Systems and methods for trailer angle detection
A trailer angle identification system includes an imaging device configured to capture an image. A controller is configured to receive steering angle data corresponding to a steering angle of a vehicle, receive vehicle speed data corresponding to a vehicle speed, and estimate an angle of the trailer relative to the vehicle by processing the image, the steering angle data, and the vehicle speed data in at least one neural network.
1. A trailer angle identification system comprising:
an imaging device configured to capture an image of a trailer;
a controller configured to:
receive steering angle data corresponding to a steering angle of a vehicle;
receive vehicle speed data corresponding to a vehicle speed; and
estimate an angle of the trailer relative to the vehicle by processing the image, the steering angle data, and the vehicle speed data in at least one neural network that includes a first neural network and a second neural network, the first neural network operable to process the image and the second neural network operable to process the steering angle data and the vehicle speed data, wherein an output of the second neural network is an estimated range of a change of the angle.
2. The trailer angle identification system of claim 1 , wherein the at least one neural network comprises:
a third neural network operable to process a first output of the first neural network and a second output of the second neural network.
3. The trailer angle identification system of claim 2 , wherein the third neural network is a fully-connected neural network.
4. The trailer angle identification system of claim 2 , wherein the controller is further configured to:
train the third neural network to estimate the angle.
5. The trailer angle identification system of claim 4 , wherein the third neural network is further trained to limit estimation of the angle.
6. The trailer angle identification system of claim 4 , wherein training of the third neural network is based on a difference between the first output and the second output.
7. The trailer angle identification system of claim 4 , wherein the third neural network is trained by comparing the angle to the first output and the second output.
8. The trailer angle identification system of claim 1 , wherein the controller is further configured to:
compare the angle to a previous estimation of the angle estimated by the at least one network.
9. The trailer angle identification system of claim 8 , wherein the controller is further configured to:
limit estimation of the angle based on the comparison of the angle to the previous estimation of the angle.
10. A method of identifying a trailer angle comprising:
capturing an image of a trailer in a field of view;
receiving steering angle data corresponding to a steering angle of a vehicle;
receiving vehicle speed data corresponding to a vehicle speed; and
estimating an angle of the trailer relative to the vehicle by processing the image, the steering angle data, and the vehicle speed data in at least one neural network that includes a first neural network and a second neural network, the first neural network operable to process the image and the second neural network operable to process the steering angle data and the vehicle speed data, wherein an output of the second neural network is an estimated range of a change of the angle.
11. The method of identifying a trailer angle of claim 10 , wherein the at least one neural network comprises:
a third neural network operable to process a first output of the first neural network and a second output of the second neural network.
12. The method of identifying a trailer angle of claim 11 , wherein the third neural network is a fully-connected neural network.
13. The method of identifying a trailer angle of claim 11 , further comprising:
training, via a controller, the third neural network to estimate the angle.
14. The method of identifying a trailer angle of claim 11 , wherein the third neural network is further trained to limit estimation of the angle.
15. The method of identifying a trailer angle of claim 11 , wherein training of the third neural network is based on a difference between the first output and the second output.
16. The method of identifying a trailer angle of claim 11 , wherein the third neural network is trained by comparing the angle to the first output and the second output.
17. The method of identifying a trailer angle of claim 10 , wherein the method further comprises:
comparing the angle to a previous estimation of the angle estimated by the at least one network.
18. The method of identifying a trailer angle of claim 17 , further comprising:
limiting estimation of the angle based on the comparison of the angle to the previous estimation of the angle.
19. A trailer angle identification system comprising:
an imaging device configured to capture an image;
an angle sensor configured to measure a steering angle of a vehicle;
a speed sensor configured to measure a vehicle speed; and
a controller configured to:
process the image in a first neural network trained to estimate a first angle of the trailer relative to the vehicle;
process the steering angle and the vehicle speed in a second neural network trained to determine an estimated change in the trailer angle;
estimate the trailer angle based on the first angle and the estimated change; and
update the trailer angle by comparing the estimated change to a difference of the trailer angle and a previous estimation of the trailer angle.
20. The trailer angle identification system of claim 19 , wherein the estimated change in the trailer angle includes a range having a minimum angle change and a maximum angle change.