![]() |
Re: Open Source IMU designed for FIRST robotics
I think if you are using 10% of the cpu to keep receiving updates is a lot. I think 50Hz will be fine for pretty much anything.
My team wants to take 2 of them. But the only thing stopping me from getting them is that there is no labview code for them. I could to hook it up to an arduino and use i2c, but thats a lot of work and adds a little lag to it. The reason why we wanted to use one is because we tried gyros and they dont work or require a lot of tuning and drift a lot, so we thought it would just be easier to read the rates from this IMU, but if I have to try to work labview into it, we might have to just stick with tuning a gyro If you can get labview code out, I will buy two |
Re: Open Source IMU designed for FIRST robotics
cRio RS232 is done by NI VISA and does not involve the FPGA. As we (33) already do a lot of at 100hz I don't imagine RS232 adding much to it. I imagine the RS232 port is a hardware UART on the PowerPC but am not entirely sure.
|
Re: Open Source IMU designed for FIRST robotics
In order to allow any team concerned about the 100Hz update rate to choose a slower update rate, a simple option to modify the update rate (between 4 and 100Hz) has been added. This should allow anyone concerned about CPU utilization the option to make the trade off between CPU usage vs. response time.
Once we get concrete measurements on CPU usage, we'll make that information available, too. |
Re: Open Source IMU designed for FIRST robotics
We're working on locating a partner willing to develop a LabView VI for the nav6. When that happens, I'll be sure to let you know.
We will provide a free nav6 IMU and cables ($95 value) to whomever is first to step up and agree to develop and test a LabView VI for it - as long as they're willing to allow us to make that LabView VI available to others, and to allow us to modify it to include the nav6 'advanced' modes of operation. If you know of anyone interested, please have them contact me (scott@kauailabs.com). |
Re: Open Source IMU designed for FIRST robotics
Quote:
1) Have you put together that package you noted above? 2) If so, do you have a link to "high level" specifics? 3) What was the final price? 4) How do we place the order (direct or via the AndyMark website)? 5) How long after the order is placed will it be before we receive the kit? Thanks |
Re: Open Source IMU designed for FIRST robotics
Quote:
An alternative to the Nav6 board is this "Flight controller" for $35. I believe your code would be a straight drop in, although a minor tweak may be needed. It has a slightly better controller (AtMega2560), same IMU6050 and HMC5883L integrated on the board already. Thoughts? |
Re: Open Source IMU designed for FIRST robotics
Quote:
Here's some thoughts on how take the various open source parts of nav6 and approach porting them to this board, and getting it running on a FRC robot: - Based on the MultiWiiPro2 Atmega2560, MPU6050 and the HMC5883L, the nav6 open source firmware should run on this flight controller - as long as each IC''s interrupt line is connected to the Atmega microcontroller. The I2C addresses might be different, based upon the schematic. I haven't found the schematics, but if you do find them, keep an eye out for the interrupt lines. The MultiWiiPro2 code I found supports many different gyros/accels, and doesn't appear to utilize the MPU6050's DMP (which requires interrupts to tell the microcontroller when the next chunk of filtered pose data [quaternions] and sensor data is available). - Based on the MultiWiiPro2 code (I haven't been able to locate the schematics), serial communications is supported. I didn't see a RS-232 transceiver listed on the specs, so a different approach than interfacing to the CRio serial port, or some external circuitry for that, might be required. I did see that there's a header for I2C, so if you were to add I2C slave code, and this I2C bus was separate from the I2C bus used to communicate w/the onboard sensors, this could be an option. - I'm not sure what kind of voltage regulation the MultiWii2 has onboard. It appears (from looking at the silkscreen of the board from photos I've found) to require require 5VDC; for nav6 we found it simplest to handle a range from 6 to 14VDC, so that the IMU can be powered directly from an unregulated output from the FRC Power Distribution Board. - So in summary, and assuming you'd use I2C to interface to it, I believe you'd need to (a) ensure the MPU6050/HMC5883 interrupts are available to the processor, (b) adapt the nav6 firmware to add a I2C slave interface, (c) adapt the nav6 serial protocol to be command/request over the I2C bus and (d) power the board from the Digital sidecar. I recommend reviewing the digital sidecar bandwidth (CRio<-> sidecar) to ensure it's sufficient to transfer the IMU data at the update rates you want. It's a key part of the nav6 mission to enable students to modify the firmware, circuitry, interface software, and enclosure of the nav6. All of this is available at http://code.google.com/p/nav6. So if you decide to port the nav6 firmware to this board, please let me know if there's anything I can do to help. Aloha, - scott |
Re: Open Source IMU designed for FIRST robotics
Quote:
- Full documentation, and source code for firmware, board schematics, bill of materials, and STL files to 3d-print an enclosure are on the wiki, at http://code.google.com/p/nav6. - We will ship out the nav6 board the day after we receive the order. If there are any questions you can contact me at scott@kauailabs.com. |
Re: Open Source IMU designed for FIRST robotics
Quote:
|
Re: Open Source IMU designed for FIRST robotics
Order 37 processed today, TORC.
Scott. |
Re: Open Source IMU designed for FIRST robotics
Quote:
|
Re: Open Source IMU designed for FIRST robotics
Did anyone use a Nav6 in competition? What were your thoughts?
|
Re: Open Source IMU designed for FIRST robotics
We did not use an IMU in competition yet but the testing we did do last summer indicated that with a game like this years game with the heavy defense and high G impacts would have yielded less than acceptable field centric control. IMU's that are perfect for our needs are out of our price range.
|
Re: Open Source IMU designed for FIRST robotics
Quote:
So, I'm curious what IMUs you do think are perfect for your needs, or more specifically what specifications you're shooting for. The Nav6 goal is teleop Field-oriented drive for any team that wants it, without requiring "zeroing the yaw" during the game. Thanks for any input you may have, - scott |
Re: Open Source IMU designed for FIRST robotics
This nav6 thread was recently "found" by some of my students who are interested in giving this guy a shot. However, I'm concerned about its compatibility with the new roboRIO controller. Does the RS232 on the nav6 require any pins other than the RXD and TXD pins - these are the only 2 RS232 pins available on the roboRIO (other than a ground reference).
Thanks! -Danny |
| All times are GMT -5. The time now is 05:50. |
Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi