Q:  I assume we have the latitude/longitude coordinates of the last waypoint which is near the target building. How far could this be from the actual location of the building. In other words, what's the maximum error for the position of the building?
A:  For Level 1 we just give a terminal orbit center. For Level 2, we want to get you into the "ball park" but without uniquely identifying the building so your sensors must be be deployed to identify the specific structure. In the 3 scenarios, it is assumed that a general knowledge of the building is known, but not with differential accuracies or especially carrier phase accuracy. So with GPS information you should find yourself within regular unaugmented GPS range of the building. Obviously the teams will be able to see the building and visually identify its location, so to keep teams from mapping the actual location and using GPS to identify the structure without using sensors to home in on the provided visual cues (lights/symbol), we will provide a final GPS waypoint that is within a 100m circular error probable (CEP) of the actual target structure. 100m CEP defines the maximum offset, it actual waypoint could be closer. The final GPS location may be changed from run to run so teams' aerial robots don't memorize it and compute an offset to be used in subsequent runs (this way, every run will be unique and will demonstrate the robustness of your sensor systems).

Q:  If my understaning is right, once someone has demonstrated the fact that they can fly waypoints within a radius of 3km, one does not have to do it again? And one may also takeoff within 100m of the target and perform mission? Say for Level 3 or 4 qualification?
A:  No, Level 4 requires performance of all aspects of the mission:  3km in, and performance of the mission. The behaviors that pertain to entry of the structure need not involve the 3 km ingress as one progresses through the levels prior to No. 4.   Basically people can show portions of the overall mission, but having done so, should then (at Level 4) show that they can integrate the separate behaviors.
 

Q:  Will there be portals that show up but are really not open (i.e, is it guaranteed that every portal is open)?
A:  No, some portals (i.e., "windows" or "doors") may be closed. Part of the problem is determining if a window is open or closed. If it is open, it will be wide open, and if it is closed, it will be totally closed. This is a remote sensing task. Acoustics or radar will work here, but there my be optical techniques that can be envisioned as well. An open portal will be a gaping hole of at least the guaranteed minimum size as specified.
 

Q:  Once we enter the structure, all we have to do is take a picture and get out, right? Or is there some sensor provided by IARC that must be carried and dropped off or used?
A:  Right, except that you don't even need to get out. Look at the mission examples. There is really no need to retreive the vehicle that goes in. In fact, it says in the rules that it is desirable for the vehicle NOT to return due to radiation hazard/bioharzard/cost-benefit.
 

Q:  I think the 1m portal requirement and the range requirements are at two extremes and the optimal solution will not satisfy both. For range need size/endurance => bigger. For compactness => size small. I don’t think any solution is possible with the 1m portal AND 3km range requirement.
A:  This is by design... So you must ask yourself how can this problem with two extremes be solved?
 

Q:  Am I reading this right? The "openings" in the building(s) will be 1 m x 1 m. This means that all flight vehicles developed by previous teams will have to be re-designed if not totally built from scratch. I figure that a helicopter would have to have a rotor span of less than 0.75 m to get through the window safely, which basically forces us to use "micro-aircraft".
A:  Analyze the mission more closely. A larger vehicle can easily go the 3 km in about 3 minutes (leaving 12 minutes for the rest of the mission), but as you point out it isn't guaranteed to be able to enter a structure (besides, in your example, you wouldn't want an open rotor inside a building anyway). A micro air vehicle that can fit through a window won't go the 3 km (at least no one has ever done that to date)... so that could drive you to a "mother ship"-subvehicle combination. Of course that is just one solution. You are free to be innovative in the solution of the mission. Think "outside the box". Don't limit yourself to an existing vehicle that you may have, but also try to build on what you've already done.
 

Q:  Will there be a landing strip at the target building site?
A:  There will probably be an area to land as most buildings have streets associated with them, however this is not being guaranteed. Even for helicopters, there is no guarantee of level ground, though one would expect to find such, even if it were the roof of the structure. Again, no guaranteed landing area-- not because we don't want to specify such, but because even if one were known to exist right in front of the structure based on prior reconnaissance data (e.g., it is known that there is a street right out in front of the embassy in the first example misson), one would not know if a car had parked in the street (therefore blocking it) 10 minutes before your aerial robot arrived in the area.
 

