HobbyKing.com New Products Flash Sale
Closed Thread
Thread Tools
This thread is privately moderated by Mick Molloy, who may elect to delete unwanted replies.
Old Dec 18, 2011, 05:04 PM
OpenUAS
openuas's Avatar
Amsterdam
Joined May 2009
68 Posts
The files are not accessible for me either. I can send them to you, but I think it is better to get them from official URL just in case something has changed in the files. Maybe they are updated.
openuas is offline Find More Posts by openuas
Sign up now
to remove ads between posts
Old Dec 18, 2011, 11:26 PM
Peter Yu
peteryu's Avatar
United States, CA, San Jose
Joined Dec 2011
6 Posts
Quote:
Originally Posted by openuas View Post
The files are not accessible for me either. I can send them to you, but I think it is better to get them from official URL just in case something has changed in the files. Maybe they are updated.
Thanks! But I've already made my own. Unless they've changed the coordinates (which hopefully did not happen), there wouldn't be anything new, right?

I'll contact them again and hopefully someone from the OBC is reading this.

On a different note, is it true that last year the teams only received the waypoints a few days before the competition? That's what sUAS says:

http://www.suasnews.com/outback-challenge-2010/
peteryu is offline Find More Posts by peteryu
Old Dec 19, 2011, 03:50 AM
Registered User
Joined Mar 2011
56 Posts
teams using APM ?

I'm curious to know how many teams will be using an ArduPilotMega with ArduPlane (or even ArduCopter!) for their entry. I've been adding a few features to ArduPlane to make it more suitable for the OBC S&R challenge and it would be nice to hear from other teams using it and what extra features may be needed by those teams.

Most recently I've added geo-fencing support, and used the OBC 2010 boundary as the test suite. That allows an APM to form the core of the failsafe code for the OBC.

Cheers, Tridge
tridge is offline Find More Posts by tridge
Old Dec 19, 2011, 08:39 AM
MINE!
Mick Molloy's Avatar
Australia, VIC, Point Cook
Joined Jul 2003
3,106 Posts
Quick reply as Mick is really busy.

I'll inform the web side that a link is broken and see what can be done about it.

I would only use the official file(s) from the Outback challenge website. It can change in respect to previous years, assume no year is the same as the ones gone by.

Teams gets their waypoints when they pass D2, not days before as reported elsewhere, however during team briefings on the event days, you will be given the latest information on the rescue of Joe. This may include any waypoint changes (as in previous years) that are necessary for the competition to commence. The waypoints are not set in stone, and reflect a real life situation where any updated information may increase or decrease the search area.

The tech committee is mainly on holidays (it's that time of year), so some answers on insurance etc, may not be able to be provided till our next meeting early next year. I'm sure Mick or myself will check in from time to time here etc

Happy xmas

Cheers

Mark
Mick Molloy is online now Find More Posts by Mick Molloy
Old Dec 19, 2011, 12:11 PM
Gaftopher
Gary Mortimer's Avatar
Nottingham Road South Africa/Bedford UK
Joined Feb 2007
3,624 Posts
@peteryu that wording was from the original Outback site, it was the brief from there.

For this time look at the recent stuff as Mark says.
Gary Mortimer is online now Find More Posts by Gary Mortimer
Old Dec 19, 2011, 12:12 PM
Gaftopher
Gary Mortimer's Avatar
Nottingham Road South Africa/Bedford UK
Joined Feb 2007
3,624 Posts
@peteryu that wording was from the original Outback site, it was the brief from then.

For this time look at the recent stuff as Mark says.
Gary Mortimer is online now Find More Posts by Gary Mortimer
Old Dec 20, 2011, 08:07 AM
Registered User
Joined Dec 2010
28 Posts
Team TGIF, from Hackerspace Brisbane, are using APM+ArduPlane for a significant part of our flight platform. In case you are interested.... We currently have three experimental plane/s flying using this electronics, and one quadcopter. The quad copter is for fun., not for the OBC. :-)

I assume you'll be connecting your geofence implementation to a commercial "termination unit" ( which then interrupts the RC signals appropriately on command ) .. and this commection will be by signaling it with an I/O pin from the primary flight APM?

