[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] 4-Dimensional Trajectories (4DT)

From: David Conger
Subject: Re: [Paparazzi-devel] 4-Dimensional Trajectories (4DT)
Date: Tue, 23 Oct 2012 20:15:15 -0700

Hi Reto,
I am very interested in glider flying myself. Glad to hear of another pilot. My 
thoughts changed dramatically regarding flying autonomous things when I started 
flying full scale. It is so easy to not see full sized planes up there when 
they are right in front of you. So UAV even big ones are nearly invisible to 
the eye at times. 

So many things to work on. However I suppose like anything if you do them one 
by one eventually you will complete them all. 

I saw that about the Level 2 interactions with ATC. This is all very exciting. 
Is there plans for Paparazzi teams to do this? Any teams in Southern California 

-David Conger
On Oct 22, 2012, at 11:13 PM, Reto Büttner wrote:

> Hi David
> I completely agree to your thoughts. I am a full scale glider and
> motor pilot myself (no IFR rating though).
> Flying IFR with an UAV is a lot easier that VFR. Unfortunately most
> useful tasks for small UAVs must be flown VFR.
> The first one certifying a (small and affordable) UAV for VFR flights
> in NAS including collision avoidance with non-cooperative
> (non-ADS-B-equipped) airspace participants like small aircrafts or
> paragliders will get rich. That is the holy grail of small UAVs.
> Exactely what is requested by NASA in the Level 2 competition ...
> Regards, Reto
> 2012/10/23 David Conger <address@hidden>:
>> All,
>> As a new pilot I'm becoming more familiar with FAA rules and pilotage. 
>> Commercial planes fly IFR with filed and closely controlled flight plans. 
>> The flights are not point to point. That is changing with GPS and WAAS. 
>> Airspaces are a consideration when making a flight plan. Also before a plane 
>> is released for takeoff there needs to be a place and gate for it to land 
>> at. If planes are not leaving San Francisco due to weather planes leaving 
>> for that destination are going to be held up. Fuel is another consideration 
>> with a flight plan path and altitude.
>> Commercial air traffic is pretty complex but highly controlled and monitored.
>> Given all those planes are on pre-determined paths and staying on them with 
>> little variance (maybe due to turbulence) it's really the VFR pilots that 
>> pose the big challenge. When I fly cross country I call ATC And fly "flight 
>> following". So they give me a transponder code and follow my flight. They 
>> tell me of traffic warnings etc. I am under their control for the flight but 
>> I retain the responsibility for flying safely and following the VFR flight 
>> rules.
>> I'm curious if there are other pilots out there with Paparazzi? This will 
>> make more sense to them but it's fairly easy to understand. Given all 
>> commercial traffic is IFR let's start there. Make the UAV fly IFR. The 
>> reason is a pilot flying IFR maybe in zero visibility for all but take off 
>> and landing. So they are no better off than a UAV without eyes. What sets 
>> them apart is that they copy the instructions carefully and follow them. 
>> Speech recognition is good enough now. If the UAV (even for now the ground 
>> station) could file IFR flight plans and follow directions from ATC there is 
>> most of the problem solved there. Even better if the UAV can use speech 
>> recognition to turn ATC commands into flight commands. Example maybe ATC 
>> says change to heading 090, descend to four thousand feet (MSL) and say the 
>> barometer at a nearby airport is 29.92 inches of mercury. That is important 
>> as a pilot must keep setting their barometer value to that so every pilot in 
>> that area is using the same value. IFR flight would be far and above the 
>> easiest and I'd imagine the preferred method for UAV flight. No need for 
>> Radar, no need for optical recognition and target (visual, IR, radar) 
>> detection. I can see now why all UAV pilots are IFR rated. I hope to get my 
>> IFR rating next.
>> Anyone trying to make a UAV fly VFR in the NAS is really going to have 
>> nearly an impossible challenge.
>> -David Conger
>> On Oct 22, 2012, at 2:06 PM, Bernard Davison wrote:
>>> Guys don't forget about the obvious...
