Method for determining unstable behavior of a trailer and method for stabilizing a trailer, as well as evaluation unit and vehicle combination

Description

[0001] The invention relates to a method for determining unstable behavior of a trailer, in particular a trailer of a vehicle combination, and a method for stabilizing a trailer, an evaluation unit for carrying out the method, and a vehicle combination with at least one trailer.

[0002] With a vehicle combination, especially a commercial vehicle combination, certain driving situations can cause the towing vehicle and / or the towed vehicle or trailer to begin oscillating or swaying, or to break away laterally, potentially leading to critical driving situations. Such oscillations or lateral breaks can occur, for example, while cornering, during evasive or overtaking maneuvers, due to wind conditions, or similar circumstances. If the vehicle combination is traveling on a wet or slippery surface, the oscillations or breakaway can be intensified.

[0003] To react to such unstable driving situations, driver assistance systems are known to evaluate the actual driving dynamics of the towing vehicle and use this information to infer oscillation behavior or a potential loss of control. They then intervene accordingly or inform the driver. Information about the oscillation behavior or lateral movement of the trailer is either not considered or only very imprecisely assessed, as the trailer's movement itself is not monitored. At best, the system may activate a trailer braking system to, for example, initiate a stretch braking maneuver or brake the entire vehicle combination, thereby minimizing oscillation or counteracting a loss of control. Such intervention can be controlled by a control unit of an electronic stability control (ESC) system in the towing vehicle, which includes an evaluation unit for monitoring the actual driving dynamics of the towing vehicle.

[0004] A disadvantage of this approach is that determining the oscillation behavior of the vehicle combination or the lateral movement of the trailer relies solely on the actual driving dynamics parameters of the towing vehicle. Deviations in driving behavior, such as those caused by different loads or weight ratios between the towing vehicle and trailer, are not taken into account. Therefore, a situation might arise where the towing vehicle is much heavier than the trailer, causing the trailer to sway and exhibit a certain degree of oscillation, and / or even a single instance of trailer breakaway. However, due to the weight difference, this oscillation or breakaway is not transmitted to the towing vehicle, or only to a very limited extent, meaning the stability control system may not detect it at all, detect it too late, or detect it to an insufficient degree.This can lead to instabilities in the entire vehicle combination, for example, if the trailer exhibits excessive oscillation or lateral movement, causing the towing vehicle to also begin oscillating or veering sideways, which is then noticed very late or too late. Furthermore, an oscillating or veering trailer can also pose a danger to other road users in the vicinity, especially if this goes unnoticed and no countermeasures are taken.

[0005] WO 2010 / 087022 describes a stability control system in which the actual yaw rate of the towing vehicle, as a measured dynamic characteristic of the towing vehicle, is compared with the actual yaw rate of the trailer, as a dynamic characteristic of the trailer. If these deviate from each other by a defined limit, the brakes are applied accordingly to counteract this deviation. A disadvantage of this system is that it only monitors when a limit is exceeded and not the precise course or dynamic behavior of the trailer during such an exceedance.

[0006] Furthermore, US 9,573,589 B2 describes a stability control system for a vehicle combination consisting of a towing vehicle and a towed vehicle or trailer. This system uses an unspecified vehicle combination model to determine a deceleration factor that indicates the time delay after which the trailer is expected to react to a steering action by the towing vehicle. After measuring the current actual yaw rate of the towing vehicle as a measured dynamic characteristic of the towing vehicle, a target yaw rate or reference yaw rate for the trailer can be determined using the deceleration factor. This target yaw rate, or reference yaw rate, is a dynamic characteristic of the trailer and, based on the vehicle combination model, indicates how the trailer should react to the actual yaw rate of the towing vehicle.By comparing the reference yaw rate with the current actual trailer yaw rate as a driving dynamics parameter of the trailer, it can be determined whether an intervention in the brakes and / or the steering of the trailer is necessary to maintain the reference yaw rate.

[0007] Each trailer is controlled separately by a control unit that includes a yaw rate determination unit and implements the vehicle combination model. Thus, the actual yaw rate of the towing vehicle and the actual yaw rate of the trailer, as well as the delay factor, are processed in an evaluation unit for each trailer in order to subsequently determine and set the target yaw rate or reference yaw rate for the trailer.

[0008] EP 2 773 544 B1 describes a comparable method in which lateral acceleration is considered as a vehicle dynamics parameter instead of yaw rate. The evaluation and control for maintaining a reference lateral acceleration is similar to US 9,573,589 B2 and also uses a deceleration factor.

[0009] In this case, too, the system adjusts the trailer's driving dynamics to the towing vehicle's dynamics with a time delay, assuming that the trailer will approach the target value in response to a corresponding intervention in the driving dynamics. The timing of this adjustment and whether further instabilities, such as swaying or lateral skidding, occur as a result of this adjustment are not monitored. Consequently, escalating oscillations of the trailer or lateral skidding may not be detected in time, as adjustments are made for any deviation towards the target value.

[0010] DE 10 2016 105 916 A1 relates generally to systems for controlling vehicle parameters while towing a trailer, such as a trailer reversing assistance system. In particular, various systems for controlling the speed of a vehicle while using a trailer reversing assistance system are disclosed.

[0011] A further disadvantage of the described methods is that the geometric dimensions or the type of coupling between the vehicles in the vehicle combination are not sufficiently taken into account when comparing the actual state with the target state. As a result, the respective methods cannot be easily applied to any vehicle combination without adaptation, or they must be reprogrammed accordingly for a newly assembled vehicle combination.

[0012] The object of the invention is therefore to provide a method by which an unstable state or unstable behavior of the trailer can be reliably and simply determined. Furthermore, the object of the invention is to provide a method for subsequent stabilization, in particular of the trailer, an evaluation unit, and a vehicle combination with or in which the methods can be carried out.

[0013] This problem is solved by a method according to claim 1 and by a method, an evaluation unit, and a vehicle combination according to the further independent claims. The dependent claims specify preferred embodiments.

[0014] According to the invention, at least one actual driving dynamics parameter of the trailer is first determined, characterizing the current driving dynamics state of the trailer and derived from a measurement taken by at least one sensor in the trailer. This actual driving dynamics parameter is obtained directly or indirectly from a measurement in the trailer. It is also possible to convert one driving dynamics parameter into another to obtain the driving dynamics state of the trailer. Furthermore, at least one target driving dynamics parameter of the trailer is determined, derived from the actual driving dynamics parameters of the towing vehicle by applying a kinematic model that models the movement of the entire vehicle combination as a function of geometric parameters of the vehicle combination, starting from the towing vehicle.Finally, at least one of the actual driving dynamics parameters of the respective trailer is compared with at least one target driving dynamics parameter of the respective trailer determined via the kinematic model. This comparison aims to identify any deviation of the trailer's driving dynamics behavior from that predicted by the kinematic model. If the at least one actual driving dynamics parameter of the respective trailer deviates from the at least one determined target driving dynamics parameter of the respective trailer by more than a defined reference value, the conclusion is drawn that the trailer exhibits unstable behavior. This determines the extent to which the actual state still corresponds to the target state.

[0015] This offers the advantage of using actual vehicle dynamics measurements taken in the trailer to determine unstable behavior. This allows for the detection of trailer movements that are not transmitted to the towing vehicle, or only minimally, or too late, and which are therefore not detected, or not detected at all, by conventional systems that only measure and evaluate the kinematics of the towing vehicle.

[0016] Furthermore, to determine unstable behavior, the target behavior of the trailer is also taken into account. This target behavior is derived simply and reliably from the actual behavior of the towing vehicle using a kinematic model. Advantageously, the exact dimensions of the vehicle combination, i.e., the geometric parameters, are used. These parameters are known for a vehicle combination whose trailer is equipped with an electronic braking system and has its own CAN bus, and are provided on the CAN bus via the trailer interface according to ISO 11992. Thus, for the currently configured vehicle combination, the difference between the expected stable state of the trailer and its actual driving state can be determined with good accuracy.