The way I read the OBC rules, the "termination system" must be completely independent from the on-board flight systems, which means that having a signal from the primary flight APM that controls when to terminate ( like you've implemented, if I understand it ), is no longer completely independant, is it? eg: the commercial termination unit will not be triggered if the primary APM fails for some reason ( eg battery goes flat, system reboots, locks up, or even just a software bug, whatever).. I know it's a low risk issue, because without the primary APM, the plane is not long for this world, but it might mean the difference between a plane that terminates, and one that doesn't, yes?

The only way I can see that the rule/s could be obeyed 100% is that the "geofence calculator" ( whatever that is) be separate from the other flight systems to the point of having its own battery ( eg sharing the battery with the termination unit, or having these two unit/s be one. )

I'm also looking at the rule which basically says "must be able to terminate the plane on-demand, from the ground" , and wondering if this means we must have an entirely separate out-of-band long-range radio link of some sort, just for the termination system? (Eg, would it be acceptable to use the existing telemetry link to send termination instructions, because it's already a "must have" with regard to 1hz the heart-beat signal and what-not guarantees that have been specified around it? ) .

Buzz.


Quote:
Originally Posted by tridge View Post
I'm curious to know how many teams will be using an ArduPilotMega with ArduPlane (or even ArduCopter!) for their entry. I've been adding a few features to ArduPlane to make it more suitable for the OBC S&R challenge and it would be nice to hear from other teams using it and what extra features may be needed by those teams.

Most recently I've added geo-fencing support, and used the OBC 2010 boundary as the test suite. That allows an APM to form the core of the failsafe code for the OBC.

Cheers, Tridge
davidbuzz is offline Find More Posts by davidbuzz
Old Dec 20, 2011, 03:10 PM
Registered User
Joined Mar 2011
56 Posts
Quote:
Originally Posted by davidbuzz View Post
Team TGIF, from Hackerspace Brisbane, are using APM+ArduPlane for a significant part of our flight platform. In case you are interested.... We currently have three experimental plane/s flying using this electronics, and one quadcopter. The quad copter is for fun., not for the OBC. :-)
The quads are certainly fun :-)

Quote:
I assume you'll be connecting your geofence implementation to a commercial "termination unit" ( which then interrupts the RC signals appropriately on command ) .. and this commection will be by signaling it with an I/O pin from the primary flight APM?
yes, except we are building our own termination unit, not using a commercial one.

Quote:
The way I read the OBC rules, the "termination system" must be completely independent from the on-board flight systems, which means that having a signal from the primary flight APM that controls when to terminate ( like you've implemented, if I understand it ), is no longer completely independant, is it?
Section 5.5.5 says "The autopilot is allowed to detect the mission boundary crossing and activate the flight termination device". We have interpreted the "completely independent" requirement as meaning that the termination device won't fail due to failure of other components (so separate power and CPU, as listed in the requirements).

Quote:
eg: the commercial termination unit will not be triggered if the primary APM fails for some reason ( eg battery goes flat, system reboots, locks up, or even just a software bug, whatever)
In our design the termination device also monitors a heartbeat from the autopilots, so it knows if the autopilot has locked up. The rules cover what to do when that happens.

Quote:
I'm also looking at the rule which basically says "must be able to terminate the plane on-demand, from the ground" , and wondering if this means we must have an entirely separate out-of-band long-range radio link of some sort, just for the termination system? (Eg, would it be acceptable to use the existing telemetry link to send termination instructions, because it's already a "must have" with regard to 1hz the heart-beat signal and what-not guarantees that have been specified around it? ) .
Our interpretation is that using the existing telemetry link is OK for sending a termination command. In our design we have 2 radio links (one high speed, one low speed), and we can terminate via either link.

