I would like to design an autopilot controller for my openrov (using a state-space model), but I can’t find the coefficients needed to get started (friction matrix/drag coef, added mass, coriolis forces…).
Are they somewhere on the forum and if yes, do they come from simulations or tank-testing?
Thanks a lot for your help!
A very interesting question indeed. This is something that me and my small team of students are working on at the moment, hope to have it done by mid-year… just so we’re on the same page here I assume you’re looking for the 6x6 matricies of the named forces? Also, have you considered the thruster model you want to employ?
We’ve already done some analysis to work out the center of gravity & buoyancy and their respective magnitudes, and we’re continuing work on the hydrodynamic characterisation… We’re employing CFD along with an experiment in a single degree of freedom for validation. Will certainly let you know the results.
Would you care to elaborate more on your project? ie. what is the motivation, what do you hope to achieve, how far have you gotten so far, what is your background etc. Sounds like a most interesting endeavour and we’d be pleased to assist in any way
Hi! Thanks for your interest!!
Yes this is exactly what I am looking for. I am not an expert on hydrodynamics though…
Well my goal is to create a docking system for ROV (doing a PhD on underwater connections), and I would like to start with this small openrov. So the project is to build an autopilot on one side and an optical positionning tool using green lasers on the other side (additionnally to an acoustic system for longer ranges but this one is too heavy for the openrov).
About the thruster model I was also planning to build one if nobody has done it before, but haven’t looked at it yet. It would be nice to cooperate on this!!! When do you plan to have results?
What is the software you used for the CFD hydrodynamic characterization? Maybe I should rather focus on the thruster model while you work on the matrices.
Sorry for the delayed response, I haven’t checked the forums in a while. That sounds like a very interesting project indeed. So you need the model for your autopilot controller? I am also doing a PhD at the moment. It is regarding the effect of underwater currents on underwater Simultaneous Localisation and Mapping (SLAM). Part of my program will entail using a 3D dynamic simulation of the OpenROV in order to study the effect of currents. By the sounds of it, this would be equally valuable to you. Perhaps you could develop your autopilot on the simulation before implementing it in real life. In addition it seems like there are other areas of cross relevance, for example, you will need image processing in order to carry out the optical positioning, and I for my SLAM, so we can share knowledge regarding this aspect too. So it seems like a lot of overlap and I’d definitely be interested in collaborating in any way.
For the CFD characterisation I’m using StarCCM+, I’ve already created the simulation and am now trying to get to grips with the method to actually obtain the coefficients.
– Regarding the thruster model I can recommend a relatively simple and straight forward one, if you refer to the paper ‘The Influence of Thruster Dynamics on Underwater Vehicle Behavior and Their Incorporation Into Control System Design’ by Yoerger et al. 1990.
So if you know the torque of the motor, through the first differential equation you can model the thruster speed. Then with the thruster speed you can work out the thrust. (Relevant constants can be found through experimentation).
As for the torque given a control voltage, I refer to ‘Modeling and Identification of Open-Frame Variable
Configuration Unmanned Underwater Vehicles’ by Caccia et al. 2000
So they are assuming torque is proportional to the square of the voltage.
Combining these two you can create a model for the thrusters, using a simple experiment to determine the appropriate quantities.
Do keep us updated with your interesting project!
@flore.remouit, Do you get the friction matrix/drag coef, added mass, coriolis forces…of Openrov 2.8 right now?would you mind give me the hydrodynamic coefficient of Openrov 2.8 to me.Thank you very much.
I moved from OpenROV to BlueROV without doing those tests. So no
I have a model of my forward thruster though, even if it’s not perfect.
Would you like to have it?
As per your explanation it looks you have a very clear research plan. I like your explanations.
If you need any help/assitance, just count on me.
Congratulations and regards.
I am newly added to the forum.
I am doing PhD where I am designing a combine control system for the swarm of AUVs. An Open ROV 2.8 has been purchased and will be with me soon.
I am really happy that you have already worked out the thruster model. Could you please share the data also, could you please tell the methodology.
Is there anyone done the thruster model of OpenROV2.8?