[0017] This means that the measurement parameters and geometric characteristics of the vehicle combination, already provided in the vehicle (e.g., via the CAN bus and the trailer interface), are used. This eliminates the need for additional sensors and also eliminates the need for manual parameterization when assembling or modifying the vehicle combination, especially when attaching a new trailer. This is because the geometric data can be automatically transferred from the trailer (with its electronic braking system and its own CAN bus) via the trailer interface to the CAN bus, e.g., via ISO 11992, and used for the kinematic model. Therefore, the method can be easily applied to different vehicle combinations or changing vehicle configurations without requiring any reprogramming.

[0018] The kinematic model is based on the assumption that the vehicle combination consists of N links, with one link being formed by the towing vehicle and at least one other link by the trailer, which is pulled by the towing vehicle either indirectly via further links or directly. Using the geometric parameters, the kinematics of the entire N-link vehicle combination can be determined via equations of motion, thus allowing for the simple extraction of a target state for the respective trailer that depends on the kinematics of the towing vehicle.

[0019] According to a preferred further development, a time-resolved difference is calculated between at least one actual driving dynamics parameter of the trailer and at least one target driving dynamics parameter of the trailer for comparison purposes. This allows the deviation to be easily determined and analyzed over time in order to conclude that an unstable state exists.

[0020] Preferably, at least one target trailer dynamics parameter, a target trailer yaw rate and / or a target trailer track offset, can be determined via the kinematic model, and at least one actual trailer dynamics parameter, an actual trailer yaw rate and / or an actual trailer track offset, can be determined. These parameters can each be determined based on measurements taken by sensors in the trailer. This allows the actual lateral dynamics of the trailer to be determined, from which unstable behavior can be easily derived by comparison with the target lateral dynamics. For example, a deviation in the yaw rate can be used to estimate whether the trailer is yawing more than modeled and / or whether its track offset relative to the towing vehicle is greater than expected. Based on this behavior and considering reference values, unstable behavior can be easily concluded.

[0021] Preferably, the system is designed so that a time-oscillating deviation of the actual trailer yaw rate and / or track offset from the target trailer yaw rate and / or track offset, with at least one recurring differential yaw rate amplitude as a reference value and / or differential track offset amplitude as a reference value, indicates the presence of oscillatory behavior of the trailer as an unstable characteristic. Thus, a permissible differential amplitude is defined as a reference value, which is undershot by the respective actual vehicle dynamics parameter in stable conditions. However, if the respective actual vehicle dynamics parameter repeatedly deviates by more than this defined amplitude as a reference value, e.g., periodically, then an instabilizing oscillatory behavior of the trailer is inferred.

[0022] By appropriately defining the difference amplitude as a reference value, small or acceptable oscillations can advantageously be permitted, which do not necessarily lead to instability. For example, assuming driving on a rural road, a recurring deviation of the actual trailer track offset from the target trailer track offset can be permitted with a maximum offset of up to 25 cm, whereby a risk to vehicles in the vicinity can be avoided up to this deviation. This allows for minor oscillations, e.g., due to ruts or measurement and calculation inaccuracies. Instead of a maximum permissible track offset, a maximum permissible articulation angle can also be defined, given knowledge of the geometric dimensions.

[0023] According to a preferred embodiment, it can also be provided that a single deviation of the actual trailer track offset from the target trailer track offset, with at least one predefined limit difference track offset, is used to infer lateral fishtailing of the trailer as an unstable behavior. Thus, a driving situation in which a single lateral fishtailing of the trailer leads to unstable behavior can be easily identified by comparing the target state with the actual state. The limit difference track offset is preferably larger than the difference track offset amplitude, enabling a clear distinction between the two unstable driving situations. Furthermore, it is expected that lateral fishtailing requires larger deviations in the track offsets between the towing vehicle and trailer to result in an unstable state.

[0024] According to a preferred further development, it is also provided that the actual trailer yaw rate is determined as a function of the trailer's wheel speeds measured by wheel speed sensors. Additionally or alternatively, the actual trailer yaw rate can be determined as a function of a direct measurement of the trailer's actual yaw rate via a yaw rate sensor in the trailer.

[0025] When determining yaw rate via wheel speeds, it is advantageous to use the sensors already present in the trailer. In trailers with an electronic braking system and their own CAN bus, the signals from these sensors are transmitted to the towing vehicle's CAN bus via the trailer interface according to ISO 11992 and are thus readily available. The yaw rate can then be reliably estimated from the wheel speeds. To save computational effort, a yaw rate sensor in the trailer can also be used, the signals from which can likewise be transmitted to the CAN bus via the trailer interface.

[0026] According to a preferred embodiment, the actual track offset of the trailer is derived from the actual total articulation angle between the trailer and the towing vehicle. Advantageously, this actual total articulation angle can be derived from the sum of the actual articulation angles between the individual components of the vehicle combination. These actual articulation angles are determined based on the trailer's wheel speeds measured by wheel speed sensors in the trailer and / or based on a direct measurement of the trailer's actual yaw rate via a yaw rate sensor in the trailer, and based on the towing vehicle's actual yaw rate. Thus, this determination can also be easily based on actual parameters already measured in the trailer and towing vehicle, enabling a reliable determination of the trailer's actual state.

[0027] The target trailer track offset is then preferably derived from a target total articulation angle modeled via the kinematic model from the actual vehicle dynamics parameters of the towing vehicle, whereby the target total articulation angle results from a sum of modeled target articulation angles between the links of the vehicle combination. Thus, in this case as well, a target state can be easily estimated via the kinematic model in order to derive a vehicle dynamics parameter from which unstable behavior can be extracted by comparing the actual and target states.

[0028] According to a further preferred embodiment, it is provided that the determination of the at least one target driving dynamics parameter of the trailer by applying the kinematic model and the comparison of the at least one actual driving dynamics parameter of the respective trailer with the target driving dynamics parameter of the respective trailer determined via the kinematic model takes place in the towing vehicle.

[0029] This allows the calculation to advantageously take place at a central location within the vehicle combination, where the required signals can be easily provided. This is the case, for example, in the control unit of the towing vehicle's stability control system, which can access the CAN bus. The relevant trailer data is transmitted via this bus, eliminating the need for complex wiring or additional control units in the trailer. The method can therefore be implemented with minimal effort by, for example, incorporating an evaluation unit according to the invention into the control unit. This unit extends the control unit's software and / or hardware capabilities and enables the execution of the method according to the invention.The stability system can then preferably intervene in the driving dynamics of the vehicle combination in order to counteract a detected oscillation or skidding, the means for which are already available, since a stability system can intervene automatically and in a stabilizing manner in the driving process and / or is able to issue a warning to the driver.

[0030] Preferably, it is further provided that at least some of the geometric parameters of the vehicle combination are transmitted to the towing vehicle via a trailer interface, preferably in accordance with ISO 11992. This eliminates the need for any further adjustments, as the relevant parameters are already provided. The processing of this provided data is therefore possible with minimal effort using the method according to the invention.

[0031] According to a preferred embodiment, it is further provided that at least one actual vehicle dynamics parameter measured in the trailer and / or at least a part of the geometric parameters of the vehicle combination are made available from the trailer interface to a CAN bus in the towing vehicle. This enables a simple supply of the relevant parameters to the evaluation unit, since this only needs to be connected to the CAN bus.

[0032] According to a preferred further development, the actual driving dynamics parameters of the towing vehicle are derived from measurements taken by sensors in the towing vehicle. Thus, the actual parameters of the towing vehicle, from which the target parameters for the trailer are derived via the kinematic model, are also measured directly in the towing vehicle and are therefore directly available to the evaluation unit, e.g., via the CAN bus.

[0033] Preferably, the geometric parameters of the vehicle combination used for the kinematic model are the trailer drawbar length and / or the wheelbase of the towing vehicle and / or the trailer wheelbase. This allows for a simple definition of the vehicle combination's geometric configuration, as these parameters are responsible for its kinematics, and the target parameters vary for each vehicle combination. By using these available parameters, a very precise kinematic description of the vehicle combination can be achieved, thus enabling an accurate determination of the trailer's target dynamic state.

