Method for checking whether a target trajectory is suitable for vehicle trajectory control

Abstract

This invention relates to a method for checking whether a target motion trajectory is suitable for vehicle (1) motion trajectory control, wherein the target motion trajectory includes route information about the direction of the route to be traversed and dynamic information about the dynamics that can be used to traverse the route, wherein the slope (λ), lateral inclination (η), and coefficient of friction (μ) of the road along the target motion trajectory are included. max The target values ​​of the horizontal target tire force, target traction force, target traction power, and target steering power at each wheel of the vehicle (1) are determined and / or estimated using sensors and / or from map data, based on a model formula of the vehicle (1) from the route information and dynamic information of the target trajectory and from the determined slope (λ) and lateral tilt (η), and are calculated as required for driving the target trajectory. The target trajectory is evaluated as suitable for trajectory control under the following conditions: - the target traction force and target traction power are lower than a predetermined traction characteristic curve or traction power characteristic curve; - the target steering power is lower than a predetermined steering system power limit; and - the horizontal target tire force at each wheel is determined by the friction coefficient (μ). max Within the determined friction coefficient limit, otherwise the target trajectory is evaluated as inappropriate.

Description

Technical Field

[0001] This invention relates to a method for checking whether a target motion trajectory is suitable for vehicle motion trajectory control. Background Technology

[0002] In manually driven vehicles, the driver is responsible for trajectory planning; that is, determining the path, time, and speed at which they intend to move. During trajectory planning, the driver considers the road characteristics ahead and learned knowledge of anticipated vehicle reactions. Using this information, the driver plans a trajectory that the vehicle, based on experience, will follow; that is, the trajectory is drivable.

[0003] DE 100 50 421 A1 discloses a method for controlling the driving dynamics of a four-wheeled motor vehicle, in which the potential value of the adhesion force available between the wheels and the road is determined and the adhesion margin is determined by comparing it with the current adhesion utilization rate, which is also determined. According to this invention, in addition to the forces acting in the horizontal plane, the vertical movement of the vehicle body relative to the wheels is also considered when determining the adhesion margin. Preferably, the maximum horizontal force transmitted by the tire is determined by multiplying the wheel load along the vertical orientation by the estimated or sensor-determined coefficient of friction between the tire and the road, from which the estimated actual values ​​of the longitudinal and lateral forces suitable for acting in the horizontal plane are used to determine the adhesion margin in the longitudinal, lateral, and vertical directions of the vehicle. The determined adhesion margin can then be transmitted to a so-called adhesion controller, which facilitates the advantageous use of the actually supplied adhesion force while considering the desired driving action. Summary of the Invention

[0004] The objective of this invention is to provide an improved method for checking whether a target motion trajectory is suitable for vehicle motion trajectory control.

[0005] According to the present invention, this task is accomplished by a method for checking whether a target motion trajectory is suitable for vehicle motion trajectory control, as described below.

[0006] Advantageous designs of the present invention are also described below.

[0007] In the method according to the invention for checking whether a target trajectory is suitable for vehicle trajectory control, the target trajectory includes route information about the direction of the route to be traversed and dynamic information about the dynamics that can be used to traverse the route. According to the invention, the road gradient, lateral inclination, and coefficient of friction along the target trajectory are determined and / or estimated using sensors and / or from map data. If some values ​​cannot be determined by sensors, they are taken from map data, for example. If a value cannot be directly measured or obtained directly from map data, the value is estimated, for example, based on sensor information or map data. Based on vehicle model formulas, target values ​​required for traversing the target trajectory are calculated from the route information and dynamic information of the target trajectory, as well as from the determined gradient and lateral inclination, wherein the target trajectory is evaluated as suitable for trajectory control under the following conditions:

[0008] - The target traction force and target traction power are lower than the predetermined traction force characteristic curve or traction power characteristic curve.

[0009] - The target steering power is lower than the predetermined power limit of the steering system, and

[0010] - The horizontal target tire force on each wheel is within the friction coefficient limit determined by the friction coefficient.

[0011] Otherwise, the target's trajectory is evaluated as inappropriate.

[0012] In one implementation, the method is used on each target motion trajectory from a predetermined set of target motion trajectories.

[0013] In one implementation, each target trajectory deemed unsuitable is considered invalid and discarded.

