Inertial sensors exploit inertial forces acting on an object to determine its dynamic behavior. The basic dynamic parameters are acceleration along some axis and the angular rate. External forces acting on a body cause an acceleration and/or a change of its orientation (angular position). The rate of change of the angular position is the angular velocity (angular rate). A speedometer is not an inertial sensor because it is able to measure a constant velocity of a body that is not exposed to inertial forces. An inertial sensor is unable to do so; however, if the initial conditions of the body are known, their evolution can be calculated by integrating the dynamic equation on the basis of the measured acceleration and rate signals.

In the overwhelming majority of practical applications, such as vibrational measurements, active suspension systems, crash-detection systems, alert systems, medical activity monitoring, safety systems in cars, and computer-game interfaces, the short-term dynamic changes of the object are of interest. But there are also many applications where inertial sensors are used for determination of the positions and orientations of a body, as in robotics, general machine control, and navigation. to the necessity of integrating the corresponding dynamic equations, the accuracy requirements in these applications are usually higher because the measurement errors and instabilities of the sensors are accumulated over the integration time.