[0034] According to a preferred further development, the driving dynamics of the towing vehicle, characterized by its actual driving dynamic parameters, are modeled on the respective trailer using the kinematic model, taking into account the geometric parameters of the vehicle combination. By applying the kinematic model, it is thus possible to calculate in a simple and reliable manner how the individual components will react depending on the geometric dimensions and the movement of the towing vehicle, allowing for a precise determination of the target state for the respective vehicle combination.

[0035] According to a preferred embodiment, the application of the kinematic model involves the time-resolved, successive solving of equations of motion for each link of the vehicle combination. Each link has a coupling point, and adjacent links are rotatably connected to one another via these coupling points. The yaw rate of the i-link of the vehicle combination is estimated from at least one actual driving dynamics parameter of the towing vehicle using the equations of motion. The at least one target driving dynamics parameter of the respective trailer is then determined as a function of the estimated yaw rate of the i-link of the vehicle combination. The equations of motion thus advantageously take into account the driving dynamics behavior of each individual link, preferably also considering the relative motion between the links in order to model the entire combination and the influence of the links on one another.

[0036] For this purpose, the equations of motion are preferably dependent on the geometric parameters. This allows the kinematics of the entire train to be calculated easily, taking into account the currently available dimensions.

[0037] According to the invention, an evaluation unit according to independent claim 18 is further provided for carrying out the method according to the invention, wherein the evaluation unit is configured to read in the geometric parameters and the actual vehicle dynamic parameters of the towing vehicle as described and, using the kinematic model, to determine the at least one target vehicle dynamic parameter of the respective trailer and to compare it with the at least one actual vehicle dynamic parameter of the trailer, and, depending on this, to conclude whether the respective trailer exhibits unstable behavior. Thus, the advantages of the method can be ensured by the evaluation unit, which can preferably be implemented as software and / or hardware in a control unit of the stability system, and which is configured to carry out the method.

[0038] According to the invention, this evaluation unit is installed in a vehicle combination consisting of a towing vehicle and at least one trailer, wherein at least part of the geometric parameters and the at least one actual driving dynamic parameter of the trailer measured in the trailer can be transferred from the trailer to the evaluation device, which is advantageously located in the towing vehicle, via a trailer interface in order to detect unstable behavior of the trailer originating from the towing vehicle.

[0039] According to the invention, a method for stabilizing the trailer is further provided, comprising at least the following steps: Determining unstable behavior of the at least one trailer in the method according to independent claim 1, and controlling a drive system to reduce the engine torque and / or a braking system and / or a steering system in the towing vehicle and / or in the trailer, if unstable behavior of the at least one trailer is present, in order to stabilize the trailer, wherein the control is carried out depending on the determined deviation of the at least one actual driving dynamics parameter of the respective trailer from the at least one determined target driving dynamics parameter of the respective trailer, and / or issuing a warning to the driver.

[0040] This allows for a targeted response to the actual deviation by automatically applying the brakes via the stability control system or manually by the driver at the appropriate moment, when the driver is notified by a warning message that an instability exists which they might not notice because it does not transmit to the towing vehicle. Preferably, the drive system and / or the braking system in the towing vehicle and / or the trailer can be controlled in such a way that braking effect is increased when there is a maximum deviation between the actual state and the target state, and reduced when there is a minimal deviation. This allows for the targeted damping of oscillations or one-sided fishtailing of the trailer.

[0041] The invention will be explained in more detail below with the aid of figures. These show: Figs. 1a, 1b, 1c show different versions of a vehicle combination; Fig. 2 shows a detailed view of a front axle of a trailer according to the Figuren 1a, 1b ; and Fig. 3 a detailed view of a vehicle combination with a drawbar trailer according to Fig. 1a ; Fig. 4 a time course of a difference yaw rate or a difference track offset; and Fig. 5 a flowchart of the method according to the invention.

[0042] In the Figuren 1a,1b and 1c Different versions of a vehicle combination 100a, 100b, 100c, in particular a commercial vehicle combination, are shown, wherein the respective vehicle combination 100a, 100b, 100c consists of a towing vehicle 1a, 1b, 1c (tracting vehicle) and a trailer 2a, 2b, 2c (towed vehicle). The vehicle combination in Fig. 1a und Fig. 1b Each is a truck combination 100a, 100b consisting of a truck 1a, 1b as the towing vehicle and a drawbar trailer 2a or central axle trailer 2b as the trailer and in Fig. 1c a semi-trailer truck 100c consisting of a tractor unit 1c as the towing vehicle and a semi-trailer 2c as the trailer.

[0043] In the Fig. 1a und 1b The drawbar trailer 2a or central axle trailer 2b is connected to the truck 1a, 1b by a drawbar 3 via a trailer coupling 4. The drawbar 3, with a drawbar length DL, is in the version in Fig. 1a with a front axle 5A of the drawbar trailer 2a rigidly connected via a turntable, the front axle 5A being rotatably mounted on the drawbar trailer 2a at a third drawbar point P3, thus enabling steering of the drawbar trailer 2a. Fig. 1b The drawbar 3 is rigidly connected to the non-steerable ("front") axle 5A of the central axle trailer 2b. Fig. 1c The semi-trailer 2c is connected to the tractor unit 1c in the usual way via a kingpin 7 at a second towing point P2.

[0044] Thus, in the three depicted vehicle combinations 100a, 100b, 100c, a train consisting of N elements Yi, with i = 1, ..., N, is formed, wherein in Fig. 1a a three-part train (Y1: truck 1a - Y2: drawbar 3 with turntable - Y3: drawbar trailer 2a) and in the Figuren 1b and 1cA two-unit vehicle combination (Y1: truck 1b - Y2: center-axle trailer 2b or Y1: tractor unit 1c - Y2: semi-trailer 2c) is formed. The number N of units Yi of a vehicle combination 100a, 100b, 100c is defined by a number k of drawbar points Pi, where i = 1, ..., k, and the first drawbar point P1 is located, as shown in the figures, on the front axle 5Z of the respective tractor unit 1a, 1b, 1c. Drawbar points Pi are understood to be those points on the respective i-th unit Yi of the vehicle combination 100a, 100b, 100c over which the respective i-th unit Yi is pulled. For the towing vehicle 1a, 1b, 1c, it is specified that it is "pulled" by the steerable front axle 5Z, since the movement of the towing vehicle 1a, 1b, 1c is determined by the movement of the steerable front axle 5Z.

[0045] In principle, it is also possible for a vehicle combination 100a, 100b, 100c to have more than three links Yi (N>3) because the towing vehicle 1a, 1b, 1c, for example, pulls more than one trailer 2a, 2b, 2c, e.g., in the case of tractor combinations or extra-long trucks with several drawbar trailers 2b, 2c and / or dollies or a semi-trailer 2c with an additionally attached drawbar trailer 2b, 2c and / or with a dolly.

[0046] In order to detect unstable behavior IV, such as oscillation behavior O or one-sided skidding A, of the respective trailer 2a, 2b, 2c in such N-unit vehicle combinations 100a, 100b, 100c, it is planned to determine an actual driving dynamics parameter of the respective trailer 2a, 2b, 2c and to compare this with a modeled or estimated target driving dynamics parameter for the respective trailer 2a, 2b, 2c. Based on a comparison result E, conclusions can be drawn about unstable behavior IV, O, A of the respective trailer 2a, 2b, 2c. This is carried out as follows: First, signals S are exchanged between the trailer 2a, 2b, 2c and the towing vehicle 1a, 1b, 1c via a trailer interface 8. The signals S here particularly indicate the actual driving dynamic parameters of the respective trailer 2a, 2b, 2c determined in the respective trailer 2a, 2b, 2c or related parameters.

[0047] The actual driving dynamics parameters of the trailer 2a, 2b, 2c include, among others, the wheel speeds nAL, nAR of the wheels 5AL, 5AR of the front axle 5A and / or the wheel speeds nBL, nBR of the wheels 5BL, 5BR of the rear axle 5B (if present) of the respective trailer 2a, 2b, 2c. Furthermore, the actual driving dynamics parameter of the respective trailer 2a, 2b, 2c can also be understood as the trailer's actual yaw rate GAlst, which can be measured, for example, by a yaw rate sensor 9A in the trailer 2a, 2b, 2c.