[0014] In one implementation, the traction force characteristic curve and traction power characteristic curve are obtained from the corresponding lookup table.

[0015] In one implementation, the traction force characteristic curve and traction power characteristic curve take into account the powertrain degradation phenomenon.

[0016] In one implementation, a quantitative assessment of the potential utilization rate of the friction coefficient is performed.

[0017] In one implementation, the degradation of the switching actuator is also considered.

[0018] In one implementation, the vehicle model formula is based on the quasi-static model method.

[0019] In one implementation, in the first step, the parameters of the target trajectory, as well as the slope and lateral tilt, are converted into vehicle parameters using model formulas.

[0020] In one implementation, the target traction force and target traction power are calculated with respect to the vehicle's center of gravity.

[0021] The proposed invention solves the problem of determining the drivability of autonomous vehicles.

[0022] The present invention relates to a method for checking whether a target motion trajectory is suitable for vehicle motion trajectory control, wherein the target motion trajectory includes route information (target position, target curvature, target curvature change) about the direction of the route to be traversed and dynamic information (target speed, target acceleration) about the dynamics that should be used to traverse the route.

[0023] According to the present invention, the slope λ, lateral inclination η, and friction coefficient μ of the road along the target motion trajectory max (Based on sensors or map data) It is determined or estimated. Furthermore, based on vehicle model formulas, the driving force (target traction force F) required to traverse the target trajectory is calculated from the route and dynamic information of the target trajectory and from the determined slope λ and lateral tilt η. traction,demand ), drive power (target traction power P) traction,demand ), steering power (target steering power P) steering,max ) and the horizontal target tire force (longitudinal target tire force F) on each wheel XT,i Lateral target tire force F YT,i The target value is determined based on the premise that the target motion trajectory is suitable for motion trajectory control, and the inspection shows that:

[0024] - The target traction force and target traction power are lower than the predetermined traction force characteristic curve or traction power characteristic curve.

[0025] - The target steering power is lower than the predetermined steering system power limit.

[0026] - The horizontal target tire force on each wheel is determined by the coefficient of friction μ max Within the defined limits of the coefficient of friction (maximum usable adhesion potential, adhesion ellipse curve).

[0027] Otherwise, the target trajectory is evaluated as unsuitable for trajectory control.

[0028] This method is advantageously applied to each target motion trajectory from a predetermined set of target motion trajectories. Each target motion trajectory deemed unsuitable is considered invalid and discarded. The traction force characteristic curve and traction power characteristic curve are advantageously derived from a lookup table and take into account any potential degradation.

[0029] The present invention also relates to an apparatus configured to perform the method described above. This apparatus may, in particular, include a data processing device, such as an in-vehicle controller. Attached Figure Description

[0030] The embodiments of the present invention will be explained in detail below with reference to the figures shown:

[0031] Figure 1 A side view schematic diagram of a vehicle with multiple different coordinate axis systems is shown.

[0032] Figure 2 A rear-view diagram of a vehicle with multiple different coordinate axis systems is shown.

[0033] Figure 3 A top view schematic diagram of a vehicle with multiple different coordinate axis systems is shown.

[0034] Figure 4 This illustrates a schematic rotation of the coordinate axis system.

[0035] Figure 5 A schematic diagram of the vehicle is shown, illustrating the roll axis, front roll center, and rear roll center.

[0036] Figure 6 A schematic diagram of the vehicle is shown.

[0037] Figure 7 A schematic diagram showing the vehicle's forces and moments, including those from the front and rear axles.

[0038] Figure 8 The diagram shows the front axle, which has tire force, chassis reaction force, and chassis reaction torque.

[0039] Figure 9 A schematic diagram of the lookup table is shown.

[0040] Figure 10 A schematic diagram of the tire force ellipse curve is shown.

[0041] Figure 11 A detailed schematic diagram of the functional architecture of a device for model-based drivability checks is shown.

