Uncovering drone intentions using control physics informed machine learning
Abstract Unmanned Autonomous Vehicle (UAV) or drones are increasingly used across diverse application areas. Uncooperative drones do not announce their identity/flight plans and can pose a potential risk to critical infrastructures. Understanding drone’s intention is important to assigning risk and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-02-01
|
| Series: | Communications Engineering |
| Online Access: | https://doi.org/10.1038/s44172-024-00179-3 |
| _version_ | 1797199426201059328 |
|---|---|
| author | Adolfo Perrusquía Weisi Guo Benjamin Fraser Zhuangkun Wei |
| author_facet | Adolfo Perrusquía Weisi Guo Benjamin Fraser Zhuangkun Wei |
| author_sort | Adolfo Perrusquía |
| collection | DOAJ |
| description | Abstract Unmanned Autonomous Vehicle (UAV) or drones are increasingly used across diverse application areas. Uncooperative drones do not announce their identity/flight plans and can pose a potential risk to critical infrastructures. Understanding drone’s intention is important to assigning risk and executing countermeasures. Intentions are often intangible and unobservable, and a variety of tangible intention classes are often inferred as a proxy. However, inference of drone intention classes using observational data alone is inherently unreliable due to observational and learning bias. Here, we developed a control-physics informed machine learning (CPhy-ML) that can robustly infer across intention classes. The CPhy-ML couples the representation power of deep learning with the conservation laws of aerospace models to reduce bias and instability. The CPhy-ML achieves a 48.28% performance improvement over traditional trajectory prediction methods. The reward inference results outperforms conventional inverse reinforcement learning approaches, decreasing the root mean squared spectral norm error from 3.3747 to 0.3229. |
| first_indexed | 2024-03-07T14:58:49Z |
| format | Article |
| id | doaj.art-c94dfbb8ffe34384b5281ab3163e9bcb |
| institution | Directory Open Access Journal |
| issn | 2731-3395 |
| language | English |
| last_indexed | 2024-04-24T07:15:34Z |
| publishDate | 2024-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Engineering |
| spelling | doaj.art-c94dfbb8ffe34384b5281ab3163e9bcb2024-04-21T11:20:35ZengNature PortfolioCommunications Engineering2731-33952024-02-013111410.1038/s44172-024-00179-3Uncovering drone intentions using control physics informed machine learningAdolfo Perrusquía0Weisi Guo1Benjamin Fraser2Zhuangkun Wei3School of Aerospace, Transport and Manufacturing, Cranfield UniversitySchool of Aerospace, Transport and Manufacturing, Cranfield UniversitySchool of Aerospace, Transport and Manufacturing, Cranfield UniversitySchool of Aerospace, Transport and Manufacturing, Cranfield UniversityAbstract Unmanned Autonomous Vehicle (UAV) or drones are increasingly used across diverse application areas. Uncooperative drones do not announce their identity/flight plans and can pose a potential risk to critical infrastructures. Understanding drone’s intention is important to assigning risk and executing countermeasures. Intentions are often intangible and unobservable, and a variety of tangible intention classes are often inferred as a proxy. However, inference of drone intention classes using observational data alone is inherently unreliable due to observational and learning bias. Here, we developed a control-physics informed machine learning (CPhy-ML) that can robustly infer across intention classes. The CPhy-ML couples the representation power of deep learning with the conservation laws of aerospace models to reduce bias and instability. The CPhy-ML achieves a 48.28% performance improvement over traditional trajectory prediction methods. The reward inference results outperforms conventional inverse reinforcement learning approaches, decreasing the root mean squared spectral norm error from 3.3747 to 0.3229.https://doi.org/10.1038/s44172-024-00179-3 |
| spellingShingle | Adolfo Perrusquía Weisi Guo Benjamin Fraser Zhuangkun Wei Uncovering drone intentions using control physics informed machine learning Communications Engineering |
| title | Uncovering drone intentions using control physics informed machine learning |
| title_full | Uncovering drone intentions using control physics informed machine learning |
| title_fullStr | Uncovering drone intentions using control physics informed machine learning |
| title_full_unstemmed | Uncovering drone intentions using control physics informed machine learning |
| title_short | Uncovering drone intentions using control physics informed machine learning |
| title_sort | uncovering drone intentions using control physics informed machine learning |
| url | https://doi.org/10.1038/s44172-024-00179-3 |
| work_keys_str_mv | AT adolfoperrusquia uncoveringdroneintentionsusingcontrolphysicsinformedmachinelearning AT weisiguo uncoveringdroneintentionsusingcontrolphysicsinformedmachinelearning AT benjaminfraser uncoveringdroneintentionsusingcontrolphysicsinformedmachinelearning AT zhuangkunwei uncoveringdroneintentionsusingcontrolphysicsinformedmachinelearning |