Re: [Paparazzi-devel] No throttle in Auto2

[cory barton]

Uttam, thanks for the help. I am continuing to dig into this. I did
try everything I could find to resurrect the throttle to no avail. The
resurrect button did nothing, and setting kill_throttle to 0 in the
settings tab also failed.

I was completely unable to get out of the holding point block.

I found this in var/include/messages.h:
#define DOWNLINK_SEND_BAT(throttle, voltage, flight_time,
kill_auto_throttle, block_time, stage_time, energy){ \
        if (DownlinkCheckFreeSpace(DownlinkSizeOf(0+2+1+2+1+2+2+2))) {\
          DownlinkCountBytes(DownlinkSizeOf(0+2+1+2+1+2+2+2)); \
          DownlinkStartMessage("BAT", DL_BAT, 0+2+1+2+1+2+2+2) \
          DownlinkPutInt16ByAddr((throttle)); \
          DownlinkPutUint8ByAddr((voltage)); \
          DownlinkPutUint16ByAddr((flight_time)); \
          DownlinkPutUint8ByAddr((kill_auto_throttle)); \
          DownlinkPutUint16ByAddr((block_time)); \
          DownlinkPutUint16ByAddr((stage_time)); \
          DownlinkPutUint16ByAddr((energy)); \
          DownlinkEndMessage() \
        } else \
          DownlinkOverrun(); \

Here is a BAT message from the log. All the BAT messages have the
fourth token equal to 1.
5.316 1 BAT 432 79 0 1 13 0 1

If I understand the above code correctly this is what BAT means
BAT 432=throttle 79=voltage 0=flight_time 1=kill_auto_throttle

I only found a reference to kill_auto_throttle in /var/include/messages.h

I did find references to kill_throttle(seems to be the same as
kill_auto_throttle) in:

sw/airborne/main_ap.c
sw/airborne/coaxial/tl_control.c
var/nexstarep50/flight_plan.h
var/nexstarep50/main_ap.lst

from main_ap.c
these lines seem to be relevant(starting at line 423):

    static uint8_t t = 0;
    if (vsupply < LOW_BATTERY_DECIVOLT) t++; else t = 0;
    kill_throttle |= (t >= LOW_BATTERY_DELAY);
    kill_throttle |= launch && (dist2_to_home > Square(KILL_MODE_DISTANCE));

I have <define name="CATASTROPHIC_BAT_LEVEL" value="6.0" unit="V"/> in
my airframe config, and the battery is reading 79 decivolts, so I
think that is ok.
Also, I am very close to the home waypoint when I am trying this out,
so I think dist2_to_home should be less than the square of
KILL_MODE_DISTANCE
I have max_dist_from_home="700" in my flight plan.

I do not think that sw/airborne/coaxial/tl_control.c is used in my case(?).

from var/nexstarep50/flight_plan.h

static inline void auto_nav(void) {
  switch (nav_block) {
    Block(0) // Wait GPS
    ; // pre_call
    switch(nav_stage) {
      Stage(0)
        kill_throttle = 1;
        NextStageAndBreak();
        break;
...
<snip>
...
    Block(2) // Holding point
    ; // pre_call
    switch(nav_stage) {
      Stage(0)
        kill_throttle = 1;
        NextStageAndBreak();
        break;
....
<snip>
...
    Block(3) // Takeoff
    ; // pre_call
    if ((estimator_z>(ground_alt+25)) && (nav_block != 4)) {
GotoBlock(4); return; }
    switch(nav_stage) {
      Stage(0)
        kill_throttle = 0;
...
<snip>

That all looks good to me. I am at a loss now. I am not sure where else to look.

Please let me know where the problem may be.

Thanks,

Cory

On Wed, Aug 20, 2008 at 11:29 PM, Uttam Chandrashekhar
<address@hidden> wrote:
> Cory, did you try resurrect throttle on the GCS? Look at the BAT:
> kill_auto_throttle message.
>
>
> -Uttam
>
> -
>
> ________________________________
> Unlimited freedom, unlimited storage. Get it now
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>
>