RoboSavvy Forum

Bluetooth and MR-C3024

2007-01-22T17:38:47+01:00

This thread seems quite old but just in case anyone is still watching....

came across this company and wondered...

http://www.industrialethernet.com/setowi.html

The firefly pair seem quite interesting as they auto-sense the 2 baud rates, 9600 and 15200 which is coincidental.

Also they claim to power off the RS323 port..

Incidentally, as has been mentioned loads before, RS232 was originally spec'd as +13v and -13v but with the advent of PC's and more importantly laptops and palmtops it has come to be +5v and 0v or even +3.3v and 0v.

A lot of people have WiFi and one of the devices on the site is an RS232 to WiFi unit. Could mean robots roaming the house and garden but still in range of the PC.

Statistics: Posted by PaulP — Mon Jan 22, 2007 5:38 pm

Bluetooth and MR-C3024

2006-08-28T20:55:35+01:00

thanks for that so the only reason to have bluetooth is purely to make use of other devises for sending commands.ie a pda or smartphone.Thats a shame because the one ache i have is plugging Rn into the pc lead every time i code (or try at least).

Statistics: Posted by rep001 — Mon Aug 28, 2006 8:55 pm

Bluetooth and MR-C3024

2006-08-28T17:32:43+01:00

The problem is not just with the rate change. My hacked code operates OK at a constant 115K. However the delays introduced over bluetooth even at 115K due to the byte by byte transfer make it unusable.

Statistics: Posted by i-Bot — Mon Aug 28, 2006 5:32 pm

Bluetooth and MR-C3024

2006-08-28T17:14:15+01:00

Right, the big problem is that the bluetooth, regardless of which unit you are using, has issues switching baud rate, and thus you cannot use it to program robo-basic.

Statistics: Posted by hivemind — Mon Aug 28, 2006 5:14 pm

Bluetooth and MR-C3024

2006-08-28T17:05:50+01:00

so am i reading this correctly,you cannot send program code via bluesmirf only run commands ??

Statistics: Posted by rep001 — Mon Aug 28, 2006 5:05 pm

Bluetooth and MR-C3024

2006-06-11T18:12:44+01:00

OK, thanks for the clarification. I suspected replacing the serial cable may not be so easy.

Statistics: Posted by mpthompson — Sun Jun 11, 2006 6:12 pm

Bluetooth and MR-C3024

2006-06-09T18:04:56+01:00

I take it back, What you want is to replace the programming cable with bluetooth. You have to connect the bluesmirf some how to the stereo jack not ERX and ETX.

Statistics: Posted by RubikCube — Fri Jun 09, 2006 6:04 pm

Bluetooth and MR-C3024

2006-06-09T17:57:22+01:00

Yes it will work, but when you want to transfer program Robobsic you still need to use the programming cable.

Statistics: Posted by RubikCube — Fri Jun 09, 2006 5:57 pm

Bluetooth and MR-C3024

2006-06-08T22:42:12+01:00

I've read this thread and I am hoping to clarify adding Bluetooth control to the MR-C3024. I basically want to control the robot remotely from a PC using the low-level serial commands to the MR-C3024. Unfortunately, I know virtually nothing about Bluetooth.

If I get the SparkFun BlueSMiRF bluetooth link and connect it to the ETX and ERX pins and get a bluetooth dongle for my PC will I be able to talk to the robot just as I can with a hardwired serial link? I realize the switching between 115K baud and 9600 baud could cause problems, but I'm not interested in downloading code. I basically just want to keep the serial port at 9600 baud and issue the commands as documented here.

I basically want to clarify that this will work as I intend before spending the money on the bluetooth hardware.

Thanks,

-Mike

Statistics: Posted by mpthompson — Thu Jun 08, 2006 10:42 pm

Bluetooth and MR-C3024

2006-06-07T10:35:08+01:00

Finally got it working, it was a bad gyro the whole time :/ LED's worked but nothing else apparently! I had an other gyro working fine last month, a Futaba model, and tonight I went back to it and it worked. Got me thinking, I got my friend's GWS and it worked great right off the bat. Lots of tweaking to do but it works just like in Pev's video #3.

I must thank all of you for your help on this, I know i've been hounding you all for some time now, but alas who could have guessed it was a bum gyro.

Thanks again guys, I'll post a video tomorrow night.

Statistics: Posted by H3dude — Wed Jun 07, 2006 10:35 am

Bluetooth and MR-C3024

2006-06-07T07:15:42+01:00

H3Dude,

I know this is a bit off topic for this thread but I wanted to make sure we all understood where you are with the gyro setup. The three movies below show without and with gyro. This is the setup I still run now to improve stability and I haven't re-zero'd the ankles I turned the gain on the PWS down to make it work as a gyro but not give me the shakes (or rather the RN1). If you still having trouble PM me and I'll see what I can do to help.

http://robosavvy.com/Builders/Pev/RN1-NoGyro.wmv
http://robosavvy.com/Builders/Pev/Rn1-Gyro.wmv
http://robosavvy.com/Builders/Pev/Pushme.wmv

The full set of code I use is published in the thread I mentioned previously except the gyro off routine, which is very similair to the one posted by Bullit.

Pev

Statistics: Posted by Pev — Wed Jun 07, 2006 7:15 am

Bluetooth and MR-C3024

2006-06-07T03:41:05+01:00

The baudrate switch from 9600 to 115k in the command interface makes it difficult to implement bluetooth for the command interface. I also have seen no way to interface control characters from the command port also.
Unfortunate. It would be nice to program snd contol the robot from one bluetooth serial port.