[0048] The actual yaw rate GAlst of the trailer can also be determined from the wheel speeds nAL, nAR, nBL, nBR of the trailer 2a, 2b, 2c which are measured and transmitted anyway via wheel speed sensors 13A, 13B in the trailer 2a, 2b, 2c. It is assumed that the actual trailer yaw rate GAlst, taking corrections into account, results from a difference in the wheel speeds nBL, nBR of the wheels 5BL, 5BR of the non-driven, unbraked rear axle 5B of the drawbar trailer 2a, or the wheel speeds nAL, nAR of the wheels 5AL, 5AR of the non-driven, unbraked front axle 5A of the center-axle trailer 2b or the semi-trailer 2c, since these wheel speeds nAL, nAR, nBL, nBR differ when the trailer 2a, 2b, 2c yaws. From the equation GAIst = 2 πr nAR , nNBR − nAL , nBL SW Knowing the track width SW and the rolling radius r of each wheel 5AR, 5BR, 5AL, 5BL, the actual yaw rate GAlst of the trailer is calculated as a first approximation. To compensate for an induced error in the rolling radius r due to a lateral acceleration qa acting on the trailer 2a, 2b, 2c during yaw, the actual yaw rate GAlst of the trailer can be corrected by a linear lateral acceleration-dependent term, resulting in: GAIst , korr = 2 πr nAR , nNBR − nAL , nBL SW − C vz aq with a constant C that can be empirically or analytically determined in driving tests for different loads, the measurable vehicle speed vz and the measurable lateral acceleration aq. By determining the actual trailer yaw rate GAlst in this way, an additional yaw rate sensor 9A in the trailer 2a, 2b, 2c can be dispensed with.

[0049] From the aforementioned measured variables, further actual driving dynamics parameters of the trailer 2a, 2b, 2c can also be determined, for example a trailer actual track offset VAlst, which indicates the offset of the track of the trailer 2a, 2b, 2c relative to the track of the towing vehicle 1a, 1b, 1c (see Fig. 3 The track of the trailer 2a, 2b, 2c is given, for example, by a longitudinal center axis 15A of the trailer 2a, 2b, 2c, and the track of the towing vehicle 1a, 1b, 1c by a longitudinal center axis 15Z of the towing vehicle 1a, 1b, 1c. The actual trailer track offset VAlst can be determined, as described below, from the actual articulation angles γiIst, which exist between the individual links Yi and which result from the yaw rates Gi of the individual links Yi.

[0050] The signal transmission between the trailer 2a, 2b, 2c and the towing vehicle 1a, 1b, 1c can be carried out, for example, according to ISO 11992, if the respective trailer 2a, 2b, 2c has an electronic braking system and is equipped with its own CAN bus. According to ISO 11992, it is provided that, in particular, the wheel speeds nAL, nAR, nBL, nBR of the front axle 5A or the rear axle 5B (if present) of the respective trailer 2a, 2b, 2c, or a difference in the wheel speeds nAL, nAR, nBL, nBR of an axle 5A, 5B, are transmitted from the trailer 2a, 2b, 2c to the towing vehicle 1a, 1b, 1c via a trailer interface 8 and thus made available to a CAN bus 6 in the vehicle combination 100a, 100b, 100c.The wheel speeds nZL, nZR of the steerable front axle 5Z of the towing vehicle 1a, 1b, 1c, which are measured via wheel speed sensors 13Z in the towing vehicle 1a, 1b, 1c, can also be made available via the CAN bus 6 in order to determine a towing vehicle actual yaw rate GZIst as a driving dynamic actual parameter of the towing vehicle 1a, 1b, 1c.

[0051] In order to infer unstable behavior IV, O, A of trailer 2a, 2b, 2c from the actual driving dynamics parameters nAL, nAR, nBL, nBR, GAlst, VAlst of trailer 2a, 2b, 2c, a target driving dynamics for trailer 2a, 2b, 2c must be estimated and, depending on the actual driving dynamics parameters nAL, nAR, nBL, nBR, GAlst, VAlst of trailer 2a, 2b, 2c, a deviation between the actual state and the target state must be recorded and evaluated. This is done by referring to the actual driving dynamics parameters of the towing vehicle 1a, 1b, 1c and by applying a kinematic model M. The actual driving dynamics parameters of the towing vehicle 1a, 1b, 1c are understood to be, in particular, the actual yaw rate GZIst and a towing vehicle speed vZ, which are derived from the wheel speeds nZR, nZL of the wheels 5ZR, 5ZL of the towing vehicle 1a, 1b, 1c as described for the trailer 2a, 2b, 2c, or directly via a yaw rate sensor 9Z, e.g.of the stability system 11 (ESC), in the towing vehicle 1a, 1b, 1c or a speed sensor, and are provided via the CAN bus 6.

[0052] Using the kinematic model M, the current driving dynamics of the towing vehicle 1a, 1b, 1c, taking into account the geometry of the respective vehicle combination 100a, 100b, 100c, are used to estimate the expected driving dynamics for the trailer 2a, 2b, 2c, assuming stable driving behavior. The driving behavior of the towing vehicle 1a, 1b, 1c is thus extrapolated to the trailer 2a, 2b, 2c by considering the assumed stable motion (kinematics) of the N-unit vehicle combination 100a, 100b, 100c. This is done using the following equations of motion B1, B2, B3, where the index i ranges from 1 to the respective number N of links Yi of the respective vehicle combination 100a, 100b, 100c:

[0053] In these equations of motion B1, B2, B3, γi denotes a Fig. 2 The angle of articulation between the i-link Yi and the i + 1-link Y(i+1) is represented, Gi is the yaw rate of the i-link Yi, vi is the velocity of the i-link Yi, li is the distance between the point of application Pi of the i-link Yi and the non-steerable axis of the i-link Yi, and ui is the distance between the non-steerable axis of the i-link Yi and the subsequent point of application P(i+1). If two non-steerable axes are provided, an effective distance ui, li is to be assumed accordingly, which lies between these two axes. In the first equation of motion B1, a buckling angular velocity is considered in order to determine the buckling angle γi (see...). Fig. 2 ) to determine. The buckling angular velocity follows directly from the difference in the yaw rates Gi between the i-link Yi and the i + 1-link Y(i+1) and thus indicates the relative motion between the two links Yi, Y(i+1).

[0054] The aforementioned equations of motion B1, B2, B3 apply equally to the target and actual states of the individual links Yi of the respective vehicle combination 100a, 100b, 100c. Therefore, given knowledge of the respective variables, both actual and target values ​​for the articulation angle γi and / or the respective yaw rate Gi can be calculated.

[0055] The index i, which refers to the respective member Yi, is in the explanations of the Figuren 1a, 1b , 1c defined as follows: i = 1: Y1: Zugfahrzeug 1a, 1b, 1c (alle Ausführungen); i = 2: Y2: Drawbar 3 with turntable ( Fig. 1a ), central axle trailer 2b ( Fig. 1b ), semi-trailer 2c ( Fig. 1c ), i = 3 Y3: Deichselanhänger 2a ( Fig. 1a )

[0056] If the subsequent pull point P(i+1) of the i. member Yi lies as in Fig. 1a For the drawbar 3 with turntable (i=2) shown on the single axis (5A) of the second link Y2, or if there is no subsequent third draw point P3 for the third link Y3, then the quantities u2 and u3 are equal to zero. Accordingly, in Fig. 1c The size u3 is negative because the subsequent draw point P2 of the first link Y1 is located in front of the non-steerable rear axle of the semi-trailer 1c.

