Two-phase multi-expert knowledge approach by using fuzzy clustering and rule-based system for technology evaluation of unmanned aerial vehicles

Unmanned Aerial Vehicles are expected to create enormous benefits to society, but there are safety concerns in recognizing faults at the vehicle’s control component. Prior studies proposed various fault detection approaches leveraging heuristics-based rules and supervised learning-based models, but there were several drawbacks. The rule-based approaches required an engineer to update the rules on every type of fault, and the supervised learning-based approaches necessitated the acquisition of a finely-labeled training dataset. Moreover, both prior approaches commonly include a limit that the detection model can identify the trained type of faults only, but fail to recognize the unseen type of faults. In pursuit of resolving the aforementioned drawbacks, we proposed a fault detection model utilizing a stacked autoencoder that lies under unsupervised learning. The autoencoder was trained with data from safe UAV states, and its reconstruction loss was examined to distinguish the safe states and faulty states. The key contributions of our study are, as follows. First, we presented a series of analyses to extract essential features from raw UAV flight logs. Second, we designed a fault detection model consisting of the stacked autoencoder and the classifier. Lastly, we validated our approach’s fault detection performance with two datasets consisting of different types of UAV faults.

Download Full-text

On Analyzing Routing Selection for Aerial Autonomous Vehicles Connected to Mobile Network

Sensors ◽

10.3390/s21020399 ◽

2021 ◽

Vol 21 (2) ◽

pp. 399

Author(s):

Jordi Mongay Batalla ◽

Constandinos X. Mavromoustakis ◽

George Mastorakis ◽

Evangelos K. Markakis ◽

Evangelos Pallis ◽

...

Keyword(s):

Unmanned Aerial Vehicles ◽

Autonomous Vehicles ◽

Mobile Network ◽

Radio Link ◽

Learning Tools ◽

Two Phase ◽

Flight Duration ◽

Aerial Vehicles ◽

Mobile Network Operators ◽

Selection Of

This paper proposes a two-phase algorithm for multi-criteria selection of packet forwarding in unmanned aerial vehicles (UAV), which communicate with the control station through commercial mobile network. The selection of proper data forwarding in the two radio link: From UAV to the antenna and from the antenna to the control station, are independent but subject to constrains. The proposed approach is independent of the intra-domain forwarding, so it may be useful for a number of different scenarios of Unmanned Aerial Vehicles connectivity (e.g., a swarm of drones). In the implementation developed in this paper, the connection is served by three different mobile network operators in order to ensure reliable connectivity. The proposed algorithm makes use of Machine Learning tools that are properly trained for predicting the behavior of the link connectivity during the flight duration. The results presented in the last section validate the algorithm and the training process of the machines.

Download Full-text

MedExt: combining expert knowledge and deep learning for medication extraction from French clinical texts (Preprint)

10.2196/preprints.17934 ◽

2020 ◽

Cited By ~ 1

Author(s):

Jordan Jouffroy ◽

Sarah F Feldman ◽

Ivan Lerner ◽

Bastien Rance ◽

Anita Burgun ◽

...

Keyword(s):

Language Processing ◽

Expert Knowledge ◽

Rule Based ◽

Clinical Text ◽

Clinical Notes ◽

Rule Based System ◽

Related Information ◽

Medication Information ◽

Deep Recurrent Neural Network ◽

F Measure

BACKGROUND Information related to patient medication is crucial for health care. However, up to 80% of the information resides solely in unstructured text. Manual extraction may be difficult and time-consuming. Many studies have shown the interest of natural language processing for this task but only a few on French corpus. OBJECTIVE We aim at developing a system to extract medication-related information from French clinical text. METHODS We developed a hybrid system combining an expert rule-based system (RBS), contextual word embedding (ELMo) trained on clinical notes and a deep recurrent neural network (BiLSTM-CRF). The task consists in extracting drug mentions and their related information (e.g. dosage, frequency, duration, route, condition). We manually annotated 320 clinical notes extracted from a French clinical data warehouse, to train and evaluate the model. We compared the performances of our approach to standard approaches: rule-based or machine learning only, and classic word embeddings. We evaluated the models using token level recall, precision and F-measure. RESULTS Models including RBS, ELMo and BiLSTM reached the best results: overall F-measure of 89.9%. F-measures per category were 95.3% for the medication name, 64.4% for the drug class mentions, 95.3% for the dosage, 92.2% for the frequency, 78.8% for the duration, and 62.2% for the condition of the intake. CONCLUSIONS Associating expert rules, deep contextualized embedding (ELMo) and deep neural networks improves medication information extraction. Our results reveal a synergy when associating expert knowledge and latent knowledge.

Download Full-text

Development of thickened semi-synthetic engine oil M-5z / 20 AERO for four-stroke gasoline engines aircraft piston of unmanned aerial vehicles (UAVs)

World of OIL Products the Oil Companies Bulletin ◽

10.32758/2071-5951-2020-0-6-44-47 ◽

2020 ◽

Vol 06 ◽

pp. 44-47

Author(s):

A.A. Moykin ◽

◽

A.S. Medzhibovsky ◽

S.A. Kriushin ◽

M.V. Seleznev ◽

...

Keyword(s):

Unmanned Aerial Vehicles ◽

Engine Oil ◽

Motor Oil ◽

Gasoline Engines ◽

State Research Institute ◽

Technical Requirements ◽

Aerial Vehicles ◽

State Research ◽

Industrial Batch ◽

The Russian Federation

Nowadays, the creation of remotely-piloted aerial vehicles for various purposes is regarded as one of the most relevant and promising trends of aircraft development. FAU "25 State Research Institute of Chemmotology of the Ministry of Defense of the Russian Federation" have studied the operation features of aircraft piston engines and developed technical requirements for motor oil for piston four-stroke UAV engines, as well as a new engine oil M-5z/20 AERO in cooperation with NPP KVALITET, LLC. Based on the complex of qualification tests, the stated operational properties of the experimental-industrial batch of M-5z/20 AERO oil are generally confirmed.

Download Full-text