System and method for recalibration of an uncalibrated sensor

1. A system comprising:

a sensor within an autonomous or a semi-autonomous vehicle, the sensor comprising a Lidar, a radar, or a camera;

and one or more processors configured to:

determine a frequency of a validation of the sensor, wherein the validation determines whether the sensor is uncalibrated;

perform the validation at the frequency, wherein the validation comprises a comparison between historical data captured by the sensor and current data captured by the sensor, or a comparison between historical data captured by an other sensor of a same modality as the sensor and the current data captured by the sensor;

in response to performing the validation, and determining that the sensor is uncalibrated, determining a type or a mode of an error handling procedure based on a frequency of a recalibration error and based on a historical rate of success associated with the type or the mode of the error handling procedure; and

perform the type or the mode of the error handling procedure.

2. The system of claim 1 , wherein the determination of the frequency of the validation is based on at least two of:

a history of the sensor;

a density of moving objects previously detected;

and a weather condition.

3. The system of claim 1 , wherein the determination of the frequency of the validation is based on an amount of computation load consumed by performing the validation.

4. The system of claim 1 , wherein the determination of the type or mode of the error handling procedure comprises determining whether to perform a recalibration of the sensor based on an availability of a redundant sensor and on traffic and environmental conditions.

5. The system of claim 1 , wherein the determination of the type or mode of the error handling procedure comprises determining whether to perform a recalibration of the sensor based on a probability of the sensor being uncalibrated.

6. The system of claim 5 , wherein the determination of whether to perform the recalibration comprises eliminating erroneous data if the probability is below a threshold.

7. The system of claim 1 , wherein the determination of the type or mode of the error handling procedure comprises determining a type of recalibration based on an identification of a type of a sensor error being of an offset, a sensitivity, or a linearity.

8. The system of claim 1 , wherein the validation comprises determining a parameter of the sensor and comparing the parameter to a second threshold, wherein the parameter comprises a skew, a depth of field, an angle of view, a beam angle, or an aspect ratio.

9. The system of claim 1 , wherein the performing of the validation comprises determining whether a particular feature captured by the sensor, at successive frames, is increasing or decreasing at a consistent rate.

10. The system of claim 1 , wherein the processors are further configured to adjust the frequency of the validation based on an age of the sensor.

11. The system of claim 1 , wherein the error handling procedure comprises a sequential series including a first comparison between a sensor parameter of the sensor and a known parameter, a second comparison between the historical data captured by the sensor and the current data captured by the sensor, and a third comparison between the historical data captured by the other sensor of the same modality as the sensor and the current data captured by the sensor.

12. The system of claim 1 , wherein the performing the type or the mode of the error handling procedure is based on a landmark or a type of a landmark.

13. The system of claim 1 , wherein the performing the type or the mode of the error handling procedure is based on a historical rate of success associated with the type or the mode of the error handling procedure.

14. An error handling method for a sensor of a system, the sensor being within an autonomous or a semi-autonomous vehicle, the sensor comprising a Lidar, a radar, or a camera, the error handling method comprising:

determining a frequency of a validation of the sensor, wherein the validation determines whether the sensor is uncalibrated;

performing the validation at the frequency, wherein the validation comprises a comparison between historical data captured by the sensor and current data captured by the sensor, or a comparison between historical data captured by an other sensor of a same modality as the sensor and the current data captured by the sensor;

in response to performing the validation, and determining that the sensor is uncalibrated, determining a type or a mode of an error handling procedure based on a frequency of a recalibration error and based on a historical rate of success associated with the type or the mode of the error handling procedure; and

performing the type or the mode of the error handling procedure.

15. The method of claim 14 , wherein the determination of the type or mode of the error handling procedure comprises determining whether to perform a recalibration of the sensor based on an availability of a redundant sensor and on traffic and environmental conditions.

16. The method of claim 15 , wherein the determination of the type or mode of the error handling procedure comprises determining whether to perform a recalibration of the sensor based on a probability of the sensor being uncalibrated.

17. The method of claim 16 , wherein the determination of whether to perform the recalibration comprises eliminating erroneous data if the probability is below a threshold.

18. The method of claim 14 , wherein the determination of the type or mode of the error handling procedure comprises determining a type of recalibration based on an identification of a type of a sensor error being of an offset, a sensitivity, or a linearity.

19. The method of claim 14 , wherein the performing of the validation comprises determining whether a particular feature captured by the sensor, at successive frames, is increasing or decreasing at a consistent rate.