Cheers, Tridge
tridge is offline Find More Posts by tridge
Old Dec 23, 2011, 08:59 PM
Peter Yu
peteryu's Avatar
United States, CA, San Jose
Joined Dec 2011
6 Posts
Quote:
Originally Posted by tridge View Post
Our interpretation is that using the existing telemetry link is OK for sending a termination command. In our design we have 2 radio links (one high speed, one low speed), and we can terminate via either link.
The rules aren't terribly clear on what exactly an independent FTS is, but I'm inclined to believe that a separate radio link is necessary.

Here's why:

The 1 Hz "heartbeat" that monitors the telemetry link is supposed to result in two scenarios.

1. All the systems are functioning normally and a FTS command via the telemetry link would instantly activate the FTS.

2. The link is lost and the UAV enters Loss-of-Comms mode autonomously.

The problem is that in the second scenario, there is no way to trigger the FTS if the plane is clearly malfunctioning. The only possible thing that can trigger the FTS in Loss-of-Comms mode is the Geofencing system. If the plane was flying out of control but still within the boundaries, there would be no way for trigger the FTS.

Thus, I believe that when the rules state that an independent FTS is necessary, they're referring to an independent board, battery, and radio.

Hopefully someone official can comment on this, because I'm just speculating.
peteryu is offline Find More Posts by peteryu
Old Dec 23, 2011, 09:05 PM
Peter Yu
peteryu's Avatar
United States, CA, San Jose
Joined Dec 2011
6 Posts
Quote:
Originally Posted by Mick Molloy View Post
I would only use the official file(s) from the Outback challenge website. It can change in respect to previous years, assume no year is the same as the ones gone by.

Teams gets their waypoints when they pass D2, not days before as reported elsewhere, however during team briefings on the event days, you will be given the latest information on the rescue of Joe. This may include any waypoint changes (as in previous years) that are necessary for the competition to commence. The waypoints are not set in stone, and reflect a real life situation where any updated information may increase or decrease the search area.
Okay, I'm a little confused here.

To make this absolutely clear, the waypoints in the Competition Rules PDF are not the final waypoints?

So when the teams pass D2, they'll be given another set of coordinates, which can change yet again in the hours leading up to the competition?
peteryu is offline Find More Posts by peteryu
Old Dec 23, 2011, 09:18 PM
Registered User
Joined Mar 2011
56 Posts
Hi Peter,

Quote:
Originally Posted by peteryu View Post
1. All the systems are functioning normally and a FTS command via the telemetry link would instantly activate the FTS.

2. The link is lost and the UAV enters Loss-of-Comms mode autonomously.

The problem is that in the second scenario, there is no way to trigger the FTS if the plane is clearly malfunctioning. The only possible thing that can trigger the FTS in Loss-of-Comms mode is the Geofencing system. If the plane was flying out of control but still within the boundaries, there would be no way for trigger the FTS.
I think that is covered adequately by the 5.5.2 loss of data link wording. If you have lost comms but are within the mission boundary then the plane must follow a series of steps to try and re-establish comms. Those steps have time limits -- 10 sec initial timeout on link loss, flight to comms hold, 2 minutes loiter at comms hold, flight to airfield home, 2 minute loiter at airfield home, then if you still don't have control you can do a "controlled termination" which can include an automated landing.

So in the situation you describe safety is maintained by 2 mechanisms:

1) the geo-fencing is active the whole time. If it goes outside the fence it terminates
2) the time limits to re-establish comms mean that it will eventually recover or terminate

Even if all of the comms logic didn't work the plane will eventually run out of fuel/battery and crash within the mission boundary. That would be annoying, but wouldn't be a safety risk.

So I think its reasonable to assume that a separate radio link dedicated to flight termination isn't needed, which is what we will assume unless one of the organisers posts a clarification.

Cheers, Tridge
tridge is offline Find More Posts by tridge
Old Dec 29, 2011, 03:57 PM
Peter Yu
peteryu's Avatar
United States, CA, San Jose
Joined Dec 2011
6 Posts
Quote:
Originally Posted by tridge View Post
Even if all of the comms logic didn't work the plane will eventually run out of fuel/battery and crash within the mission boundary. That would be annoying, but wouldn't be a safety risk.

