Safety Enhancement

Diamond DA42 in automatic flightPicture: © Diamond Aircraft

 


 

FlySmart

Automatic Diamond DA42

The LUFO IV.4 project FlySmart was aimed at achieving fully automatic take-off and landing (ATOL) for EASA CS-23 aircraft without any ground based facilities. FlySmart was carried out as a collaboration between Diamond Aircraft, Airbus Defence & Space, and University of Stuttgart (iFR - Institute of Flight Mechanics and Controls, ILS - Institute for Aircraft Systems).

The planning, guidance, and control functions were developed at the Institute of Flight Mechanics and Controls to safely operate a CS-23 aircraft in all phases of a mission. The overall flight plan is defined with splines in a UTM-based cartesian coordinate system, allowing precise landings on the whole globe even with single precision variables. Flyability and comfort are guaranteed by the planning algorithm through the use of aircraft-specific data.

Planned trajectory with holding pattern

The control loops are formulated as monolithic multiple input multiple output (MIMO) controllers, in conjuction with gain scheduling and a dedicated anti-windup implementation, to address the aircraft dynamic characteristics throughout the whole envelope and various configurations.
The design and verification process included Software-in-the-Loop and Hardware-in-the-Loop setups, including an extensive MonteCarlo campaign aimed at verifying the robustness of the algorithm design. The success criteria were derived from EASA's CS-AWO limitations for CS-25 aircraft and readapted to the CS-23 class considered in the project.

A DA42, registration OE-FMP, was used for the final test campaign. This took place in Wiener Neustadt (A) during the summer 2015 and has led to several successful automatic landings, the first of which on August 26th 2015. This was followed by an automatic takeoff demonstration performed on September 18th 2015.

The video bellow shows a full automatic landing until full stop during the test campaign in September 2015 in Wiener Neustadt.

  • Stephan, Johannes and Walter Fichter. 2016. Nonlinear flight controller design with LPV analysis for general aviation aircraft (to be published)
  • Federico Pinchetti, Alexander Joos and Walter Fichter. 2016. Efficient G2 Path Planning and Tracking Algorithms with Pseudo-normalized Algebraic Splines (to be published)
  • Federico Pinchetti, Johannes Stephan, Alexander Joos and Walter Fichter. 2016. FlySmart - Automatic Take-Offand Landing for an EASA CS-23 Aircraft (to be published)
  • Reimund Kueke, Peter Mueller, Sebastian Polenz, Reinhard Reichel, Federico Pinchetti, Johannes Stephan, Alexander Joos and Walter Fichter. 2016. Fly-By-Wire for CS23 Aircraft - Core Technology for General Aviation and RPAS

 


 

Pilot Support Application for Emergency Landing

Case study for different scenarios

Failure or loss of the powerplant is one of the most common cause of non-fatal and fatal personal flying accidents in general aviation. If problems associated with fuel management and loss of the powerplant are considered jointly, failure of the propulsion system accounts for the largest share of overall accidents. In case of a total loss of thrust quick availability of an effective glide path to the runway is essential.
Within this project a novel approach to the obstacle-free emergency landing problem was developed that provides a fast solution, in order to be easily real-time implementable. The proposed method uses a generic landing scheme composed of a Dubins Path and a subsequent final approach based on best glide. The scheme is described by a set of algebraic, nonlinear systems of equations, which can be quickly solved using an adapted Newton's method. The result is a feasible landing trajectory with maximum final approach length, which is a reasonable target in case of critical emergency landings.
The algorithm was implemented in a mobile Android Application for pilot support, which is developed at the Institute of Flight Mechanics and Controls. In order to obtain the initial aircraft position and course, the integrated GPS measurement unit of the mobile device is utilized. Using the current Java-based code on a Samsung Galaxy Tab S 10.5 (2.3 GHz quad-core Krait 400 SoC processor) the path generation requires 6ms in total. In case of loss of power, the calculated path can be visualized to the pilot (e.g. via a tunnel in the sky) to guide him safely to the runway.

Screenshot of mobile Application (Android) Testing the Application in the Flight Simulator
After testing the Application in the simulator, a first real world test took place on 9th December 2015 at Hahnweide airfield near Stuttgart using a Diamond HK36. Within this flight the Pilot Support Application was successfully demonstrated, guiding the pilot to a virtual runway with engine idle as shown in the video below.
 
Publication:
  • Stephan, Johannes and Walter Fichter. 2016. Fast Generation of Landing Paths for Fixed-Wing Aircraft with Thrust Failure. AIAA Guidance, Navigation, and Control Conference AIAA 2016-1874. AIAA Guidance, Navigation, and Control Conference.

 

@2015 iFR - Flight Mechanics and Controls Lab

Address

Flight Mechanics and Controls Lab
Pfaffenwaldring 27
70569 Stuttgart
Germany

Contact

Phone:   +49 711 685-67061

Fax:        +49 711 685-66670

Mail:       ifr@ifr.uni-stuttgart.de