNSS-denied environments, but many existing approaches focus primarily on accuracy and lack mechanisms to monitor solution integrity or handle abnormal operating conditions, limiting their suitability for safety-critical autonomous UAV operations. This paper presents an enhanced LiDAR/inertial navigation framework based on the well-known KISS-ICP pipeline, tailored for UAV applications in indoor and urban outdoor environments. The proposed approach integrates a priori knowledge in form of public or custom point cloud data into the local voxel-hash-map used for scan-map registration. A tagging and weighting mechanism distinguishes global and local map points, enabling weighted point-to-point ICP optimization that prioritizes consistency with the global map while retaining adaptability through configurable local map updates. Several map-handling modes are supported, including global-only registration, fixed global–local updates, and temporary local map augmentation with controlled decay. To account for UAV dynamics, a loosely coupled inertial measurement unit (IMU) integration is employed for pre-leveling LiDAR scans prior to registration, improving robustness during dynamic maneuvers. The use of global map information further enables integrity monitoring through metrics such as a global-point fitness score and keypoint residual-based Root Mean Square Error (RMSE). The system is evaluated using real flight data from indoor and outdoor UAV platforms equipped with lightweight Livox LiDAR sensors. To demonstrate improved robustness and integrity checks, manual fault scenarios are introduced.
Published in:	Proceedings of the ION 2026 Pacific PNT Meeting April 13 - 16, 2026 Hilton Waikiki Beach Honolulu, Hawaii
Pages:	243 - 257
Cite this article:	Polz, Anne-Sophie, de Haag, Maarten Uijt, "Navigation Performance Analysis of LIDAR-Based KISS-ICP Enhanced with Dynamically Updated Global Maps and Integrity Monitoring for UAV Operations in GNSS-Denied Environments," Proceedings of the ION 2026 Pacific PNT Meeting, Honolulu, Hawaii, April 2026, pp. 243-257. https://doi.org/10.33012/2026.20629
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In