Procedure for monitoring driving conditions

Abstract

Method for estimating the sideslip angle for use in a control system influencing the driving state in a vehicle, comprising the following method steps: - Measuring vehicle accelerations (a x , a y , a z ) in all three spatial directions and the rotation rates (ω x , ω y , ω z ) at least around the vehicle’s longitudinal and vertical axis, - Estimating the vehicle speed (v x,e ) as a support variable in at least one spatial direction as a function of vehicle state variables (v w,i , δ S , ω z ), - Calculate vehicle speeds (v x , v y , v z ) in all three spatial directions and the attitude angle (φ, θ, ψ) at least around the vehicle's longitudinal and transverse axis by integrating a system of differential equations from the measured vehicle accelerations (a x , a y , a z ) and rotation rates (ωx , w y , ω z ), where the support size (v x,e ) to correct the corresponding calculated vehicle speed (v x ) in at least one spatial direction, - Calculate the sideslip angle (β) from the ratio of the calculated vehicle lateral speed (v y ) to the vehicle's longitudinal speed (v x ), where the difference between support size (v x,e ) and calculated vehicle speed (v x ) in a Kalman filter to calculate a corrected vehicle speed (v x ) which is taken into account when integrating the system of differential equations.

Description

[0001] The invention relates to a method for monitoring driving conditions for use in a control system influencing the driving condition in a vehicle. State of the art

[0002] DE 102 47 991 A1 describes a method for determining the slip angle of a motor vehicle, which is the angle between the vehicle speed at the center of gravity and the vehicle's longitudinal axis. According to this method, the direction of the speed at the center of gravity of the motor vehicle is determined by frequency analysis of the signals received by a GPS receiver located in the vehicle. The current yaw rate of the motor vehicle is determined using a yaw rate sensor. The slip angle is calculated by time integration of the difference between the yaw rate and the angular velocity of the vehicle, which is determined from the speed direction determined by the GPS receiver.

[0003] Implementing the method requires a GPS receiver in the vehicle. It should also be noted that the sideslip angle must be determined with a relatively high degree of accuracy to be considered for use in a vehicle control system that can influence the current vehicle condition, such as active chassis systems, steering systems, or braking systems.

[0004] US 2005 / 0 080 542 A1 describes how to determine the sideslip angle from the ratio of lateral to longitudinal speed. The longitudinal speed is determined from wheel speeds, while the lateral speed is determined from wheel speeds and vehicle acceleration.

[0005] DE 10 2006 026 937 A1 discloses the measurement of vehicle accelerations and yaw rates, the estimation of the vehicle speed on the basis of wheel speed sensors and the calculation of the vehicle speed by integrating the measured vehicle accelerations, whereby the estimated vehicle speed is used as a correction variable. Disclosure of the invention

[0006] The invention is based on the object of determining the sideslip angle in a vehicle with high estimation accuracy using simple measures. According to an advantageous embodiment, the sensors of an electronic stability program (ESP) are to be included.

[0007] This object is achieved according to the invention with the features of claim 1. The subclaims specify expedient further developments.

[0008] In the method according to the invention for determining the sideslip angle, various vehicle state variables are measured using a vehicle's own sensor system, which is at least partially present in vehicles with ESP. These variables are then further used in a differential equation system. The advantage of this approach is that existing sensor platforms in the vehicle can be used, and their signals are evaluated in a novel way.

[0009] The method is preferably implemented in a control unit in the motor vehicle, whose control signals are fed to a vehicle control system to achieve the desired setting and bring about a change in the current vehicle state. The vehicle control system may be, for example, active steering or braking systems or active chassis systems.

[0010] In the method according to the invention, the translational vehicle accelerations in all three spatial directions and also the yaw rates at least around the vehicle's longitudinal and vertical axes are measured. These measurements are preferably performed using the sensor platform belonging to the ESP system. For example, a 5D sensor system is used, which is part of the ESP system. The 5D sensor system provides the translational vehicle accelerations as well as the yaw rates around the vehicle's longitudinal and vertical axes. If necessary, the sensor system also provides the yaw rate around the transverse axis; in this case, it is a 6D sensor system.

[0011] In the next step, a vehicle speed is estimated, which is then used as a reference variable to correct a calculated vehicle speed. The vehicle speed, which serves as a reference variable, is determined in at least one spatial direction as a function of other vehicle state variables.

[0012] In the next step, the corresponding derivatives are calculated by integrating a system of differential equations, taking into account the measured vehicle accelerations and the rotation rates at least around the vehicle's longitudinal and vertical axes. This includes the vehicle speeds in all three spatial directions, as well as the roll and pitch angles, i.e., the attitude angles around the vehicle's longitudinal and transverse axes, and, if applicable, also the yaw angle. The system of differential equations is specifically designed as a kinematic system of differential equations, so that knowledge of kinematic parameters is advantageously sufficient for setting up and solving the system of differential equations, and parameters or vehicle parameters such as masses, moments of inertia, or tire stiffness are not absolutely necessary.The numerical solution of the differential equation system is initially a so-called open integration, which is carried out continuously and continuously during vehicle operation.