[0057] In the aforementioned equations of motion B1, B2, B3, the yaw rate G1 for the first element Y1 (i=1), i.e., for the respective towing vehicle 1a, 1b, 1c, is known or can be measured as described above using the yaw rate sensor 9Z in the towing vehicle 1a, 1b, 1c as the actual towing vehicle yaw rate GZIst, or calculated from the wheel speeds nZR, nZL of the towing vehicle 1a, 1b, 1c. Furthermore, the velocity v1 for the first element Y1 (i=1) is also given by the respective towing vehicle velocity vZ, which can be measured directly or determined from the wheel speeds nZR, nZL of the towing vehicle 1a, 1b, 1c. These quantities G1 and GZlst, v1 thus represent measured quantities in the above formulas, which are transmitted as actual driving dynamics parameters of the towing vehicle 1a, 1b, 1c, for example via the CAN bus 6.

[0058] Furthermore, geometric parameters gK of the vehicle combination 100a, 100b, 100c are also transmitted via CAN bus 6, from which the parameters li and ui can be determined for the equations of motion B1, B2, B3 described above. According to Fig. 1a For example, I1 is the wheelbase RSZ of the towing vehicle 1a, I2 the drawbar length DL, and I3 the wheelbase RSA of the drawbar trailer 2a. This applies accordingly to the center-axle trailer 2b and the semi-trailer 2c. All these parameters are transmitted from the trailer 2a, 2b, 2c to the CAN bus 6 in the towing vehicle 1a, 1b, 1c via the trailer interface 8, in accordance with ISO 11992, so that these geometric parameters gK can be accessed. The wheelbase RSZ of the towing vehicle 1a, 1b, 1c is stored in a control unit 11a of a stability control system 11 (ESC) of the towing vehicle 1a, 1b, 1c. The drawbar length DL and the wheelbase RSA of the drawbar trailer 2a and 2c, respectively, are stored in the control unit 11a of the stability control system 11 (ESC) of the towing vehicle 1a, 1b, 1c.The drawbar length DL of the central axle trailer 2b and other geometric parameters gK of the respective vehicle combination 100a, 100b, 100c can be obtained via the trailer interface 8 from the respective trailer 2a, 2b, 2c and provided via the CAN bus 6 to a central evaluation unit 10 in the towing vehicle 1a, 1b, 1c, provided that the trailer 2a, 2b, 2c has an electronic braking system with its own CAN bus.

[0059] The following applies to the sizes ui: Fig. 1a that u1 is equal to the distance from the rear axle of the towing vehicle 1a to the second towing point P2 or to the trailer coupling 4, u2 for the "drawbar link" 3 is zero, since the third towing point P3 is located in the front axle 5A (as shown), and u3 is also zero because there is no fourth towing point P4. These variables are all known for the respective vehicle combinations 100a, 100b, 100c and can therefore be used directly for the calculation.

[0060] The sizes or distances li and ui in the vehicle combinations 100b and 100c are comparable to those in the Figuren 1b and 1c represented, defined and derivable from the data transmitted via the CAN bus 6.

[0061] Thus, in the above equations of motion B1, B2, B3, all parameters are known, allowing them to be recalculated repeatedly for each time step dt. It is assumed that a stable state exists at the beginning of a journey or calculation, and therefore the initial yaw rates Gi for i>1, i.e., the yaw rates of the attached links Yi with i>1, correspond to the measurable yaw rate G1, GZIst of the towing vehicle 1a, 1b, 1c. Furthermore, it is assumed that the vehicle combination 100a, 100b, 100c has already moved stably over a certain distance with certain articulation angles γi between the individual links Yi, so that the trailer 2a, 2b, 2c follows the stable movement of the towing vehicle 1a, 1b, 1c.

[0062] Thus, the equations of motion B1, B2, B3 can be solved successively for each time step dt, and the yaw rate Gi of the respective i. link Yi, and therefore also of the respective trailer 2a, 2b, 2c, can be determined based on the type of vehicle combination 100a, 100b, 100c or the number N of links Yi in the train. Fig. 1a Therefore, the yaw rate G3 is derived from the equations of motion B1, B2, B3, and in the Figuren 1b and 1c The yaw rate G2 with the respective geometric parameters ui, li for this vehicle combination 100a, 100b, 100c can be determined from the above equation in a time-resolved manner.

[0063] These yaw rates G3 determined via this ( Fig. 1a ) or G2 ( Fig. 1b , 1cThe modeled or estimated target yaw rates GASoll for the respective trailers 2a, 2b, 2c are then represented, indicating the target behavior of the trailers 2a, 2b, 2c assuming stable driving. To determine unstable behavior IV, e.g., oscillation behavior O or a single lateral stall A of the trailers 2a, 2b, 2c, as per the original question, the actual state must be compared with the modeled target state, which can be done in different ways:

[0064] To determine an oscillation behavior O, a time-resolved difference yaw rate dG = GASoll - GAlst can be easily calculated, which indicates the deviation of the actual trailer motion from the modeled trailer motion. If this difference yaw rate dG exhibits, as in Fig. 4 As an example, if a time-oscillating behavior is shown, the trailer 2a, 2b, 2c moves relative to the towing vehicle 1a, 1b, 1c in an oscillating motion, i.e., the trailer 2a, 2b, 2c begins to sway. As a comparative result E, the presence of an oscillating behavior O of the trailer 2a, 2b, 2c can thus be concluded if the oscillating difference yaw rate dG repeatedly reaches or exceeds a predetermined difference yaw rate amplitude dGA, which in Fig. 4 for example, it can be assumed after the second minimum point T.

[0065] The predetermined difference yaw rate amplitude dGA is preferably chosen such that non-critical oscillations are permitted. For example, an oscillation or a recurring maximum deviation of the actual trailer yaw rate GAlst from the modeled target trailer yaw rate GASoll can be permitted if this results in an actual trailer track offset VAlst deviating by no more than a maximum track offset VMax of, for example, 25 cm from a target trailer track offset VASoll, assuming driving on a country road.

[0066] The determination of the track offsets VAlst, VASoll is described below using the following: Fig. 3 The example of the drawbar trailer 2a is explained. Fig. 3 It is assumed that the towing vehicle 1a moves straight ahead along its longitudinal center axis 15Z, so that a target trailer track offset VASoll, i.e., the expected track offset for the trailer 2a, lies on this longitudinal center axis 15Z of the towing vehicle 1a. The actual trailer track offset VAlst can be determined by the equation VAIst = I 3 × sin γ 1 Ist + γ 2 Ist + DL × sin γ 1 Ist The formula describes the actual articulation angles γ1Ist and γ2Ist, where the current actual values ​​are used. These values ​​are derived from the first equation of motion B1, as described above, depending on the other actual vehicle dynamic parameters nAR, nAL, nBR, nBL, GAlst, and GZlst for the drawbar 3, the trailer 2a, and the towing vehicle 1a. The sum γ1Ist and γ2Ist corresponds to the total actual articulation angle ΓIst between the trailer 2a and the towing vehicle 1a. The target trailer track offset VASoll is also derived from the formula. VASoll = I 3 × sin γ 1 Soll + γ 2 Soll + DL × sin γ 1 Soll , where the target articulation angles γ1Target and γ2Target are now represented by the modeled target values ​​derived from the first equation of motion B1 via the kinematic model M. The sum γ1Target + γ2Target in this case accordingly gives a target total articulation angle ΓTarget between the drawbar trailer 2a and the towing vehicle 1a, which is determined in the Fig. 3 The depicted case of driving straight ahead corresponds to zero.

[0067] Thus, by setting the maximum track offset VMax, a safety-neutral oscillation can be permitted or accepted, which can also be set independently of the geometric dimensions of the vehicle combination 100a, 100b, 100c and thereby avoids endangering surrounding traffic, since a lateral deviation of 25 cm of a vehicle combination approximately 2.5 m wide is possible on a roadway width of, for example, 3 m without affecting oncoming traffic. Alternatively, the actual track offset of the trailer VAlst, as a driving dynamics parameter of the trailer 2a, 2b, 2c, can also be used directly to determine the oscillation behavior O of the trailer 2a, 2b, 2c. Accordingly, the actual track offset of the trailer VAlst can be compared with the target track offset of the trailer VASoll.For comparison purposes, a differential track offset dV = VASoll - VAlst can be calculated, which indicates the deviation of the actual track offset from the modeled track offset. Thus, a comparison between an actual state and a target state takes place here as well. From a time-oscillating curve (see...) Fig. 4 Here too, an oscillation behavior O of the drawbar trailer 2a can be inferred if the oscillating differential track offset dV exceeds a differential track offset amplitude dVA. As with the consideration of the differential yaw rate amplitude dVA, this can take into account that a certain degree of oscillation is permitted, for example, with less than a recurring maximum track offset VMax of the drawbar trailer 2a of, say, 25 cm.

