A gyroscope measures angular velocity — how fast the device is rotating around each of its axes — in degrees per second or radians per second. A 3-axis gyroscope reports angular velocity around three perpendicular axes. Typical consumer gyroscopes cap out at something like .
The defining problem of gyroscopes is drift. If we integrate angular velocity over time to recover orientation, small errors in each instantaneous reading accumulate. After a few minutes, the integrated orientation is meaningfully wrong — degrees off, then tens of degrees. This is why a gyroscope alone is not enough to track orientation over any extended period.
The standard remedy is Sensor fusion: combine the gyroscope with an Accelerometer (which senses the direction of gravity, fixing the pitch and roll components of orientation) and a Magnetometer (which senses magnetic north, fixing the yaw component). The fused estimate doesn’t drift, because the absolute references from gravity and the magnetic field continually correct the gyroscope’s accumulating error.
MEMS gyroscopes work on the Coriolis principle: a small mass is driven into resonant vibration along one axis, and when the device rotates the mass experiences a Coriolis force perpendicular to both the vibration direction and the rotation axis. That force shows up as a secondary vibration in the perpendicular direction, read out capacitively. There is no spinning rotor — this is not the same as the classroom-demo gyroscope with a heavy disk on gimbals, and the analogy breaks if pushed. The most common MEMS architecture pairs two counter-vibrating proof masses (a tuning-fork gyroscope) to reject linear acceleration as common-mode, but ring-resonator and vibrating-wheel designs exist too.
The gyroscope is one of the three sensors inside an IMU.