[0042] Corresponding parts are labeled with the same reference numerals in all the drawings. Detailed Implementation

[0043] This invention relates to the estimation of the drivability of all desired motion trajectories from a number of candidate motion trajectories, presented in the form of horizontal motion trajectories in ground-fixed coordinates, wherein the drivability limits for the motion trajectory are generally given by vehicle parameters and road parameters. For example, these are the friction potential defining the maximum achievable horizontal tire force and the driving force and power limit defining the possible acceleration. To check these limits, the corresponding vehicle parameters need to be calculated from the desired motion trajectory. Here, a quasi-static (QSS) model method is chosen to calculate the tire force, driving force, and power requirements from a given desired motion trajectory. For example, 1000 desired candidate motion trajectories, each 10 s long, are used here, evaluated every 100 ms. Therefore, transit time is an important factor and a major reason for choosing the quasi-static model method instead of, for example, a dynamic model, which requires more precise sampling and solving the common difference formulas of that model.

[0044] The coordinate axis system used below conforms to ISO 8855:2011. The most important coordinate axis systems are specified below. Figure 1 , Figure 2 and Figure 3 The diagram shows a graphical overview of the relevant orthogonal coordinate system, kinematic properties, forces, and moments.

[0045] Vehicle coordinate system (X V ,Y V Z V Let X be a coordinate system defined in the reference frame of the vehicle's sprung mass, such that X... V The axis is substantially horizontally forward-oriented (when vehicle 1 is stationary) and parallel to the longitudinal plane of symmetry of vehicle 1. V The axis is perpendicular to the longitudinal plane of symmetry of vehicle 1 and points to the left, where Z... V The axis points upwards. Belongs to the vehicle coordinate system (x... V ,y V ,z V The starting point of the load is located at the center of the front axle under static reference load conditions.

[0046] Earth-fixed coordinate system (X E ,Y E Z E X is a coordinate system defined in an inertial reference frame. E and Y E Parallel to the ground plane. Z E Pointing upwards and based on the gravity vector. Belonging to the Earth-fixed coordinate system (x... E ,y E ,z E The starting point is in the ground plane.

[0047] The horizontalized coordinate system (X, Y, Z) set in the middle is as follows: its X-axis and Y-axis are parallel to the ground plane, wherein the X-axis is perpendicular to the ground plane. V The vertical projection of the axis onto the ground plane is taken as the reference. The starting point of the coordinate system (x, y, z) coincides with the starting point of the vehicle coordinate system.

[0048] Road plane-coordinate system (X R ,Y R Z R ) is a coordinate system with the following x-axis. R and Y R The axis is parallel to the road plane, where X R The axis is X V The vertical projection of the axis onto the road plane is used as the reference. The relevant road coordinate system (x...) R ,y R ,z R The starting point of the road coincides with the starting point of the vehicle's coordinate system. The road plane is the best fit achieved by the four contact points of the tire.

[0049] Tire coordinate axis system for right front wheel 1R (X) T,1R , Y T,1R Z T,1R ), for the tire coordinate axis system (X) of the left front wheel 1L T,1L , Y T,1L Z T,1L ), the tire coordinate axis system (X) for the right rear wheel 2R T,2R , Y T,2R Z T,2R ) and the tire coordinate axis system (X) for the left rear wheel 2L T,2L , Y T,2L Z T,2L ) is a coordinate system with the following x-axis. T and Y T The axis is parallel to the road plane, where Z T The axis is oriented perpendicular to the road plane, where X... T The orientation of the axle is determined by the intersection of the wheel plane and the road plane, where the positive Z-axis is... T The axis points upwards. Figure 1 It also shows the inclination angle θ (denoted as negative) and road slope angle λ (denoted as negative) according to ISO 8855:2011 standard. F X R,cg F Y R,cg and F Z R,cg M is an arbitrary force vector component, defined relative to the road-coordinate system and effective at the center of gravity of vehicle 1. X R,cg M Y R,cg M Z R,cg It is an arbitrary torque vector component, aligned with the road-coordinate system.

[0050] Figure 2 It also shows the rocking angle φ (denoted as positive, around X) according to ISO 8855:2011 standard. V (shaft rotation), vehicle sway angle / roll angle φ V (represented as positive, rotating about the X-axis) and road plane - wheel camber angle η (represented as positive, rotating about the X-axis).

[0051] Figure 3 It also shows the yaw angle ψ (denoted as positive, from X) according to the ISO 8855:2011 standard. E From the axis to the X-axis, around the Z-axis E and left and right front steering angles δ 1L δ 1R (represented as positive, X) V The angle from the axis to the wheel plane, around Z V axis).

