DE102006029366A1 - Driving resistance determining method for motor vehicle, involves determining model error from reference model by comparing calculated actual-vehicle longitudinal acceleration signal with reference vehicle longitudinal acceleration signal - Google Patents

Abstract

The method involves determining a model error from a reference model by comparing a calculated actual-vehicle longitudinal acceleration signal with a reference vehicle longitudinal acceleration signal. The model error is adjusted at a vehicle longitudinal acceleration component that is referred on a roadway inclination. The acceleration component is determined by subtracting the calculated actual vehicle longitudinal acceleration signal from an actual vehicle acceleration signal measured by an acceleration sensor. An independent claim is also included for a controller for executing a method for determining a driving resistance.

Description

State of the art

The The invention relates to a method for determining a driving resistance in a motor vehicle according to the preamble of claim 1 and a control device for performing a Such method according to claim 11th

at the usual Method for driving resistance detection becomes one of the wheel speeds or wheel speeds calculated actual vehicle longitudinal acceleration with one from a balance of power modeled desired vehicle longitudinal acceleration compared, with the forces required for this purpose usually from read sizes of engine and transmission control (driving forces) and from the measured or estimated brake Press (Braking forces) be won. Simple friction forces, such as the Air resistance, are also modeled. The result of this Vehicle longitudinal acceleration comparison is referred to as a model error. This model error corresponds to the superimposed one Effect of all inaccuracies or unknowns within the model acceleration determination, So an unknown road gradient (slope or gradient), one unknown resistance of a deformable roadway (deep sand, Ballast bed, etc.), a different vehicle mass than assumed in the model as well as other modeling inaccuracies, e.g. inaccurately adopted Friction, inaccurate geometric vehicle data, inaccurate motor drive torque or Transmission values, etc. The model error is a measure of the driving resistance.

One increased Travel resistance is detected when the resulting from the comparison Model error exceeds a certain predetermined value. Analog becomes one decreased driving resistance detected when a second predetermined Value is fallen short of. This known method is, for example in the Off-Road ABS (Anti-lock Braking System) of the ESP (Electronic Stability Program) used by doing this very steep road gradient be recognized. Also suspected of trafficking a highly deformable Road surface, such as deep sand, is detected here. It is not possible with the known method the detection of a climb in the deep sand, For example, when driving on sand dunes, because the influences in the Driving resistance detection counteracted. A mountain journey gives a model error with the sign of an acceleration, a Tiefsandfahrt against a mistake with the sign of a Delay. The errors thus reduce each other and can even thoroughly cancel.

Disclosure of the invention

Technical task

Of the Invention is based on the object, the known method for Determining a driving resistance to improve so that reliable Even hill driving on deformable roads can be detected.

Technical solution

These The object is achieved with the features of claim 1.

advantageous Further developments of the invention are specified in the subclaims.

Of the Invention is based on the idea of the model error, the from a comparison of a calculated actual vehicle longitudinal acceleration signal with a determined from a reference model desired vehicle longitudinal acceleration signal results and a measure of the driving resistance represents compensating for the road incline influence. It Thus, a model error compensated for the inclination component remains in development of the invention, in particular by low-pass filtering can be prepared. The adjusted model error or the processed, adjusted model error is permanently determined and is a measure of one Driving resistance independent from the slope of the road is busy, being the quantitative remaining dominating influences of the model error a deformable Road with high rolling resistance or increased vehicle mass, in particular a heavy pendant, are. Thus, the adjusted or the adjusted and prepared Model error for detecting low-level or detection trips a heavy trailer be used. To decide, whether in an ABS braking at the same time recognized high driving resistance an increase in Sollschlupfschwellen has to be done or not, can additional Decision criteria are used. Similarly, a decision whether a detected high driving resistance comes from a heavy trailer over additional Criteria, such as a minimum vehicle speed, be secured.

The Actual vehicle longitudinal acceleration signal can be derived from at least one vehicle wheel speed. In particular, the determination is made by the time derivative the vehicle wheel speed. The target vehicle wheel acceleration is which is based on a reference model, in particular due to a driver's default (e.g., accelerator pedal travel) expected longitudinal acceleration of the vehicle.

