ENAC

EUROCAE Symposium and 56th General Assembly held in ENAC on 25-26 April 2019. With a participation of more than 200 guests, chaire drone is represented by Elgiz Baskaya as one of the moderators of the fruitful sessions of the symposium.

The main focus was on automatization both for airborne and ground systems. 

Most of the parties agreed the fact that machine learning systems will have a bigger part in the years to come but not many people are working on the certification of such adaptive systems. 

Automatization was also mentioned largely for manned aviation such as for the purpose of decreasing the required number of pilots onboard while still keeping them in the loop. Since automatization was the main topic of the event, it was almost agreed by all parties that ICAO’s definition for autonomy was not sufficient or precise anymore. So there is a need for autonomy levels for aviation as is the case for car industry.

Automatization in case of safety critical situations is discussed by European Defence Agency. They stated that they are interested in fault detection and safe ditch systems for drones, which lies parallel to the topics of interests of drone chair.

For further conclusions and highlights of the conference, please refer to link.

Our member Elgiz Baskaya will defend her thesis on 16/05/2019 at 10:00 am in room G11 in ENAC. We welcome you all.

Titre : ‘Fault detection and diagnosis for drones using machine learning’

Resume :

This new era of small UAVs currently populating the airspace introduces many safety concerns, due to the absence of a pilot onboard and the less accurate nature of the sensors. This necessitates intelligent approaches to address the emergency situations that will inevitably arise for all classes of UAV operations as defined by EASA (European Aviation Safety Agency). Hardware limitations for these small vehicles point to the utilization of analytical redundancy, rather than to the usual practice of hardware redundancy in manned aviation. In the course of this study, machine learning practices are implemented in order to diagnose faults on a small fixed-wing UAV to avoid the burden of accurate modeling needed in model-based fault diagnosis. A supervised classification method, SVM (Support Vector Machines) is used to classify the faults. The data used to diagnose the faults are gyro and accelerometer measurements. The idea to restrict the data set to accelerometer and gyro measurements is to check the method’s classification ability, with a small and inexpensive chip set and without the need to access the data from the autopilot, such as the control input information.

This work addresses the faults in the control surfaces of a UAV. More specifically, the faults considered are the control surface stuck at an angle and the loss of effectiveness. First, a model of an aircraft is simulated. This model is not used for the design of Fault Detection and Diagnosis (FDD) algorithms, but is instead utilized to generate data. Simulated data are used instead of flight data in order to isolate the probable effects of the controller on the diagnosis, which may complicate a preliminary study on FDD for drones. The results show that for simulated measurements, SVM gives very accurate results on the classification of the loss of effectiveness faults on the control surfaces. These promising results call for further investigation so as to assess SVM performance on fault classification with flight data. Real flights were arranged to generate faulty flight data by manipulating the open source autopilot, Paparazzi. All data and the code are available in the code sharing and versioning system, Github. Training is held offline due to the need for labeled data and the computational burden of the tuning phase of the classifiers. Results show that from the flight data, SVM yields an F1 score of 0.98 for the classification of control surface stuck faults. For the loss of efficiency faults, some feature engineering, involving the addition of past measurements is needed in order to attain the same classification performance. A promising result is discovered when spinors are used as features instead of angular velocities. Results show that by using spinors for classification, there is a vast improvement in classification accuracy, especially when the classifiers are untuned. Using spinors and a Gaussian Kernel, an untuned classifier gives an F1 score of 0.9555, which was 0.2712 when gyro measurements were used as features. In summary, this work shows that SVM gives a satisfactory performance for the classification of faults on the control surfaces of a drone using flight data.

Jury members:

The Chair will present its scientific work at the World ATM Congress in Madrid, as part of the World ATC Congress. This presentation will take place on March 12, 2019 at 2:20 p.m. in the Tower Theatre. The conference, entitled « The ENAC – Groupe ADP – Sopra Seria research Chair on drone systems: a new way to explore drones integration into airspace » will be presented by the Chairholder Yannick Jestin. The Chair will also be presented throughout the exhibition on the ENAC – DSNA booth (n°480).

Presentation abstract :

« In the world of Unmanned Air Systems (UAS), the future is almost here; but it can, and must, be shaped still, by working on both the vehicles themselves and the organization of the airspace in which they will be deployed. Added to that, new trends keep emerging, pulling the future in different possible directions (Urban Air Mobility, driverless personal air vehicles …).

A study of ongoing efforts to shape the future of UAS shows that these efforts are almost always divided in two categories moving in parallel: work on the vehicle (trying to increase its Technology Readiness Level: its size, endurance, autonomy level, improve its software’s assurance level …); and work on the concept of operations (EASA, EUROCONTROL, SESAR …).

The Ecole Nationale de l’Aviation Civile (ENAC) – Groupe ADP – Sopra Steria Research Chair on Drone Systems is helping to bridge that gap, by providing inputs for both vehicle technology and the environment in which they operate.

The Chair has a three-year-old background in Detect And Avoid and fault detection and management. Embedded in the ENAC’s Research Lab on Small UAS, it benefits from an environment rich in skill and testing facilities, providing research on topics such as Vertical Take Off and Landing fixed wing aircraft. Added to that, the Chair works on the development of UAS operations within the airport environment, for example for the inspection of radio navigation systems.

As a result of this approach mixing both vehicle and environment, the Chair contributes to building some of the blocks of U-space. »

More info on the new Chair’s leaflet >>>