[0052] The accelerations and velocities of candidate motion trajectories provided by motion trajectory planning are described as projections of the desired rear-bridge motion onto the XY plane at the midpoint. For these properties to be usable as model input parameters, they should be converted into accelerations and velocities that describe the center of gravity in the X-plane. R -Y R Motion within the defined road plane. In the coordinate system (x, y, z) and (x... R ,y R ,z R The corresponding coordinate transformations between ) are derived in this section.

[0053] Figure 1 The rotation is shown. It is noted that the angles provided as input parameters are the road plane slope λ (slope angle) and the road plane-wheel roll angle η (lateral tilt), which are not equal to the rotation angle required for the axle rotation. This can be seen from... Figure 1 The following characteristics were observed:

[0054]

[0055] Therefore, we can conclude that:

[0056]

[0057] Figure 4 The rotation of the shaft is shown, which relates the road plane slope angle λ and the road plane wheel camber angle η:

[0058] 1. A horizontal coordinate system (X, Y, Z) is set in the middle and rotated around X by an angle η. X R To generate an auxiliary coordinate axis system (X′, Y′, Z′).

[0059] 2. The road coordinate system (X′, Y′, Z′) is obtained by rotating the auxiliary coordinate system (X′, Y′, Z′) around Y (!) by an angle λ. R ,Y R Z R It is noted that rotation about the initial Y-axis is used instead of rotation about Y′, which differs from the common Euler angle rotation or Tet-Brien angle rotation that causes rotation about the generated coordinate axis system. λ and η are in X R Y R The angles between their respective projections in the horizontal plane. The projections do not form orthogonal vector pairs.

[0060] To limit the range to (x,y,z) and (x... R ,y R ,z R Coordinate transformations (or passive transformations, e.g., see MJ Benacquista and JD Dromano, Classical Mechanics, Springer, Switzerland, 2018, p. 193) between the two bodies can be distinguished as internal and external rotation matrices (see MJ Benacquista and JD Dromano, Classical Mechanics, Springer, Switzerland, 2018, p. 200), which are defined by an axis determined relative to the body of revolution (internal rotation) or an axis determined in space (external rotation). Figure 4 The rotation in the middle is derived as a series of external rotations:

[0061] 1. Rotation angle η around X X R ,

[0062] 2. Rotate around Y by an angle λ.

[0063] Using MJ Benacquista and JDromano's *Classical Mechanics* (Springer, Kamm, Switzerland, 2018), this is equivalent to an internal rotation:

[0064] 1. Rotate about Y by an angle λ.

[0065] 2. Rotate by an angle η around X0 (!). X R .

[0066] Therefore, from (x,y,z) to (x R ,y R ,z R Coordinate transformation T IR This can be derived using two rotation matrices for the internal rotation, for example:

[0067]

[0068] T is obtained by combining two transformation matrices. IR :

[0069]

[0070] Replace η with formula (12) X R Therefore, we can conclude that:

[0071]

[0072] Earth-fixed coordinates (x) E ,y E ,z E (x) to road plane coordinates (x) R ,y R ,z R Coordinate transformation T ER This can be obtained by extending formula (14):

[0073]

[0074] In this case, T ZE Represents the relationship between Earth-fixed coordinates (x) and Earth-fixed coordinates (x) E ,y E ,z E The coordinate transformation from the intermediate coordinates (x, y, z) to the Z-axis is related to the rotation around the center. E Right turn angle ψ (see) Figure 3 The matrix is ​​as found in D.T. Greenwood's *Principles of Dynamics* (Prentice Hall, Upper River, NJ, 2nd ed., 1988, p. 357). T is obtained by combining all the transformation matrices in this way. ER :

[0075]

[0076] The matrix in formula (18) includes line breaks for better readability. Here, the first column before the line breaks and the second and third columns after the line breaks are shown.

[0077] For the quasi-static vehicle description, the moving dual-track model is assumed as a model hypothesis.

