[Ovmsdev] Locations and scripts
Michael Balzer
dexter at expeedo.de
Tue Dec 3 20:32:16 HKT 2019
Although filtering now I just got another false alert while parking with perfect GPS situation:
Currently at 51.322430;7.344803 (GPS lock; 10 satellites)
Alert:
Vehicle is being transported while parked - possible theft/flatbed (@51.322502;1.166667)
So it's not a small jitter, it's complete nonsense. That's with GPS from the modem, may be different for car GPS.
We'll need a different approach. Maybe ignore/drop GPS coordinates too far from the last known position?
Regards,
Michael
Am 26.11.19 um 21:31 schrieb Michael Balzer:
> Depends on the GPS source. On the modem it's 5 seconds.
>
> If you're using GPS from the car, it depends on the sample frequency when parked / off.
>
> But the low pass filter will pass a large step in distance also if it's a single one, so on an actual movement it will trigger before 5 samples have passed.
>
> Regards,
> Michael
>
>
> Am 26.11.19 um 21:24 schrieb Stephen Casner:
>> Michael,
>>
>> Good idea. How much time does 5 samples represent?
>>
>> -- Steve
>>
>> On Tue, 26 Nov 2019, Michael Balzer wrote:
>>
>>> I've had my first false alert last night, from the log it seems caused by a single false GPS reading.
>>>
>>> I've added smoothing over 5 samples to the distance check, that should eliminate the false alerts unless the GPS errors are huge or persistent.
>>>
>>> I've also added the coordinates to the notification so we can see how large the error can become if it still occurs.
>>>
>>> Regards,
>>> Michael
>>>
>>>
>>> Am 21.08.19 um 23:43 schrieb Greg D.:
>>>> Hmmpf.
>>>>
>>>> Precession of the GPS constellation, perhaps? Folks in northern Nevada
>>>> or Idaho should be next...
>>>>
>>>> Greg
>>>>
>>>>
>>>> Stephen Casner wrote:
>>>>> Greg,
>>>>>
>>>>> Mine has been quiet for the past couple of months without increasing
>>>>> the alarm distance beyond the default 500 meters.
>>>>>
>>>>> -- Steve
>>>>>
>>>>> On Wed, 21 Aug 2019, Greg D. wrote:
>>>>>
>>>>>> A quick follow-up to this thread...
>>>>>>
>>>>>> I've noticed that since this past spring that the number of false alerts regarding the
>>>>>> flatbedding of my Roadster have taken a sharp spike upward. Neither my car's normal
>>>>>> parking spot in the garage, nor the location of my house / garage have changed. The
>>>>>> difference in frequency did not appear to coincide with a firmware update, so I'm at a
>>>>>> bit of a loss to explain it.
>>>>>>
>>>>>> Has anyone else noticed this?
>>>>>>
>>>>>> Did the Government make a change to the GPS satellite fleet?
>>>>>>
>>>>>> Not an issue, really, but definitely something odd going on.
>>>>>>
>>>>>> Greg
>>>>>>
>>>>>>
>>>>>> Stephen Casner wrote:
>>>>>>
>>>>>> Greg,
>>>>>>
>>>>>> Attached are graphs of your data. gregd.all.png shows all of the
>>>>>> data, and gregd.peak.png is the highest distance deviation of 369
>>>>>> meters, which is less than the default 500 meter threshold.
>>>>>>
>>>>>> It looks like there is a diurnal pattern to the data with increased
>>>>>> deviations around midnight local time (the X axis times are UTC).
>>>>>>
>>>>>> It also looks like your home elevation must be about 380 meters since
>>>>>> the altitude variable drops down to negative that amount frequently,
>>>>>> which would occur when that value is taken as zero (there is a special
>>>>>> case in the code to treat some values as zero).
>>>>>>
>>>>>> I tried overlaying my data and yours, but there seems to be no
>>>>>> correlation between the peaks in my data and the peaks in yours.
>>>>>> Interestingly, my data does not show a similar diurnal pattern.
>>>>>>
>>>>>> -- Steve
>>>>>>
>>>>>> On Tue, 4 Jun 2019, Greg D. wrote:
>>>>>>
>>>>>> 200mb of logs sent to Steve for analysis....
>>>>>>
>>>>>> Greg
>>>>>>
>>>>>>
>>>>>> Stephen Casner wrote:
>>>>>>
>>>>>> Thanks. For me, today's log was much quieter. The maximum excursion
>>>>>> was 115 meters and it was roughly coincident with loss of lock.
>>>>>>
>>>>>> -- Steve
>>>>>>
>>>>>> On Fri, 31 May 2019, Greg D wrote:
>>>>>>
>>>>>> Ok. I'll let it run for a few days and see what happens. Not planning to drive the ca
>>>>>> r until next week.
>>>>>>
>>>>>> Greg
>>>>>>
>>>>>>
>>>>>> On May 31, 2019 2:11:31 PM PDT, Stephen Casner <casner at acm.org> wrote:
>>>>>>
>>>>>> On Fri, 31 May 2019, Greg D. wrote:
>>>>>>
>>>>>> Tried this, and am not getting anything written to the file system
>>>>>>
>>>>>> (vfs
>>>>>>
>>>>>> ls /sd shows zero bytes in gps.crtd). Did I miss some sort of setup
>>>>>>
>>>>>> step?
>>>>>>
>>>>>> Nothing gets written (or at least the file size as reported is not
>>>>>> updated) until the recording is stopped with "can log off".
>>>>>>
>>>>>> -- Steve
>>>>> _______________________________________________
>>>>> OvmsDev mailing list
>>>>> OvmsDev at lists.openvehicles.com
>>>>> http://lists.openvehicles.com/mailman/listinfo/ovmsdev
>>>> _______________________________________________
>>>> OvmsDev mailing list
>>>> OvmsDev at lists.openvehicles.com
>>>> http://lists.openvehicles.com/mailman/listinfo/ovmsdev
>>> --
>>> Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
>>> Fon 02333 / 833 5735 * Handy 0176 / 206 989 26
>>>
>>> _______________________________________________
>>> OvmsDev mailing list
>>> OvmsDev at lists.openvehicles.com
>>> http://lists.openvehicles.com/mailman/listinfo/ovmsdev
>>>
>>>
>> _______________________________________________
>> OvmsDev mailing list
>> OvmsDev at lists.openvehicles.com
>> http://lists.openvehicles.com/mailman/listinfo/ovmsdev
--
Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal
Fon 02333 / 833 5735 * Handy 0176 / 206 989 26
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openvehicles.com/pipermail/ovmsdev/attachments/20191203/c0ae43a5/attachment.html>
More information about the OvmsDev
mailing list