 
Q:  Given that the smallest entrance to the target building will be 1 meter x 1 meter, will there be bigger entrances? Or does our vehicle have to fit through 1m x 1m (atmleast the one going inside)?
A:  The "mimimum" is 1m x 1 m. Open entrances could be bigger, but will be no smaller.
 

 
Q:   Are some of the entrances going to be at ground level so that we can drive into the building?
A:  Some could be (like a door), but you can't bet on this to be the case. You might only find windows open.
 

Q:  Will there be any stairs inside the building that we will have to go up or down?
A:  There could be, but "to keep it simple" ; ) I'm thinking that all items of interest would be on the first floor... besides, it isn’t likely that the ancients would have built mausoleaums with an upstairs loft...

 
Q:  Is the target room going to be just inside the main entrance or are we going to have to maneuver inside the building and find the right room?
A:  You will be presented with several rooms once inside (e.g., "is it behind door number one, door number two, or door number three?").
 

Q:  Is the building going to be made out of cement, metal or wood? (this is for wireless transmission, from inside the building, if the building will block signals than we may have to use a wire) ? 
A:  Masonry of some sort common to Embassies, nuclear plants, and ancient mausoleums.
 

Q:  The non autonomous parts of the mission are the takeoff and identification of the target inside the building, we only have to get a picture inside the building, not do analysis on it???  
A:  No. Although the takeoff doesn't have to be autonomous (but it could), the identification of targets inside the structure must be autonomous. All you have to do is relay a picture (still or motion) back to the starting point (3 km distant) with enough resolution or time on target (if live video) so that the judges can read a) how many hostages are in the room, or b) the readings on some major control panel dials, or c) inscriptions on the tapestry. Image analysis will be done by the judges. If the picture is too brief or fuzzy to read, then the required information has not successfully been gathered.
 

Q:  In reference to the sub-vehicles: A multi-mode vehicle must be launched from the main air vehicle but does the sub-vehicle have to "fly" in any way? Our tenative plan is to deliver a non-flying ground vehicle through the opening by some means. In other words, it will not be able to generate any kind of lift, but it will travel along an arc such that it can get in the opening.
A:  OK. Beware of obstacle inside the structure that are common to the scenarios (people, chairs, sarcophagus)
 

Q:  Will desired target that needs to be photographed be illuminated such that a camera can take the picture without additional light?
A:  Low light levels common to the scenarios (bulb hanging from ceiling, or a fluorescent tube, or a "trouble light")
 

Q:  In any of the scenarios, will there be some obstruction that will prevent a ground-based vehicle from obtaining an acceptable picture? For example, for a horizontal nuclear control panel, a ground camera will not be able to get a picture of the panel unless the gauges are on the wall.
A:  All targets would be visible to a person walking into the room. Unusual viewing angles (e.g., from one inch off the floor) could experience occlusion. For example if you had a camera two inches from the base of the sarcophagus, you might night not be able to look up and see the tapestry hanging above the sarcophagus.
 

Q:  Is any post processing of the reconassance pictures needed? Will a series of stills or a video that will include the target at SOME point be acceptable, or does a computer have to pick out the target(s) from the video and/or stills to report to the judges (rescuers, technicians, translators)?
A:  Only "human processing" is necessary. The judges will have to be able to see and understand your reconnaissance information. A series of 6 stills showing 4 walls and the floor and the ceiling of a cubical room would obviously contain the desired information provided that sufficient resolution is provided to read the principal gauges, read symbols on the tapestry, or to count people. In the case of the hostage situation, multiple pictures might be required to see everyone in the room because they might not be lined up against the wall. Similarly, nothing has been said to gurantee that the tapestry is not centered on the room at a 45 degree angle to the walls (for example). Also, which side of tapestry contains the symbols? The control room might have panels on two walls. Which one has the required information?

 
Q:  I know that the rules prohibit pyrotechnic means of vehicle termination, but is there a problem with pyrotechnic means for other purposes? Can we consider rocket power as a means of propulsion?
A:  There's a good chance that the rocket system might not pass the safety inspection. How can you terminate a rocket were it to malfunction and head for the spectators?
 

Last updated on February 5, 2005