Real time effective mass and moment of inertia measurement

1. A computer-implemented system for controlling an autonomous or semi-autonomous device, the computer-implemented system comprising:

a processor configured to:

receive a command signal representative of a desired acceleration;

instruct the autonomous or semi-autonomous device to accelerate according to the command signal;

receive a measurement signal representative of an actual acceleration of the autonomous or semi-autonomous device;

determine an initial control response based on the actual acceleration via a prediction model;

apply a mathematical transform to the command signal and the measurement signal;

determine a mathematical model of the autonomous or semi-autonomous device based on the transformed command signal and transformed measurement signal so as to output a plurality of parameters representing properties inherent of the autonomous or semi-autonomous device;

smooth the parameters of the mathematical model;

determine a transfer function of the mathematical model;

invert the transfer function;

determine updated control responses based on the mathematical model and the inverted transfer function; and

control the autonomous or semi-autonomous device according to the updated control responses, the mathematical model being updatable based on subsequent command signals and subsequent measurement signals such that the parameters and the transfer function of the mathematical model are dynamic.

2. The computer-implemented system of claim 1 , wherein the mathematical transform is selected from the group consisting of a Fourier transform, a Laplace transform, a wavelet transform, and an empirical mode decomposition.

3. The computer-implemented system of claim 1 , wherein the processor is further configured to incorporate noise into the initial control response so as to flatten an overall noise power spectrum of the autonomous or semi-autonomous vehicle.

4. The computer-implemented system of claim 1 , wherein the processor is configured to determine the mathematical model according to least-square fitting or Newton's method.

5. The computer-implemented system of claim 1 , wherein the processor is further configured to determine a mass and drag of the autonomous or semi-autonomous device based on the actual acceleration and an applied force resulting in the actual acceleration.

6. The computer-implemented system of claim 1 , wherein the actual acceleration is an angular acceleration, the processor being further configured to determine a moment of inertia and drag of the autonomous or semi-autonomous device based on the actual acceleration and an applied torque resulting in the actual acceleration.

7. The computer-implemented system of claim 1 , wherein the mathematical model is expressed in tensor or geometric algebraic equations.

8. The computer-implemented system of claim 1 , wherein the control response is determined according to a proportional integral derivative (PID).

9. The computer-implemented system of claim 1 , wherein the processor is further configured to pass the inverted transfer function through a Kalman filter.

10. The computer-implemented system of claim 9 , wherein the filter has a time delay or phase ambiguity.

11. A computer-implemented method of controlling an autonomous or semi-autonomous device, the method comprising the steps of:

receiving a command signal representative of a desired acceleration;

instructing the autonomous or semi-autonomous device to accelerate according to the command signal;

receiving a measurement signal representative of an actual acceleration of the autonomous or semi-autonomous device;

determining an initial control response based on the actual acceleration via a prediction model;

applying a mathematical transform to the command signal and the measurement signal;

determining a mathematical model of the autonomous or semi-autonomous device based on the transformed command signal and transformed measurement signal so as to output a plurality of parameters representing properties inherent of the autonomous or semi-autonomous device;

smoothing the parameters of the mathematical model;

determining a transfer function of the mathematical model;

inverting the transfer function;

determining updated control responses based on the mathematical model and the inverted transfer function; and

controlling the autonomous or semi-autonomous device according to the updated control responses, the mathematical model being updatable based on subsequent command signals and subsequent measurement signals such that the parameters and the transfer function of the mathematical model are dynamic.

12. The computer-implemented method of claim 11 , wherein the mathematical transform is selected from the group consisting of a Fourier transform, a Laplace transform, a wavelet transform, and an empirical mode decomposition.

13. The computer-implemented method of claim 11 , further comprising the step of incorporating noise into the initial control response so as to flatten an overall noise power spectrum of the autonomous or semi-autonomous vehicle.

14. The computer-implemented method of claim 11 , wherein the mathematical model is determined according to least-square fitting or Newton's method.

15. The computer-implemented method of claim 11 , further comprising the step of determining a mass and drag of the autonomous or semi-autonomous device based on the actual acceleration and an applied force resulting in the actual acceleration.

16. The computer-implemented method of claim 11 , wherein the actual acceleration is an angular acceleration, the method further comprising the steps of determining a moment of inertia and drag of the autonomous or semi-autonomous device based on the actual acceleration and an applied torque resulting in the actual acceleration.

17. The computer-implemented method of claim 11 , wherein the mathematical model is expressed in tensor or geometric algebraic equations.

18. The computer-implemented method of claim 11 , wherein the control response is determined according to a proportional integral derivative (PID).

19. The computer-implemented method of claim 11 , further comprising the step of passing the inverted transfer function through a Kalman filter.

20. A computer-implemented system for controlling an autonomous or semi-autonomous drone, the computer-implemented system comprising:

a processor configured to:

receive a command signal;

instruct the autonomous or semi-autonomous drone to accelerate according to the command signal;

receive a measurement signal representative of an actual acceleration of the autonomous or semi-autonomous drone;

determine an initial control response according to a proportional integral derivative (PID) based on the actual acceleration via a prediction model;

apply a mathematical transform to the command signal and the measurement signal;

determine a mathematical model of the autonomous or semi-autonomous drone according to least-square fitting or Newton's method based on the transformed command signal and transformed measurement signal, the mathematical model being expressed in tensor or geometric algebraic equations so as to output a plurality of parameters representing properties inherent of the autonomous or semi-autonomous device;

smooth the parameters of the mathematical model;

determine a transfer function of the mathematical model;

invert the transfer function;

pass the inverted transfer function through a Kalman filter;

determine a mass, moment of inertia, and phase responses based on the mathematical model and inverted transfer function;

determine updated control responses based on the mathematical model, inverted transfer function, mass, moment of inertia, and phase responses; and

control the autonomous or semi-autonomous drone according to the updated control responses, the mathematical model being updatable based on subsequent command signals and subsequent measurement signals such that the parameters and the transfer function of the mathematical model are dynamic.