Countering Unmanned Aerial Vehicle (UAV) has become a major global problem in recent years and its significance is expected to increase as national and local governments plan for beyond visual line of sight operations for UAV s. Determining if a UAV has a hostile intent is central to Counter UAV (CUAV) operations and can be inferred from UAV kinematic tracking data. This paper provides a Model-based Systems Engineering (MBSE) approach for automated high-level state estimation of small UAV s (e.g., climbing, cruising, loitering, etc.) using kinematic tracking data provided by a sensing infrastructure. For UAV surveillance and situational awareness, it imperative that systems monitoring UAV operations can determine UAV states because these states and corresponding changes can be directly used to infer UAS missions. For example, a UAV loitering over protected area may be on a reconnaissance mission or a rapidly descending UAVs towards an infrastructure may have a nefarious intent. Nevertheless, it is not trivial for a human operator to extract this information from the kinematic state (such as speed and velocity etc.) due to the large amount of data, variability, and underlying uncertainties in sensing and tracking data. This paper first develops a UAV state estimation model with System Modeling Language (SysML) for automated high level state estimation, then links this model with a kinematic sensor simulation in MATLAB for verification. The results demonstrate a live integration between the MBSE/SysML environment and MATLAB that can provide a surveillance dashboard where high-level UAV states and behavior are depicted based on sensing and tracking information. Furthermore, we implement multiple sensors with track-to-track fusion to demonstrate how reducing uncertainty in kinematic state estimation improves the automated hiah-level state estimation.
Model-based Automated Intent Estimation of Unmanned Aerial Vehicles with Sensor Fusion
2024-09-29
1354425 byte
Conference paper
Electronic Resource
English
Automated Cinematography with Unmanned Aerial Vehicles
Eurographics | 2016
|POSE ESTIMATION OF UNMANNED AERIAL VEHICLES BASED ON A VISION-AIDED MULTI-SENSOR FUSION
DOAJ | 2016
|Automated Cinematography with Unmanned Aerial Vehicles
DataCite | 2016
|