[0068] Additionally or alternatively, a lateral slide A can also be detected from the differential track offset dV if, for example, it is determined once that the differential track offset dV exceeds a limit differential track offset dVG. In such a case, it can be assumed that the trailer 2a will slide laterally, since a return to the trailer's target track offset VASoll is unlikely with constant driving dynamics.

[0069] Thus, by considering the actual driving dynamics parameters of the trailer 2a, 2b, 2c, an actual state can be defined and compared with a target state modeled from the movement of the towing vehicle 1a, 1b, 1c. This is done using the equations of motion B1, B2, B3 as a function of the relative movement between the individual links Yi of the train, thereby monitoring the temporal behavior of the train and allowing conclusions to be drawn about an unstable state.

[0070] In principle, the above considerations and equations of motion B1, B2, B3 can also be carried out as a function of the lateral acceleration instead of the yaw rate, which can also be used to describe the lateral dynamics of the trailer 2a, 2b, 2c. However, a more reliable statement about the stability of the trailer can be made from the yaw rate.

[0071] The calculations and comparisons described above can be performed centrally, for example, in an evaluation unit 10 in the respective towing vehicle 1a, 1b, 1c, which has access to the CAN bus 6. The evaluation unit 10 can, for example, be part of the control unit 11a of the stability system 11 (ESC) in the towing vehicle 1a, 1b, 1c, thus extending the functionality of the stability system 11. This can be achieved through hardware and / or software modifications that allow access to the signals S transmitted via the trailer interface 8 according to ISO 11992 and their additional inclusion in the evaluation according to the procedure described above. The evaluation unit 10 is therefore implemented in software and / or hardware.

[0072] As part of the stability system 11 in the towing vehicle 1a, 1b, 1c, the evaluation unit 10 is therefore designed to assess the stability of the vehicle combination 100a, 100b, 100c. For this purpose, the sizes or distances ui and li are determined from the geometric parameters gK, RSZ, RSA, DL, ... read in via the CAN bus 6, and the comparison result E is determined using the kinematic model M with the transmitted or determined actual vehicle dynamic parameters nAL, nAR, nBL, nBR, GAlst, VAlst of the trailer 2a, 2b, 2c and the measured actual vehicle dynamic parameters nZR, nZL, GZlst, vZ of the towing vehicle 1a, 1b, 1c. This result indicates whether there is a deviation of the actual state from the target state for the trailer 2a, 2b, 2c and whether this deviation leads to unstable behavior IV, O, A of the trailer 2a, 2b, 2c.

[0073] Depending on the comparison result E, the detected oscillation behavior O, or the lateral skidding A of the trailer 2a, 2b, 2c, a corresponding action can then be carried out as part of an assistance function. This can, for example, take the form of a warning signal W to the driver, allowing them to intervene manually by means of a steering maneuver and / or a braking maneuver and / or a drive maneuver, e.g., reducing engine torque. Alternatively or additionally, the evaluation unit 10 or the control unit 11a can also issue automated control commands to a drive system 20 and / or a braking system 30A, 30Z and / or a steering system 40A, 40Z of the towing vehicle 1a, 1b, 1c and / or the trailer 2a, 2b, 2c, so that automated intervention can be carried out to counteract the oscillation behavior O or the lateral skidding A.For the sake of clarity, the drive system 20, the brake system 30A, 30Z and the steering system 40A, 40Z are only for the vehicle combination 100a in . Fig. 1a These can also be shown in Fig. 1b and Fig. 1c be provided for in an appropriate manner.

[0074] The automated intervention can be achieved, for example, by the evaluation unit 10 and / or the control unit 11a controlling the drive system 20 to reduce the motor torque and / or the braking system 30A, 30Z to initiate braking in such a way that the vehicle combination 100a, 100b, 100c is braked in pulses. This allows for pulsed stretch braking to be effected if oscillation behavior O is present, thereby selectively stretching the train and thus reducing the oscillations. The pulsed braking intervention via the drive system 20 and / or the braking system 30A, 30Z can be controlled by a system depending on the currently existing differential yaw rate dG and / or the differential track offset dV. For example, a braking effect BW can always be increased via the drive system 20 and / or the braking system 30A, 30Z when the differential yaw rate dG and / or the differential tracking error dV approaches a high point H orAs the differential yaw rate dG and / or differential tracking error dV approach a zero crossing X, a reduction in braking effect BW via the drive system 20 and / or the braking system 30A, 30Z occurs. This allows the oscillation behavior O to be efficiently dampened. Targeted control intervention is also possible in the event of a one-sided lateral breakaway A, when the differential tracking error dV approaches the predetermined limit differential tracking error dVG, in order to counteract it.

[0075] According to the invention, Fig. 3For example, it is planned that, after an initialization step St0, the actual vehicle dynamics parameters nAL, nAR, nBL, nBR, GAlst (C, r), VAlst of the trailer 2a, 2b, 2c are determined in a first step St1 with time resolution, as described above, whereby these characterize the current vehicle dynamics state of the trailer 2a, 2b, 2c. The actual trailer track offset VAlst is derived from the measured or determined actual values ​​of the yaw rates Gi for each link Yi of the vehicle combination 100a, 100b, 100c, as well as the determined actual articulation angles γiIst between the respective links Yi.

[0076] Subsequently, in a second step St2, the actual vehicle dynamic parameters nZL, nZR, GZlst, vZ of the towing vehicle 1a, 1b, 1c are determined, and in a third step St3, a target vehicle dynamic parameter GASoll, VZSoll for the trailer 2a, 2b, 2c is modeled or determined from these parameters, taking into account the kinematic model M as described above, using the equations of motion B1, B2, B3. This is done considering the geometric parameters gK, RSA, RSZ, DL of the vehicle combination 100a, 100b, 100c, from which the distances ui, li are derived. Target values ​​for the yaw rates Gi and target articulation angles γiSoll for the respective links Yi are determined from the kinematic model M.

[0077] In a fourth step (St4), a comparison is made between the actual vehicle dynamics parameter, preferably the actual trailer yaw rate GAlst and / or the actual trailer track offset VAlst, and the target vehicle dynamics parameter, preferably the target trailer yaw rate GASoll and / or the target trailer track offset VASoll, by, for example, calculating a difference yaw rate dG or a difference track offset dV. This allows the deviation between the behavior modeled by the towing vehicle 1a, 1b, 1c on the respective trailer 2a, 2b, 2c and the actual behavior of the respective trailer 2a, 2b, 2c to be determined. In the case of a time-oscillating deviation or a time-oscillating difference dG, dV with at least a defined difference amplitude dVA, dGA, the comparison result E is to conclude an oscillation behavior O and in the case of exceeding a limit difference track offset dVG to a lateral breakaway A of the trailer 2a, 2b, 2c.The difference amplitudes dVA, dGA can be determined taking into account a permissible maximum track offset VMax between the towing vehicle 1a, 1b, 1c and the respective trailer 2a, 2b, 2c.

