by i-Bot » Fri Mar 11, 2011 12:12 pm
by i-Bot
Fri Mar 11, 2011 12:12 pm
The schematics are also not the complete set. Enough to enable development of firmware though.
It will be interesting to see if this is the extent of the "open" part from Robotis. So far I found not license or copyright information in the Robotis doc or code files. Does anyone know about this ?
Pleased to see on the schematics they changed from analog to SPI Gyro and accel, I find these much better, and low cost too.
The dynamixel bus seems to support both ttl and 485 servos "almost". The dx bus messages are broadcast to both drivers, and the receive paths are or'ed together. It may be possible to mix ttl and RS485 servos, but they appear to have common power supplies and have an single ttl/485 selection for all servos ( easy mods though).
Good to see the power control to the DX bus. This is pretty essential on linux based robotis bots. Often need to power of servos to reset without restaring the main processor.
Sort of odd having the FTDI chip, when there is USB on the STM32. Advantage is that is makes writing firmware more simple. Disadvantage is some loss of performance in Main/sub controller comms.
Overall very nice and much enjoyed.
Even nicer is the main controller C++ source code. It is not made clear from the documentation of the files visible at the top level on sourceforge pages that the svn is updated regularily with the code corresponding to the Robotis documentation:
http://darwinop.svn.sourceforge.net/viewvc/darwinop/
This code is very much more complete with working software and utilities. Even a great interface from Roboplus.
I have successfully run up most of the code using a micro to perform the CM730 functions, though most of the code will run on a USB2Dynamixel with a few minor code changes to ignore the CM730 specific functions(DXL power, gyro, accel, leds, buttons).
So great work from Robotis and "Open" as promised.
The schematics are also not the complete set. Enough to enable development of firmware though.
It will be interesting to see if this is the extent of the "open" part from Robotis. So far I found not license or copyright information in the Robotis doc or code files. Does anyone know about this ?
Pleased to see on the schematics they changed from analog to SPI Gyro and accel, I find these much better, and low cost too.
The dynamixel bus seems to support both ttl and 485 servos "almost". The dx bus messages are broadcast to both drivers, and the receive paths are or'ed together. It may be possible to mix ttl and RS485 servos, but they appear to have common power supplies and have an single ttl/485 selection for all servos ( easy mods though).
Good to see the power control to the DX bus. This is pretty essential on linux based robotis bots. Often need to power of servos to reset without restaring the main processor.
Sort of odd having the FTDI chip, when there is USB on the STM32. Advantage is that is makes writing firmware more simple. Disadvantage is some loss of performance in Main/sub controller comms.
Overall very nice and much enjoyed.
Even nicer is the main controller C++ source code. It is not made clear from the documentation of the files visible at the top level on sourceforge pages that the svn is updated regularily with the code corresponding to the Robotis documentation:
http://darwinop.svn.sourceforge.net/viewvc/darwinop/
This code is very much more complete with working software and utilities. Even a great interface from Roboplus.
I have successfully run up most of the code using a micro to perform the CM730 functions, though most of the code will run on a USB2Dynamixel with a few minor code changes to ignore the CM730 specific functions(DXL power, gyro, accel, leds, buttons).
So great work from Robotis and "Open" as promised.