[0013] The vehicle speed reference variable is then used to correct at least one vehicle speed component calculated from the system of differential equations. This results in a correction of the vehicle state variables obtained by integration, based on a reference variable that can be determined with high accuracy, in particular the vehicle's longitudinal speed. The result is a stabilized solution of the system of differential equations.

[0014] In the final step, the sideslip angle is calculated from the ratio of the calculated vehicle lateral speed to the vehicle longitudinal speed. This angle is then used to adjust or control the vehicle positioning system. The sideslip angle is calculated, in particular, from the arctangent of the ratio of the vehicle's lateral speed to the vehicle longitudinal speed.

[0015] Preferably, the vehicle's longitudinal speed is used as the reference variable, although other vehicle speed components along the transverse and / or vertical axis may also be used if necessary. The vehicle's longitudinal speed can be determined with a high degree of certainty and the required accuracy, making this variable particularly suitable as a reference variable for correcting the results from the solution of the differential equation system.

[0016] The vehicle's longitudinal speed can be determined from the wheel speeds, possibly also taking into account the steering angle and the rotational speed. For the vehicle's longitudinal speed reference variable, the wheel speeds are determined from speed sensor signals. To also correct the vehicle's longitudinal speed, which is intended as a reference variable, it may be advantageous to perform a comparison with the vehicle speed obtained from a signal from a Global Positioning System (GPS). Consideration of the steering angle and the rotational speed for determining the vehicle's longitudinal speed reference variable can be omitted if the vehicle is traveling straight ahead.

[0017] The method uses a Kalman filter, in which the difference between the vehicle speed reference variable and the calculated vehicle speed is incorporated into a recursive algorithm to determine a corrected vehicle speed, which is then taken into account when integrating the system of differential equations. The application of the Kalman filter, i.e. a stochastic state estimator for dynamic systems, is state of the art and is described in detail in the literature. The values provided by the Kalman filter serve to support the numerical solution of the system of differential equations, which can be problematic without such support due to the errors contained in the sensor signals, such as offset and noise, and the boundary-stable nature of the differential equations. The Kalman filter stabilizes the numerical solution.

[0018] To support the six state variables of the Kalman filter—the longitudinal, vertical, and high speeds, as well as the pitch, roll, and yaw angles—the calculated vehicle speed support variable is used, particularly the longitudinal vehicle speed. The Kalman state and the vehicle speeds calculated in the kinematic differential equation system are supported by weighting the speed support variable, with the weighting factor being adjustable.

[0019] As a beneficial effect, in addition to determining the support size, the algorithm also outputs the number of stable running wheels, which serves as a criterion for the reliability and weighting of this support size compared to other support sizes in the Kalman feedback.

[0020] It can be advantageous to formulate boundary conditions for certain vehicle states that contribute to stabilizing the solution of the differential equation system. For example, when driving straight ahead, the yaw rate is very small, with the result that the lateral velocity is not coupled with the derivative of the longitudinal velocity. This means that the observability of errors in the Kalman filter state is not guaranteed and support for the solution for the lateral velocity cannot be guaranteed. Therefore, when driving straight ahead, the vehicle lateral velocity, which serves as a reference variable, is expediently set to zero. If the vehicle is not driving straight ahead, the vehicle lateral velocity reference variable can be calculated from a vehicle model if necessary, provided that a rotation rate, in particular the yaw rate, is still below an assigned limit value and the model is still valid.

[0021] The vehicle speed in the direction of the vehicle's vertical axis can be used as a further supporting variable, which is set to a constant zero and is included as a weighted variable in the Kalman filter and in the kinematic differential equation system.

[0022] Further advantages and practical embodiments can be found in the further claims, the description of the figures, and the drawings. They show: Fig. 1 a block diagram with various components to be implemented in a control unit, which are required to carry out the method for estimating the sideslip angle, Fig. 2 a block diagram of a modified version.

[0023] In the procedure according to Fig. 1 To estimate the sideslip angle, the translational vehicle accelerations a x , a y , a zin all three spatial directions and the rotation rates ω x , ω y and ω z measured around all three vehicle axes. The measurement signals are subjected to signal conditioning in Block 1, in which the raw data of the measurement signals are processed, for example, signal peaks are smoothed. The measured yaw rate ω', which was processed in Block 1, z is fed to a block 2, in which, taking into account the wheel speeds v w,i and the current steering angle δs from a kinematic relationship the vehicle speed v x,e in the vehicle's longitudinal direction. The estimated vehicle's longitudinal speed v x,e serves as a support variable that is used to correct further state variables that are calculated in the further course of the process.

