Actuator monitoring system using inertial sensors

1 . A system, comprising:

a processor; and

a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which, when executed, cause the processor to:

receive vehicle state information and at least one actuator command from a vehicle, wherein the vehicle is symmetric around a longitudinal axis of symmetry and the vehicle includes a first rotor and a second rotor that are symmetric with respect to each other around the longitudinal axis of symmetry;

determine a plurality of models associated with vehicle failure;

determine a subset of the plurality of models associated with vehicle failure to identify a malfunctioning rotor, including by:

including a first model, associated with the first rotor failing, in the subset of the plurality of models;

excluding a second model, associated with the second rotor failing, from the subset of the plurality of models; and

sorting at least some of the plurality of models based on similarity to at least one of an attitude and an attitude rate;

determine an observed flight characteristic and an observed flight characteristic rate of the vehicle based at least in part on sensor data from an inertial measurement unit on the vehicle;

determine an expected flight characteristic and an expected flight characteristic rate of the vehicle based at least in part on the determined plurality of models;

identify, from a plurality of rotors associated with the vehicle, the malfunctioning rotor based at least in part on the observed flight characteristic, the observed flight characteristic rate, the expected flight characteristic, and the expected flight characteristic rate; and

in response to identifying the malfunctioning rotor, perform a responsive action.

2 . The system of claim 1 , wherein:

the observed flight characteristic includes an observed attitude;

the observed flight characteristic rate includes an observed attitude rate;

the expected flight characteristic includes an expected attitude; and

the expected flight characteristic rate includes an expected attitude rate.

3 . The system of claim 1 , wherein:

the vehicle state information includes historical vehicle state information; and

the at least one actuator command includes at least one historical actuator command.

4 . The system of claim 3 , wherein the vehicle state information includes at least one of: an attitude, an attitude rate, a position, a velocity, a wind velocity, or a geometry of the vehicle.

5 . The system of claim 3 , wherein the vehicle state information identifies the vehicle failure associated with a set of data.

6 . The system of claim 5 , wherein determining the subset of the plurality of models includes at least one of grouping or sorting data.

7 . The system of claim 1 , wherein at least one model of the plurality of models includes an expected attitude associated with the vehicle failure.

8 . The system of claim 1 , wherein at least one model of the plurality of models includes an expected attitude rate associated with the vehicle failure.

9 . The system of claim 1 , wherein:

the vehicle includes at least one tilt wing; and

determining at least one model of the plurality of models is based at least in part on at least one of a direction or a mode of the at least one tilt wing.

10 . The system of claim 9 , wherein determining the at least one model includes selecting a model to use during flight based at least in part on at least one of a direction or a mode of the at least one tilt wing.

11 . The system of claim 1 , wherein the processor is further configured to update at least one model of the plurality of models in real time as the vehicle is flying.

12 . The system of claim 11 , wherein updating the at least one model in real time as the vehicle is flying includes collecting flight data during flight to update the at least one model which the vehicle is airborne.

13 . The system recited in claim 1 , wherein the responsive action includes updating a geometry matrix, which is used to generate a plurality of actuator commands for the plurality of rotors, including by updating a previous geometry matrix with a precomputed geometry matrix.

14 . The system recited in claim 13 , wherein:

the malfunctioning rotor is associated with a first region of the vehicle; and

updating the geometry matrix includes increasing an authority of a second region of the vehicle in response to detection of the malfunctioning rotor in the first region.

15 . The system of claim 1 , further comprising modeling the vehicle failure independently from at least another one of the plurality of models including by applying a respective function for the vehicle failure to determine a respective output value based on a same set of inputs provided to all functions.

16 . A method, comprising:

receiving vehicle state information and at least one actuator command from a vehicle, wherein the vehicle is symmetric around a longitudinal axis of symmetry and the vehicle includes a first rotor and a second rotor that are symmetric with respect to each other around the longitudinal axis of symmetry;

determining a plurality of models associated with vehicle failure;

determining a subset of the plurality of models associated with vehicle failure to identify a malfunctioning rotor, including by:

including a first model, associated with the first rotor failing, in the subset of the plurality of models;

excluding a second model, associated with the second rotor failing, from the subset of the plurality of models; and

sorting at least some of the plurality of models based on similarity to at least one of an attitude and an attitude rate;

determining an observed flight characteristic and an observed flight characteristic rate of the vehicle based at least in part on sensor data from an inertial measurement unit on the vehicle;

determining an expected flight characteristic and an expected flight characteristic rate of the vehicle based at least in part on the determined plurality of models;

identifying, from a plurality of rotors associated with the vehicle, the malfunctioning rotor based at least in part on the observed flight characteristic, the observed flight characteristic rate, the expected flight characteristic, and the expected flight characteristic rate; and

in response to identifying the malfunctioning rotor, performing a responsive action.

17 . The method of claim 16 , wherein:

the vehicle includes at least one tilt wing; and

determining at least one model of the plurality of models is based at least in part on at least one of a direction or a mode of the at least one tilt wing.

18 . The method of claim 17 , wherein determining at least one model of the plurality of models includes selecting a model to use during flight based at least in part on at least one of a direction or a mode of the at least one tilt wing.

19 . The method of claim 16 , further comprising updating at least one model of the plurality of models in real time as the vehicle is flying.

20 . A computer program product embodied in a non-transitory computer readable storage medium and comprising computer instructions which when executed by a processor cause the processor to be configured for:

receiving vehicle state information and at least one actuator command from a vehicle, wherein the vehicle is symmetric around a longitudinal axis of symmetry and the vehicle includes a first rotor and a second rotor that are symmetric with respect to each other around the longitudinal axis of symmetry;

determining a plurality of models associated with vehicle failure;

determining a subset of the plurality of models associated with vehicle failure to identify a malfunctioning rotor, including by:

including a first model, associated with the first rotor failing, in the subset of the plurality of models;

excluding a second model, associated with the second rotor failing, from the subset of the plurality of models; and

sorting at least some of the plurality of models based on similarity to at least one of an attitude and an attitude rate;

determining an observed flight characteristic and an observed flight characteristic rate of the vehicle based at least in part on sensor data from an inertial measurement unit on the vehicle;

determining an expected flight characteristic and an expected flight characteristic rate of the vehicle based at least in part on the determined plurality of models;

identifying, from a plurality of rotors associated with the vehicle, the malfunctioning rotor based at least in part on the observed flight characteristic, the observed flight characteristic rate, the expected flight characteristic, and the expected flight characteristic rate; and

in response to identifying the malfunctioning rotor, performing a responsive action.