BROWSE
Detail
Conference Paper
Development of a Simulated UAV Platform for Sensor-Fusion-Based SLAM
Cited - time in 
Share 
Share
- Authors
- Muhammad Fairuz Mummtaz, Jaejun Yoo, Ida Bagus Krishna Yoga Utama, Irzal Zaini, Yeong Min Jang
- Citation
- International Conference on Artificial Intelligence in Information and Communication (ICAIIC) 2026, pp.1-5
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- Accurate positioning is crucial for autonomous navigation of unmanned aerial vehicles (UAVs), as it provides reliable pose information to the perception and planning-control modules. However, when global navigation satellite system (GNSS) signals are unavailable or degraded, the accumulated drift of deadreckoning and inertial estimates significantly reduces localization accuracy, especially in complex 3D environments. To address this issue, this paper proposes a simulation-based sensor fusion based simultaneous localization and mapping (SLAM) framework for robust UAV positioning without relying on GNSS. The framework directly utilizes sequential 3D LiDAR scans to estimate the UAV pose and incrementally build a dense map of the environment, while high-resolution camera streams provide contextual information for the segmentation task. In addition, a modular simulation pipeline is constructed to support configurable sensor models, flight trajectories, and scene geometries, enabling systematic evaluation under GNSS-denied scenarios. The proposed approach is validated using a simulation experiments on a complex environment, and the results demonstrate that the 3D LiDAR SLAM system achieves accurate and stable localization in the simulation environment.
- Keyword
- UAV, Simultaneous Localization and Mapping, Segmentation, Sensor Fusion, Deep Learning
- KSP Keywords
- 3D Lidar, 3D environment, Accurate positioning, High resolution, Simulation Environment, Simulation and experiment, UAV Positioning, autonomous navigation, complex environment, deep learning(DL), global navigation satellite systems(GNSSs)