30 December 2011

NXT Wifi Sensor by Dexter Industries

So, after some midterm and final tests are out of the way, I'm back and ready to play.

Well, for Christmas, I got the Dexter Industries Wifi Sensor.

I'll will be using it to start a new project in swarm robotics, and will post every now and then with updates on my project. When everything is done and under wraps, I will post it as a new project under the Projects page. But for now, the details.

I am going to dabble in User-Based Swarms, where the user will have access to all the coordinates and sensor feeds from the robots in the swarm. The user will be able to readily control positions, rotation, and other actions of all of the robots either individually, or collectively.

Step 1: Following a tutorial. I will be using NXC for this project, as my RobotC subscription is already half done and lasts only for a year, and because NXC is free. I will create a simple web server and learn all the essential commands required for communicating with the web server through the NXT.

Step 2: Start building a Ruby-on-Rails application. This application will be the user interface, in which anyone who I have given the password can log in and mess around with the position of the robot and see how it works. Later on through this project I will write a post concerning the specifics of this user interface, and maybe include a couple of diagrams.

Step 3: Launch. Simply launch the website, and let my family be the debuggers, looking for ways to hack/break the code I have written (my family is pretty good with computers).

Step 4: Purchase more Wifi Sensors. As of right now, I am using only 1 robot because I have only 1 Wifi Sensor. Later on, I will purchase more, and hopefully connect all of them to the application I have built. Then a true swarm will be born

Step 5: Setup Bluetooth connections between the multiple NXT's. This way, while there is no user logged in, I can let the NXT's roam freely in the environment I have given them, and let them do the multitasking in order to not hit objects, or play follow the leader, or any simple random task. This will create an autonomous swarm, which I will run, again, when no one is logged in.

Well, those are the basics. I will start the tutorial this week and pick up on the basics. If this project proves to be too long (I am off to college next year), then I will just post some theory about it, and diagram what I was thinking of doing-stuff like that.

Till the tutorial is done, then.

P.S. I got Battlefield 3 and MW3 for Christmas too, in case anyone was wondering...

05 November 2011

Lateral Thinking

I read a book by Edward de Bono called Lateral Thinking and it really gets a guy thinking about the smallest of problems that can be solved by looking at things, well, laterally. The problem I want to talk about is the 3 cups and 3 knives problem, in which 3 knives have to be placed on 3 cups in such a way that another cup can be place on top of the 3 knives. The solution is easily found on YouTube, but what I wanted to address is a slight deviation. I am a big juggler, and found myself alone with 3 juggling clubs and a big exercise ball. Thus, I took two minutes, and created a solution to the problem with clubs on a sphere by getting the center of gravity perfect. Here are the goods:

And, a little video to prove it is standing by itself, with no extra aid.


The main reason I brought this up, however, is due to the FTC(First Tech Challenge) competition my team just competed in today. My team's robot had the most LEGO, which most people disagree with. But I think laterally, and tell others to picture it like a multistage rocket. A part falls off after being used, and can be recycled if necessary. In the end, we came in 4th, with some of the most unique strategies used.

To get anywhere in robotics, creative thinking has got to be used. Big companies are already on top of the ideas of today, but the next generation has to think laterally to help along the ideas of tomorrow.

25 October 2011

Robotics Roach to the Rescue

As we all well know, cockroaches can reach just about anywhere. They are small, quick, and stable, as well as able to crawl, jump, and, for some, fly. For this reason, the researchers at U.C. Berkeley's Biomimetic Milli System Lab built DASH Plus Wings, an interesting little robot that tries to imitate the stability, speed, and, in some ways, flight, of our friendly roaches. Take a look (courtesy of UC Berkeley News Center):

Very effective, I must say. Especially when looking at the evolutionary "top down" or "bottom up" science of flight.

 I have notice that generally, robotics branches into two ideas.
  1. Robots try to become more and more like natural organisms living today. Achieving the ability to make efficient decisions is the main topics discussed in artificial intelligence and behavior based robotics. Trying to make robots with the ability to fly, swim, and walk are also falls under achieving organism-based actions.
  2. Robots extend to places humans can not go (i.e mars rovers, deep undersea autonomous robot) and extend the grasp of human knowledge.
This article obviously leans to the number one theory. However, it rightly does so, addressing the rescue potential of these robots, which has nothing to do with grasping knowledge. The need for wings to conquer any terrain is correctly approached using the first theory, in which birds have an easy command over both land and air. Next, of course, comes all three: land, air, and sea.

Oh, and it throws in a little reference to swarm robotics, saying that fleets of "50 or 100" of these could go out on rescue operations.

Video and article from the U.C. Berkeley News Center. Article by Sarah Wang (again).

Boy, do these Berkeley guys get it right.

24 October 2011

Back again, after some time

Once again, almost a year since I posted. Yes, I neglected this, and yes, I'm sorry. It is just that, every time I come to this blog, I see the updated blogs on the sidebar, and go straight to them. But from here on out, I'll create two posts a week, scout's honor. Now, back to the real news.

I recently read in the "Popular Science" magazine about students at UC Berkeley who created an exoskeleton for Austin Whitney, a paraplegic, so he could walk to receive his diploma for graduation. Here is the video, courtesy of Sarah Yang, who wrote the article for the UC Berkeley News Center (http://newscenter.berkeley.edu/2011/05/14/paraplegic-student-stands-walks-at-graduation/)

The students at the UC Berkeley Human Engineering Laboratory did a fine job. The article quotes Whitney, saying

 "The second I pressed the button and stood up, I was flooded with a series of emotions."

However, I can't imagine the engineers who worked day in and day out to build the exoskeleton having any other kind of feeling as well.

Getting a little more technical, I am very interested in the mechanics behind this simple exoskeleton. For my high school, we are required to, as seniors, pick a social issue to try and fix. I have picked a topic in the field of prostheses and exoskeletons for veterans, and plan to contact these very engineers to see if I can do the hours of experience required with them in their lab, giving me some valuable insight into mechanics behind the exoskeleton.

This accomplishment is an excellent show of the potential at Berkeley, and is one of the main reasons I would like to go there. Mind you, any college that offers robotics and has its stuff covered in news makes me want to go there, but the point is that this piece of work covers not only the robotics aspect of colleges, but the human perspective many of the projects try to achieve as well. Kudos to those students; they did a great job.

Again, thanks to "Popular Science" and UC Berkeley News Center for such awesome articles and the video.