Quote:
Originally Posted by Rockin Robbins
[/INDENT]The ideal way to fix the problem would be to have a scalable error envelope for each observation method with your measured number modified by a random number within the error envelope. That way you could have radar accurate to within 30 yards, visual within a certain percentage, stadimeter the same, sonar within a certain percentage with error envelopes tailored for each data acquisition method.
Then there would be the random errors in torpedo performance and user input that you've dealt with.
Too bad we can't do anything about the first part, data acquisition and measurement errors.
|
It would not be a difficult calculation to have random errors different for each mode of detection. All you need is the variance of each method and a random number generator.
What I would like to see over and above what you said is the error reduce as the target is observed for longer. So if you shadow it from a distance for a few hours even with visuals only you can get a much better idea of the course and speed.