[0078] The following aspects were taken into consideration:

[0079] Lateral acceleration in curves

[0080] •Longitudinal acceleration caused by driving and braking

[0081] • Aerodynamic traction

[0082] • Aerodynamic lift

[0083] • Distribution of roll moment between the front and rear axles

[0084] • Static effects of lateral tilt and slope

[0085] • Roll drag (effect of required engine torque only)

[0086] The following aspects can be ignored:

[0087] • Suspension roll and pitch movements

[0088] • Rapid changes in lateral tilt and slope

[0089] • Divide into total mass and sprung mass

[0090] • Asymmetric transverse load conditions

[0091] • The effects of ABS, ASR, and ESP (i.e., this model is a slightly conservative description of the vehicle's capabilities).

[0092] • Distinguish between left and right driving forces and braking forces (i.e., no torque vectoring and no μ-splitter)

[0093] • Slow cornering and stopping maneuvers.

[0094] Figure 5 It is in X R -Z R A schematic diagram of vehicle 1 in a plane, showing the roll axis RA, the front roll center FRC, and the rear roll center RRC. cg Indicates the height of the center of gravity, l f This indicates the distance from the center of gravity to the front axle; l represents the wheelbase. F X R,f F Z R,f F X R,r F Z R,r It is axial force, F X R,R1 F Z R,R1 F X R,L1 F Z R,L1 F X R,R2 F Z R,R2 F X R,L2 F Z R,L2 It's tire force. F X R,cg and F Z R,cg It is an arbitrary force vector component, which is located at the vehicle's center of gravity relative to the road plane coordinate system. M Y R,cg It is an arbitrary torque vector component, which is based on the road plane coordinate system.

[0095] From Figure 5 The following is obtained along Z: R Equilibrium forces in different directions:

[0096]

[0097] On the front axle, at a horizontal height around the ground, rotate Y RThe torque balance is as follows:

[0098]

[0099] From (20):

[0100]

[0101] And by substituting into formula (19), we get:

[0102]

[0103] Credibility can be easily observed by substituting the following conditions:

[0104]

[0105] Note: Wheel load transitions during acceleration and braking are assumed to be evenly distributed between the left and right wheels due to symmetry. Therefore, pitch center and pitch moment distribution need not be considered in (21) and (22). However, roll moment distribution is considered in the following sections.

[0106] Figure 6 Shown in X R -Y R A schematic diagram of vehicle 1 in a plane.

[0107] F X R,f F Y R,f F X R,r F Y R,r It is axial force. F X R,cg F Y R,cg and F Z R,cg It is an arbitrary force vector component, which is based on the road plane-coordinate system and takes effect at the center of gravity of vehicle 1. X R,cg M Y R,cg M Z R,cg It is any torque vector component, which is based on the road plane-coordinate system.

[0108] Available from Figure 6 As you can see, the following is obtained along Y R Equilibrium forces in different directions:

[0109]

[0110] Around the center of the front axle (Z) R The torque balance is

[0111]

[0112] The braking and driving forces on both sides are assumed to be the same and therefore do not contribute to the rotation around Z. R The torques are balanced. Therefore, from (24):

[0113]

[0114] And by substituting this formula into (25), we obtain:

[0115]

[0116] In X R The direction that applies is:

[0117]

[0118] F X R,r and F X R,f It is determined by the distribution of driving and braking torques.

[0119] Figure 7 It is in Y R -Z R A schematic diagram of vehicle 1 in a plane, which has front and rear axle forces F. Y R,f F Y R,r and front and rear axle torque M X R,f M X R,r F Y R,cg and F Z R,cg It is an arbitrary force vector component, which is based on the road plane coordinate system and takes effect at the center of gravity of vehicle 1. M X R,cg It is an arbitrary torque vector component, based on the road plane coordinate system. h cg h represents the height of the center of gravity. rcg It is at the center of gravity x R RC at the position cg The height.

[0120] From Figure 7 Obtaining X R The diagram shows the torque balance of the shaft, with the front and rear axle forces F. Y R,f F Y R,r Front axle torque, rear axle torque M X R,f M X R,r And any forces and moments that take effect at the center of gravity. If x is used at the center of gravity... R RC at the position cg To express around X R The torque balance is as follows:

[0121]

[0122] With M X R,f+r With the addition of [the element], this can be converted to:

[0123]

[0124] Figure 8 It is in Y R -Z R A schematic diagram of the front axle in a plane, which has tire force F. Y R,1L F Z R,1L F Y R,1R F Z R,1R Chassis reaction force F Y R,1 F Z R,1 and chassis reaction torque M X R,1 h rf Indicates the height of the front axle roll center (FRC), b f It's the front axle track. People noticed that the tire force F... Y R,1L and F Y R,1R It is always given as a vector component in the road plane coordinate system, which is not related to the F component in the tire coordinate system. Y T,1L and F Y T,1R coincide.

