Spacecraft are tricky … and engineering is an art form

In March of this year the first of two ExoMars missions was successfully launched from Baikonur. After months of travelling through a hazardous environment we received a first image showing the destination of the mission: the Red Planet! At that point I was finally convinced that the goal of having a European mission orbiting Mars to look for life could be achieved. Last Sunday, the mother ship and its tiny companion, Schiaparelli, approached the distance from the planet at which they were scheduled to separate before going on to complete the immediate tasks assigned to them. The job of the Trace Gas Orbiter mother ship (TGO) was to find its way into orbit around Mars, while that of the Schiaparelli entry, descent and landing demonstrator module (EDM) was to descend to the planet’s surface and gather as much information as possible on the way to be used for the next European Mars mission to be launched in 2020.

ExoMars 2016 separation. Credit: ESA.

ExoMars 2016 separation. Credit: ESA.

On Wednesday, 19 October, TGO was successfully inserted into its intended orbit. This is the precondition for it to carry out its assigned task of gathering data from the Martian atmosphere to investigate, in particular, methane and its origin. The open question to which we are seeking an answer concerns the existence of life on Mars today or in the past.

While TGO was achieving orbit insertion around Mars, the test lander was making its way to the planet’s surface. It then had to perform a very complex set of procedures that included atmospheric entry, parachute deployment, heat shield separation, parachute separation, starting its thrusters and finally, a semi-soft touchdown.

During the descent, signal transmission from the lander suddenly stopped and could no longer be received. Once TGO’s orbit insertion had been confirmed, all the focus switched to the situation with the lander. Another function of TGO beyond that of scientific investigation is to serve as a communication relay station for the 2020 ExoMars mission in which a rover is planned to perform science on the ground on Mars. That same relay function is now of crucial importance as it allows us to understand the different stages of Schiaparelli’s descent with many sensors having been installed on the lander to get as much data as possible from the complex manoeuvres it would have to perform. These data were transmitted to TGO and received on Earth hours later.

ExoMars Trace Gas Orbiter (TGO), Schiaparelli EDM and Rover. Credit: ESA.

ExoMars Trace Gas Orbiter (TGO), Schiaparelli EDM and Rover. Credit: ESA.

From the data received we have already learned that contact stopped very late in the descent. This means we will obtain information from a close analysis of the data that Schiaparelli was built for, notably on the performance of elements such as the heat shield, parachute, radar, thrusters and so on. This information can subsequently be used to improve the design of the 2020 Exomars mission, since in that mission the survival of the descent module will be of real scientific relevance.

Some people have asked about the success of the 2016 mission so far. For the number-crunchers among them the following observation may be helpful:

The importance of TGO and EDM can be described as 80% vs. 20%, respectively. Since we obtained at least 80% of the data during the descent, the overall success rate can be calculated as follows: 80+20*0.8 = 96%. All in all, a very positive result.

So to sum up where we stand, we have achieved a successful orbit insertion of TGO, ready to perform science and act as a relay station for ExoMars 2020 ground science. Not only that, but we have received a large quantity of data from the lander giving us crucial information to help us perform a successful landing of the next mission.



  • Alexis Lavail says:

    Thanks for this blog entry, which I appreciate. However, my goodness, what is that success rate computation?

    Is the EDM 80% successful simply because you obtain 80& of descent data? Don’t you silence that EDM was testing Entry, Descent, and Landing. Landing failed. We won’t get DREAMS data, neither do we get Descent Camera images, neither do we get the satisfaction of a soft landing on Mars.

    Mars landing is hard, failure happens and we need to accept that, understand the problems and improve for the next mission.

    • Jesus Raymat says:

      All serious reactions to what happened and the way ESA is managing it for the public startles me. I can’t but agree 100% with all commentators in this sense

    • gruel says:

      What about transonic oscillation ? .
      MSL descent scenario planned that wrist mode oscillations be mitigated by using little thrusters of its reaction control system to counteract these oscillations.
      Contact was lost at 2 to 4 km altitude in the crtical area of transonic speed inducing resonant wave with possible parachute twist,.disturbing attitude control system and leading to a crtically reduced rocket burst.

  • Marcin says:

    Europe and Russia sent the largest spacecraft ever to the Mars, and everyone focus on a landing failure of EDM. While the core of the mission, enormously important for the search of life on Mars TGO is doing great.

