Anomaly detection systems and methods

1. A method for detecting an anomaly of a sensor of a vehicle, comprising:

storing information of a plurality of sensor correlation groups based on vehicle dynamics, wherein each of the plurality of sensor correlation groups includes at least two sensors that are correlated based on the vehicle dynamics;

sensing, by the at least two sensors of all of the plurality of sensor correlation groups, a subset of signals;

processing the subset of signals by computing a mean absolute error (MAE) based on the plurality of sensor correlation groups to determine when an anomaly exists;

processing the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous; and

generating notification data to a vehicle system, the notification data indicating the sensor that is anomalous.

2. The method of claim 1 , wherein the plurality of sensor correlation groups includes a position group, a speed group, an acceleration group, a heading group, and a yaw rate group.

3. The method of claim 2 , wherein the subset of signals includes a position signal, a speed signal, an acceleration signal, a heading signal, and a yaw rate signal.

4. The method of claim 1 , wherein the processing the subset of signals to determine when the anomaly exists is based on a defined batch size.

5. The method of claim 1 , wherein the processing the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous is based on a computed mean of a predicted condition vector associated with a sensor correlation group of the plurality of correlation groups.

6. The method of claim 1 , further comprising training a plurality of thresholds based on anomalous signal data, and wherein the processing the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous is based on the plurality of trained thresholds.

7. The method of claim 1 , further comprising training a plurality of thresholds based on anomalous signal data, and wherein the processing the subset of signals to determine when a sensor of the plurality of sensor correlation groups is anomalous is based on the plurality of trained thresholds.

8. A computer implemented system, the system comprising:

a plurality of sensors configured to generate a subset of signals: and

an anomaly detection module that comprises one or more processors configured by programming instructions encoded in non-transitory computer readable media, the anomaly detection module configured to:

store information of a plurality of sensor correlation groups based on vehicle dynamics, wherein each sensor correlation group of the plurality of sensor correlation groups includes at least two sensors of the plurality of sensors that are correlated based on the vehicle dynamics;

process the subset of signals by computing a mean absolute error (MAE) based on the plurality of correlation sensor groups to determine when an anomaly exists;

process the subset of signals to determine which sensor of the sensor correlation groups is anomalous; and

generate notification data to a vehicle system, the notification data indicating the sensor that is anomalous.

9. The computer implemented system of claim 8 , wherein the plurality of sensor correlation groups includes a position group, a speed group, an acceleration group, a heading group, and a yaw rate group.

10. The computer implemented system of claim 8 , wherein the subset of signals includes a position signal, a speed signal, an acceleration signal, a heading signal, and a yaw rate signal.

11. The computer implemented system of claim 8 , wherein the anomaly detection module is configured to process the subset of signals to determine when the anomaly exists based on a defined batch size.

12. The computer implemented system of claim 8 , wherein the anomaly detection module is configured to process the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous based on a depth first search method and the plurality of sensor correlation groups.

13. The computer implemented system of claim 8 , wherein the anomaly detection module is configured to process the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous based on a computed mean of a predicted condition vector associated with a sensor correlation group of the plurality of correlation groups.

14. The computer implemented system of claim 8 , wherein the anomaly detection module is configured to train a plurality of thresholds based on anomalous signal data, and process the subset of signals to determine which sensor of the plurality of sensor correlation groups is anomalous based on the plurality of trained thresholds.

15. The computer implemented system of claim 8 , wherein the anomaly detection module is configured to train a plurality of thresholds based on anomalous signal data, and process the subset of signals to determine when a sensor of the plurality of sensor correlation groups is anomalous based on the plurality of trained thresholds.

16. A method for detecting an anomaly of a sensor of a vehicle, comprising:

storing information of a plurality of sensor correlation groups based on vehicle dynamics, wherein each of the plurality of sensor correlation groups includes at least two sensors that are correlated based on the vehicle dynamics;

sensing, by the at least two sensors of all of the plurality of sensor correlation groups, a subset of signals;

processing the subset of signals to determine when an anomaly exists based on the plurality of sensor correlation groups;

determining which sensor of the plurality of sensor correlation groups is anomalous based on a depth first search method and the plurality of sensor correlation groups; and

generating notification data to a vehicle system, the notification data indicating the sensor that is anomalous.