[0125] If one incorporates the roll moment distribution that integrates the effects of suspension stiffness and stabilizer stiffness, then the roll moments of the front and rear axles caused by the misalignment between the center of gravity and the roll axis can be obtained:

[0126]

[0127] Because the roll moment distribution and axial force are available, it can be done as follows: Figure 8 The vertical tire force is calculated as shown.

[0128] In Z R and Y R The forces in the direction of equilibrium are:

[0129]

[0130] The equilibrium of moments about the axis of symmetry at the horizontal level of the road plane yields the following:

[0131]

[0132] Using Newton's third law, the chassis reaction force and axial force are related as follows:

[0133]

[0134] By using it together with (32) to (37), we arrive at the following conclusion:

[0135]

[0136]

[0137] By targeting F Z R,1L F Z R,1Rand F Z R,2L F Z R,2R Solving (41) and (43) and then substituting the results into (45) and (46) yields the following:

[0138]

[0139]

[0140] Substituting (22), (26), and (30) into (49) and (52), we obtain:

[0141]

[0142]

[0143] Substituting (21), (25), and (31) into (55) and (58), we obtain:

[0144]

[0145]

[0146] In the formula given above, F ZR,1L F ZR,1R F ZR,2L F ZR,2R It should be limited to a value >= 0 because negative force cannot be achieved due to the wheel jumping.

[0147] To calculate the lateral tire force distribution and maximum horizontal tire force, the concept of effective wheel load is introduced. The maximum horizontal tire force does not increase linearly with wheel load, but rather exhibits a slightly decreasing characteristic. This means that two wheels under the same load exert a greater maximum horizontal tire force than two wheels under different loads, even if they have the same sum of vertical forces. Various approaches exist to simulate this behavior. Here, an approach following R. Orend's "Modeling and Control of Vehicles with Single-Wheel Chassis Actuators" (IFAC Proceedings, Vol. 38, 1st Edition, pp. 79-84, 2005) is adopted, which incorporates the effective wheel load F as follows: ZT,eff,i :

[0148]

[0149] Where, 0≤k FZ ≤1 indicates the empirical wheel load regression coefficient, F ZT,N,i Indicates the nominal wheel load.

[0150] If only longitudinal tire force or only lateral tire force is considered, the maximum force can be simulated as follows:

[0151]

[0152] Where, μ max μ represents the latent value of the friction coefficient. ql It represents the ratio of lateral grip to longitudinal grip, simulating the force characteristics of anisotropic tires.

[0153] To calculate along X R The horizontal tire force in the direction of travel is defined as follows:

[0154]

[0155] In addition to (27), the following can be obtained:

[0156]

[0157] When people join the standard The distribution of driving and braking torque γ d and γ b At that time, along X R The front and rear axle forces in the following directions are expressed as follows:

[0158]

[0159] Assuming that the left and right driving forces and braking forces are the same as described above, it can be concluded that:

[0160]

[0161] To calculate along Y R The directional tire force is expressed as a function described in D. Ammon's "Modeling and System Development of Vehicle Dynamics" (Teubner, Stuttgart, Germany, 1997):

[0162]

[0163] α i Indicates the sideslip angle, s i The longitudinal slip is represented as defined in D. Ammon's *Modeling and System Development of Vehicle Dynamics* (Teubner, Stuttgart, Germany, 1997). The normalized shape function saturates at 1 for larger sideslip angles, which is the only region of interest for further design. α max,N It is a metric, but it has no further significance in the current work.

[0164] Assuming lateral tire force F Y T,i Next, according to

[0165] ,

[0166] Approaching the limit and therefore at F Y T,i In the saturated region, the following assumptions can be made:

[0167]

[0168] Using the aforementioned assumptions and (76), the ratio of the left and right tire forces on each axle can be estimated as follows:

[0169]

[0170] This means that lateral tire force is distributed based on the effective wheel load ratio.