Statistics: Posted by Bullit — Wed Jun 07, 2006 3:41 am

Bluetooth and MR-C3024

2006-06-07T03:07:58+01:00

Here's my code - mostly from other's postings here.
I use "F" and "G" keys on the IR remote to turn the gyro on and off manually and then I have some routines that turn it on and off at different places. This allows me to experiment with how the gyro effects things like the standard_pose. My gyro is mounted on the right shoulder right now so that if responds to back and front motion.
I zero my robot with the gyro off. I then turn on the gyro and adjust the gain and center on the gyro. When you increase the gyro's gain pot the center shifts so you need to re adjust the center to get back to the zero position. Once this is all done it doesn't seem to drift much. I find more gain on carpet is better. On carpet its better to have the robot shake a little but not fall. The default punch move is very cool with the gyro on. He's locked and loaded rock solid on any surface. Have fun.

gyro_off:
GYROSET G6A, 0, 0, 0, 0, 0, 0 ' set Gyro to use (0=no Gyro)
GYROSET G6D, 0, 0, 0, 0, 0, 0 ' set Gyro to use (0=no Gyro)
GYROSET G6C, 0, 0, 0, 0, 0, 0 ' set Gyro to use (0=no Gyro)
GYROSET G6B, 0, 0, 0, 0, 0, 0 ' set Gyro to use (0=no Gyro)

GYROSENSE G6A, 0,0,0, 0, 0, 0
GYROSENSE G6D, 0,0,0, 0, 0, 0
GYROSENSE G6C, 0,0,0, 0, 0, 0
GYROSENSE G6B, 0,0,0, 0, 0, 0
RETURN

gyro_on:
GYROSET G6A, 0, 1, 1, 0, 0, 0
GYROSET G6D, 0, 1, 1, 0, 0, 0
GYROSET G6C, 0, 0, 0, 0, 0, 0
GYROSET G6B, 0, 0, 0, 0, 0, 0

GYRODIR G6A, 1, 0, 0, 0, 0, 0
GYRODIR G6D, 1, 0, 0, 0, 0, 0
GYRODIR G6C, 1, 0, 0, 0, 0, 0
GYRODIR G6B, 1, 0, 0, 0, 0, 0
GYROSENSE G6A, 0,200,200, 0, 0, 0
GYROSENSE G6D, 0,200,200, 0, 0, 0
RETURN

Statistics: Posted by Bullit — Wed Jun 07, 2006 3:07 am

Bluetooth and MR-C3024

2006-06-07T02:37:23+01:00

Bullit wrote:
Use the accelerometer as a tilt sensor. It can tell which side is up once fallen. The gyro is to stabilize. Its a rate gyro. I have my GWS gyro working. Its definately more stable with the gyro on. I have it switch on on off for different moves. It does get a bit of the shakes when you have the gain high. I can definately get mine to run on the carpet (stock 0 move repeated with small modifications). I'm still working on improving the gait. The stock gait is a little awkward I think. Mine runs better on the carpet then on a smooth surface with the gyro. Every so often it still falls flat on its face and then get convultions due to the gyro. Overall the gyro is definately an improvent. I use it to control 4 servos, ankle and knee both 200 gain. I think I'll add the accelerometer to have mine able to adjust for tilt and walk up or down a small incline also. That would be neat. I was thinkng about adding another gyro to correct for side to side swager when it walks but perhaps the accelerometer would be better for swager, some is necessary.

Can you do me a solid favor and post your gyro init code, heck, all your code! I've messed with the gyrosense & gyrodir values but I get nothing that even resembles what most folks are talking about.

I'm using this right above the MAIN block:

GYROSET G6A,0,1,1,0,0,0
GYROSET G6D,0,1,1,0,0,0
GYRODIR G6A,0,1,1,0,0,0
GYRODIR G6D,0,1,1,0,0,0
GYROSENSE G6A,0,200,200,00,0,0
GYROSENSE G6D,0,200,200,00,0,0

I have it calibrated with the red and green leds centered, the gain is all the way up, and I have reset the zero positions on the ankle servos to compensate for the downhill skiing look, but it still seems like the servos are not reacting.

I'll try and post a photo of my wiring when I get home. Perhaps I have wired it wrong.

Statistics: Posted by H3dude — Wed Jun 07, 2006 2:37 am

Bluetooth and MR-C3024

2006-06-07T02:24:59+01:00

Use the accelerometer as a tilt sensor. It can tell which side is up once fallen. The gyro is to stabilize. Its a rate gyro. I have my GWS gyro working. Its definately more stable with the gyro on. I have it switch on on off for different moves. It does get a bit of the shakes when you have the gain high. I can definately get mine to run on the carpet (stock 0 move repeated with small modifications). I'm still working on improving the gait. The stock gait is a little awkward I think. Mine runs better on the carpet then on a smooth surface with the gyro. Every so often it still falls flat on its face and then get convultions due to the gyro. Overall the gyro is definately an improvent. I use it to control 4 servos, ankle and knee both 200 gain. I think I'll add the accelerometer to have mine able to adjust for tilt and walk up or down a small incline also. That would be neat. I was thinkng about adding another gyro to correct for side to side swager when it walks but perhaps the accelerometer would be better for swager, some is necessary.

Statistics: Posted by Bullit — Wed Jun 07, 2006 2:24 am