studywolf / control Goto Github PK

Hi, Travis,

When I run the code on Mac as this post says, an error
ImportError: dlopen(/Users/cl/Downloads/control-master/studywolf_control/arms/three_link/py3LinkArm.so, 2): no suitable image found. Did find: /Users/cl/Downloads/control-master/studywolf_control/arms/three_link/py3LinkArm.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x00
occurs.

Then I went the directory, deleting py3LinkArm.so file and run the command python setup.py build_ext -i, but many errors and warnings occurs,
21 warnings and 11 errors generated. error: command 'clang' failed with exit status 1.

So, how can I solve this problem?

I'm studying dmp through your posts.

I've read the first dmp post and code.
Those are really helpful to me and so, thank you!

For the 2nd dmp post, I read and to see the code I download this git.
However, I am wondering if there is no dmp code you wrote, in this git you just use the library(pydmps), right?

Hi,
I am currently using your python remote API to control a kinova Jaco 2 in vrep and it seems to be moving towards the target position at first but then it looks like it is moving randomly about.

I have tried changing the parameters of the PID controller and it seems to be helping but once it reaches the target the behavior becomes random again.

Do you still have the parameters which you used for the PID controller ?

Thank you in advance and best regards,
Mattia

Mr.studywolf,
I am a master student and right now my direction is related to DMP, and the code you wrote is so nice.
But when I tried to run "python run.py arm3 dmp write ", I got this error"TypeError: __init__() takes at least 3 arguments (1 given)"
I followed the procedure as you said in the wiki. And I tried to debug this code, In this interface

control_shell, runner_pars = task(
    arm, controller_class,
    sequence=args['--sequence'], scale=args['--scale'],
    force=float(args['--force']) if args['--force'] is not None else None,
    write_to_file=bool(args['--write_to_file']))

I got the error and get inside, in here

control_shell = controller_class.Shell(controller=controller,

                                       **control_pars)

and get inside

super(Shell, self).__init__( **kwargs)

I once tried to manully add the parametes like gain, tau, trajectory, but it does not work.
I am sorry that my logical maybe seems bad, I really appreciate if you can have a look!

Hi,

Thanks for sharing the code.

I have a quick question about iLQR.
You seem to be running the iLQR optimization function each time step, but my understanding was that in iLQR, we optimise at the start and then run the trajectory, and the optimize based on that trajectory? Surely if we step by one step and then run the optimisation again, we are doing a MPC based approach? Though I may just be completely lost.

Thanks in advanced!

Hello, I'm trying to understand the meaning of this function in your iLQR code. I wonder why there is np.cos and np.sin involved.

BTW, thanks for such a great code.

I am running the code using the following command:

python run.py arm3 dmp write --sequence 333 --scale 123

But this is not scaling the trajectory. I am attaching the plot here for your reference.

File "D:\Anaconda\envs\tlyf\lib\ctypes_init_.py", line 451, in LoadLibrary
return self.dlltype(name)
File "D:\Anaconda\envs\tlyf\lib\ctypes_init.py", line 373, in init
self._handle = _dlopen(self._name, mode)
OSError: [WinError 193] %1 不是有效的 Win32 应用程序。

We have been following your pymdps and the current (control) GitHub repositories to understand how DMPs work, and we have a few questions we have not been able to answer yet:

• For multi-joint arms, what is the input trajectory for training?
• We noticed the code (in this repository) only involves trajectories for the end-effector in world coordinates. However, it is able to train the DMPs for all joints. Is this where OSC was involved? If yes, then can we instead give all joint coordinates (in angles, instead of world coordinates) as training input and train the DMPs using the same code (without major adjustments)? We are thinking of doing this so that it is easier to directly apply forcing functions (as qdotdot) to the robot joints, and perhaps not involve OSC to infer those from only the end-effector motion in world coordinates.
• We were assuming that only the end-effector trajectory is passed in for training the DMPs. However, in that case, when are joint trajectories calculated, if our interpretation of OSC, as above, is incorrect?

Any help or pointers on this will be highly appreciated!

I am getting the error on running the following code inside the two_link arm folder:

python setup.py build_ext -i

Error: numpy/arrayobject.h: No such file or directory

Whe I apply the iLQR algorithm in order to smooth a trajectory I have (I try to reduce the constrols while not diverging too much from the trajectory to follow) if the trajectory I am considering has many steps the Quu results in infinite and the algorithm breaks. How could it be solved?

On your related blog post you stated that running the Python version with a command like

python2 run.py arm2_python ilqr reach

should work. Running this command throws errors about importing the py2LinkArm.so file.

Hi Travis,

I couldn't find the setup.py file in the repository. I am on a mac and and getting the following error.
~/control/studywolf_control/arms/two_link/py2LinkArm.so: unknown file type, first eight bytes: 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x00, I guess I will have to build it. Hence asked.