TY - GEN
T1 - Real-Time Sensor Fault Detection in Drones
T2 - 35th International Conference on Principles of Diagnosis and Resilient Systems, DX 2024
AU - Roshanski, Inbal
AU - Roshanski, Magenya
AU - Kalech, Meir
N1 - Publisher Copyright:
© Inbal Roshanski, Magenya Roshanski, and Meir Kalech.
PY - 2024/11/26
Y1 - 2024/11/26
N2 - Drones, or unmanned aerial vehicles (UAVs), are becoming increasingly vital across various industries, where their reliable operation is crucial for safety and efficiency. Ensuring this reliability requires the early detection of sensor-related faults, which are critical for maintaining the performance and safety of UAVs. This study addresses this challenge by leveraging real-world data from an Aero-Sentinel Military UAV Sentinel G2 quadcopter. The data was collected through a collaboration with Maris-Tech Ltd, using their advanced Mercury Nano system to capture detailed communication between the drone and its control unit. A set of correlation-based algorithms was developed and evaluated, specifically tailored to address the unique complexities of drone sensor data, which is often influenced by environmental factors. Among the algorithms tested, two novel methods emerged as particularly effective, demonstrating significant improvement compared to previous methods, in fault detection accuracy. These methods, designed to accurately identify and predict sensor malfunctions, offer a robust solution for enhancing the reliability and safety of UAV operations.
AB - Drones, or unmanned aerial vehicles (UAVs), are becoming increasingly vital across various industries, where their reliable operation is crucial for safety and efficiency. Ensuring this reliability requires the early detection of sensor-related faults, which are critical for maintaining the performance and safety of UAVs. This study addresses this challenge by leveraging real-world data from an Aero-Sentinel Military UAV Sentinel G2 quadcopter. The data was collected through a collaboration with Maris-Tech Ltd, using their advanced Mercury Nano system to capture detailed communication between the drone and its control unit. A set of correlation-based algorithms was developed and evaluated, specifically tailored to address the unique complexities of drone sensor data, which is often influenced by environmental factors. Among the algorithms tested, two novel methods emerged as particularly effective, demonstrating significant improvement compared to previous methods, in fault detection accuracy. These methods, designed to accurately identify and predict sensor malfunctions, offer a robust solution for enhancing the reliability and safety of UAV operations.
KW - Anomaly Detection
KW - Correlation-Based Algorithms
KW - Data-Driven Fault Detection
KW - Drones
KW - Sensor Data Analysis
KW - Sensor Fault Detection
UR - https://www.scopus.com/pages/publications/85211891140
U2 - 10.4230/OASIcs.DX.2024.17
DO - 10.4230/OASIcs.DX.2024.17
M3 - Conference contribution
AN - SCOPUS:85211891140
T3 - OpenAccess Series in Informatics
BT - 35th International Conference on Principles of Diagnosis and Resilient Systems, DX 2024
A2 - Pill, Ingo
A2 - Natan, Avraham
A2 - Wotawa, Franz
PB - Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing
Y2 - 4 November 2024 through 7 November 2024
ER -