[0078] In the event of any unstable behavior IV, O, A, a warning W can subsequently be issued to initiate manual intervention, and / or automated intervention in the driving operation can be carried out to counteract the unstable behavior IV, O, A, for example by selectively adjusting a braking effect BW by controlling the drive system 20 to reduce the motor torque and / or the brake system 30A, 30Z to effect braking. Reference symbol list (part of the description)

[0079] 1a, 1b Truck (tractor unit of 100a, 100b) 1c Tractor unit (tractor unit of 100c) 2a Drawbar trailer 2b Center axle trailer 2c Semi-trailer 3 Drawbar with turntable 4 Trailer coupling 5A Front axle of the trailer 5AL Left wheel of the front axle 5A 5AR Right wheel of the front axle 5A 5B Rear axle of the trailer 5BL Left wheel of the rear axle 5B 5BR Right wheel of the rear axle 5B 5Z Front axle of the towing vehicle 5ZL Left wheel of the front axle 5Z 5ZR Right wheel of the front axle 5Z 6 CAN bus 7 Kingpin 8 Trailer interface 9Z Yaw rate sensor in the towing vehicle 9AG Yaw rate sensor in the trailer 10 Evaluation unit 11 Electronic Stability Control (ESC) 11a Control unit of the Stability system 13A, 13B Wheel speed sensors in trailer 13Z Wheel speed sensors in towing vehicle 15A Longitudinal center axle of trailer 15B Longitudinal center axle of towing vehicle 20 Drive 30A Brake system of trailer 30Z Brake system of towing vehicle 40A Steering system of trailer 40Z Steering system of towing vehicle 100A,100b truck 100c semi-trailer truck , Lateral breakaway aq Lateral acceleration B1 First equation of motion B2 Second equation of motion B3 Third equation of motion BW Braking effect C Constant dt Time step dGA Difference yaw rate amplitude dG Difference yaw rate dVA Difference track offset amplitude dVD Difference track offset dVG Limit difference track offset DL Drawbar length E Comparison result F Amplitude γii Articulation angle γi Actual articulation angle γi Target articulation angle Γ Actual Actual Total articulation angle Γ Target Target Total articulation angle GAI Actual Trailer Actual Yaw Rate GAS Target Trailer Target Yaw Rate GZI Actual Tractor Actual Yaw Rate Gi Yaw rate of i.Link Yi gK geometric parameter H high point i index IV unstable behavior k number of draw points li, ui distances M kinematic model NA number of links Yi nAL, nAR wheel speeds on the front axle 5A of the trailer nBL, nBR wheel speeds on the rear axle 5A of the trailer nZL, nZR wheel speeds on the towing vehicle OO oscillation behavior Pi draw point i r rolling radius of the wheel RSZ wheelbase of the towing vehicle 1a, 1b, 1c RSA wheelbase of the trailer 1a, 1b, 1c S signal SWS track width T low point vi speed of the i. link Yi VAI actual trailer track offset VA target trailer track offset VMax maximum track offset vZ towing vehicle speed WWar warning X zero crossing Y.ii. Link Y1 first link Y2 second link Y3 third link St0, St1, St2, St3, St4 Steps of the procedure.

Claims

1. Method for determining unstable behavior (IV, A, O) of a trailer (2a, 2b, 2c) of a vehicle combination (100a, 100b, 100c), the vehicle combination (100a, 100b, 100c) being designed so as to be N-membered and one of the members (Yi, i=1,...,N) being formed by a towing vehicle (1a, 1b, 1c) and at least one further member (Yi) being formed by the trailer (2a, 2b, 2c), the unstable behavior (IV, A, O) of the trailer (2a, 2b, 2c) being determined depending on an actual driving dynamics parameter (nZL, nZR, GZlst, vZ) of the towing vehicle (1a, 1b, 1c), characterized by at least the following steps: - determining at least one actual driving dynamics parameter (nAL, nAR, nBL, nBR, GAlst, VAlst) of the trailer (2a, 2b, 2c) in the form of an actual track offset (VAlst), the actual driving dynamics parameter characterizing the current driving dynamics state of the trailer (2a, 2b, 2c) and resulting depending on a measurement via at least one sensor (9A, 13A) in the trailer (2a, 2b, 2c) (St1); - determining at least one target driving dynamics parameter (GASoll, VASoll) of the trailer (2a, 2b, 2c) in the form of a target track offset (VASoll) (St3), the target driving dynamics parameter resulting from the actual driving dynamics parameters (nZL, nZR, GZlst, vZ) of the towing vehicle (1a, 1b, 1c) by applying a kinematic model (M) depending on geometric parameters (gK, RSA, RSZ, DL) of the vehicle combination (100a, 100b, 100c) (St2); and - comparing the at least one actual driving dynamics parameter (GAlst, VAlst) in the form of the actual track offset (VAlst) of the relevant trailer (2a, 2b, 2c) with the at least one target driving dynamics parameter (GASoll, VASoll) in the form of the target track offset (VASoll) of the relevant trailer (2a, 2b, 2c) determined via the kinematic model (M), the presence of unstable behavior (IV, A, O) of the relevant trailer (2a, 2b, 2c) being deduced if the at least one actual driving dynamics parameter (GAlst, VAlst) in the form of the actual track offset (VAlst) of the relevant trailer (2a, 2b, 2c) deviates from the at least one determined target driving dynamics parameter (GASoll, VASoll) in the form of the target track offset (VASoll) of the relevant trailer (2a, 2b, 2c) by a specified reference value (dVG, dVA, dGA) in the form of a maximum track offset (Vmax) (St4); - issuing a warning to the driver of the vehicle combination if the deviation exceeds the specified maximum track offset (VMax).

2. Method according to claim 1, characterized in that, for the comparison, a differential (dG, dV) is established between the at least one actual driving dynamics parameter (GAlst, VAlst) of the trailer (2a, 2b, 2c) and the at least one target driving dynamics parameter (GASoll, GASoll) of the trailer (2a, 2b, 2c).

3. Method according to claim 1 or claim 2, characterized in that a target trailer yaw rate (GASoll) is determined via the kinematic model (M) as a target driving dynamics parameter of the trailer (2a, 2b, 2c), and an actual trailer yaw rate (GAlst) is determined depending on a measurement via sensors (9A, 13A) in the trailer (2a, 2b, 2c) as an actual driving dynamics parameter of the trailer (2a, 2b, 2c),.

4. Method according to claim 3, characterized in that from a temporally oscillating deviation of the actual trailer yaw rate (GAlst) and / or of the actual trailer track offset (VAlst) from the target trailer yaw rate (GASoll) and / or the target trailer track offset (VASoll) by at least one recurring differential yaw rate amplitude (dGA) and / or by one differential track offset amplitude (dVA), the presence of oscillation behavior (O) of the trailer (2a, 2b, 2c) as unstable behavior (IV) is deduced.

5. Method according to claim 4, characterized in that in order to specify the differential yaw rate amplitude (dGA) and / or a differential track offset amplitude (dVA), a permissible maximum track offset (VMax) of in particular 25 cm is taken into account.

6. Method according to any of claims 3 to 5, characterized in that from an actual trailer track offset (VAlst) which deviates non-recurrently from the target trailer track offset (VASoll) by at least one predetermined limit differential track offset (dVG), a lateral breakaway (A) of the trailer (2a, 2b, 2c) as unstable behavior (IV) is deduced.

7. Method according to any of claims 3 to 6, characterized in that the actual trailer yaw rate (GAlst) is determined depending on wheel speeds (nAR, nAL, nBR, nBL) of the trailer (2a, 2b, 2c) measured via wheel speed sensors (13A, 13B) in the trailer (2a, 2b, 2c) and / or depending on a direct measurement of the actual trailer yaw rate (GAlst) via a yaw rate sensor (9A) in the trailer (2a, 2b, 2c).

8. Method according to claim 7, characterized in that the wheel speeds (nAR, nAL, nBR, nBL) measured in the trailer (2a, 2b, 2c) and / or the actual trailer yaw rate (GAlst) measured directly in the trailer (2a, 2b, 2c) are transmitted to the towing vehicle (1a, 1b, 1c) via a trailer interface (8), preferably in accordance with ISO 11992, and are provided by the trailer interface (8) to a CAN bus (6) in the towing vehicle (1a, 1b, 1c).

9. Method according to any of claims 3 to 8, characterized in that the actual trailer track offset (VAlst) results from an actual total articulation angle (ΓIst) between the trailer (2a, 2b, 2c) and the towing vehicle (1a, 1b, 1c).