    EDM was a tough sell – media complain that it failed, but media would also complain if it’d succeed because it had no cameras to take landscape pictures for the articles or battery life to provide topics for more than a few days. In either way people would be unhappy.

    The brutal truth about EDM failure is that it was a bigger loss in Public Relations than it was from scientific or financial point of view.

    • Mike says:

      Yes, a PR disaster. I don’t think it was sufficiently clear that the lander was not the maiin point of the mission. Delighted to see you refer to “a landing failure”, though: the press conference where ESA all desperately avoided theword and pretended to believe that a test of landing technology was a success because they “got a lot of data” was embarrassing. It may have provided essential data for a succesful landing in 2020, but the test was a failure.

  • PaulT says:

    We often learn more from the things that go wrong than the things that go right. Here’s to a long and successful mission for TGO and the new rover when it gets there.

  • daniel says:

    I hope that we won’t have another “80%” of landing success in 2020…

  • Don says:

    It’s not an art form. It’s systems engineering, systems engineering, and systems engineering. When something fails, it means it wasn’t built right and it wasn’t tested right.

  • Ray Johannson says:

    Europe had a 0% success rate in Mars landings.

    NASA is 7/8. Why re-invent the wheel? Next time, ask NASA to teach you, and focus on new things that add value.

    Besides, this parachute/crash-landing technique you attempted will soon be completely outmoded by Musk’s precision landing technology using retro-propulsion.

    The landing part of the mission was a complete waste of taxpayer money. Which is sadly probably why it was approved.

    • It’s a very small sample. The US have had 7 successfull landings -the two vikings, Pathfinder, Phoenix, Spirit and Opportunity, and Curiosity and one failure, the Mars Polar lander.

      Suppose that each one was actually a 50/50 chance of success. Then there would be a 3.5% probability that just through pure chance you get 7 out of 8 successes So the US long run of successes doesn’t give that conclusive a reason for supposing that they have actually cracked landing rovers on Mars.

      Perhaps they have. But they might also just be lucky. It was with good reason that they called it “seven minutes of terror” when Curiosity landed. It could have crashed, easily, and the all heaved a huge sigh of relief when it landed. There is still a significant risk that Curiosity 2020 crashes on landing.

      Beagle 2 nearly succeeded, with just one panel failing to open. What was the chance that it stuck like that? Who knows, suppose it is 50/50 then it might be that in an alternative past the UK has a success for Beagle 2.

      The ESA have done things right by testing the landing first. And with so much data they can then improve on the systems for the next landing.

    • Jorn Helbert says:

      There actually was a partnership for ExoMars with NASA – and NASA pulled out of it. That’s why ESA is partnering up with ROSCOSMOS.
      I agree it would be great if all would play together. However since they don’t it’s good to be independent.
      So – try, learn, repeat.. That’s how it always has been for space

  • Philippe Labrot says:

    Your calculation is not correct. Firstly, TGO has only managed to get into orbit, and to relay EDM telemetry. The probe will have to handle a year of aerobraking, and we didn’t receive a signal bit of scientific data yet. 80 % ? No way.

    And for EDM, this is a real, embarrassing failure. The probe didn’t reach the Martian surface ! Without the data from the scientific payload, and the images from DECA, 80 % of success is a gross exaggeration. Unless you have taken in account the fact that Schiaparelli left a nice, fresh, impact crater on Meridiania Planum for Opportunity ?

  • GH Green says:

    Dear fellow commenters,

    what more can you do than calling a simple lander module ‘entry descent and landing DEMONSTRATOR’ – doesn’t this and the way the ESA has communicated about it make perfectly clear that the whole purpose was to prepare a successful landing of the rover in 2020?? From an engineering standpoint an ‘unsuccessful’ landing, i.e. the way it happened is a successful landing, with 80% of the telemetry data gathered. With Curiosity NASA did nothing than also benefit from it previous experiences. And please stop praising Musk: The EDM had ‘real’ retro rockets (take the time and watch the video!), but yes, they didn’t fire long enough – and even this insight will help for 2020. Keep in mind so far nothing from Musk/SpaceX arrived at Mars. All he and his company do is PR.