[0024] The measurement signals of the vehicle accelerations a processed in block 1 x , a y , a z and the rotation rate ωx , ω y , ω z are then used as input variables a' x , a' y , a' z and ω' x , ω' y , ω' z fed to a block 4, which represents a kinematic differential equation system in which relationships for the vehicle speeds v x , v y , v z in all three spatial directions and the attitude angle φ, θ, ψ around all vehicle-fixed axis directions, of which the attitude angle φ denotes the roll angle, the attitude angle θ the pitch angle, and the attitude angle ψ the yaw angle. The kinematic differential equation system in Block 4 can be expressed in the following form: [φ˙θ˙ψ˙]=[1sinφtanθcosφtanθ0cosφ−sinφ0sinφcosθcosφcosθ]⋅[ωxωyωz] [v˙xv˙yv˙z]=−[0−ωzωyωz0−ωx−ωyωx0]⋅[vxvyvz]+[axayaz]−g⋅[−sinθsinφcosθcosφcosθ], where g denotes the acceleration due to gravity.

[0025] The information about the vehicle speeds v obtained from the numerical solution of the kinematic differential equation system x , v y , v z and the attitude angles φ, θ, ψ are then fed to a block 5, in which a transformation of the calculated or estimated vehicle state variables is performed. This transformation is intended to enable a comparison of the estimated variables with measured variables fed to the system.

[0026] Following process block 5, the kinematic vehicle state variables v x , v y , v z and φ, θ, ψ are fed to a control system 7 in which further processing takes place, in particular the calculation of the float angle β according to the relationship β=arctan(vyvx) from the ratio of the vehicle lateral speed v y to the vehicle longitudinal speed v XThe sideslip angle β is used in the control system, for example, to control a vehicle control system, such as a braking or steering system or an active system for chassis adjustment.

[0027] The processed data for vehicle speeds v x , v y , v z and attitude angles φ, θ, ψ are also fed back in a closed loop, where, in addition to the other supporting variables, the vehicle's longitudinal speed v x from the vehicle longitudinal speed support variable v x,e subtracted and the difference value is fed to a Kalman filter 3, which is arranged in the feedback loop to block 4 with the kinematic differential equation system. The Kalman filter 3 receives as an additional input variable Stab w the information about the number of stable running vehicle wheels from block 2, in which the calculation of the vehicle speed reference variable vx,e is performed. The Kalman filter 3 serves to support the open integration of the kinematic differential equation system performed in block 4.

[0028] The block diagram also contains a block 6, via which boundary conditions and process constants can be introduced into the process, in particular with regard to the vehicle longitudinal speed support variable v x,e . The inertial variables, i.e. the longitudinal accelerations and the yaw rates, as well as the vehicle speed reference variable v, are fed to block 6 as input variables. x,e and the current steering angle δ S .

[0029] The block diagram according to Fig. 2 corresponds essentially to that according to Fig. 1, so that with regard to the matching blocks 1 to 7, the description Fig. 1 can be referred to. In Fig. 2, however, the control unit is contained in the separate block 9.

[0030] In Fig. 2, two additional blocks 8 and 9 are inserted, of which block 8 represents an alternative for calculating the sideslip angle β and block 9 symbolises the control unit into which either the sideslip angle β calculated according to block 7 or the sideslip angle β calculated according to block 8 is entered as an input variable.

[0031] In block 8, the sideslip angle β is calculated for the case that a longitudinal speed limit value v x,limit from the vehicle's longitudinal speed v x is exceeded. The calculation according to block 8 is carried out using a vehicle dynamics model, as an alternative to the calculation method according to block 7. The background is that for small longitudinal speed values v x the calculation of the sideslip angle β according to block 7 is subject to uncertainty, since the estimated longitudinal speed is included in the denominator of the calculation formula according to block 7, which is described in the description of Fig.1. In order to nevertheless achieve a suitable speed v in the subsequent control unit 9 even for small longitudinal speeds x In order to provide a valid sideslip angle value, below the longitudinal speed limit v x,limit The sideslip angle value is determined not according to block 7, but according to block 8 via the vehicle dynamics model. At the output of block 8, a sideslip angle value β is available, which is fed as an input variable to the control unit 9.

[0032] To calculate the sideslip angle according to Block 8, various driving state variables are required, from which the sideslip angle value β is calculated in the driving dynamics model. These driving state variables are, for example, the speed values v x , v y , v z , the attitude angles φ, θ, ψ and the rotation rates ω x , ω y , and ω z. In principle, however, other state variables of the system can also be used to calculate the slip angle β, which is indicated by the dashed arrow, which is fed as an input variable to block 8.

