Moving from flight simulation to vehicle dynamics/simulation

Post here information about your own engineering projects, including but not limited to building your own car or designing a virtual car through CAD.
Post Reply
Xwang
29
Joined: 02 Dec 2012, 11:12

Moving from flight simulation to vehicle dynamics/simulation

Post

Hi to all,
I'm an aerospace engineer with 8 years of expertise in the development of flight simulations. In particular I've always developed the dynamic aspects of the simulation (equation of motions, flight control systems, aerodynamic models integration and validation, inertia, landing gear dynamics and so on).
However my greatest passion is motorsport and I would like to move to a motorsport company as a vehicle dynamic or simulation expert.
I've seen that a lot of such companies require the knowledge of multibody systems and modelica, so I'm thinking that maybe I can start a my own project of developing a vehicle dynamic model in openmodelica or jmodelica (I'm a linux user) just to learn the basis of the language and in order to use the multibody aspect of such a language.
Do you have any expertise with such tools? Which is the best between the two?
Moreover I would like to learn more about vehicle dynamics and, since I have some books regarding such a matter, I would like to know which one do you suggest me as a basis for the development of a multibody parametric vehicle dynamic simulation. The books are the following:

1) FUNDAMENTALS OF VEHICLE DYNAMICS (THOMAS D. GILLESPIE)
2) RACE CAR VEHICLE DYNAMICS (WILLIAM F. MILLIKEN & DOUGLAS L. MILLIKEN)
3) VEHICLE DYNAMICS: THEORY AND APPLICATION (REZA N. JAZAR)
4) MULTIBODY SYSTEMS APPROACH TO VEHICLE DYNAMICS (MICHAEL BLUNDELL & DAMIAN HARTY)
5) TYRE AND VEHICLE DYNAMICS (H. B. PACEJKA)
6) RACE CAR AERODYNAMICS: DESIGNING FOR SPEED (JOSEPH KATZ) (this is more on aerodynamics)

If you have any other to suggest please do not hesitate.
Thank you,
Xwang
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

User avatar
Tim.Wright
330
Joined: 13 Feb 2009, 06:29

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Milliken for sure. They were the ones who introduced aircraft style stability and control type analyses into the automotive industry. Given your background, you will get the most from that book.

The key to understanding vehicle dynamics is understanding the complexities of the tyres, so I'd put the Pacejka book (5 on your list) as the second book to get. The tyres are covered in a simple way in Milliken which is sufficient to analyse the funadmentals of vehicle dynamics but the pacejka book goes much further into detail which is required for a more "high performance" analysis involving limit conditions, non-linearities etc.

For software, if you want to start out in the general automotive world (i.e. not motorsport) then Adams is the tool to learn. I don't know for sure, but I suspect its not used so often in F1 likely because the simulation time is too slow. Motorsport is not really my industry, but going on various job advertisements Modelica is a big plus, but you also need to knowledge in a language like C or Matlab to write customisations and post processing.

Matlab, (or its free open source copy - Octave) are massively helpful for post processing since pretty much no vehicle dynamics software does an adequate job for of this.

Good luck
Not the engineer at Force India

Xwang
29
Joined: 02 Dec 2012, 11:12

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Thank you for your answer.
The order of the books was not the one I was thinking of and in reality I was already thinking to first study Milliken and then Pacejca, so your opinion confirms mine.
As for the software side, my working experience is with matlab/simulink, so I already know very well such tools.
Is ADAMS available for personal non commercial use? Is there any free alternative to ADAMS?
Xwang
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

User avatar
Tim.Wright
330
Joined: 13 Feb 2009, 06:29

Re: Moving from flight simulation to vehicle dynamics/simula

Post

I don't know if MSC give out copies for personal use. They have (at least in the past) supported universties involved in FSAE but I think if you take one of their intro training courses, they might be more likely to help you out with an education license for a few months

Taking the courses would be a good idea anyway. The user interface and general useability of adams is absolutely the worst of any program I have ever used in my life. Add to that various mistakes in the documentation and you will likely waste a lot of time and get very frustrated if you try to learn it yourself. Its very powerful software, but its UI is just terribly written.

There aren't any free alternatives to adams unfortunately. Though if you want the same experience as using Adams, you could try running barefoot through a room covered in broken glass and lego pieces while punching yourself in the balls.
Not the engineer at Force India

Xwang
29
Joined: 02 Dec 2012, 11:12

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Ok for the moment I'll study the books and I'll try to learn modelica language. Meanwhile I'll contact MSC to see if there is any sort of educational licence.
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

Jersey Tom
166
Joined: 29 May 2006, 20:49
Location: Huntersville, NC

Re: Moving from flight simulation to vehicle dynamics/simula

Post

If you have 8 years experience in aerospace - particularly with a reasonably well known or reputable organization - you may be able to jump into motorsport straight away, as a simulation specialist. You'd probably have the best chances at a larger pro team or a satellite / supporting organization - a place where you can contribute without shouldering responsibility of pure vehicle dynamics knowledge.

IMO work experience is valued. Experience in time management, organizing and leading projects, developing software, validating simulations to something physical, etc - these presumably are all things you have and can leverage. Learning about fundamentals of vehicle dynamics on the side can't hurt, but I'm not sure it's going to be much of a gain. If I were recruiting I'd see that as hobbyist/amateur level experience and not really something concrete that I know is going to add value right away.