>>> Usually Commercial aircraft fly to schedules. They take off at a time and 
>>> arrive at their destination at a time. (We hope. How annoyed are we when 
>>> our flight is late)
>>> Cheers,
>>> Bernie.
>>> Sent from my iPad mini.
>>> On 22/10/2012, at 11:16 PM, Reto Büttner <address@hidden> wrote:
>>>> Hi Chris
>>>> The ATC will work from prior knowledge. The draft rules say on page 7:
>>>> "Prior to each mission, competitors must declare several details about
>>>> their aircraft and how they intend to operate it. Chief among these is
>>>> their preferred cruise speed for their aircraft."
>>>> Regards, Reto
>>>> 2012/10/22 Chris Gough <address@hidden>:
>>>>> Im only just starting to learn about  ADS-B. When my budget SDR kit 
>>>>> arrives (hacked tv tuner dongle) I'll start sniffing the ADS-B and TCAS 
>>>>> traffic with a copy the protocol specs, to try to get my head around them.
>>>>> With regard to commanding 4D trajectories, obviously the ATC should only 
>>>>> command trajectories that the aircraft is capable off, and where 
>>>>> possible, minimize disruption to it's mission. Do you imagine there would 
>>>>> be an naive exchange of messages to discover the vehicle's 
>>>>> characteristics (speed limits, remaining range, altitude ceiling etc), 
>>>>> environment (wind, temp, humidity etc) and objectives? Or do you think 
>>>>> that the ATC should work from prior knowledge about aircraft 
>>>>> characteristics and telemetry logs?
>>>>> Chris Gough
>>>>> On 22/10/2012, at 8:33 PM, Reto Büttner <address@hidden> wrote:
>>>>>> Hi Chris
>>>>>> The telemetry message you are imagining does exist. It is the ADS-B
>>>>>> message. It is part of the collision avoidance (separation, TCAS)
>>>>>> functionality.
>>>>>> The 4DT part is a simpler and is therefore the starting point of the
>>>>>> contest (L1C). That basic functionality can and should be added to
>>>>>> paparazzi straight forward. Once that is mastered, the a lot harder
>>>>>> collision avoidance functionality can be addressed (L2C).
>>>>>> Regards, Reto
>>>>>> 2012/10/22 Chris Gough <address@hidden>:
>>>>>>> I read the draft rules quickly, and needed to google quite a few terms 
>>>>>>> that went familiar to me. 4D trajectory was one of then, and it didn't 
>>>>>>> throw up anything that looked like a conventional meaning. I had 
>>>>>>> Imagined (probably wrongly) that it was a telemetry message that other 
>>>>>>> planes (and ATC) could use as a basis for smart avoiding (or directing) 
>>>>>>> behavior. If it's a telecommand as you suggest, in some ways that's a 
>>>>>>> lot simpler (it's just an extension or adaption of existing 
>>>>>>> flight-planning constructs).
>>>>>>> Either way, the harder part of the rules seems to be robustness to 
>>>>>>> spoofing (both GPS and "ghost planes"). Has anybody got experience 
>>>>>>> integrating paparazzi with light weight affordable radar? :) I've 
>>>>>>> started reading up on gnuradio, and passive (SDR) radar solution looks 
>>>>>>> to me like Mount Everest. Maybe _just_ possible for ground/terrain 
>>>>>>> sensing, but other air traffic?
>>>>>>> Has anyone used the paparazzi TCAS code recently? Is it up to date with 
>>>>>>> all the code changes from the last year or so?
>>>>>>> Chris Gough
>>>>>>> On 22/10/2012, at 4:56 PM, Reto Büttner <address@hidden> wrote:
>>>>>>>> Here are the draft rules:
>>>>>>>> They say:
>>>>>>>> Page 3: "Competition missions will be defined by Four-Dimensional
>>>>>>>> Trajectories (4DTs), which will be comprised of a series of
>>>>>>>> three-dimensional waypoints in space and a specific time of arrival
>>>>>>>> for each waypoint."