    And please, don’t overstate the few science experiments on Schiaparelli. That was nothing more than a mere weather station.

    • Kevin D says:

      I realy can’t understand the comments.
      Yes it’s frustrating that the lander didn’t achiev histoire goal but it was a test and they told that since the begining. Like a rocket engineer told when the Space x rocket exploded “now they can learn somthing” better it happend now then in 2020 With a real science relevant payload !
      Greatings to Esa ! And Good luck

    • daniel says:

      Whell, it was a demonstrator of HOW NOT to land on Mars. A failure is a failure. Unfortunately, on Mars you can not bounce as on a comet

  • Jeronimo Vargas says:

    You assumed 80% and 20% with 80% success for EDM.
    Come on, that’s a bit ridiculous.
    The purpose of EDM was: 1) entry 2) descend 3) landing

    Entry was successfully. Descend was partially completed because the chute jettisoned too son. Landing was a miserable fail. If each of these has a 33% proportion, the EDM was only successfully around 30+30/2+0=45%.

    Together with the TGO it makes: 80+20*0.45=89% … not bad, but this assumes the 80-20 proportion you have put into the two elements. And that is respectable.

    For the public and european in general… that proportion might be somewhere around 50-50… so in their eyes the mission failed miserable… using their money !

    ESA…. you’ve got a communication problem ! Celeberate before time (i.e. philae).
    Disaster ! Disaster !

  • Matthias says:

    Hey, this is a great achievement! TGO is in orbit around mars and ready for science. And the lander, yes it crashed, but it hits exactly the landing zone so the math was right. This mission is a success in my opinion – not perfect, but next time will be better.

  • gruel says:

    What about transonic oscillation ? .
    MSL descent scenario planned that wrist mode oscillations be mitigated by using little thrusters of its reaction control system to counteract these oscillations.
    Contact was lost at 2 to 4 km altitude in the crtical area of transonic speed inducing resonant wave with possible parachute twist,.disturbing attitude control system and leading to a crtically reduced rocket burst.

  • rw says:

    The next lander will be build by ROSCOSMOS, which do not have any experience in space exploration beyond Earth Orbit. Please do not use their 40 years old space probes as heritage. The know-ledge has faded away over the past decades and ISS and Soyuz are different to planetary space engineering. The last space probe (Phobos Grunt) was a huge fail. ROSCOSMOS wants to do the most difficult step first: a Mars landing. They better should implement a program which increases mission complexity step by step (e. g. like the Chinese). The Russian-lead lander for ExoMars 2020 will be a failure again (together with the 300 Mio EUR from ESA)

    • Dear Reiner,

      actually, the ExoMars rover is developed and built by ESA and Airbus Defence and Space.

    • Dr.Madhav says:

      rw – ‘The Russian-lead lander for ExoMars 2020 will be a failure again’

      How can you predict that for sure? They are a lot smarter than you think. Failure can happen to anyone. Did you forget Space Shuttles? Please don’t be a pessimistic fool, if you do not have anything positive to write about.

      I wish ExoMars2020 a great success from the mankind perspective.

  • daniel says:

    I have read that maybe the radar altimeter failed on telliing the correct altitude to the on-board computer, which then, thinking that the module had landed, switched off the engines and hanged up comunications as it was programmed. So I suggest the use of some kind of Artificial intelligence in the 2020 mission, in order to avoid such problems

  • MBB says:

    Thank you for writing about this mission.

    I am glad to read that the communication relay function works. Having a laser connection to Mars should really reduce the travel time of information from hours to minutes, right?

    What I am wondering, is other then for specific missions like this one for finding methane, does Mars need special satellites of would copies of Earth satellites work too?
    Because ESA and the EU developed the high-quality Copernicus fleet with it’s design paid for and tested in Earth orbit.
    Can ESA not not simply mass-produce the relevant satellites from them for sister-fleets around Mars and Venus? The ground infrastructure to analyse the data is already here.
    And it might even get Arianespace enough flights to consider re-usability.
    Or are the launch and packing costs to Mars high enough to require customized satellites and independent missions?

  • Terry Hankins says:

    Hey, you could always get NASA’s info! They’ve done it successfully multiple times.

Leave a Reply

Your email address will not be published. Required fields are marked *


Comment moderation is enabled. Your comment may take some time to appear.