[Elphel-support] Connecting the IMU and GPS adapter boards to 10369board
Andrey Filippov
support-list at support.elphel.com
Wed Oct 12 09:06:30 PDT 2011
Lee,
That's a smart solution - I did not think of it myself.
Andrey
On Wed, Oct 12, 2011 at 10:02 AM, Lemay, Lee C CIV SPAWARSYSCEN-PACIFIC,
52340 <lee.lemay at navy.mil> wrote:
> **
> Andrey,
>
> That's fine. Please let me know when you add this functionality.
>
> In the meanwhile, we can program the ADIS separately and save the program
> to the ADIS's flash. Then when we place it on the camera it will be
> correctly configured.
>
> Lee
>
> ------------------------------
> *From:* elphel at gmail.com [mailto:elphel at gmail.com] *On Behalf Of *Andrey
> Filippov
> *Sent:* Tuesday, October 11, 2011 6:06 PM
>
> *To:* Lemay, Lee C CIV SPAWARSYSCEN-PACIFIC, 52340;
> support-list at support.elphel.com
> *Subject:* Re: [Elphel-support] Connecting the IMU and GPS adapter boards
> to 10369board
>
> Lee,
>
> I'm sorry - it seems this will have to wait until I'll get to the FPGA
> programming again - the easy solution I told you about will not work. I just
> looked in the code - only 6 bits of the 16 bit sent over SPI are
> programmable in the current FPGA code. I forgot to implement writing (just
> missed the fact it is not possible with possible to do slowly in software
> anymore) as I was not counting much on the in-IMU filtering - logging raw
> data at full 2560samples/sec should be enough to play with the filtering at
> post-processing.
>
> But that's definitely the flaw of the current code - I'll fix that, but not
> right now.
>
> Andrey
>
> On Tue, Oct 11, 2011 at 6:28 PM, Lemay, Lee C CIV SPAWARSYSCEN-PACIFIC,
> 52340 <lee.lemay at navy.mil> wrote:
>
>> **
>> Andrey,
>>
>> This is excellent. Thank you for the information.
>>
>> I have one request, which would be to be able to configure the ADIS IMU
>> sample decimation rate through the use of imu_setup.php. For
>> completeness, one would also able to select the FIR filter in default
>> configuration, A-D, and enable or disable the filter.
>> I understand if this is not a priority. It would definitely enhance the
>> usability of the ADIS IMU.
>>
>> Lee
>>
>> ------------------------------
>> *From:* elphel at gmail.com [mailto:elphel at gmail.com] *On Behalf Of *Andrey
>> Filippov
>> *Sent:* Tuesday, October 11, 2011 3:50 PM
>> *To:* Lemay, Lee C CIV SPAWARSYSCEN-PACIFIC, 52340;
>> support-list at support.elphel.com
>> *Subject:* Re: [Elphel-support] Connecting the IMU and GPS adapter boards
>> to 10369board
>>
>>
>>
>> On Tue, Oct 11, 2011 at 4:24 PM, Lemay, Lee C CIV SPAWARSYSCEN-PACIFIC,
>> 52340 <lee.lemay at navy.mil> wrote:
>>
>>> **
>>> Thank you Andrey. I have a few more questions.
>>>
>>> Where do you set the baud rate for GPS?
>>>
>>
>> To write to EEPROM you may use this page http://community.elphel.com/<remove-this>bootblock353/
>> , this page also allows to read EEPROM and bootblock records on the camera
>> boards
>>
>>
>>> Which directory is the XML configuration file located in?
>>>
>>
>> /etc/imu_logger.xml
>>
>> You may also try to open the page the http://camera-ip/imu_setup.php(without parameters) - it shows how to control the logger (I'll attach the
>> output below)
>>
>>
>>
>>> If I want to use "auto" baud rate, will the process be:
>>> - power on Elphel
>>> - wait until GPS gets a lock
>>> - type 'reboot' into the console ?
>>>
>> "reboot -f"
>> or open url in the browser
>> http://camera-ip/phpshell.php?command=reboot+-f
>>
>> just make sure you close that page - otherwise if the browser (or the
>> computer) will crash, it will try to open that page when you run the
>> browser, rebooting camera unintentionally.
>>
>>>
>>> Because if I hard reboot the Elphel by removing power, the GPS will lose
>>> power too. So typing reboot seems like the only way to maintain GPS lock and
>>> reboot the camera. Correct me if I am wrong.
