Zero Robotics Update

So, with new shipment going to the International Space Station, the deadline for the semifinals have been moved back by a week.  Instead of a final code, they decided to make an optional code due last week to see how every team is doing.  17 out of 24 alliances, including our team, participated.

The Pirate Squad (our team) came in 6th, having scored 247 points (the leader scored 266 points).  Not bad, but we still have plenty of work to do before I can feel reasonably confident that we have a shot at making it to the finals.

As always, I will keep you updated as the results for the semifinals are released!

Job with the Mayor of Hartsville

A couple days ago, my friend Mike L. (the same guy who's part of the zero robotics team with me) and I were contacted by the mayor of Hartsville, South Carolina.  He asked if we could develop an iPhone app for a funeral home. 

I have never developed an iPhone app before, but the mayor's going to sponsor us with the money to get us started, so we said, "Why not?"

This seems like a great opportunity.  I'll be doing my best to get him the app he needs.  I'll keep you updated!

Pterodactyl Love Song

Here's something interesting that I've dug up from my hard drive.

This song, humorously entitled "Pterodactyl Love Song", is a goth(?)-techno-themed song composed and put together by me on FL Studio when I was in 10th grade.  I did this over two hour and half class periods at my old high school (before moving to SCGSSM).  I won the best song of the class award and a $10 gift card for iTunes!  The song contains an analog drum set, a wide variety of synthesizers, and a few white and pink noises for effect.

Anyways, I hope you enjoy this little piece from the past!

Music Supplement

This post contains recordings of my performances for my music supplement.


Guitar:

Drums:

 

 (Really a bit of everything):

Videos

For the past two days, I've been working hard to get the videos up and running!  I was updating all my videos from its horrible quality to a version streamed directly from YouTube.  That means you should be able to see all my videos in pretty good quality!

Big Blue Guitar Cover

One day during the summer vacation before my senior year, I was really bored and also felt like rocking out, so I decided hook my amp up to my computer and start jamming on my electric guitar.

On top of that, I also felt like making a video and putting it up on YouTube.  I couldn't see an easier way of doing that other than putting together a video of me playing my guitar.  This video was the result.

The song is Big Blue, one of great songs from the video game series, F-Zero.  I really love this music and because I grew up playing F-Zero, gives me a wave of nostalgia every time I play it.  It's quite lovely.

Anyways, I hope you enjoy my performance and the music!  All instruments (guitar and drums) were recorded and programmed, respectively, by me.

Close Combat AI in Armed Assault 2

Have I mentioned that I really love working with AIs?  I had so much fun working on RCBOT 2 AI.  As much as I love them, they can also be frustrating.  The default AI in Armed Assault 2, a military simulator, had flaws that really bothered me.

For those of you who don't know what Armed Assault 2 is about:  It is a military simulation game developed by Bohemia Interactive Studios.  Their game engine has been under development for over a decade now and has gathered a large fan base over the years, thanks to their high mod-ability.  Its AI, however, has been its weak point since day one.  Although BIS has been improving the AI and is now quite intelligent, one problem still remains:  the AI is poor in close combat situations.

The following video is a demonstration of the default AI at work in close combat situations:

The reaction time is not that bad as it used to be.  The developers did a good job in making the AI smoother.  However, this is still too easy for experienced players to find near challenging.  Those players want AIs that can react EVEN FASTER.  They want to be kept on the edge instead of leaning back on their chairs and shooting down slow AIs.

My journey to create such an AI began there. 

It took me a day of brainstorming and crunching out the vector math and two days of testing prototypes.

My AI, which will be referred to as CQB AI from here on, first checks for hostile targets that are nearby.  Once a hostile target has been spotted, the CQB AI will lock onto that target unless another hostile target that has been spotted poses a greater threat than its current target (this is determined using a complicated formula that involves the distance to the target and the orientation of the target in relation to the AI).  Once a target has been acquired, the AI will continuously fire an invisible bullet toward this target.  If this bullet hits the target, then the AI has a line of sight to the target and will thus fire at the target.  Now, the AI doesn't actually fire this bullet.  This bullet, called the LOSShot, is spawned at the position of the AI and the physics engine applies a velocity of great magnitude toward the target.  Armed Assault 2 engine stores different types of bullets in the form of classes.  LOSShot is a bullet class on its own that, unlike the other bullet classes, does no damage and has no visual feedback.

The method through which the CQB AI acquires its target is the same as that of the default AI.  Enemy acquisition of the default AI is good enough to simulate the reaction of a human.  The problem of the default AI lay in the fact that the process after acquiring its target was not fast enough.  Naturally, I only had to improve that portion.  The mechanism through which I make the AI fire its weapon is not conventional.  As with the LOSShot, I do not make the AI fire the weapon.  Instead, I use what is called an event handler to simulate firing.  An event handler is a conditional code that is triggered every time a certain action occurs.  In my case, I added an event handler that handles firing.  Every time the AI fires, my event handler is called.  This event handler gets the projectile that was fired from the and redirects its velocity to the target (the engine lets me obtain the projectile and changes its velocity).  This portion required an extensive use of vector math because I had to make so many adjustments to the bullet's velocity as the game engine simulates bullet drop (which is dependent on the distance) and windage (which requires me to adjust the sway of the bullet).  Fortunately, the game engine already had many math functions such as the cross multiplication of the vectors already built in, so I did not have to create new math functions.  However, I still had the daunting and tedious tasks of finding the golden adjustment constant and calling the appropriate math functions throughout the code.

Now, why did I have to simulate the AI fire instead of making him actually fire his weapon?  There are limitations to the scripting engine of Armed Assault 2.  While I can manually adjust the AI's orientation as a whole, I cannot change the orientation of his weapon.  Therefore, while I can make the unit face its target, I cannot make his weapon face the target.  As a result, the bullets have a large chance of not hitting the target.  In order to fix this problem, I had to simulate firing and redirect the bullets' velocities toward the target.  It was an additional step I shouldn't have had to take, but it was a crucial one nonetheless for my purposes.

The most difficult part of this, however, was definitely figuring out a way to check for the AI's line of sight to his target.  Many modders were unable to figure out a way to do this, so I was left on my own.  I brainstormed for two days, and finally I decided on creating a bullet class.  Programming it was not that easy either.  The bullet traveled so fast that it was really hard to detect when it came close to its target.  But I persisted, and by the end of the day, I had a very reliable line of sight checker.  Designing and brainstorming was definitely the hardest part of this whole ordeal.

After three days of working on the CQB AI, my journey came to an end.  The result is a quite skilled and fast AI that is lethal enough to keep the players on their toes.  The following video is the CQB AI at work:

Now, that is pretty fast!  Note that in both situations, the AI was facing towards me before they spotted me.  At this rate, even the experienced players should be careful when they are going into close combat situations.  There is no more need for placing a bunch of enemies when one skilled AI will do.

MIT Zero Robotics Competition Update - We're going to the semifinals!

At first, I really didn't think we were going to be doing this well in this competition, but after the 2D competition, we came in 21st!  To make things even better, we came in 6th after the 3D competition #1!  This puts us at position 7 and Pirate Squad is safely into semifinals!

I'm really proud of what my team managed to achieve.  We managed to make something that actually worked from a group of five ordinary high school students.  But it's not over yet, and we'll be working hard to optimize our code so that we'll be fit for the semifinals and hopefully, finals.

You can see the results of the 3D Competition #1 here and the semifinalist listing here.  You can also browse and watch simulations of our robot here.  Please note that our team name is Pirate Squad. 

Thanks for reading and I promise to keep you updated on Pirate Squad and Zero Robotics!

Pirate Squad Members:
Paul P. (The author) - Code, Physics, Team Lead
Karl S. - Code, Physics
Mike L. - Code
Josh B. - Management
Max F. - Physics
Mrs. Bunn - Mentor

Genome3D

Last summer, I had the pleasure of working for Dr. Jijun Tang of the department of computer science at the University of South Carolina for my summer research. The title of my research was A Viewer Model Framework for Visualizing Multi-scale Three-dimensional Genome with Online Integration.

I worked on a program called Genome3D over a six-week period. The program, coded in C#, was designed to render, in 3D, genome models that are stored in XML format.

The following are some screenshots of the program in action:


A "nucleo" segment displayed by Genome3D.


A segment of "fiber" displayed by Genome3D.


Genome3D workspace displaying a "gloop" segment.

For more details, check out the research paper here.

Somewhere Only We Know Recording

Hello! It's been a while. I have a few things I've been writing about, but they've all been put on drafts, but now that MIT deadline is here (Yay?) it's time for me to put them all on the blog!

Today, I would like to share something that's quite different from the rest. It's not computer science, nor is it a video game of some sort. Today, I would like to share a song that I've recorded on my own, my take on an excellent song, Somewhere Only We Know, by Keane.

All instruments were recorded by me. They include bass, guitar, piano, drums, and vocals. Enjoy!