Certainly learning new software and languages is a good thing. Multibody simulation is indeed common in vehicle dynamics work - but it's not the only thing. There are more simple, parametric models and tools used as well. In any event I'd find it highly doubtful you'd be able to score an ADAMS license - but Modelica would be good to learn.

The books you list are good. RCVD and Tire & Vehicle Dynamics will be a deep dive just in and of themselves. Can be a bit academic though. You may be able to find some good SAE papers which are easier to digest and show some more practical application of vehicle simulation work. But again, all of that is just going to be a little helpful for yourself but probably not very concrete on a resume. As I said earlier, you could apply to some places now and leverage your existing simulation experience - might go a long way.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.

Greg Locock
230
Joined: 30 Jun 2012, 00:48

Re: Moving from flight simulation to vehicle dynamics/simula

Post

I scored a working demo of Working Model 3D (no longer available) before I started in vehicle dynamcis, and basically taught myself the fundamentals of MBD in that. That was very useful when it came to ADAMS. So I think learning modelica first is a good move. Unfortunately my experience with the modelica community has not been positive so I have not pursued it. You should be aware that although ADAMS is probably the most commonly used of the MBD/vehicle dynamics packages there are several others out there, so depending on where you want to work a huge knowledge of ADAMS in particular is not necessarily a good thing.

As Tim and Tom have noted the tires are the important thing (roughly 75% of the cpu time running a full adams model is tires) so a logical person might wonder if a simpler vehicle model but a more complex tire model might not be a better strategy, and indeed that is the way the industry is running, because a real time model is much better for HiL testing and of course for playstation experiments and evaluation. As such the ~15 dof model is getting a lot more attention than the 500 dof model. Equally the use of flex bodies ~5000 dof in models is also getting more attention, so there are distinct pulls in model complexity in different directions depending on what you want (that of course is the first thing you should decide, not the last).

A couple more books to go with the excellent ones above. Dixon is essential. Haney seems to drive some people ballistic, so i suppose I'd describe it as thought provoking rather than authoritative.

The Racing and High Performance Tire – Paul Haney
The Shock Absorber Handbook – John Dixon

You should also read Unfair Advantage and Olley's biography as background. Do not get the coffee table Donohue book.

Xwang
29
Joined: 02 Dec 2012, 11:12

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Thanks to all of you.
As far as the "meantime apply to a position approach", of course I'm doing it and I will continue to do it (when the opportunity arises) leveraging my existing simulation experience.
However I think that meantime it is important for me to acquire some specific knowledge by studying this books and, since I have often found useful to immediately apply what I learn, I will try to start to implement something related on my pc in my spare time so that to consolidate the matter (at first I'm thinking of implementing something that, given the main characteristics of a car, will estimate the lap time of such a car on a given track).
So multibody is not really useful at first, but since I would like to know more about it also for my actual working duties, I will give a look at it (using modelica or this tool I've found on the net https://www.mbdyn.org developed by an Italian University).
Meanwhile I'll continue to read and post on this beautiful forum :-)
Xwang

PS I have also the Dixon book so I presume the order in wich I'll study them is: Milliken, Pacejca and Adams. Is it right?
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

Greg Locock
230
Joined: 30 Jun 2012, 00:48

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Laptime simulation is a very interesting field in its own right, and requires detailed knowledge of the tire characteristics. However it is more similar to the 15 dof model (or less)than the typical MBD thingo. If Bosch's Lapsim doesn't do it pretty well I'd be surprised.

Xwang
29
Joined: 02 Dec 2012, 11:12

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Yes I was thinking to create something similar from scratch as a test of what I'm going to learn from the books.
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

User avatar
Tim.Wright
330
Joined: 13 Feb 2009, 06:29

Re: Moving from flight simulation to vehicle dynamics/simula

Post

Why not give this a crack?

Red Bull: HR13051401 – SIMULATION ENGINEERS, REAL-TIME SYSTEMS – SENIOR & JUNIOR POSITIONS AVAILABLE
The main role is to develop and maintain our vehicle simulators in order to deliver to our drivers and engineers a state of the art platform to support vehicle performance development. The role will require a strong collaboration with the various groups within the Vehicle Dynamic group: Simulation, Vehicle Modelling and Control.

The ideal candidates, degree qualified (or equivalent) in Engineering or Science, will have a strong background in robotics or industrial automation. Good knowledge of real-time systems, control and embedded software is essential. Experience with 3d vision, computer graphics, VHDL would additionally be desirable. The candidates must be fluent in Matlab/Simulink/RTW and C++ and a working knowledge of Dymola would be extremely beneficial.

Candidates are not required to have previous F1 experience as we value talented people with different backgrounds. However, a good knowledge of Atlas/System Monitor telemetry system and single ECU code would be useful.
You've got until next friday...
Not the engineer at Force India

Xwang
29
Joined: 02 Dec 2012, 11:12

Re: Moving from flight simulation to vehicle dynamics/simula

Post

I've sent my application three hours ago. Thank you.
I'm still learning English so please excuse me if my English is not good enough and feel free to correct me via PM if you want.

MadMatt
125
Joined: 08 Jan 2011, 16:04
Contact:

Re: Moving from flight simulation to vehicle dynamics/simula

Post

They forgot to add "the idea candiate should expect to be paid less than your local supermarket cashier." :)

Post Reply