>>>
>>
>> Usually, if the GPS was locked, it will re-lock after short power outage
>> before the camera will try to detect it.
>>
>>>
>>> I appreciate your help!
>>>
>>> Lee
>>>
>>> ------------------------------
>>>
>>> Usage of IMU/GPS logger (/imu_setup.php)
>>
>> This script controls IMU/GPS logger. When the camera starts, the script
>> looks for the */etc/imu_logger.xml*" and if it is found, the settings
>> from this file are applied to the logger. If there is no such file (i.e. at
>> first boot after reflashing the camera) script looks for the optional
>> extension boards 103695 <http://192.168.0.9/imu_setup.php> and 103696<http://wiki.elphel.com/index.php?title=103696>and uses the information from their flash memory (not erased during reflash)
>> combined with the defaults to create initial */etc/imu_logger.xml* file
>> that can be modified later
>>
>> Script receives parameters as HTTP GET request, there are five special
>> parameters:
>>
>> - *source <http://192.168.0.9/imu_setup.php?source>* - returns the
>> source of this script
>> - *list <http://192.168.0.9/imu_setup.php?list>* - returns an XML
>> containing current logger settings, the names of the nodes can be used to
>> modify specific settings
>> - *save* - overwrite */etc/imu_logger.xml* with the current settings -
>> they will be used next time camera will boot up
>> - *msg=url-encoded-text-up-to-56-bytes* - set the "odometer" message
>> that will be logged next time the external input is toggled
>> - *smsg=url-encoded-text-up-to-56-bytes* - (not yet implemented) set
>> the "odometer" message and trigger logging immediately
>>
>> Other parameters are just *name=value* pairs with colons separating names
>> and subnames
>>
>> - *imu_period* - set the IMU logging period in SPI SCLK periods, maybe
>> set to *auto* (default value, same as 0xff000000) to log when the IMU
>> data is ready. Value 0 turns the IMU logging off
>> - *sclk_freq* - set SPI clock frequency, in Hz. Default is 5000000
>> (5MHz)
>> - *stall* - set SPI 'stall' time, in microseconds. Default is 2 usec
>> - *baud_rate* - set baud rate for the serial GPS, default is 19200
>> - *imu_slot* - specify the IMU board (103695<http://wiki.elphel.com/index.php?title=103695>)
>> connection to the 10369 <http://wiki.elphel.com/index.php?title=10369>board 0 - none, 1 - J9, 2 - J10, 3 - J11. Default is 1 (J9)
>> - *gps_slot* - specify the GPS board (103696<http://wiki.elphel.com/index.php?title=103696>)
>> connection to the 10369 <http://wiki.elphel.com/index.php?title=10369>board 0 - none, 1 - J9, 2 - J10, 3 - J11. Default is 2 (J10)
>> - *gps_mode* - synchronization mode for the GPS: 0 - pulse-per-sample
>> external input, negative front, 1 - pps input positive front, 2 - start of a
>> first sentence after pause, 3 - start of each individual sentence. Mode 2 is
>> the default, modes 0 and 1 require that GPS has a dedicated synchro output
>> - *msg_conf* - GPIO number and polarity (+16 - invert polarity) uses
>> to log external events (i.e. odometer pulses from a vehicle wheel sensor).
>> Timestamp uses leading edge, software may write to the 56-byte buffer (see
>> 'msg=' command above) before the trailing edge to attach the message to the
>> event after the event occured. In the case of short external pulses the
>> logged message will be the last written before the event
>> - *img_sync* - 0 - disable logging of the images, 1 - enable. In the
>> case of the multiple synchronized cameras this provides a way to match
>> external (master) camera timestamps transmitted over the synchronization
>> cable, with the local camera timestamps used by the logger - both are
>> logged. Default is 1
>> - *extra_conf* - Development signals (0 - config_long_sda_en, 1
>> -config_late_clk, 2 - config_single_wire, should be set for 103695 rev A).
>> Default is 4
>> - *slow_spi* - may be set to 1 for slow SPI devices (other changes may
>> be needed, i.e. changing power from 3.3V to 5.0V selected by a resistor
>> divider). Default is 0
>> - *imu_sa* - i2c slave address modifier (matches hardware jumpers).
