Sorry for the short notice, but upon discussion with several group members the consensus was reached that the Monday (19/11) 1PM meeting should be switched to Tuesday (20/11).
I will send an email around also but please spread the word so we are clear on the meeting time.
Thanks all.
PS. For convenience meet on MSN to discuss the amalgamation of the documentation or email it.
Here is the current version of the walking GA if anyone wants to have a look.
Pressing ENTER toggles the graph on/off.
Pressing BACKSPACE toggles drawing of the creatures (it runs substantially faster with drawing turned off).
They can also turn now. Hold down LEFT or RIGHT arrow keys to see.
Plenty more to do.
By the way, we’re meeting up tomorrow right? 1pm? Or am I remembering incorrectly?
I quickly coded up a scroll wheel zoom feature a couple of days ago. Nick mentioned that we should probably keep the code versions as consistent as possible to avoid a big mess when combining it all at the end. So to that end here is the short snippet of code that you can all add to your sketches. I’m not posting it together with the AI because it’s not ready yet and this will just keep things consistent in the meantime.
Place this at the end of your Setup() method, right after the “//Load Sounds” stuff.
//Mouse Wheel Zoom addMouseWheelListener(new java.awt.event.MouseWheelListener() { public void mouseWheelMoved(java.awt.event.MouseWheelEvent evt) { int notches = evt.getWheelRotation(); if(notches!=0){ float notchTemp = notches; zoom -= (notchTemp/100); } } }); |
Ok here is a new version of the program, just like last time you will need the Ess sound plug-in to run this properly. This version is the same as Ben’s last version with some slight updates to the sound, it is also a lot bigger nearly 4 meg now, sorry
These updates include…
– Changes to the sound play section to include a test function to see if that sound is already playing. This is to stop unneeded calling of the Ess.play and stop functions. (thanks Igor)
– The addition of a timer so that the sound update function is only called every ‘soundCooldown’ frames. This helps lower the lag caused by the sounds but it can lead to missed sound events, depending on how long the timer is.
– Fixing the sound sets for creatures so that it now plays the correct sound based on the animals actions and size instead of just calling random sounds.
– Added a full set of sounds for ‘small creatures’(creatures with a massGene of 0.2 or less). The sounds are now all part of a 2d array. The array is build in the form [Creatures size][event which causes the sound].
I have a small scale spill code that is easily modifiable.
The transparent black circle is the coffee cup and the red stuff is the spill. Click on the red stuff, hold the mousebutton, to simulate the creature drinking from the pool. Release the mouse button to let it restore.
Change “spilllines” boolean to true, it shows what the coordinates for curveVertex is based on. The coodinates are basically random distances at random angles from the center (the spot where the lines meet) The ‘coffee cup’ is set a little off the centre so that when the spill shrinks it will won’t go away from the cup. The variables that starts the spill class is spillname=new spill(centerX,centerY,RotateSpillAngle, AverageSpillSize)
Oh yeah, and it detects if the cursor is in the spill by the colour of the pixel.
Here’s the latest Code. This is Jacks + Nicks + new AI..
Doesn’t really concern anyone besides Igor. Although it is kinda cool if you wanna look. Hold down the ‘f’ key to see a new colour scheme, where the blue guys are fleeing, and the red guys are attacking. This is using a new fleeing evaluator. Atm for each animal it’s set randomly between 0 and 1, so some animals are mega courageous, others are mega cautious. Eventually though, they’ll learn/evolve appropriately. Until Igor gets the learning system going though, don’t expect much. :p
Anyway, the fleeing is determined by the ‘cautiousness’ value. Which ranges from 0 to 1.
I’ve also put in a flockSpeed value, which also ranges from 0 to 1. Hopefully, different animals will learn to flock at appropriate speeds.. determining how far away they like to explore before returning to food/water.
I’ve put in one other value, maxFoodTravel.. but this one is kinda experimental, and is too complicated to bother explaining.
Anyway, Igor.. I’ve done these 3, because they had to be calculated within the methods themselves. Everything else (when to eat? when to drink?) can be calculated outside, in the determineTarget() method. So when you go about making some sorta system of learning (maybe a genetic algorithm?) where these ‘learning variables’ are altered.. make sure to include the 3 I’ve already done.
I’ve put a whole lot of comments in there to try make my old, experimental, messy code easier to understand. You should be good, but if you’ve got any questions or wanna delve deeper into the methods, just lemme know.
Enjoy! 😀
~Ben
Dear Jack,
It all sounds very promising. The group’s project should be fine. You can use the university’s projectors, but be mindful that if you want to rig up a projector from the ceiling we must have the correct craddle due to OH&S considerations. Alternatively place it on a plinth. You will need to supply your own computers however: maybe talk to people in the faculty ICT? You don’t need to pass everything by us: remember its your work so get the group together and start determining the equipment you need. 🙂 Maybe a meeting some time next week would serve us better?
Best,
Tim.
Hey all, seems most people can do wednesday. And wednesday is the day we usually go in anyway.. AND it’s ASAP.. so let’s meet up tomorrow.
If we need to, we can always meet up again later in the week. But like Igor suggested, tomorrow would be a good time to reassign tasks, etc..
Same time? 1ish?
Ben
Guys, It is really important that you contact Tim, Nicole or Mercedes ASAP to submit something for the catalogue and arrange for the facilities and space that you’ll need for the exhibition. Although it is called the Graduate Exhibition the organiser’s have generously opened it up to graduating students as well. This will be a great opportunity for you to show off what you’ve done.
I just posted a comment on this, but to repeat. You should document the project (as it currently stands) this week and submit a document by next Monday (19th). Each person should contribute 1000-2000 words documenting their contribution as a section/chapter of the whole document, with the author clearly marked.
Get this done first! You can continue to work on the project after this week, and I hope you’ve already contacted Tim or Mercedes about exhibiting the ecosystem at the graduate show.
-RoB-