METHOD FOR DETERMINING A BRAKE PRESSURE CHANGE

Abstract

A method for determining a brake pressure change for a wheel of a vehicle to optimize a braking operation. The method includes: supplying a current wheel status of the wheel, wherein the wheel status includes a plurality of status parameters; determining at least one status parameter whose value deviates from a target wheel status; determining a change direction of the brake pressure change depending on a deviation of the at least one status parameter from the target wheel status; supplying a brake pressure characteristic map for determining a value of the brake pressure change, wherein the brake pressure characteristic map associates a brake pressure change with the plurality of status parameters and is specific to the determined change direction of the brake pressure change and status parameter change; determining a value of the brake pressure change using the current wheel status and the supplied brake pressure characteristic map.

Claims

1-15. (canceled)

16. A method for determining a brake pressure change for a wheel of a vehicle in order to optimize a braking operation, comprising the following steps: supplying a current wheel status of the wheel, wherein the wheel status includes a plurality of status parameters; determining at least one status parameter whose value deviates from a target wheel status; determining a change direction of the brake pressure change depending on a deviation of the at least one status parameter from the target wheel status; supplying a brake pressure characteristic map for determining a value of the brake pressure change, wherein the brake pressure characteristic map associates a brake pressure change with the plurality of status parameters and is specific to the determined change direction of the brake pressure change; determining a value of the brake pressure change using the current wheel status and the supplied brake pressure characteristic map.

17. The method as recited in claim 16, wherein at least one previous wheel status is supplied, and the at least one supplied brake pressure characteristic map is dependent on a change direction of the at least one status parameter whose value deviates from a target wheel status.

18. The method as recited in claim 17, wherein a gradient of the at least one status parameter is determined using at least the previous wheel status and a current value of the status parameter; and the determination of the change direction of the brake pressure change is additionally determined by way of the gradient.

19. The method as recited in claim 16, wherein the determination of the change direction of the brake pressure change is determined by way of a multiplicity of the status parameters.

20. The method as recited in claim 16, wherein the determination of the change direction of the brake pressure change is determined depending on a delay time of an entire system for changing the brake pressure.

21. The method as recited in claim 16, wherein the at least one status parameter is determined from the multiplicity of status parameters in accordance with a prioritization order.

22. The method as recited in claim 16, wherein the plurality of status parameters includes a wheel slip and/or an acceleration of the wheel and/or a gradient of the wheel slip and/or an acceleration of the wheel and/or a jerk of the wheel and/or a wheel acceleration relative to the acceleration of the vehicle.

23. A method for determining a brake pressure characteristic map, comprising the following steps: supplying a current wheel status, wherein the wheel status includes a plurality of status parameters; supplying reward rules for a reinforcement learning method; determining a reward using the reward rules and the current wheel status; and where the reward has been determined for the reinforcement learning method: determining a most recently completed brake pressure change in terms of value and change direction and an associated brake pressure characteristic map; determining a correction value for the associated brake pressure characteristic map in accordance with the reinforcement learning method.

24. The method as recited in claim 23, wherein the reward rules determine the reward depending on whether a slip value drops below a limit value for the slip and/or on whether a slip value drops below zero and/or on a modulation frequency of a pressure change.

25. The method as recited in claim 16, further comprising controlling the brake pressure on the wheel depending on the determined brake pressure change.

26. The method as recited in claim 23 wherein the method is used for optimizing a performance of a brake pressure controller for a wheel of the vehicle.

27. The method as recited in claim 16, further comprising supplying a control signal for actuating an at least partially automated vehicle based on the determined brake pressure change, and/or supplying a warning signal for alerting a vehicle occupant based on the determined brake pressure change.

28. A braking system configured to determine a brake pressure change for a wheel of a vehicle in order to optimize a braking operation, the braking system configured to: supply a current wheel status of the wheel, wherein the wheel status includes a plurality of status parameters; determine at least one status parameter whose value deviates from a target wheel status; determine a change direction of the brake pressure change depending on a deviation of the at least one status parameter from the target wheel status; supply a brake pressure characteristic map for determining a value of the brake pressure change, wherein the brake pressure characteristic map associates a brake pressure change with the plurality of status parameters and is specific to the determined change direction of the brake pressure change; determine a value of the brake pressure change using the current wheel status and the supplied brake pressure characteristic map.

29. A non-transitory machine-readable storage medium on which is stored a computer program for determining a brake pressure change for a wheel of a vehicle in order to optimize a braking operation, the computer program, when executed by a computer, causing the computer to perform the following steps: supplying a current wheel status of the wheel, wherein the wheel status includes a plurality of status parameters; determining at least one status parameter whose value deviates from a target wheel status; determining a change direction of the brake pressure change depending on a deviation of the at least one status parameter from the target wheel status; supplying a brake pressure characteristic map for determining a value of the brake pressure change, wherein the brake pressure characteristic map associates a brake pressure change with the plurality of status parameters and is specific to the determined change direction of the brake pressure change; and determining a value of the brake pressure change using the current wheel status and the supplied brake pressure characteristic map.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Exemplary embodiments of the present invention are illustrated with reference to FIGS. 1 to 9 and explained in more detail below.

[0071] FIG. 1 shows a curve of a slip value and the corresponding wheel speed over time.

[0072] FIG. 2 shows a braking system.

[0073] FIG. 3 is a status chart showing a progression of status values.

[0074] FIG. 4 is a graph showing status parameters.

[0075] FIG. 5 shows a characteristic map comprising a matrix for changing the values of the characteristic map.

