Steering angle calibration
Among other things, techniques are described for steering angle calibration. An autonomous vehicle receives a steering angle measurement and a yaw rate measurement, and estimates a steering angle offset using the steering angle measurement, the yaw rate measurement, and a wheel base of the autonomous vehicle. An estimated yaw rate is determined based on a yaw rate model, the steering angle measurement and the estimated steering angle offset. The yaw rate measurement and the estimated yaw rate are compared and an action is initiated on the autonomous vehicle in response to the comparing.
1. A method comprising:
receiving, using a processing circuit of an autonomous vehicle (AV), a steering angle measurement;
receiving, using the processing circuit, a yaw rate measurement, wherein the steering angle measurement and the yaw rate measurement are taken when a set of operating conditions is met, wherein the set of operating conditions includes a forward speed of the AV less than a threshold value and presence of one or more road surface conditions,
estimating, using the processing circuit, a steering angle offset based on the steering angle measurement, the yaw rate measurement, and a wheel base of the AV;
determining, using the processing circuit, an estimated yaw rate equal to
V
x
L
tan
(
δ
m
+
δ
ˆ
o
)
where V x is a forward speed of the AV, L is the wheel base of the AV, δ m is the measured steering angle, and {circumflex over (δ)} o is the estimated steering angle offset; and
operating, using a control circuit, the AV based on the estimated yaw rate.
2. The method of claim 1 , further comprising:
comparing, using the processing circuit, the yaw rate measurement, and the estimated yaw rate; and
wherein operating the AV based on the estimated yaw rate comprises initiating, using the processing circuit, an action on the AV based on a result of the comparing.
3. The method of claim 2 , wherein the action comprises:
calibrating, by the processing circuit, the yaw rate measurement using the estimated yaw rate.
4. The method of claim 1 , wherein receiving the steering angle offset comprises:
receiving, using the processing circuit, a left-turn component of the steering angle measurement when the AV is making a left turn;
receiving, using the processing circuit, a right-turn component of the steering angle measurement when the AV is making a right turn; and
determining, using the processing circuit, the steering angle measurement based on the left-turn component and the right-turn component.
5. The method of claim 4 , wherein the determined steering angle measurement is an average of the left-turn component and the right-turn component.
6. The method of claim 1 , wherein the estimated steering angle offset is determined using a recursive least squares estimator formulation that includes the steering angle measurement, the yaw rate measurement, and the wheel base of the AV.
7. The method of claim 1 , wherein the estimated steering angle offset is determined using a Kalman filter formulation that includes the steering angle measurement, the yaw rate measurement, and the wheel base of the AV.
8. The method of claim 1 , wherein the set of operating conditions further includes the yaw rate measurement less than 0.5 rad/s.
9. The method of claim 1 , further comprising:
determining, using data from one or more sensors of the AV, whether a wheel of the AV is in static contact with a road surface; and
in accordance with a determination that the wheel of the AV is not in static contact with the road surface,
excluding, using the processing circuit, the steering angle measurement from the steering angle offset estimating or augmenting the steering angle measurement.
10. The method of claim 1 , further comprising:
estimating a plurality of steering angle offsets over a period of time or a driving distance; and
filtering the plurality of steering angle offsets to remove outlier estimated steering angle offsets.
11. The method of claim 1 , further comprising:
estimating a plurality of steering angle offsets over a period of time or a driving distance; and
determining a variable steering angle offset based that is continuously adjusted based on the plurality of steering angle offsets.
12. The method of claim 1 , wherein operating the AV based on the estimated yaw rate comprises:
presenting an alert to an output device of the AV or a personal device of a passenger of the AV.
13. The method of claim 1 , wherein operating the AV based on the estimated yaw rate comprises:
sending, using a wireless transmitter of the AV, at least one of steering angle measurements or yaw rate measurements to a network-based computing platform;
receiving, using a wireless receiver of the AV, a maintenance alert from the network-based computing platform; and
presenting, using an output device of the AV, the maintenance alert.