In Embodiment of the invention is provided that for cleanup of a comparison of the calculated actual vehicle longitudinal acceleration signal with the desired vehicle longitudinal acceleration signal resulting model error a measured by means of at least one longitudinal acceleration sensor Actual vehicle longitudinal acceleration signal is used. This measured actual acceleration signal sets composed of two components, namely one of the vehicle movement stemming Acceleration component and one from one of the traveled Road originating Tilt component. This implies, conversely, that when out the sensor signal of the acceleration sensor, the vehicle longitudinal acceleration is eliminated, the vehicle longitudinal acceleration component attributable to the road inclination remains. These vehicle longitudinal acceleration component due to the roadway inclination can then follow be taken out of the model error, whereby a to the vehicle longitudinal acceleration component due to the roadway inclination compensated or adjusted model error remains. This adjusted model error or the edited adjusted model error is proportional to a road resistance adjusted to the road inclination.

One increased Travel resistance, i. a ride on a deformable surface and / or a trailer ride, or a decreased driving resistance is detected when the adjusted model errors or processed, corrected model errors exceeds or falls below a predetermined threshold. Prefers is not immediately after a single overrun or undercut the Threshold increased or decreased driving resistance, but only a counter elevated or humiliated. If a preset counter reading is reached, it will open an elevated one or lowered driving resistance closed, with the result that the Behavior of the vehicle dynamics control system is modified. Especially takes place at increased driving resistance, if over Additional criteria for trailer operation is detected, an activation of driving stabilization interventions, preferably of driving stabilization operations for the purpose of reduction especially for trailer operation occurring instabilities; if over Additional criteria is detected on a deformable roadway, a Adaptation of the ABS control strategy, preferably an adaptation of the ABS slip waves for the purpose of improving braking performance by wedge formation due to wheels with high brake slip.

Prefers is only on a trailer operation closed if additional Criteria such as exceeding a minimum speed Fulfills are.

Brief description of the drawings

Further Advantages and expedient designs The invention are the other claims, the description of the figures and to take the drawing.

The only 1 shows a schematic flow diagram of a vehicle dynamics control method.

Embodiment of the invention

In a first method step, the data required for determining the driving resistance are read. In this _exemplary embodiment, these are the vehicle acceleration _signal a _xSensor measured by an acceleration _sensor , the wheel speeds n _wheel , the traveled accelerator pedal _travel , and optionally further input values or driver specifications for a reference model.

In a subsequent second method step, the desired vehicle _{longitudinal acceleration} a _{x setpoint is determined} as a function of the input variables of the reference model, in particular of the driver specifications. In this _exemplary embodiment, the desired longitudinal vehicle acceleration a _x target is _ascertained, inter alia, as a function of the gas pedal _travel and other unspecified variables. The target vehicle _{longitudinal acceleration} a _xsoll is the expected vehicle _{longitudinal acceleration} .

In a subsequent third method step, the actual vehicle longitudinal acceleration is calculated as a function of the wheel speeds n _wheel .

In a next, fourth method step, the desired vehicle _{longitudinal acceleration} a _xsetpoint and the actual vehicle _{longitudinal acceleration} a _xactual are compared. This results in a model _error a _xoff , which corresponds to the superposition of all inaccuracies or unknowns. The model _error a _xoff includes the influences of the road _inclination , the condition of the roadway, the vehicle mass and other model inaccuracies. In the _exemplary embodiment shown, the model _error a _{xoff is} determined by subtracting the calculated actual longitudinal vehicle _acceleration a _xIst from the nominal vehicle _{longitudinal acceleration} a _xset determined in the reference model including filtering this difference, wherein the filtering serves to eliminate short-term deviations due to the different phase responses of the values to be compared ,

In a substantial, fifth method step, an adjusted model _error a _{xFw, roh is} determined. For this purpose, the model _error a _{xoff is adjusted} by the influences of the roadway _inclination . In the concrete embodiment, this is done by the Modellfeh The difference is _subtracted from the acceleration _signal a _xSensor measured by means of an acceleration _sensor and the calculated actual vehicle _longitudinal acceleration a _xIst . The difference between the measured vehicle _{longitudinal acceleration signal} a _xSensor and the calculated actual vehicle _{longitudinal acceleration signal} a _xIst corresponds to the vehicle _{longitudinal acceleration} component a _xNeig due to the roadway _inclination .

