Sensors

[robject]

I'm thinking about sensors -- orbital sensors, for instance -- and what kind of performance characteristics they range from, for the sake of Traveller. High tech, low tech, and appropriate tech, abstractly defined.

I'd like your advice and opinions.


The Example

When a friend was playing Traveller, he stumped the referee with sensors, which has nagged at me off and on for some time. He wanted to scan a planet from orbit for a smallish building. He proposed moving the ship in orbit in such a way that he could 'paint' the surface to a reasonable resolution and have the ship's computer buzz him when it found something that matched certain criteria.


The Solution

I invoke the Burrito Principle as a basis for approximation (80% of the meat is in 20% of the burrito). A first approximation is all I need.

The Mars Global Surveyor basically posits that, at TL8 with scientific sensors, one satellite can map the surface of a world to a resolution of 1.4m in under two years -- call it 1m resolution in 3 years. Thus a relatively clean table is:

</font><blockquote>code:</font><hr /><pre style="font-size:x-small; font-family: monospace;">Scientific sensor abilities.
(Civilian are 2 TLs lower)

TL Area Mapped (million sq km)
-- ---------------------------
7 1 per 6 months
8 1 per 45 days
9 1 per 15 days
10 1 per 5 days
11 20 per month
12 2 per day
13 6 per day
14 20 per day
15 60 per day

Data Points: each million sq km at 1m yields
1,000,000,000,000,000,000 data points.</pre>[/QUOTE] Caveat

This heuristic is for data collection only, and assumes analysis can keep up with the flow of data coming in. For some surveys, a large computer may be required for analysis. To this I defer to T4's computer rules in the Central Supply Catalog.


The Application

Assuming standard sensors in Traveller can do this, how long does it take to scan a planet from orbit?

Assume: a resolution of 1m is required
Assume: surface area is 20 million square km (world diameter = 5000km)
Assume: we have 10 survey satellites to collect data (divides scan time by 10)

At TL12 it takes 10 days to scan 20M sq km with one ship; with the satellites, it takes only one day.

Amount of data gathered = 1 data point per square meter, or 20 000 000 000 000 000 000 data points.

Does that sound about right for TL12, non-military, non-scientific, ordinary civilian sensors?

Military and scientific sensors are up to 2 TLs better than standard.[/CODE]

 

[Anthony]

Originally posted by robject:
The Heuristic

The human eye can scan about one square meter at a distance of about one meter in about one second at a resolution of around 1 square cm.

Hm..interesting claim. The actual resolution of the human eye (standard 20/20 vision) is the ability to distinguish objects separated by 1 minute of arc, which is about 0.35mm. This implies an actual 'pixel size' of about 0.2mm. This is the resolution of central vision, and it's not possible to scan a steradian in a second at that resolution, but there's enough detectors to allow a resolution of 1-2mm over the entire area.

A standard 4 megapixel digital camera will have a resolution of 0.5mm; it lacks the center focus effects of the human eye.

 

[Andrew Boulton]

You need Bruce's Definitive Sensor Rules

http://home1.gte.net/res04u7k/Traveller/hrules/sensor.rules.html

 

[robject]

Originally posted by Andrew Boulton:
You need Bruce's Definitive Sensor Rules

http://home1.gte.net/res04u7k/Traveller/hrules/sensor.rules.html

They are fine, but do not specifically solve the problem of "how long does it take to find a particular structure which is somewhere on the surface of a world", although it looks like that can be inferred. Perhaps I'm pursuing a rabbit trail rather than focusing on more important things, eh?

 

[TheDS]

Well, your question is at least semi-important, and not really addressed outside a couple sources.

I recall that World Tamer's Handbook (TNE) does a little in this regard, something like Xd6 days to get a full map of a world, where X is the world's size digit. I don't know how good the map you get is supposed to be, but I would have to suppose that it would be equivalent to a road atlas; showing major areas of settlement, roads, rails, stuff like that. Not to the level of a street map, mind you, just like a state map.

I would have to suppose that GT: First In does a similar thing, since it covers the subject of exploring new planets. GT:FI has a much more modern and much better starsystem and world generation design system, and is a bigger book, so I am pretty sure that it will do a better job of answering your questions, and it's probably still in print.

 

[clementk]

Originally posted by TheDS:
GT:FI [...is] probably still in print.

No. It's one of the way out of print Gurps Traveller books and one * not * scheduled to be reprinted this April. I was able to get a copy at last year's Origins though from the Atlas Games / SJG / etc. booth. When I get the chance I'll take a look at how it handles (or doesn't) this.

Casey

 

[Richard Saunders]

Hello.
Sorry am i missing something.
You want to find a building that you dont know what it looks like from orbit.
I assume its an inhabited planet, if not its the only building on the planet, look for a heat source or energy signature.
If there are towns or even worse cities you have bugger all change of spotting a specific building even if you know what it looks like, imagine trying to pick a friends house from amoung thousands of other houses from orbit.
Mapping planets from orbit would be done one layer at a time.
1 - planetary map (scale in miles).
2 - arbitrary country scale (100meters).
3 - town scale (1 meter).
Each scale would be done only on the part of the planet you are interested in (why would you town scale the polar ice fields).
Also it would depend on the type of survey being done,weather,mineral,agricultural,biodiversity or any other type you like to name, And most would take feet on the surface to do.
The feet on the planet is what GT/First in is about.
Bye.

 

[Andrew Boulton]

Mapping the UK at .25m resolution took about 3 years.

http://www.millennium-map.com/

"I can see my house from here!"