>>>>>>>> Page 7: "The five distinct segments of a mission are: aircraft launch,
>>>>>>>> pre-4DT loiter, 4DT flight, post-4DT loiter, and aircraft recovery."
>>>>>>>> There will be an air traffic control ("central puppent master"), as
>>>>>>>> they want to be able to create specific scenarios for the competitors
>>>>>>>> with surrounding air traffic using a combination of real and virtual
>>>>>>>> aircraft working synchronously.
>>>>>>>> Managing air traffic might be the next step in development of
>>>>>>>> paparazzi. I would start out with 4DT. That would be a great new
>>>>>>>> feature!
>>>>>>>> Regards, Reto
>>>>>>>> 2012/10/21 Chris Gough <address@hidden>:
>>>>>>>>> I had imagined the 4d trajectories would be chirped about between 
>>>>>>>>> vehicles
>>>>>>>>> to enhance 'autonomous sense and avoid' with vehicles at potentially 
>>>>>>>>> very
>>>>>>>>> different speeds. I need to read the rules more carefully, but I 
>>>>>>>>> didn't get
>>>>>>>>> the impression that a central puppet master was involved. That 
>>>>>>>>> wouldn't
>>>>>>>>> scale well.
>>>>>>>>> Chris Gough
>>>>>>>>> On 21/10/2012, at 9:50 PM, Gerard Toonstra <address@hidden> wrote:
>>>>>>>>> I'd expect this to be interpreted as a starting time when a uav is 
>>>>>>>>> *allowed*
>>>>>>>>> to be in some location, not so much when
>>>>>>>>> it *must* be in some location. My implication is that it's more about
>>>>>>>>> devising a strategy where the uav can be kept
>>>>>>>>> safely in waiting until it's time to move on. The use case here is 
>>>>>>>>> that this
>>>>>>>>> allows atc to keep an area void of
>>>>>>>>> other traffic until the landing of special craft X has taken place or 
>>>>>>>>> other
>>>>>>>>> use cases alike, or that a specific mission
>>>>>>>>> may only commence at time Z.
>>>>>>>>> There may also be an additional requirement where a NoFlyZone has a
>>>>>>>>> particular time range. You may cross the zone
>>>>>>>>> prior or after, but not during, otherwise you have to go around. 
>>>>>>>>> These NFZ's
>>>>>>>>> may pop up at any time during a trajectory
>>>>>>>>> and may require substantial replanning of the flight itself.
>>>>>>>>> Replanning flights isn't necessarily bad, as long as it's clear to the
>>>>>>>>> operator why it is necessary and what will happen in the
>>>>>>>>> new plan. It should also be clear what will happen if the new plan is 
>>>>>>>>> not
>>>>>>>>> accepted, because sometimes the old plan becomes
>>>>>>>>> totally incompatible with the new situation.
>>>>>>>>> Anyway... I'm speculating  :).  The actual rules will define how this 
>>>>>>>>> should
>>>>>>>>> be interpreted. I do think that here's an excellent opportunity to
>>>>>>>>> impress the judges by thinking ahead of the requirements and 
>>>>>>>>> demonstrating
>>>>>>>>> that beyond a technical implementation, some
>>>>>>>>> thinking has been undertaken why 4D is a necessity and how operators
>>>>>>>>> 'interact" with uav's to enable this in the best way
>>>>>>>>> possible (maintaining overview of the situation, reducing interaction
>>>>>>>>> complexity, etc.)
>>>>>>>>> G>
>>>>>>>>> On Sun, Oct 21, 2012 at 7:12 AM, Reto Büttner <address@hidden>
>>>>>>>>> wrote:
>>>>>>>>>> I do not think that the calculation of an ETA in flight will be 
>>>>>>>>>> enough for
>>>>>>>>>> NASA.
>>>>>>>>>> I expect rules similar to the following:
>>>>>>>>>> - Before flight you file a flight plan including 4D waypoints
>>>>>>>>>> (position, altitude and time). This calculation must include the
>>>>>>>>>> expected wind.