The adjusted model _error a _{xFw, crude} , is a raw value that is subject to fluctuations resulting from dynamic effects. For example, when the vehicle speed changes, the vehicle acceleration _signal a _xSensor measured by the acceleration _{sensor will have} a different phase response than the vehicle _{reference speed} differentiation. Such short-term fluctuations should _be eliminated from the raw value a _{xFw, raw} . For this purpose, the adjusted model _error a _{xFw, raw} processed accordingly, low-pass filtered in this embodiment in a sixth step.

In the exemplary embodiment shown, the query as to whether an increased or reduced driving resistance is present is shown substantially simplified. In a seventh method step, it is queried whether the processed, corrected model _error a _{xFw is} greater than a threshold value S. If this is not the case, the first method step continues. If the threshold value S is exceeded, an increased driving resistance is detected and an adaptation of the vehicle dynamics control system takes place, after which the first method step is continued.

As mentioned, is the query whether an elevated or lowered driving resistance is present, simplified. Usually will be exceeded of the threshold value S is a counter elevated and / or falls below the threshold. Upon reaching a predetermined meter reading is then raised to an or lowered driving resistance closed. By appropriate Selection of thresholds can Hysteresis effects are achieved.

Claims (11)

Method for determining a driving resistance, in particular as part of a vehicle dynamics control system, wherein a model _error (a _xOff ) is determined by a comparison of a calculated actual vehicle _{longitudinal acceleration signal} (a _xIst ) with a desired vehicle _{longitudinal acceleration signal} (a _xSoll ) from a reference model, characterized in that the model errors (a _XOFF) to a attributable to the road inclination vehicle longitudinal acceleration component (a _xNeig) is adjusted. A method according to claim 1, characterized in that the attributable to the road _inclination vehicle _{longitudinal acceleration} component (a _xNeig ) by subtracting the calculated actual vehicle _longitudinal acceleration _signal (a _xIst ) of a measured by means of at least one acceleration _sensor actual vehicle acceleration _signal (a _xSensor ) is determined. Method according to one of the preceding claims, characterized in that the model _error (a _{xFw, raw} ), which is adjusted by the vehicle _{longitudinal acceleration} component (a _xNeig ) attributable to the roadway _{inclination, is} processed, preferably filtered, in particular low-pass filtered. Method according to one of the preceding claims, characterized in that the reference model for determining the target vehicle _longitudinal acceleration (a _xSoll ) at least one input variable, in particular the accelerator pedal _travel , and / or accelerator acceleration, and / or the engine speed, and / or the current transmission ratio, and / or the brake pressure. Method according to one of claims 4 to 8, characterized in that an increased driving resistance is determined if the adjusted model _error (a _{xFw, raw} ) and / or the processed, adjusted model _error (a _xFw ) a threshold at least once, preferably several times over - or falls below. A method according to claim 9, characterized in that a reduced driving resistance is determined when the adjusted model _error (a _{xFw, raw} ) and / or the processed, adjusted model _error (a _xFw ) a threshold at least once, preferably multiple, under or over. Method according to one of claims 5 or 6, characterized that at least once, preferably multiple, exceeding a first threshold value a trailer operation detected becomes. Method according to one of claims 5 to 7, characterized that at least once, preferably multiple, exceeding a second threshold a deformable roadway is detected. Method according to claim 7, characterized in that that detection on a trailer operation only when crossing a minimum speed, in particular of 60 km / h, preferably of 70 km / h, preferably of 80 km / h, is activated. Method according to one of the preceding Claims, characterized in that upon detection of an increased or decreased driving resistance, the behavior of a vehicle dynamics control system is modified, in particular an activation of driving stabilization interventions, preferably of driving stabilization interventions for the purpose of reducing the particular occurring in trailer operation instabilities and / or adaptation of the ABS control strategy, preferably An adaptation of ABS slip thresholds for the purpose of improving the braking performance on deformable roads, takes place. control unit to carry out the method according to any one of the preceding claims.