[0033] The longitudinal speed limit v x,limit It is conveniently calculated based on other state variables and continuously updated to account for the vehicle's current driving situation. Furthermore, it is possible to consider other driving state variable limits in addition to the longitudinal speed limit, which can also be used as an alternative to the longitudinal speed limit if necessary.

Claims

[1] Method for estimating the sideslip angle for use in a control system influencing the driving state in a vehicle, comprising the following method steps: - Measuring vehicle accelerations (a x , a y , a z ) in all three spatial directions and the rotation rates (ω x , ω y , ω z ) at least around the vehicle’s longitudinal and vertical axis, - Estimating the vehicle speed (v x,e ) as a support variable in at least one spatial direction as a function of vehicle state variables (v w,i , δ S , ω z ), - Calculate vehicle speeds (v x , v y , v z ) in all three spatial directions and the attitude angle (φ, θ, ψ) at least around the vehicle's longitudinal and transverse axis by integrating a system of differential equations from the measured vehicle accelerations (a x , a y , a z ) and rotation rates (ωx , w y , ω z ), where the support size (v x,e ) to correct the corresponding calculated vehicle speed (v x ) in at least one spatial direction, - Calculate the sideslip angle (β) from the ratio of the calculated vehicle lateral speed (v y ) to the vehicle's longitudinal speed (v x ), where the difference between support size (v x,e ) and calculated vehicle speed (v x ) in a Kalman filter to calculate a corrected vehicle speed (v x ) which is taken into account when integrating the system of differential equations. [2] Method according to claim 1, characterized by that the rotation rate (ω y ) around the vehicle's transverse axis is taken into account. [3] Method according to one of claims 1 to 2, characterized by that the support variable is the vehicle's longitudinal speed (v x,e ) is. [4] Method according to claim 3, characterized by that the estimated vehicle longitudinal speed (v x,e ) from the wheel speeds (v w,i ) is determined. [5] Method according to claim 3 or 4, characterized by that the estimated vehicle longitudinal speed (v x,e ) is compared with the vehicle speed from a GPS signal (Global Positioning System) for correction purposes. [6] Method according to one of claims 3 to 5, characterized by that to determine the estimated vehicle longitudinal speed (v x,e ) the steering angle (δ S ) and the rotation rates (ω x , ω y , ω z ) should be taken into account. [7] Method according to one of claims 1 to 6, characterized by that the support variable is the vehicle lateral speed (v y,e ) is. [8] Method according to claim 7, characterized by that the vehicle lateral speed (vy,e ) is calculated from a vehicle model if the measured rotation rates (ω x , ω y , ω z ), the vehicle accelerations (a x , a y , a z ) and the steering angle (δ S ) within the model validity. [9] Method according to claim 7 or 8, characterized by that when driving straight ahead, the vehicle lateral speed (v y,e ) is set to zero. [10] Method according to one of claims 1 to 9, characterized by that the vehicle speed acting in the direction of the vehicle's vertical axis (v z,e ) is set to zero as a support variable. [11] Method according to one of claims 1 to 10, characterized by that the system of differential equations for calculating the vehicle speeds (v x , v y , v z ) and the angular positions (φ, θ, ψ) with the measured vehicle accelerations (a x , a y, a z ) and the measured rotation rates (ω x , ω y , ω z ) in the form [φ˙θ˙ψ˙]=[1sinφtanθcosφtanθ0cosφ−sinφ0sinφcosθcosφcosθ]⋅[ωxωyωz] [v˙xv˙yv˙z]=[0−ωzωyωz0−ωx−ωyωx0]⋅[vxvyvz]+[axayaz]−g⋅[−sinθsinφcosθcosφcosθ] can be represented. [12] Method according to one of claims 1 to 11, characterized by that the sideslip angle (β) is calculated from the vehicle lateral speed (v y ) and vehicle longitudinal speed (v x ) according to the context β=arctan(vyvx) is determined. [13] Method according to one of claims 1 to 12, characterized by that the sideslip angle (β) is calculated from a vehicle dynamics model. [14] Method according to claim 13, characterized by that the calculation of the sideslip angle (β) from the vehicle dynamics model is only possible below a state variable limit value (v x,Limit ) is carried out. [15] Method according to claim 14, characterized by that the state variable limit value is a longitudinal velocity limit value (v x,Limit ) is taken into account. [16] Method according to claim 15, characterized by that the longitudinal speed limit (v x,Limit ) can be represented as a function of other state variables. [17] Method according to claim 12 and any one of claims 14 to 16, characterized by that when the state variable limit value (v x,Limit ) the sideslip angle (β) is calculated from the arctangent of the ratio of the calculated vehicle lateral speed (v y ) and vehicle longitudinal speed (v x ) is calculated. [18] Control device which is arranged to carry out the method according to one of claims 1 to 17.

Images