Dynamic gravity vector estimation for memory constrained devices

A system, comprising:

• one or more sensors, which, in operation, generate rotational data and acceleration data;
• and
• processing circuitry coupled to the one or more sensors, wherein the processing circuitry, in operation:
• estimates an angular rate of change based on the rotational data;
• determines a rotational versor based on the rotational data;
• estimates a gravity vector based on the angular rate of change and the rotational versor;
• estimates a linear acceleration based on the acceleration data;
• determines an acceleration versor based on the acceleration data;
• determines a correction factor based on the linear acceleration;
• estimates an error in the estimated gravity vector based on the acceleration versor;
• and
• generates a dynamic gravity vector based on the estimated gravity vector, the correction factor and the estimated error in the estimated gravity vector.

A method, comprising:

• estimating an angular rate of change based on rotational data;
• determining a rotational versor based on the rotational data;
• estimating a gravity vector based on the angular rate of change and the rotational versor;
• estimating a linear acceleration based on acceleration data;
• determining an acceleration versor based on the acceleration data;
• determining a correction factor based on the linear acceleration;
• estimating an error in the estimated gravity vector based on the acceleration versor;
• generating a dynamic gravity vector based on the estimated gravity vector, the correction factor and the estimated error in the estimated gravity vector;
• and
• using the dynamic gravity vector to perform an operation of a software application.