Adaptive model predictive control for vehicle route planning

• US 10,060,370 B2
• Filed: 08/25/2015
• Issued: 08/28/2018
• Est. Priority Date: 10/02/2014
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: D'Amato, Anthony Mario, Filev, Dimitar Petrov, Michelini, John Ottavio, Mullen, Jonathan Thomas
• Primary Examiner(s): Vo, Hieu T
• Assistant Examiner(s): Castro, Arnold
• Application Number:
  US14/835,337
• Publication Number:
  US 20160096527A1
• Time in Patent Office: 1,099 Days
• CPC Class Codes: B60W 10/00 Conjoint control of vehicle sub-units of different type or different function for propulsion of purely electrically-propelled vehicles with power supplied within the vehicle B60L50/00 - B60L58/00
• CPC Class Codes: B60W 30/00 Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
• CPC Class Codes: F02D 2200/0625 Fuel consumption, e.g. measured in fuel liters per 100 kms or miles per gallon
• CPC Class Codes: F02D 2200/701 Information about vehicle position, e.g. from navigation system or GPS signal
• CPC Class Codes: F02D 2250/18 Control of the engine output torque
• CPC Class Codes: F02D 41/0215 in relation with elements of the transmission
• CPC Class Codes: F02D 41/1406 with use of a optimisation method, e.g. iteration

Abstract

Method and system are provided for vehicle route planning based on adaptive model predictive control. In one example, a method may include real-time online identification of the vehicle model base on the vehicle inputs and outputs; compression of the input space to increase the optimization efficiency; and optimization of the route planning based on the model parameter of the vehicle and the known road grade.

Claims

1. A method, comprising:

adjusting a vehicle parameter responsive to a determined planned route, the planned route determined for a given time horizon to reduce fuel consumption and further based on disturbances along the planned route and according to a compressed total number of parameters of an engine torque over the given time horizon, the total number of parameters of the engine torque being less than a number of data points of the given time horizon.

2. The method of claim 1, wherein the planned route is determined based on a constraint on the engine torque.

3. The method of claim 1, wherein the planned route is further determined based on a vehicle model estimated while operating a vehicle.

4. The method of claim 3, wherein the vehicle model is estimated through recursive least square estimation based on the engine torque, the disturbance, the fuel consumption, and an acceleration of the vehicle.

5. The method of claim 3, wherein the vehicle model is corrected based on an output of torque converter lock up response during discrete events when a torque converter unlocks.

6. The method of claim 1, wherein the total number of parameters of the engine torque is compressed based on kernel vector coefficients derived from a bank of basis vectors.

7. The method of claim 6, wherein the bank of basis vectors is updated in real time based on current kernel vector coefficients and future disturbance.

8. The method of claim 6, wherein the kernel vector coefficients are initialized based on the bank of basis vectors before minimizing the fuel consumption.

9. The method of claim 6, wherein the kernel vector coefficients are initialized based on the bank of basis vectors after resetting a covariance when estimating a model parameter.

10. A method for a vehicle, comprising:

during vehicle operation, estimating vehicle model parameters based on an engine torque, a disturbance along a route, a fuel consumption, and a vehicle acceleration;

determining a time horizon;

constructing a future engine torque over the time horizon, wherein the future engine torque has a number of parameters less than a length of the time horizon, wherein the length of the time horizon is a number of data points in the time horizon;

determining each parameter of the future engine torque to provide a desired fuel consumption;

operating the vehicle based on the future engine torque; and

moving the time horizon forward.

11. The method of claim 10, wherein the disturbance is a road grade.

12. The method of claim 10, wherein each parameter of the future engine torque is determined based on a constraint on the future engine torque.

13. The method of claim 10, wherein the future engine torque is constructed based on kernel vector coefficients generated from the disturbance.

14. The method of claim 13, wherein the kernel vector coefficients are generated based on a bank of basis vectors derived from the disturbance.

15. The method of claim 14, wherein the bank of basis vectors is updated based on current kernel vector coefficients and future disturbance.

16. The method of claim 15, wherein the acceleration of the vehicle is modeled based on a first model with a torque converter locked and a second model with the torque converter unlocked.

17. A vehicle system, comprising:

a torque converter for transmitting an engine torque to a drive train;

a sensor for sensing a disturbance along a route; and

a controller configured with computer readable instructions stored on non-transitory memory for;

estimating an engine torque;

estimating a fuel consumption;

estimating an acceleration of the vehicle system;

determining a time horizon;

reducing a total number of parameters of the engine torque over the time horizon based on the sensed disturbance, wherein the total number of parameters is less than a length of the determined time horizon, wherein the length of the determined time horizon is a number of data points in the determined time horizon;

estimating vehicle model parameters based on the engine torque, the estimated fuel consumption, the estimated acceleration of the vehicle system, and the sensed disturbance;

determining each parameter of a future engine torque with the reduced total number of parameters by minimizing a future fuel consumption;

operating the vehicle system based on a first parameter of the future engine torque; and

moving the time horizon forward.

18. The method of claim 17, wherein the acceleration of the vehicle system is estimated based on the estimated engine torque, and is corrected when the torque converter is unlocked.

19. The method of claim 17, wherein a compression is based on kernel vector coefficients derived from a bank of basis vectors, and the bank of basis vectors is updated in response to the sensed disturbance.