So I think its reasonable to assume that a separate radio link dedicated to flight termination isn't needed, which is what we will assume unless one of the organisers posts a clarification.
Hi Tridge!

From a competitor point of view, I totally agree.

Unfortunately, I still stand by my interpretation of a mandatory independent radio FTS link, chiefly because of what I think is the function of the manual FTS trigger.

Like you said, even in the event of a loss of comms, the geo-fencing still ensures that the UAV is within the mission boundary and the time limit for re-establishing comms gives a pretty short window for when the UAV is flying completely by itself.

However, under normal circumstances (without comms failure), those systems are still active and yet, the organizers still request a manual FTS trigger.

I guess the easier way to put it is that when all systems are up, we are required to have a manual FTS trigger. When comms go down, we should still be able to manually trigger the FTS. Just because there are safeguards in geofencing and a comms re-estabilsihing timeout doesn't change the fact that the organizers are requesting a manual FTS trigger.

As to why a manual FTS trigger is required when the UAV has autonomous FTS triggers for most dangerous circumstances, I don't really know. I assume that there are times when a UAV is clearly out of control but still within the mission boundary, has a stable comms and GPS link, etc.

Then again, I'm only interpreting what I think the organizers want. If I were you, I'd stick to whatever plan you have already made until/unless we hear from the organizers officially.
peteryu is offline Find More Posts by peteryu
Old Dec 29, 2011, 06:20 PM
Registered User
wanabigaplane's Avatar
Joined Dec 2008
24 Posts
Peteryu - The more carefully and thoroughly we read the rules, the more difficulties and conflicts we discovered. We arrived at our final interpretation at the time of writing the D1, and we considered that we had it right when our D1 was accepted.
Jack - CanberraUAV.
wanabigaplane is offline Find More Posts by wanabigaplane
Old Dec 29, 2011, 09:04 PM
Peter Yu
peteryu's Avatar
United States, CA, San Jose
Joined Dec 2011
6 Posts
Quote:
Originally Posted by wanabigaplane View Post
Peteryu - The more carefully and thoroughly we read the rules, the more difficulties and conflicts we discovered. We arrived at our final interpretation at the time of writing the D1, and we considered that we had it right when our D1 was accepted.
Jack - CanberraUAV.
Hi Jack,

For our D1, we just said that our UAV was going to have an independent FTS. That's it. This was due to (A) not knowing exactly what the system architecture was going to look like and (B) just quoting exactly what the rules said. I don't know how in-depth the organizers were in examining the finer points of the rules, but I suspect they were neither very strict nor had they sought to answer these more in-depth questions. Either way, parsing the competition rules to micro-analyze small details seems superfluous to me. It seems unlikely that the competition organizers would hand us technical rhetoric that has some deep, underlying meaning to it. If they haven't addressed it directly, it's anybody's guess. Let's wait for the organizers to reconvene and comment on this officially.

-Peter
peteryu is offline Find More Posts by peteryu
Old Jan 04, 2012, 06:08 AM
OpenUAS
openuas's Avatar
Amsterdam
Joined May 2009
68 Posts
peteryu, indeed they are not final. as someone stated "The waypoints are not set in stone, and reflect a real life situation", e.g. if for example someone decides to remodel a county road during the competition days in an area to search, the area could be moved a little or shifted. So make sure you can still adjust waypoints etc. in days before the actual event. Highly unlikely scenario, but ... still possible.
openuas is offline Find More Posts by openuas
Closed Thread


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Uav challenge outback 2011 leogarabello Australian Electric Flight Association 1 Feb 08, 2011 08:33 PM
Contest OFFICIAL - 2010 UAV Outback Challenge Forum Mick Molloy UAV - Unmanned Aerial Vehicles 589 Feb 02, 2011 05:18 PM
Discussion UAV Outback Challenge ommadawn FPV Talk 0 Sep 25, 2010 01:37 AM
Contest UAV Challenge Outback Rescue 2009 - Official Thread Mick Molloy UAV - Unmanned Aerial Vehicles 720 Nov 19, 2009 11:42 AM