3D Robotics


Remember a year ago when Randy used an ArduCopter to autonomously drop a rover on a freakin' volcano? Good times. 

Well, now the European Space Agency has done the same thing, but this time time they have ridiculous music, a massive budget and a bunch of German university researchers. They also say it's for a Mars rover, which will be quite a trick for the quadcopter, since the Martian atmosphere barely exists (it's 0.6% the density of Earth's atmosphere, which means copters wouldn't work). Let's assume they know this, and if they actually do drop a rover on Mars, the dropship will have rockets, not props. 

You can read the whole over-the-top press release below. But really, the music in that video has to be heard for full comic effect:

Watch the video HERE (it won't embed for some reason)

The dramatic conclusion to ESA’s latest StarTiger project: a ‘dropship’ quadcopter steers itself to lower a rover gently onto a safe patch of the rocky martian surface.

StarTiger’s Dropter project was tasked with developing and demonstrating a European precision-landing capability for Mars and other targets.

The Skycrane that lowered NASA’s Curiosity rover onto Mars showed the potential of this approach, precisely delivering rovers to their science targets while avoiding rock fields, slopes and other hazards.

“StarTiger is a fresh approach to space engineering,” explains Peter de Maagt, overseeing the project. “Take a highly qualified, well-motivated team, gather them at a single well-equipped site, then give them a fixed time to solve a challenging technical problem.”

This latest team was hosted at Airbus Defence & Space’s facility in Bremen, Germany, joined by engineers from the German Research Center for Artificial Intelligence, Portgual’s Spin.Works aeronautics company, and Poland’s Poznań University of Technology Institute of Control and Information Engineering.

Starting from scratch for the eight-month project, the Dropter team was challenged to produce vision-based navigation and hazard detection and avoidance for the dropship.

It has to identify a safe landing site and height before winching down its passenger rover on a set of cables.

Flying to a maximum height of 17 m, the dropship comes gently down to 10 m above the ground, where it begins lowering the rover on a 5 m-long bridle, coming lower until the rover touches down. Then it returns to a safe altitude.

Flight testing took place at Airbus’s Trauen site in northern Germany, which back in the 1940s was the scene of spaceplane pioneer Eugen Sänger’s rocket experiments.

A 40 m by 40 m Mars-scape was created, littered with hazardous rocks, where the dropship had to pick a safe spot to deliver its passenger.

Dropter team

The dropship was customised for the project from commercial quadcopter components, with a smaller drone used for preparatory indoor testing.

Using GPS and inertial systems to fly into position, it then switched to vision-based navigation supplemented by a laser range-finder and barometer to land its rover autonomously.

This demonstration having proved the concept, the dropship approach is now available for follow-on development by planetary missions to come.

About StarTiger

StarTiger, a pioneering ESA initiative in technology R&D

StarTiger stands for ‘Space Technology Advancements by Resourceful, Targeted and Innovative Groups of Experts and Researchers’ working within the Agency’s TRP Basic Technology Research Programme.

It brings team members together on a single site to work on a set challenge, aiming to produce a working prototype by the end of the project’s time limit.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • There has always been a dream of propellers on Mars, but until recently, the common sense to stick to rockets.  Now that a wink & a smile gets you a $50 million line of credit, any idea is a brilliant idea.

  • Developer

    Just one more side note, I think the vision based navigation is really starting to take-off (no pun intended).  The balloon finder (below) was one example of using vision based navigation using an odroid "companion computer" and I'm sure we're going to see much more over the next year including precision landing on a coloured landing mat.  Exciting times!

  • Developer

    Not to be outdone on the music front, below is another version of the rover drop I did with EnRoute + Tohoku University.

  • It probably is possible to make really light copters work on Mars, but the thinness of the atmosphere will require serious light weight construction and really big rotors.

    And it probably is not suitable for delivering heavy cargo like a rover, but the point wasn't the copter, the point was the vision system for locating a safe landing spot for the rover and that they did very well at.

    Possibly take 56, but probably not.

    I agree with Chris on the music.

    We have a lot of FPV videos with similar over the top music, but in some of them the flying and the scenery actually justify it.

    Possibly Also Sprache Zarathustra is getting over used a bit. (2001)

  • 3D Robotics

    Guys, I don't mean to be critical of the project, which looks really cool. I just couldn't help laughing at the music in the video, which is incredibly bombastic for some guys looking at checklists. 

  • Actually, copters can work on Mars, although a conventional rover payload would be tough. See section 3.1 here for a proposed effort:

  • Hey guys, whats the negative attitude about here???

    There is a group of researchers showing an experiment using a multicopter. You should use it as a positive example what the technology can be used for... (Even if it´s not Ardupilot :-) And it seemed to have worked quite well. Ask them about more information, if you are interested...

    And there is a little too much guessing in your writing. Do you know their budget? I don´t. Of course they won´t use a multicopter on mars, it´s a demonstration of methods. It´s written in the text.

    And what about the music? Come on, not everybody has a hollywood composer on hand :-)

    Just my thoughts...

  • They didn't really say much about the vision based navigation. I am not sure how it's really even helpful. All they showed is a bunch of people checking things off on a clipboard and a robot clumsily dropping on the ground.

  • 3D Robotics

    Sure it can. Already integrated with Project Tango for vision-based SLAM.

This reply was deleted.