>>>>>>>>>> - In flight the autopilot must control position, altitude and speed 
>>>>>>>>>> to
>>>>>>>>>> hit the filed 4D waypoints.
>>>>>>>>>> - Perhaps in flight you are allowed to request a change of the filed
>>>>>>>>>> flight plan, e.g. if a delay in departure has occured or wind is
>>>>>>>>>> completely different than expected. I am sure Air Traffic Control 
>>>>>>>>>> will
>>>>>>>>>> allow only a few changes and only for good reasons.
>>>>>>>>>> Therefore Paparazzi should accept 4D waypoints (position, altitude 
>>>>>>>>>> and
>>>>>>>>>> time) and the flight control should be enhanced to hit the time. Has
>>>>>>>>>> anybody done that in Paparazzi?
>>>>>>>>>> Regards, Reto
>>>>>>>>>> 2012/10/21 Steffen Spies <address@hidden>:
>>>>>>>>>>> I think it means, that the flightplan has position and the time. 
>>>>>>>>>>> Like
>>>>>>>>>>> "be at home at 6pm" while the plane always tells if it will be in 
>>>>>>>>>>> time or
>>>>>>>>>>> not.
>>>>>>>>>>> Am 21.10.2012 um 11:21 schrieb Chris Gough
>>>>>>>>>>> <address@hidden>:
>>>>>>>>>>>> I noticed that too, and don't really understand what it means. Is 
>>>>>>>>>>>> does
>>>>>>>>>>>> it mean the telemetry messages that say "I am here now, and expect 
>>>>>>>>>>>> to be at
>>>>>>>>>>>> that place in two minutes"?
>>>>>>>>>>>> Chris Gough
>>>>>>>>>>>> On 21/10/2012, at 6:43 PM, Reto Büttner <address@hidden>
>>>>>>>>>>>> wrote:
>>>>>>>>>>>>> Hi guys
>>>>>>>>>>>>> The newest UAS competition of NASA requests 4-Dimensional 
>>>>>>>>>>>>> Trajectories
>>>>>>>>>>>>> (4DT):
>>>>>>>>>>>>> "The Level 1 Competition (L1C) would focus on a competitors 
>>>>>>>>>>>>> ability to
>>>>>>>>>>>>> fly 4-Dimensional Trajectories (4DT) to provide a reasonable
>>>>>>>>>>>>> expectation that they will be where they are supposed to be, when 
>>>>>>>>>>>>> they
>>>>>>>>>>>>> are scheduled to be there."
>>>>>>>>>>>>> See:
>>>>>>>>>>>>> Has anyone in Paparazzi realized 4-Dimensional Trajectories, in 
>>>>>>>>>>>>> other
>>>>>>>>>>>>> words has included the time dimension in flight control?
>>>>>>>>>>>>> Regards
>>>>>>>>>>>>> Reto
>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>>>>>> address@hidden
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>>>>> address@hidden
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>>>> address@hidden
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>>> address@hidden
>>>>>>>>> --
>>>>>>>>> Gerard Toonstra
>>>>>>>>> -----------------------
>>>>>>>>> _______________________________________________
>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>> address@hidden
>>>>>>>>> _______________________________________________
>>>>>>>>> Paparazzi-devel mailing list
>>>>>>>>> address@hidden
>>>>>>>> _______________________________________________
>>>>>>>> Paparazzi-devel mailing list
>>>>>>>> address@hidden
>>>>>>> _______________________________________________
>>>>>>> Paparazzi-devel mailing list
>>>>>>> address@hidden
>>>>>> _______________________________________________
>>>>>> Paparazzi-devel mailing list
>>>>>> address@hidden
>>>>> _______________________________________________
>>>>> Paparazzi-devel mailing list
>>>>> address@hidden
>>>> _______________________________________________
>>>> Paparazzi-devel mailing list
>>>> address@hidden
>>> _______________________________________________
>>> Paparazzi-devel mailing list
>>> address@hidden
>> _______________________________________________
>> Paparazzi-devel mailing list
>> address@hidden
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]