Re: pic: Thousands of manhours of work... coming '07
 

Well Matt, I would love to take you up on your offer, but as always, I have yet another trick up my sleeve.:cool:

Upper 200s? Try 150. With a touchscreen.

See ya in 07!;)

Re: pic: Thousands of manhours of work... coming '07
 

Team 306 would definitely be interested in beta testing this...

Re: pic: Thousands of manhours of work... coming '07
 

This is a mistaken assumption. The thread you linked discusses R71, which deals with limitations on using code from the robots of previous years.

Noticed the bolded text. This rule prevents a team from using identical blocks of code from prior years. It does not prevent a team from using code developed outside of the competition. Code like Kevin Watson's or what Mike and Matt have developed is legal according to my interpretation of the 2006 rules.

Of course, the rules on this topic may change for 2007. We'll find out in a few days.

Re: pic: Thousands of manhours of work... coming '07
 

I'll have what he's having!

Seriously, though, I'd be interested in pitching this to the team if the game looks to call for it. A more complex autonomous, like in 2003, 2004, and some in 2006, would make me strongly in favor of implementing the system. But if we're just knocking off the hanging tetra, then I just can't quite justify it for this season. Ask me after Kickoff.

Re: pic: Thousands of manhours of work... coming '07
 

Matt, Team 2021 is willing to help beta test it for you.

Re: pic: Thousands of manhours of work... coming '07
 

I know how you guys said you want 2 X 20 or 4X20 LCD screens but I don't think you guys can get what you want. Here is why, you don't have lot of character options and I know how you wanted feedback from camera but what kind of feedback? A simple text based saying "Colour Green Detected" "Distance 45 meters"? or something that shows you what the camera is viewing, in other words looking from camera's point of view.

Also, With a small 2X20 LCD screen how are you guys planning on creating a navigation system. Especially the interface? I thought you guys have a GUI that goes along with your navigation system. In such circumstance you guys will be using a laptop to map the co-ordinates and than the GUI will generate a code based on co-ordinates you selected and than feed info to micro-controller. Well why go through so much trouble?

Have an embedded computer into your control system, which you can use to run your GUI while designing autonomous. Once you are done designing autonomous you can have options to save 9 different autonomous modes into EEPROM. During the autonomous have an onboard binary swtich system which will allow you to activate one of the 9 autonomous from EEPROM. At the same time you can use your computer that is embedded to get realtime feedback from the robot during a game.

-:::-Jigar Patel [1219 IRON EAGLES]-:::-

:D ;)

Re: pic: Thousands of manhours of work... coming '07
 

Matt and I never mentioned camera data.

Would you rather be required to pay $XXX more in order to purchase a PDA to attach to the bot?

$200 price limit. Why have binary switches if we already have a fully loaded/embedded computer? If the computer is on the robot, how are we supposed to see the information in realtime?

Tom, is your system plug and play? :cool:

Re: pic: Thousands of manhours of work... coming '07
 

I have a question regarding your plug and play setup. I assume the gyro and encoders are plugged into the dev. board, and you just supply power and a serial connection to the brain, right?

Re: pic: Thousands of manhours of work... coming '07
 

Now that I think about it maybe have support for serveral diffrent gyros and encoders so it would be mor eplug and play.

@magical hands: I dont think there was ever a plan to output camera info to a character lcd. Maybe to a larger oled or soemthing but not a character lcd.

Re: pic: Thousands of manhours of work... coming '07
 

Bill,

Quality over quanity.

Sincerely,
Linus

Re: pic: Thousands of manhours of work... coming '07
 

Our board will house all the sensor connections, and power, and a serial cable goes to the program port.

One misconception I'd like to clear up is the kind of data this will make available. Besides passing on raw sensor data, it will also pass on an X/Y coordinate and heading value to the robot. This allows an out of the box user to immediately have a wealth of info it takes most teams a few weeks to get to. Also, we are looking in to writing and releasing code after kickoff that will illustrate waypoint based driving and an autonomous state machine. We are also playing with the idea of an auto-calibration system to make user set up even easier.

EDIT: I wanted to add this. Once calibrated, the system will work any gyro/encoder system you use, with varying degrees of accuracy depending on the sensors used. Higher count encoders will be more accurate, lower count will be less. Higher/lower quality gyros can have a same effect. But fundamentally the code is designed to be sensor-independent, with all the variance accounted for in the initial calibration.

Re: pic: Thousands of manhours of work... coming '07
 

Thanks for the input all,

We'd like to bring this topic back into focus. The competing systems are an interesting discussion, and will be a fun topic once the full capabilities of both are revealed. Matt and I are confident in what we have developed, and are looking for others to invest in that confidence.

So, who would be willing to beta test?

Re: pic: Thousands of manhours of work... coming '07
 

For possible system beta testers:

Take everything I've said in this thread with a grain of salt. My INS is still in early stages of the design process, and I have no intention of making it public in the 2007 season.

So, beta test away. I believe all you need is encoders on each drive side, a gyro, and the AVR board. You'll probably get a gyro and gear tooth sensors in the kit, so maybe you'll just need the AVR board.

Re: pic: Thousands of manhours of work... coming '07
 

I have to confirm when my team meets on Wednesday, but those that I have spoken to on Team 540 are interested and want to do it. We will also ask 384 and 1086 - who we are working with to share resources.