| layout | date | to | from | serial | subject |
|---|---|---|---|---|---|
tindallgram |
Jul 8 1966 |
FM4/Chief, Mathematical Physics Branch |
FM/Deputy Chief, Mission Planning and Analysis Division |
66-FM1-85 |
Determination of relative CSM orbit. |
Jim, this is just a reminder of conversations with you and Emil about a job I'd like your people to do. In thinking more about this orbit determination task wherein the LEM determines the CSM orbit while sitting on the lunar surface, I wonder if perhaps MIT has lost sight of our primary objectives, thus leading them to the conclusion that they should use only range and range rate data.
The only purpose of this orbit determination, as you recall, is to determine the orientation of the CSM's orbital plane for use in targeting the LEM ascent guidance and to select a lift-off time which must be within a few seconds of optimum. It is not to obtain some sort of a precision total state vector of the CSM. Based on these ground rules, I just can't believe that the angular radar data, even with relatively large biases, cannot be useful if properly weighted, and I would think that it would provide a great strength or reliability to the process, which I would consider mandatory. That is, we are much more interested in assuring ourselves of getting a pretty good answer all the time rather than an excellent answer some of the time.
The questions to be answered are: should we or shouldn't we use the angular data, even with large biases, and how do we take maximum advantage of our external knowledge, such as the CSM's own orbit determination (though it's not with respect to the LEM). Don't forget, this data processing must be entirely automatic. The crew will never have time to learn how to operate all those statistical filters, etc., whatever they are.
Emil said he would start something here, but I wanted to make sure you were aware of it and concurred and, in particular, would give it some of your own personal attention. Perhaps these remarks belong at the top, but I'd just like to reiterate that as much as I distrust it, I'm afraid our best source of relative orbit determination for this particular mission phase may be by the LEM radar data. I doubt if the CSM will ever see the LEM on the surface, at least we'd better not count on it, and the MSFN tracking certainly can't figure out where the LEM is. Our other source is the G&N state vector T/M at LEM touchdown, which is probably the best, if the antenna are pointed at us.