Well, ISEF is finally here and it is in Los Angeles this year! I'm going to keep a blog here of all major events in chronological order. Waking up at 6:45 was a drag, but I managed. Now jumping onto the flight, it was extremely smooth! I had never been on a plane that large until this time. It had a 2-3-2 seating arrangement and also had screens mounted on the back of every seat. Among many other features, one was live flight data that gave our altitude, time and distance from destination, a topographical map, and some other useful information I enjoyed during the flight. I watched A Beautiful Mind on the way and I must say that it is a fantastic movie that I would recommend to anyone. It's somewhat similar to Good Will Hunting (geniuses struggling with psychiatric problems) so if you liked that movie, you'll like this.
When we arrived, there was a coach bus waiting for us, which was nice. On the way to the airport, we passed at least three defense contractors within a mile vicinity of each other, which was interesting. I also saw Tesla Motor's innovative and automated electric car factory.
We're staying at the JW Marriot, which is very luxurious and hospitable and it is right next to the Los Angeles Convention Center where ISEF is being held and also the Staples Center. The pool and Jacuzzi is on the roof, with an excellent view of downtown Los Angeles. We went in at night and the lighting was spectacular.
The area is incredible. There was a LA Kings hockey game tonight at the Staples Center so the city square was packed and there were many activities going on. It was alive. You could feel the energy.
We got registered today and got our badges, bags, water bottles, etc.
As for food, we ate at Smashburger for lunch and went to an Italian spot for dinner where I got a caesar salad.
Tomorrow, we are going to Disneyland, which should be an interesting experience!
Wow, I cannot believe its been a month since my last blog post. I have a lot to tell you about.
Lets start off with a software update. I am currently in the processing of coding a Android Application that will be used to control the robot. I have decided to go with an Android application rather than a full control box because after hours of consideration, I realized that my phone can do everything that the controller box can plus more. It has a speaker, microphone, accelerometer, LCD Screen, 4G network module and antenna, Lithium ion battery that can last up to 9 hours, and its about 100 times smaller than my controller box. So the choice is obvious.
So like I said I am in the process of making the application. I just finished the graphical layout for the buttons. Now I have to program what the buttons do and configure UDP network protocols. I chose to transmit through UDP sockets to ensure reliable communication that won’t get bogged down by lost packets. If I were to use TCP/IP, it would be forced to wait for those dropped packages before it could continue processing newer data. This is doubly bad because old data would be re-transmitted (that's probably for a frame that was already displayed and therefore worthless) and that new data can't arrive until after old data was re-transmitted. The goal is to have as up-to-date information as possible so UDP is the best option.
The Android IDE makes graphical layouts very easy so that didn't take too long. Its mainly drag and drop. Below is an image of what it will look like. It is not set in stone at this point. Changes are ongoing.
I am also working on testing each individual electrical system via the Beaglebone's GPIO and deriving the equations needed to convert sensor voltage output to useful data. This was actually the first time I had booted up the Beaglebone so I had to go through the process of uploading the Angstrom operating system to the EMMC internal memory on the BBB. I plugged in my Ethernet cord, set up an SSH connection through Putty, then installed the Adafruit's python library and was ready to
Once I started testing, there was a problem. When I sent a HIGH signal to the MOSFET, the fan jittered, but didn't even have enough power to make one full revolution. It was evident that there was either low voltage being supplied or not enough current. Well I checked it with my multimeter and noticed that the MOSFET was only supplying 2v to the fan rather than 12v! I was confused how this was happening so I performed multiple experiments with different resistor configurations and loads to see if that affected the transistor switching. No results. It turns out that the Vgs (gate threshold voltage) for that particular transistor (IRFZ14PBF) was 4v and the BBB GPIO can only output 3.3v! So I ordered some new MOSFETs that have a lower Vgs around 2v. That should fix it.
1) Fabricate extraction forks
2) Solder replacement transistors and get all sensors and other devices functioning individually
3) Get the VES System functioning
4) Weld support on doglegs to prevent bending
5) Program Android application
6) Replace 3/4" wide track pads with 1.5" wide track pads
7) Purchase new shock absorbers
Well, this week has been crazy! So much has happened. I'm not sure where to start...
The judging for the Twin Cities Science and Engineering Fair began Friday and at the beginning of the week, I had yet to test the robot on the ground. Not to mention, it was -10 below all week so it wasn't exactly ideal testing conditions. But I was able to test Wednesday on the ground and the tests were a success! It could pull me (160 lbs) up a 30 degree incline covered in snow and ice as you can see in the video below. I was very pleased with the test results, but there were a few issues that need to be resolved.
- The first is that the shock absorbers were too stiff. So I disassembled them to see if I could reduce the force needed to compress the shock absorber and I found that the shock absorbers were sent to me faulty. Usually there is a gas- typically nitrogen- found in the cylinder that is responsible for dampening, but the only thing that was doing the dampening was the coil itself and thats just bad. So I need to find some gas springs to replace these.
- The wire connected to the relay coils is too fragile and often tears when performing maintenance. 24 gauge wire with thicker strands (larger diameter) and a more heavy duty insulation must be used
- At high speeds, there is too much slack in the tracks so they must be tightened to prevent undulations
Check out this test video below
Well, now its time to get back down to business with the VES (Victim Extraction System) and getting the suspension system properly functioning. I just ordered the actuators from Progressive Automations for over $600! Its a large investment, but I think it will pay off.
As for the suspension, I am going to replace the current shock absorbers with gas springs for dampening in combination with torsion springs for rebound/ return. Now the difficult part is finding a product with the correct specs.