14. The method of claim 1 , wherein operating the AV based on the estimated yaw rate comprises:
sending, using a wireless transmitter of the AV, the one or more road surface conditions to a network-based computing platform; and
receiving, using a wireless receiver of the AV, a steering angle calibration parameter from the network-based computing platform, the steering angle calibration parameter determined by the network-based computing platform based on the one or more road surface conditions; and
augmenting, by the processing circuit, the steering angle measurement with the steering angle calibration parameter.
15. The method of claim 1 , further comprising:
operating, using the control circuit, an AV according to a maximum speed limit, wherein the maximum speed limit is determined based at least in part on the steering angle offset.
16. The method of claim 15 , wherein the maximum speed limit is determined based on the steering angle offset and weather conditions of the environment.
17. The method of claim 1 , further comprising:
recording supplemental data, wherein the supplemental data comprises weather conditions, passenger characteristics, passenger preferences, load or luggage characteristics, spatiotemporal location, traffic, time of day, and road conditions; and
determining the set of operating conditions based on the supplemental data.
18. The method of claim 1 , further comprising:
training a neural network with previously recorded offset data; and
predicting, using the neural network, the estimated steering angle offset.
19. The method of claim 18 , further comprising:
training the neural network with previously recorded supplemental data.
20. The method of claim 1 , wherein the steering angle measurement or the yaw rate measurement is based on processing, by the processing circuit, the steering angle measurement based on a road condition.
21. The method of claim 1 , wherein the steering angle measurement or the yaw rate measurement is based on identification, by the processing circuit, of a wheel base angle.
22. The method of claim 1 , wherein the forward speed of the AV is less than 5 m/s.
23. The method of claim 1 , wherein the one or more road surface conditions include a wet road surface, an icy road surface, or an oily road surface.
24. A steering angle calibration system for an autonomous vehicle (AV), the system comprising:
a steering angle sensor;
a yaw rate sensor;
one or more processing circuits;
one or more non-transitory storage media storing instructions which, when executed by the one or more processing circuits, cause performance of operations comprising:
receiving a steering angle measurement from the steering angle sensor;
receiving a yaw rate measurement from the yaw rate sensor, wherein the steering angle measurement and the yaw rate measurement are taken when a wheel base angle of a wheel of the AV is less than a wheel base angle threshold value;
estimating a steering angle offset based on the steering angle measurement, the yaw rate measurement and a wheel base of the AV;
determining an estimated yaw rate, the estimated yaw rate equal to
V
x
L
tan
(
δ
m
+
δ
ˆ
o
)
where V x is a forward speed of the AV, L is the wheel base of the AV, δ m is a measured steering angle based on the steering angle measurement, and {circumflex over (δ)} o is the estimated steering angle offset;
comparing the yaw rate measurement and the estimated yaw rate; and
initiating an action on the AV based on a result of the comparing.
25. The system of claim 24 , wherein the steering angle sensor is a steering column sensor.
26. The system of claim 24 , wherein the steering angle sensor is a steering wheel sensor.
27. A non-transitory, computer-readable storage medium having instructions stored thereon, that when executed by one or more processors of an autonomous vehicle (AV), cause the one or more processors to perform operations comprising:
receiving a steering angle measurement;
receiving a yaw rate measurement, wherein the steering angle measurement and the yaw rate measurement are taken when a wheel base angle of a wheel of the AV is less than a wheel base angle threshold value;
estimating a steering angle offset based on the steering angle measurement, the yaw rate measurement, and a wheel base of the AV;
determining an estimated yaw rate equal to
V
x
L
tan
(
δ
m
+
δ
ˆ
o
)
where V x is a forward speed of the AV, L is the wheel base of the AV, δ m is a measured steering angle based on the steering angle measurement, and {circumflex over (δ)} o is the estimated steering angle offset; and
operating the AV based on the estimated yaw rate.