>> Default is 3
>> - *sleep_busy* - setting for the driver. It is a sleep time (in
>> microseconds) before the driver will poll the logger buffer if it does not
>> have the full sample record (64 bytes of data) ready
>> - *imu_registers* - setting the IMU register addresses to be logged
>> (total number of logged registers is 28). You may see the complete list of
>> the registers in the IMU datasheet. The registers are addressed as
>> 'imu_registers:NN', where NN is the register sequnce number, starting with
>> 0. Below are the default values:
>> - *imu_registers:0*=0x10 - x gyro low
>> - *imu_registers:1*=0x12 - x gyro high
>> - *imu_registers:2*=0x14 - y gyro low
>> - *imu_registers:3*=0x16 - y gyro high
>> - *imu_registers:4*=0x18 - z gyro low
>> - *imu_registers:5*=0x1a - z gyro high
>> - *imu_registers:6*=0x1c - x accel low
>> - *imu_registers:7*=0x1e - x accel high
>> - *imu_registers:8*=0x20 - y accel low
>> - *imu_registers:9*=0x22 - y accel high
>> - *imu_registers:10*=0x24 - z accel low
>> - *imu_registers:11*=0x26 - z accel high
>> - *imu_registers:12*=0x40 - x delta angle low
>> - *imu_registers:13*=0x42 - x delta angle high
>> - *imu_registers:14*=0x44 - y delta angle low
>> - *imu_registers:15*=0x46 - y delta angle high
>> - *imu_registers:16*=0x48 - z delta angle low
>> - *imu_registers:17*=0x4a - z delta angle high
>> - *imu_registers:18*=0x4c - x delta velocity low
>> - *imu_registers:19*=0x4e - x delta velocity high
>> - *imu_registers:20*=0x50 - y delta velocity low
>> - *imu_registers:21*=0x52 - y delta velocity high
>> - *imu_registers:22*=0x54 - z delta velocity low
>> - *imu_registers:23*=0x56 - z delta velocity high
>> - *imu_registers:24*=0x0e - temperature
>> - *imu_registers:25*=0x70 - time m/s
>> - *imu_registers:26*=0x72 - time d/h
>> - *imu_registers:27*=0x74 - time y/m
>> - *nmea* - set format of the four logged NMEA sentences. For each
>> sentence two text fields have to be specified - 'sentence' (the three
>> letters following the comman prefix '$GP'), and 'format' that consists of a
>> sequence of 'n' (for number) and 'b' (for bytes), up to 24 total. Numbers
>> consisting of digits, minus sign and decimal point are encoded with 2 digits
>> per byte (to make sure the longest sentence fits into the 56 byte logger
>> payload buffer), followed by the 0xf nibble (half-byte). Bytes are encoded
>> with MSB cleared ( & 0x7f) for all bytes in a field but the last, the last
>> byte has it set (|0x80). Empty byte field is logged as 0xff. Below are the
>> default settings:
>> - *nmea:0:sentence*=RMC - sentence $GPRMC
>> - *nmea:0:format*=nbnbnbnnnnb - format for the $GPRMC
>> - *nmea:1:sentence*=GGA - sentence $GPGGA
>> - *nmea:1:format*=nnbnbnnnnbnbbb - format for the $GPGGA
>> - *nmea:2:sentence*=GSA - sentence $GPGSA
>> - *nmea:2:format*=bnnnnnnnnnnnnnnnn - format for the $GPGSA
>> - *nmea:3:sentence*=VTG - sentence $GPVTG
>> - *nmea:3:format*=nbnbnbnb - format for the $GPVTG
>>
>> You may record the log to a file (i.e. on external USB stick or SSD/HDD)
>> with teh following command:
>> *cat /dev/imu >log-file-path*
>>
>> Test logger and GPS:
>> *fpcf -imu 10000 | grep "1:"*
>>
>> The fpcf output starts the line (logged sample) with the source of the
>> sample:
>>
>> - *0:* - IMU (normally 2560 samples/sec)
>> - *1:* - GPS
>> - *2:* - camera images
>> - *3:* - 'odometer' - external events
>>
>> '10000' means that fpcf should output 10000 samples, that takes
>> approximately 4 seconds if IMU output is being logged.
>>
>> _______________________________________________
>> Support-list mailing list
>> Support-list at support.elphel.com
>> http://support.elphel.com/mailman/listinfo/support-list_support.elphel.com
>>
>>
>
> _______________________________________________
> Support-list mailing list
> Support-list at support.elphel.com
> http://support.elphel.com/mailman/listinfo/support-list_support.elphel.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://support.elphel.com/pipermail/support-list_support.elphel.com/attachments/20111012/9e1cdde4/attachment-0002.html>
More information about the Support-list
mailing list