[0076] FIG. 6 is an illustration of changed characteristic map values.

[0077] FIG. 7 is a graph showing a status parameter over time.

[0078] FIG. 8 shows a characteristic map comprising a list for changing the values of the characteristic map.

[0079] FIG. 9 is a graph showing a multiplicity of status parameters over time.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0080] By way of example, FIG. 1 is a graph 100 showing a curve of a slip value 120 over time, along with limits S.sub.max, S.sub.min within which the slip value should be located. In this case, the curve 160 shows a reference speed v and the curve 140 shows the speed of the wheel being observed.

[0081] FIG. 2 shows a braking system 200 configured to carry out the method for determining the brake pressure change in order to optimize a braking operation.

[0082] By way of wheel sensors and other vehicle sensors, a current wheel status 210 of the wheel is supplied, the wheel status 210 including a plurality of status parameters, for example slip s and wheel acceleration a. In the module 220 of the braking system 200, at least one status parameter whose value deviates from a target wheel status is determined, and a change direction of the brake pressure change is determined depending on a deviation of the at least one status parameter from the target wheel status. Alternatively, the brake pressure can also be kept constant.

[0083] Using this determined change direction p⬆ or p⬇, a supplied brake pressure characteristic map 250a, 250b, 251a, 251b for determining a value of the brake pressure change is selected, the brake pressure characteristic map associating a brake pressure change with the plurality of status parameters slip s, wheel acceleration a, or D for the braking, and being specific to the determined change direction of the brake pressure change and status parameter change. In other words, the brake pressure characteristic maps are picked depending on the direction of the change in the status parameters. There is thus a brake pressure characteristic map 250b that is specific to an increase in the wheel acceleration, and a brake pressure characteristic map 250a that is specific to a decrease in the wheel acceleration. The same applies to the slip status parameter: there is a brake pressure characteristic map 251a for an increase in the slip and a brake pressure characteristic map 251b for a decrease in the slip. In this case, the brake pressure characteristic maps 250a, 250b are associated with the positive brake pressure change, and the brake pressure characteristic maps 251a, 251b are associated with the negative brake pressure change. The value of the brake pressure change Dp can thus be determined using the current wheel status and the corresponding supplied brake pressure characteristic map, and in particular it can be forwarded to the braking system for the wheel. The values of the status parameters and of the pressure changes are forwarded to the module 240 so that the corresponding brake pressure characteristic maps can be modified using the above-described reinforcement learning method.

[0084] FIG. 3 is a schematic chart 300 showing a multiplicity of combinations of two status parameters slip s and wheel acceleration a, indicating a progression of status values. In this case, the target wheel status 310 for the status fields is highlighted by a thick border. If the method starts at a wheel status arranged in the top right in this chart 300, then a progression of status fields in which, in each case, the relevant status is determined within a set period of time of, for example, 5 ms can be denoted by way of the double-hatched fields 314; in this case, a positive brake pressure change is carried out without directly achieving the target wheel status 310, in particular owing to a delay time of, for example, 30 ms. The fields 310 highlighted in black denote status fields in which a negative brake pressure change occurs.

[0085] Brake pressure changes affect the change in the wheel status in the chart 300 as follows:

TABLE-US-00001 Slip (x axis) aWheel (y axis) Pressure build-up greater smaller Pressure build-up smaller greater

[0086] FIG. 4 is a schematic graph 400 showing a progression of slip s values 430 over time, which by way of example, are located outside an upper and a lower limit value in the region 440. In this case, the curve 420 indicates the associated status value S, and the curve 410 shows a variation of the resultant brake pressure p in line with the cumulative brake pressure changes. A region 450 is indicated, and the brake pressure changes therein prior to the time frame in the region 440 can be deemed to be the reason why the slip value 430 exceeds the limit in the region 440. Using reinforcement learning, therefore, the corresponding brake pressure characteristic map for the status 420 can be adjusted to achieve a control in which the slip remains within established limits.

[0087] FIG. 5 shows a characteristic map 520 for an increasing pressure change, in which the status parameter for an acceleration of the wheel is decreasing, or braking is increasing.

[0088] FIG. 6 is an illustration of changed characteristic map values that indicate, on the basis of the status parameters wheel acceleration a and slip s, how the brake pressure p is to be changed depending on the status parameters.

[0089] FIG. 7 is a graph 700 showing the slip s status parameter 720 over time; it can be seen that the slip s is below an established minimum slip value in the region 725. In addition, this graph 700 shows the variation 715 of the brake pressure p over time. The reinforcement learning method can identify this most recently completed brake pressure change and modify the associated brake pressure characteristic map accordingly, so as to prevent the slip from dropping below the limit for the corresponding wheel status in future.

[0090] FIG. 8 shows a characteristic map associated with a positive brake pressure change in the event of a decreasing wheel acceleration or increasing braking; it is updated depending on the wheel status by way of a list for changing the values of the characteristic map, the list having been determined using the reinforcement learning method.

[0091] FIG. 9 is a graph showing a multiplicity of parameters and status parameters over time: slip 946, quick pressure increase 944, slow pressure increase 945, pressure kept constant 943, quick pressure drop 941, slow pressure drop 942, status index 947, target brake pressure 948. In the region 930, which denotes a quick succession of slip changes, said change can be attributed to target brake pressure changes prior thereto in the curve 948 together with the wheel status denoted in the region 920.

[0092] Since a high frequency of changes in the slip should be avoided, a modulation frequency of the pressure change can be integrated in the reward rules of the reinforcement learning method.