Control and Automation - Aalborg Universitet

i hope everyone have had a merry christmas, and had an even better new years eve! I've spent the past week on solving a stupid bug in my code, which turned out to be a "-1" that was outside a bracket, it should have been inside! It is quite annoying to spend time on finding such minor errors - but hey, now I will check that first the next time!

Anyway, the attached figure, is a simulation of a publication i'm working on - and as seen on the figure, the "Barrier Certificate" catches the error quite well! Even before the control system begins to go nuts about it. The error is a leak in the condenser of a refrigeration system, which entails a pressure drop. I'm quite happy about the results!

That's it for now! See you next week.
/Rasmus

It's been waaay to quiet on this channel - and with the holidays coming up, i don't think anyone will be posting anything interesting! Hence, i'm just making a quick post about my project, and it's current progress.

I'm currently working on a research article, describing the findings of the first year of my PhD, where i've used Barrier Certificates to determine wether a refrigeration container is faulty or not. It is kind of like playing at a golf-course, where we want to dictate the direction that the ball travels along the field, and hopefully hit somewhere where it slopes towards the hole, and then design a rail, so it always stays there (maybe that was over simplified, but it describes it very well).

I'm working on a publication, and results from tests - so we can present it somewhere nice!

With that, i would like to wish everyone a merry christmas, and a happy new year! See you again sometime early next year, where hopefully, a bit more material will be posted.

/Rasmus

Finally we have some decent results to share. For a long time we have been working with determining an optimal trajectory into orbit, while taking both gravitational- and aerodynamic drag into account. We have not really had any doubts about what the optimal trajectory would look like, and our solution to the optimization problem has shown that the optimal trajectory is as expected. So please enjoy the beautiful pictures Kasper Hemme have generated in matlab.

// Michael - CA10

Hi Everyone

Just a little update from my master project. I have been working on making a line of sight/way-point guidance law together with a control law for my little boat. The control law is made out of a heading controller cascaded with a velocity controller. The result after some fine tuning and after a little debugging is that it works:)
The boat is following a path which is created to test the turning capabilities of the boat.
The boat switches way-point when it comes inside the circle of acceptance connected to the specific way-point as can be seen in picture two.
As you can see it has a bit of a hard time settling when switching, this is because of the rather hard switching of the way-point, which is going to be improved later on.

Thank you for your time and have a nice day!

Best Regards
Kevin (10th Semester Stud.Msc)

Hello everybody.

I have also been asked to tell you about what I am working with in my master thesis. Me and my group mate Kasper are working with design of a control system for a self-foraging rocket. In the fall 2014 we worked on setting up models and a simulation environment for the rocket. This semester we are designing an orientation and position estimation system together with a control system. The control system is meant to make the rocket follow an optimal trajectory into an orbit of 500 km altitude.
What makes this rocket different from any other rocket, is that it is meant to use its own mechanical structure as fuel. Our hope is that this makes it possible to design a "single stage to orbit" rocket, which have never been implemented before.
You will hear more when we have results to share

// Michael - CA10

Hi everyone!
I have been asked to tell you a little bit about what I am going to do on my Master Thesis. I have chosen to make a long termination Master thesis. The main focus of the thesis is to make use of active and passive methods for failure detection of an autonomous surface vessel.
At the current time in my project I am working with the design of an auxiliary signal for failure detection and a waypoint tracking control algorithm. When I get a bit further in the project and have some results, I will share them with you! :)

Best Regards
Kevin ( 10th Semester Stud.M.sc)

finally, its working!

Simon and Anna have spent their 9th semester on developing control algorithms for the AAU3 project, and today we were able to capture it working on film!

Me (one of the project "founders", are quite happy to see that it's actually working as intended), and the project group is overly excited about it too.. The exam is just a formality.

As soon as we get everything botched together nicely, we'll post a new movie, ceptured with a better camera!

On to the 3 dimensional version! (Y)
/Rasmus, Industrial PhD Student

AAU3 - Balancing Test (Feedback Linearizing) First test of the AAU3, a moment-wheel balanced inverted pendulum. The idea stems from the CUBLi project developed at ETH in Zürich, but we latered it a litt...

den balancerende kube vi har bygget på, blev igår aftes skilt ad - for at foretage målinger, så controlleren kan dimensioneres og gøres ikke-lineær og robust! imens vi havde den liggende på borde fik vi taget et billede af den, og den ser faktisk ret flot ud som den ligger der!

målet med projektet er at den kan bruges i studieprojekter samt til forsknings og udvikling af nye kontrol metoder, da den er forholdsvis simpel, men samtidig er super god til at illustrere forskellen på lineære og ikke-lineære metoder! der skal nok komme en film når vi har noget der fungerer!

forhåbentlig virker den inden ugens udgang!
/Rasmus, (snart PhD Fellow)

Som en del af mine PhD studier besøgte jeg SMART (Singapore MIT Alliance for Research and Technology), hvor jeg arbejdede med selvkørende biler. Jeg var med til at udvikle en selvkørende platform baseret på golfbiler. De to golfbiler vi byggede bliver nu testet i en offentlig park i Singapore. Se en video af det her: http://www.channelnewsasia.com/news/singapore/driverless-rides-trial-at/1425530.html

Hvis der er nogen der er interesserede i et besøg i på SMART, vil jeg gerne formidle kontakten.

- Karl

Driverless rides trial at Chinese and Japanese Gardens From Thursday, you can book a driverless buggy free of charge for as part of a smart nation initiative at Jurong Lake District.

so, it's been quite some time since an update have been made! this is what i'm working on - simulation of wakes in large off-shore wind-turbine park. This is a simulation of how the wind speed alters as a wind-field enters the park, and the wakes begin to take effect (basically working as a GIGANTIC low-pass filter on the fluctuations in the wind speed).

Short update for now, i'll try to do them on a more regular basis!
/Rasmus, Research Assistant

so, it seems that a group on the 9th semester of Control and Automation have chosen to work with the AAU3 project, John (the supervisor) and myself (the co-supervisor) are proud that the project have been elected, and hopefully we're able to develop methods of control that are of scientific contribution, as well as learn the students about non-linear control aspects!

We're currently working on finishing the software aspects of the project, so the students can focus on developing the control and estimation algorithm! This is a picture of the interface that shows up when you connect to the inverted pendulum (still under construction).

In two weeks time, we should have everything working, fine and dandy! At least, i'm getting a brush-up on my java skills!

Thanks to Benjamin and Frederik for being awesome!

/Rasmus, Research Assistant

it's been a while, but here goes!

I've spent the past month and a half, on developing a module that is able to compute the wake in a given wind-farm, and the results are somewhat pretty to look at (at least they're colorfull).

It's still not completely done, there are some minor tweaks to be looked at, but I think the general idea is quite easy to understand. The purpose is to use this module, and then compute the effective power one can get form the turbines - which can then be used in farm-wide control, to optimize the output as well as reduce structural fatigue!

/Rasmus, Research Assistant