Lane assistance for vehicles with trailers
Orientations of a vehicle with an attached trailer can be specified for obtaining respective images of the trailer. Respective images of the trailer captured by a vehicle camera at the specified orientations can then be received. The images can be input to a machine learning program trained to output trailer size metrics. The vehicle can then be controlled according to the trailer size metrics output from the machine learning program.
1. A system, comprising a computer that includes a processor and a memory, the memory storing instructions executable by the processor such that the computer is programmed to:
specify paths for a vehicle with an attached trailer to obtain respective orientations for obtaining respective images of the trailer;
receive, after the vehicle obtains at least one of the respective orientations, the respective images of the trailer captured by a vehicle camera at the respective orientations;
input the images to a machine learning program trained to output trailer size metrics; and
control the vehicle according to the trailer size metrics output from the machine learning program.
2. The system of claim 1 , wherein controlling the vehicle according to the trailer size metrics includes one or both of adjusting a bias or adjusting a gain to a parameters of a steering assist system of the vehicle based on the trailer size metrics and one or both of a direction and magnitude of wind.
3. The system of claim 2 , wherein adjusting the bias or the gain adjusts an output steering wheel torque.
4. The system of claim 2 , wherein adjusting the bias or the gain adjusts an offset from a lane center.
5. The system of claim 1 , wherein the instructions further include instructions to download from a remote server a trailer profile specifying a parameter for controlling the vehicle based on the trailer size metrics output from the machine learning program based on the images.
6. The system of claim 1 , wherein the trailer size metrics include one or more of a mass of the trailer, a surface area of the trailer, a weight of the trailer, a tongue weight of the trailer, or a trailer dimension.
7. The system of claim 6 , wherein the instructions further include instructions to obtain the tongue weight of the trailer as output from scales onboard the vehicle.
8. The system of claim 6 , wherein the instructions further include instructions to obtain the mass of the trailer as output from a vehicle mass estimation module that uses vehicle powertrain data.
9. The system of claim 1 , wherein the instructions further include instructions to input an angle of incline to the machine learning program for at least one of the images.
10. The system of claim 1 , wherein the instructions further include instructions to control the vehicle to move so that the trailer is at one of the respective orientations.
11. The system of claim 1 , wherein the instructions further include instructions to output the respective orientations to a display device.
12. A method, comprising:
specifying paths for a vehicle with an attached trailer to obtain respective orientations for obtaining respective images of the trailer;
receiving, after the vehicle obtains at least one of the respective orientations, the respective images of the trailer captured by a vehicle camera at the respective orientations;
inputting the images to a machine learning program trained to output trailer size metrics; and
controlling the vehicle according to the trailer size metrics output from the machine learning program.
13. The method of claim 12 , wherein controlling the vehicle according to the trailer size metrics includes one or both of adjusting a bias or adjusting a gain to a parameters of a steering assist system of the vehicle based on the trailer size metrics and one or both of a direction and magnitude of wind.
14. The method of claim 13 , wherein adjusting the bias or the gain adjusts one or more of an output steering wheel torque or an offset from a lane center.
15. The method of claim 12 , further comprising downloading from a remote server a trailer profile specifying a parameter for controlling the vehicle based on the trailer size metrics output from the machine learning program based on the images.
16. The method of claim 12 , wherein the trailer size metrics include one or more of a mass of the trailer, a surface area of the trailer, a weight of the trailer, a tongue weight of the trailer, or a trailer dimension.
17. The method of claim 16 , further comprising obtaining the tongue weight of the trailer as one or more of output from scales onboard the vehicle or from a vehicle mass estimation module that uses vehicle powertrain data.
18. The method of claim 12 , further comprising inputting an angle of incline to the machine learning program for at least one of the images.
19. The method of claim 12 , further comprising controlling the vehicle to move so that the trailer is at one of the respective orientations.
20. The method of claim 12 , further comprising outputting the respective orientations to a display device.