[solved] using joystick axis and using it's input to alter the sketchpad

Comments

• Edwin

  February 2013 edited February 2013

  Hi,

  1 Try this script (add it to the run phase of an inline_script, placed after a joystick item) and mess about with the joystick:

  from openexp.canvas import canvas
  from libjoystick import libjoystick
  
  # create canvas object
  canvas = canvas(exp)
  canvas.clear()
  
  # create joystick object
  joystick = libjoystick(exp,joybuttonlist=None,timeout=None)
  
  # create basic canvas
  joyinputs = joystick.input_options() # returns: [buttons, axes, balls, hats]
  joyinputtypes = ["buttons", "axes", "balls", "hats"]
  for i in range(0,len(joyinputtypes)):
      disptext = "You have <b>" + str(joyinputs[i]) + "</b> " + joyinputtypes[i]
      canvas.text(disptext, center=True, y=100+i*50,html=True)
  canvas.show()
  
  # joystick interaction
  stop = False
  timeout = 20000 # ms
  t0 = self.time()
  while not stop:
      # get input
      eventtype, value, time = joystick.get_joyinput(timeout=10)
      # present input
      if eventtype != None:
          # reset canvas
          canvas.rect(0, self.get("height")/2, self.get("width"), self.get("height")/2, fill=True, color=self.get("background"))
          # update canvas
          canvas.text(eventtype, center=True, y=self.get("height")/2+50)
          canvas.text(value, center=True, y=self.get("height")/2+100)
          canvas.show()
      # stop
      if self.time() - t0 > timeout:
          stop=True
  

  It should give you the number of buttons, axes, hats and balls you have on your joystick, as well as giving you the opportunity to mess about with the joystick for 20 seconds, while showing which buttons or axes you are moving (i.e.: when you move the stick, it could read something like "joyaxes motion [0.88989 0.122332 -1]", where the changing number in between the brackets is the axis you are currently moving).

  2 You should use an inline_script as above. Accessing the axes is easy if you use:
  eventtype, value, time = joystick.get_joyinput()

  This will provide you with an eventtype (a string, which tells you what kind of input has been given (e.g. 'joybuttonpress' or 'joyaxismotion' etc.), a value for that eventtype (e.g. 4 or [0.5 0.3 -1]) and a timestamp for the input. A little sample script (for an inline_script item), for increasing a text size:

  from openexp.canvas import canvas
  from libjoystick import libjoystick
  
  # create canvas object
  canvas = canvas(exp)
  canvas.clear()
  
  # create joystick object
  joystick = libjoystick(exp,joybuttonlist=None,timeout=None)
  axisnr = 1 # (position of the relevant axis, start counting at 0: [0 1 2])
  
  # text
  textsize = 20 # starting size
  sizegain = 10 # factor to multiply axis state with (growth speed)
  
  # joystick interaction
  stop = False
  prevvalue = 0
  while not stop:
      # get input
      eventtype, value, time = joystick.get_joyinput(timeout=10)
      # use input
      if eventtype == 'joybuttonpress':
          stop = True
      elif eventtype == 'joyaxismotion':
          textsize += value[axisnr]*sizegain # add axis value to textsize (i.e.: add a number between -1 and 1)
          prevvalue = value[:]
      elif prevvalue and eventtype == None:
          textsize += prevvalue[axisnr]*sizegain # add axis value to textsize (i.e.: add a number between -1 and 1)
      canvas.set_font(size=int(textsize))
      # prevent textsize from going under 0
      if textsize < 0:
          textsize = 0
      # present text
      canvas.text("Textsize = " + str(int(textsize)), center=True)
      canvas.show()
      # reset canvas
      canvas.clear()
  
  # save textsize variable
  exp.set("textsize", textsize)
  

  For a sample experiment (that is, this inline_script at work in an opensesame-file), see here.

  Good luck!
• koenraad

  February 2013 edited February 2013

  Thank you for the clear instructions, I've started from the .opensesame file you linked and am happily coding away. sorted my error aswell. Using t_text = exp.get("t_text") to import the not automatically available text string that is in the master loop. So now my participants (me atm) can see the stimulus text instead of something fixed
• Marie

  January 2014 edited January 2014

  Hi, I'm also extremely new to opensesame, and I'm trying to make an Approach Avoidance task.

