In 1972, while working for TRW Systems in Houston, Texas, as a contractor for the NASA Manned Spacecraft Center (the name of it at that time), my project leader came to me and asked if I wanted to work on a task to estimate the roll angle for the Skylab vehicle. He said that another engineer had worked on it, but this engineer who had more experience and more education than I did at that time was unable to solve the problem. I agreed to undertake the assignment.
After deriving the appropriate equations for the gravity gradient torques that were essential to determining this angle, I approached the problem of trying to solve the angle estimation technique. I was having difficulty solving the problem due to the complexity of the gravity gradient torque equations that I was using. On a Friday afternoon, while driving home to my apartment, it dawned on me that the solution was not to consider the entire equation but to just consider it as a vector. This was a prime example of “not seeing the forest for the trees.” I sat down at the dinette table when I got home and worked out the solution.
On Monday morning, I told my project leader that I had solved the problem. Within 15 min, we were at NASA, and I was explaining my solution to some skeptical people since the other engineer had been unsuccessful and had decided that the solution could not be attained. They liked my solution, and I then implemented a Kalman filter and did simulations to determine the parameters required to use for the covariance matrix in the Kalman filter to estimate the required variables in the angle estimation equation. Once I convinced NASA personnel that the Kalman filter could be designed so that I would not have to continually update the covariance matrix, I completed this very gratifying task. This is a good example of the words of Albert Einstein, “Everything should be made as simple as possible, but not simpler.”