10. Method according to claim 9, characterized in that the actual total articulation angle (ΓIst) results from a sum of actual articulation angles (γiIst) between the members (Yi) of the vehicle combination (100a, 100b, 100c), the actual articulation angle (γiIst) being determined depending on wheel speeds (nAR, nAL, nBR, nBL) of the trailer (2a, 2b, 2c) measured via wheel speed sensors (13A, 13B) in the trailer (2a, 2b, 2c) and / or depending on a direct measurement of the actual trailer yaw rate (GAlst) via a yaw rate sensor (9A) in the trailer (2a, 2b, 2c) and depending on the actual towing vehicle yaw rate (GZIst).

11. Method according to any of claims 3 to 10, characterized in that the target trailer track offset (VASoll) results from a target total articulation angle (ΓSoll) modeled via the kinematic model (M) from the actual driving dynamics parameters (nZL, nZR, GZlst, vZ) of the towing vehicle (1a, 1b, 1c), the target total articulation angle (ΓSoll) resulting from a sum of modeled target articulation angles (γiSoll) between the members (Yi) of the vehicle combination (100a, 100b, 100c).

12. Method according to any of the preceding claims, characterized in that the determination of the at least one target driving dynamics parameter (GASoll, VASoll) of the trailer (2a, 2b, 2c) by applying the kinematic model (M) and the comparison of the at least one actual driving dynamics parameter (GAlst, VAlst) of the relevant trailer (2a, 2b, 2c) with the at least one target driving dynamics parameter (GASoll, VASoll) of the relevant trailer (2a, 2b, 2c) determined via the kinematic model (M) takes place in the towing vehicle (1a, 1b, 1c).

13. Method according to claim 12, characterized in that at least some of the geometric parameters (gK, RSA, DL) of the vehicle combination (100a, 100b, 100c) are transmitted to the towing vehicle (1a, 1b, 1c) via a trailer interface (8), preferably in accordance with ISO 11992, and are provided by the trailer interface (8) to a CAN bus (6) in the towing vehicle (1a, 1b, 1c).

14. Method according to any of the preceding claims, characterized in that a drawbar length (DL) of the trailer (2a, 2b) and / or a wheelbase of the towing vehicle (RSZ) and / or a wheelbase of the trailer (RSA) are used as geometric parameters (gK) of the vehicle combination (100a, 100b, 100c) for the kinematic model (M).

15. Method according to any of the preceding claims, characterized in that the driving dynamics of the towing vehicle (1a, 1b, 1c) which are characterized by the actual driving dynamics parameters (nZL, nZR, GZlst, vZ) are modeled by the kinematic model (M) on the relevant trailer (2a, 2b, 2c), taking into account the geometric parameters (gK, RSA, RSZ, DL) of the vehicle combination (100a, 100b, 100c).

16. Method according to any of the preceding claims, characterized in that the application of the kinematic model (M) includes a time-resolved (dt) successive solution of motion equations (B1, B2, B3) for each member (Yi) of the vehicle combination (100a, 100b, 100c), each member (Yi) having a tow point (Pi;i=1,... ,k) and adjacent members (Yi) being rotatably connected to one another via the tow points (Pi), a yaw rate (Gi) being estimated for each member (Yi) of the vehicle combination (100a, 100b, 100c)from the at least one actual driving dynamics parameter (nZL, nZR, GZlst, vZ) of the towing vehicle (1a, 1b, 1c) via the motion equations (B1, B2, B3) and the at least one target driving dynamics parameter (GASoll, VASoll) of the relevant trailer (2a, 2b, 2c) being specified depending on the estimated yaw rates (Gi) of the members (Yi) of the vehicle combination (100a, 100b, 100c).

17. Method according to claim 16, characterized in that the motion equations (B1, B2, B3) depend on the geometric parameters (gK, RSA, RSZ, DL).

18. Evaluation unit (10) for carrying out a method according to any of the preceding claims, the evaluation unit (10) being designed to read in the geometric parameters (gK, RSA, RSZ, DL) and the actual driving dynamics parameters (nZL, nZR, GZlst, vZ) of the towing vehicle (1a, 1b, 1c) and to determine the at least one target driving dynamics parameter (GASoll, VASoll) in the form of a target track offset (VASoll) of the relevant trailer (2a, 2b, 2c) via the kinematic model (M) and to compare the at least one target driving dynamics parameter with the at least one actual driving dynamics parameter (GAlst, VAlst) in the form of an actual track offset (VAlst) of the trailer (2a, 2b, 2c), and, depending thereon, to deduce unstable behavior (IV, O, A) of the relevant trailer (2a, 2b, 2c) and to issue a warning to the driver of the vehicle combination if the comparison shows that the actual track offset (VAlst) of the trailer (2a, 2b, 2c) deviates from the target track offset (VASoll) of the trailer (2a, 2b, 2c) by more than the specified maximum track offset (VMax) (St4).

19. Evaluation unit (10) according to claim 18, characterized in that the evaluation unit (10) is arranged in the towing vehicle (1a, 1b, 1c), preferably in a control unit (11a) of a stability system (11).

20. Vehicle combination (100a, 100b, 100c) consisting of a towing vehicle (1a, 1b, 1c) and at least one trailer (2a, 2b, 2c), the combination comprising an evaluation device (10) according to claim 18 or claim 19.

21. Vehicle combination (100a, 100b, 100c) according to claim 20, characterized in that at least some of the geometric parameters (gK, RSA, DL) and at least one actual driving dynamics parameter (nAL, nAR, nBL, nBR, GAlst) of the trailer (2a, 2b, 2c) measured in the trailer (2a, 2b, 2c) can be transmitted from the trailer (2a, 2b, 2c) to the evaluation device (10) in the towing vehicle (1a, 1b, 1c) via a trailer interface (8) in order to determine unstable behavior (IV, O, A) of the trailer (2a, 2b, 2c) starting from the towing vehicle (1).

22. Method for stabilizing a trailer (2a, 2b, 2c) of an N-membered vehicle combination (100a, 100b, 100c), in particular according to claim 20 or claim 21, consisting of a towing vehicle (1a, 1b, 1c) and at least one trailer (2a, 2b, 2c), the method comprising at least the following steps: - determining unstable behavior (IV, A, O) of the at least one trailer (2a, 2b, 2c) in a method according to any of claims 1 to 17, and - controlling a drive system (20) and / or a braking system (30A, 30Z) and / or a steering system (40A, 40B) in the towing vehicle (1a, 1b, 1c) and / or in the trailer (2a, 2b, 2c) when there is unstable behavior (IV, A, O) of the at least one trailer (2a, 2b, 2c) in order to stabilize the trailer (2a, 2b, 2c), the control being carried out depending on the determined deviation (dG, dV) of the at least one actual driving dynamics parameter (GAlst, VAlst) in the form of the actual track offset (VAlst) of the relevant trailer (2a, 2b, 2c) from the at least one determined target driving dynamics parameter (GASoll, VASoll) in the form of the target track offset (VASoll) of the relevant trailer (2a, 2b, 2c), and - issuing a warning (W) to the driver in order to effect manual intervention via the drive system (20) and / or the braking system (30A, 30Z) and / or the steering system (40A, 40B) in the towing vehicle (1a, 1b, 1c) and / or in the trailer (2a, 2b, 2c) if the actual track offset (VAlst) of the trailer (2a, 2b, 2c) deviates from the target track offset (VASoll) of the trailer (2a, 2b, 2c) by more than the specified maximum track offset (VMax) (St4).

23. Method according to claim 22, characterized in that the drive system (20) and / or the braking system (30A, 30Z) in the towing vehicle (1a, 1b, 1c) and / or in the trailer (2a, 2b, 2c) are controlled in such a way that in the case of a maximum deviation (H, T) of the actual driving dynamics parameter (GAlst, VAlst) from the target driving dynamics parameter (GASoll, VASoll), a braking effect (BW) is increased and in the case of a minimum deviation (X), a braking effect (BW) is reduced.