What sensors or combination of sensors do rockets use during takeoff for their orientation?

Question

I know that after using up rockets and before reaching final orbit, satellites use Earth, Sun, Star trackers to find the absolute position depending on the application and orbit. After that, they use those sensors in combination with gyroscopes and magnetometers to stay in orbit.

But I want to know what specific types of sensors are used during the takeoff as the rocket does not change orientation that much and also has lots of vibrations from the engine.

I can guess that it has to do something with the centre of mass control but what sensor exactly is used?

Edit(More context): Some old rockets use just spin stabilization to average out any disturbances during takeoff but Modern Rockets can orient the thrust from main engine in many ways depending on the design of Engine, so it needs some input from some sensor that is sensitive to slight change and also robust to withstand vibrations.

Answer 1

As a real-world example, the Space Shuttle used a combination of Inertial Measurement Units (IMUs) and Rate Gyro Assemblies (RGAs). The IMUs sensed accelerations and the RGAs measured body rates.

The IMUs were mounted on a navigation base in the crew compartment. The RGAs were mounted in the midbody, and there were additional RGAs mounted in the Solid Rocket Boosters.

These measurements were filtered and integrated in the flight software to calculate the current vehicle position and attitude. As mentioned in comments, Kalman filtering is used.

You are correct that vibration is a major consideration in the design of these systems. Ground vibration testing at Marshall Space Center prior to the first launch resulted in software changes and relocation of some of the RGAs from the original design location.

Added for completeness: For entry, the Orbiter additionally used traditional airplane type sensors: air data probes, TACAN, MSBLS, and radar altimeters. Late in the program, GPS receivers were added and incorporated into the navigation system.

References: 1988 Shuttle press reference, personal notes.

Answer 2

The generic answer is: rockets use accelerometers, rate sensors (gyros), and/or GPS to measure their position and attitude during flight. Often these are packaged together as an "inertial reference unit" or an "inertial measurement unit".

For a more specific answer, you can look at Honeywell's SIGI, which is used by a variety of real-life rockets:

It offers triple simultaneous navigation solutions of pure Inertial, GPS-only, and blended GPS/INS as well as angular rate output for autopilot input. The inertial instruments include three single axis Honeywell Ring Laser Gyros that measure vehicle body attitude changes and three Honeywell QA3000 single axis accelerometers to measure changes in vehicle body velocities.

Old rockets relied exclusively on accelerometers and gyros. (Wikipedia has description of the V2 guidance system which was primarily a gyroscope) The term "inertial" in the device name came from the fact that the sensors would give you position and attitude based on your initial inertial (standing still) state. Both position and attitude could be derived by integrating the acceleration or the rotational rates. However, depending on the accuracy, noise, and drift of your sensor you could easily build up a cumulative error. More recently GPS has been integrated as an absolute knowledge source of position.