  I have questions :

  1. How I can import some words and/or pictures ?
  2. I would like that the position when words or pictures appears is 0 and if people pull the Joystick the score will be positive and if people push the joystick, the score will be negative. How I can have it?
  3. is it possible to have a picture as background ?

  I'm sorry maybe it's easy questions.

  All help is appreciated!
  Thank you !
• lvanderlinden

  January 2014 edited 11:37PM

  Hi Marie,

  Welcome to the forum! Building an Approach Avoidance task is certainly possible with OpenSesame.

  Did you already go through the step-by-step tutorial? I think by doing that, you'll already find an answer to the 1st and 3rd question.

  Did you already try to build something yourself? And if so, where did it go wrong?

  If you have any more specific questions, perhaps you could provide us with your experimental script, and we could have a closer look at it.

  To upload your experiment you could do the following:

  • Open the "General properties" tab, click "Script editor" and copy the whole script,
  • go to http://pastbin.com, upload your script there (simply paste and submit)
  • and provide us with the resulting link in your forum post.

  Cheers,

  Lotje

  Did you like my answer? Feel free to
• Marie

  January 2014 edited 11:37PM

  Hi Ivanderlinden,

  Thank you very much for your help.
  I read the tutorial. for the question 1, I see that I can use the function .get but I don't know where I must to put it in the script.
  I base on the script of Edwin (see his answer just before my question).
  I try to adapt this script.
  http://pastebin.com/pABPVRA1

  More precisely, in the "'define inline_script inline_script", I try to put a function .get but when I run the words list don't move.

  For the question 2, I would like that the initial position of words is 0. How is possible to have the coordinates of each words?

  Thank you !
• sebastiaan

  January 2014 edited 11:37PM

  Hi Marie,

  The tricky part in your case is to collect the joystick movement. Other than that, you probably won't need any inline coding. So let's get to the joystick part, and from there you'll probably be able to figure things out with help from the tutorial.

  Basically, you want to wait until the participant pushes the joystick up or down, and then set an experimental variable (move_joystick) to 'up' or 'down' depending on the participant's response. Once this variable has been set, you can simply use it in the OpenSesame GUI, for example in a Run-if statement that determines which sketchpad comes next. Again, that's all explained in the tutorials, the hard part is collecting the joystick response.

  So ... you can collect joystick movement with the script below. You should put this code in the run phase of an inline_script that follows the presentation of the target word. In addition, you should insert a joystick plug-in item somewhere in your experiment, because that is needed to initialize the joystick. Even if you don't use the plug-in, it still has to be there, and you can disable it by setting its run-if statement to 'never'. Does that make sense?

  # Loop until a response is given
  while True:
      # Get a joystick movement
      movement, timestamp = exp.joystick.get_joyaxes()
      # `movement` is a tuple indicating the horizontal and vertical movement.
      dx, dy = movement
      # If there is vertical movement, set the variable `move_joystick` to up or
      # down, and break the loop.
      if dy < 0:
          exp.set('move_joystick', 'up')
          break
      elif dy > 0:
          exp.set('move_joystick', 'down')
          break
  

  Cheers!
  Sebastiaan

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• Marie

  February 2014 edited 11:37PM

  Hi Sebastiaan,

  Thank you very much for your answer.

  But I need to see the size of word or picture increased when participants push and decreased when they pull.

  When I try this code, I had a error message. ("Python: too many values to unpack")

  I don't know where is the problem.

  Thank you for your help !
• Edwin

  February 2014 edited February 2014

  The error most likely comes from this line:

  dx, dy = movement
  

  As a lot of joysticks will have an output of three axes, the following line should fix this:

  dx, dy, dz = movement
  

  Alternatively, you could simply use the second axis without bothering to unpack the rest:

  # Loop until a response is given
  while True:
      # Get a joystick movement
      movement, timestamp = exp.joystick.get_joyaxes()
      # `movement` is a tuple indicating the movement over all axes,
      # but we only need the second (index 1, as Python starts counting at 0)
      dy = movement[1]
      # If there is vertical movement, set the variable `move_joystick` to up or
      # down, and break the loop.
      if dy < 0:
          exp.set('move_joystick', 'up')
          break
      elif dy > 0:
          exp.set('move_joystick', 'down')
          break
  

  Regarding your original question: my sample script already increases the text on pushing, and decreases on pulling, right? If you want to apply this to a picture, you can use the canvas' image function, which allows you to use the keyword scale.

