27 August 2019

In March 2014, a Malaysian airplane appeared to completely disappear when flying from Kuala Lumpur to Beijing. Satellite data suggested that it crashed into the Indian Ocean, far off the planned flight path. Researchers from LiU are helping in the search for the plane.

Zürich, Switzerland - December 05, 2007: Malaysia Airlines Boeing 777-200/ER departing Zurich airport. On 08 March 2014 this aircraft crashed as flight MH 370 from Kuala Lumpur to Beijing.Zürich, Switzerland - December 05, 2007: Malaysia Airlines Boeing 777-200/ER departing Zurich airport. On 08 March 2014 this aircraft crashed as Flight MH 370 from Kuala Lumpur to Beijing. Photo credit JetlinerimagesSpeculation is rife about what happened to the plane: one theory is that the captain hijacked the plane and deliberately crashed it, looking for death. It has namely been possible to use satellite data to reconstruct the path taken after the plane disappeared from radar screens, showing that it remained in the air for many hours, flying in the wrong direction. Calculations based on these data suggest that the plane crashed into the Indian Ocean west of Australia when the fuel ran out.

“Despite a huge search operation in which unmanned underwater vehicles surveyed an area of the ocean floor as large as half of Sweden, the plane has not been found”, says Fredrik Gustafsson, professor of sensor fusion in the Division of Automatic Control at LiU.

Experts in target tracking

Together with his research group, he has now been given the task by colleagues in Australia to use their specially developed methods for target tracking to estimate the position of the crash site more accurately.

“Three research groups have been asked to look at this: us in Linköping, a group in Great Britain and one in Finland. The three groups use slightly different methods for target tracking, and we hope that it will be possible to use satellite data to limit the search area”, Fredrik Gustafsson continues.

Target tracking is a method by which computers and robots can detect and follow a moving target, and research in this field is important to enable, for example, autonomous vehicles to move through traffic, while locating pedestrians and other vehicles. It is also important when robots are to work alongside humans. All types of robot must know what is in their surroundings and how these objects are moving, and be able to predict how they will move in the future. Since the researchers know where the plane was at a certain instant, it is possible to use the same technique to predict its subsequent path.

A small but challenging project

“Normally in target tracking we have lots of data: for an airplane we have its position from radar measurements essentially every second. What is unique in this case is the we only have 16 measurements, and we must use these to recreate the final six hours of the plane’s flight. This is an enormous challenge, and needs a completely new way of thinking”, says Fredrik Gustafsson.

The commission came to him during the summer, and more colleagues in the research group will become involved as soon as the initial analyses are complete.

“This is a small project for us, outside of our main line of research, but it’s very exciting. It’s fun to spend time with it in the evenings, instead of solving crossword puzzles”, says Fredrik Gustafsson, who is happy to have received the task from the Australians, and is happy to contribute the expertise of his group.

The first report should be ready as early as 1 October.

Using satellites to monitor aircraft is relatively new. It came into use after Air France Flight 447 disappeared on a flight from Rio de Janeiro to Paris in June 2009. It wasn’t until May 2011 that the black boxes were found in the Atlantic, allowing the cause of the crash to be determined. Since then, aircraft in the air have been tracked by satellites. Initially, the positions were determined once an hour, but since the disappearance of MH370 this has been increased to every 15 minutes. It is hoped that this will prevent aircraft from disappearing in the future, and enable help to be sent to the required location rapidly.

Contact

Latest news from LiU

Server room and data on black background.

Machine Psychology – a bridge to general AI

AI that is as intelligent as humans may become possible thanks to psychological learning models, combined with certain types of AI. This is the conclusion of Robert Johansson, who in his dissertation has developed the concept of Machine Psychology.

Research for a sustainable future awarded almost SEK 20 million grant

An unexpected collaboration between materials science and behavioural science. The development of better and more useful services to tackle climate change. Two projects at LiU are to receive support from the Marianne and Marcus Wallenberg Foundation.

Innovative idea for more effective cancer treatments rewarded

Lisa Menacher has been awarded the 2024 Christer Gilén Scholarship in statistics and machine learning for her master’s thesis. She utilised machine learning in an effort to make the selection of cancer treatments more effective.