  In my sample script the variable name textsize is used for the textsize, but you could easily do a similar thing with a variable you name picsize. Keep in mind that this should have a lower value that the textsize, as the scale is a proportion of the original image! So you'll likely want to use values between 0.5 (half size) and 2 (twice the size).

  Alternatively, as Sebastiaan suggests, you could use the variable move_joystick to change something via OpenSesame's graphical interface. E.g. using the show if statement of a sketchpad in which you have drawn a small image: show if [move_joystick] = 'down'.

  Does that help you at all?
• Marie

  February 2014 edited 11:37PM

  Hi Edwin,

  Thank you for your help. I have no error message. It's a interesting alternative. I add it in my script.

  In your sample script, can you say me how and where I can import automatically a text variable that is in the precedent loop ?

  I try : t_text = self.get("t_text") but It's doesn't work.
  I'm too bad :-(

  Thank you.
• Edwin

  February 2014 edited 11:37PM

  Hi Marie,

  That's good news!

  You are in fact using the right code, so I wonder if you are using the correct variable name and if it is defined in the preceding loop.

  Example:

  1. Define a variable called disptext in a loop.

  2. In the sequence attached to that loop, place the sample script above.

  3. Adjust the following line from the example:

  original:

  canvas.text("Textsize = " + str(int(textsize)), center=True)
  

  replace with:

  canvas.text(self.get("disptext"), center=True)
  

  Does that do the trick?
• Marie

  February 2014 edited 11:37PM

  Hi Edwin,

  Thank you so much. I had just a mistake in my script. Now it's ok.

  A last question, I want to specify the joystick response (button 1) and that this response is the end of my trial.

  eventtype, value, time = joystick.get_joyinput(joybuttonlist=[1], timeout=10)

  But I have this message :
  TypeError: argument of type 'int' is not iterable

  I forget someting ?
• Edwin

  February 2014 edited February 2014

  Hi Marie,

  The error suggests that the joybuttonlist is not passed as a list (i.e. in square brackets: [1]), but rather as an integer (i.e. as a single number: 1).

  I've checked the source code of the plug-in, and I don't see any reason for why this happens if you, as you indicate, pass the joybuttonlist as a list: joybuttonlist=[1]. I can't test this in practice either, since I do not have a joystick around.

  What you could try, is an alternative approach, in which you handle the recognition of the correct button yourself. See the code snippet below for an example of what I mean.

  # get the joystick input
  eventtype, value, time = joystick.get_joyinput(joybuttonlist, timeout=10)
  
  # check the eventtype for buttonpresses
  if eventtype == 'joybuttonpress':
      # check if the button is button 1
      if value == 1:
          # we have now established the button was 1,
          # so now you can do what you want, e.g.:
          exp.set("response", value)
  

  Alternatively, the more elegant solution would be to check for both conditions at once:

  # get the joystick input
  eventtype, value, time = joystick.get_joyinput(joybuttonlist, timeout=10)
  
  # check the eventtype for presses of button 1
  if eventtype == 'joybuttonpress' and value == 1:
          exp.set("response", value)
  

  Does this resolve your issue?
• Marie

  February 2014 edited 11:37PM

  Hi Edwin,

  I'll try and I'll let you know.

  Thank you very much,
• Gregor

  May 2014 edited May 2014

  Hi, this was very helpful:

  

  but i have problems to get my 'move_joystick' logged. logger is not working anymore.
  debug windows says:

  

  as u see i have figured out how to log time, but i want to log also variables like move_joystick, cr, fname... dont know how.

  pls tell asap, thanks a lot

  Best greetings
  Gregor
• Edwin

  May 2014 edited 11:37PM

  Have you defined a variable name using only numerical values? E.g. "10".

  Also, why are you using self.log in the inline_script to log a timestamp? Wouldn't it make more sense to set the timestamp as on OpenSesame variable, so that it will get picked up by the logger straight away? Like so:

  exp.set("t0", self.time())
  
  # rest of code
  
  exp.set("t1", self.time())
  exp.set("response_time", exp.get("t1") - exp.get("t0"))