[0171] If along Y R Since the directional tire force and axle force are related, we can obtain the following:

[0172]

[0173] By substituting (78) and (79), this can be transformed as follows:

[0174]

[0175] The solution derived from the tire force is as follows:

[0176]

[0177] By substituting (26) and (25), we can obtain the result along Y. R Tire force in direction:

[0178]

[0179] Finally, this is how the X-axis in the tire direction is obtained. T Y T and Z T On the tire force:

[0180]

[0181] Therefore, it is possible to check whether the tire force is within the friction limit:

[0182]

[0183] Figure 10 A schematic diagram of the so-called tire force ellipse curve is shown here, illustrating the combined force F. XT and F YT The adhesion limit.

[0184] The horizontal target tire force should be located on each wheel at a position determined by the coefficient of friction μ. maxWithin the defined limits of the coefficient of friction (maximum usable adhesion potential, adhesion elliptic curve, also see DE 100 50 421 A1).

[0185] If inequality (95) is satisfied for each wheel, then one of the three necessary conditions for drivability is satisfied; otherwise, the trajectory is not drivable.

[0186] External forces, gravity, and inertia were not initially considered. Instead, any force vector component F that takes effect at the vehicle's center of gravity is considered. X R,cg F Y R,cg and F Z R,cg and any torque vector component M X R,cg M Y R,cg and M Z R,cg All of these can be used as free parameters. The static effects of external forces, gravity, and inertia on the lateral tilt and slope are then included as terms to derive F. X R,cg F Y R,cg F Z R,cg and M X R,cg M Y R,cg M Z R,cg .

[0187] Regarding gravity, from Figure 1 and Figure 4 F was obtained from Z,cg,gravity =−m·g. Therefore:

[0188]

[0189] λ is positive when going downhill.

[0190] Regarding translational inertia, the following applies:

[0191] To indicate the inertial force, the acceleration 'a' is specified in the road plane coordinate system. X R,cg a Y R,cg and a Z R,cg The acceleration a should be considered from the expected trajectory of the center of gravity. X,cg and a Y,cg To calculate these, they are projections onto the XY plane of a coordinate system set in the middle. Under the assumption of static lateral tilt and slope, the following conditions apply:

[0192]

[0193] Therefore, what remains is the unknown value a. X R,cg and a Y R,cg The motion a of vehicle 1, which was not included as part of the motion trajectory projection, is also present. Z,cg Using (14), (15), and (99), the following can be determined:

[0194]

[0195] Regarding a Z,cg Solving (102) yields the following result:

[0196]

[0197] Substituting (103) into (100) yields:

[0198]

[0199]

[0200] Substituting (103) into (101) yields:

[0201]

[0202] Finally, the translational inertial force vector components can be determined using (99), (106), and (110) as follows:

[0203]

[0204] Regarding rotational inertia, the following applies:

[0205] To calculate the rotational torque M related to rotational inertia X R,cg,inert,rot M Y R,cg,inert,rot and M Z R,cg,inert,rot It can indicate the rotational acceleration of vehicle 1 in the road plane coordinates regarding the desired center of gravity trajectory. , and .

[0206] Using DT Greenwood's *Principles of Dynamics* (Prentice Hall, Upper River, New Jersey, 2nd ed., 1988, p. 406), the Euler angle versus angular velocity is expressed as follows:

[0207]

[0208] When using the following equivalents:

[0209]

[0210] Substituting this into the notation used in this article:

[0211]

[0212] Under the assumption of a slowly changing lateral inclination and slope under static conditions, i.e. and Therefore, we can conclude that:

[0213]

[0214] When considering the derivative of angular velocity and The following relationships exist between them, and when considering them from the perspectives of inertial and kinematic frames (see DT Greenwood, Principles of Dynamics, Prentice Hall, Upper Saddle River, New Jersey, 2nd ed., 1988, p. 392):

[0215]

[0216] Angular acceleration can be used from (122) to (124) in the application and The derivative is calculated to obtain the following result:

[0217]

[0218] Using Euler's formula for motion (DT Greenwood, Principles of Dynamics, Prentice Hall, Upper River, New Jersey, 2nd ed., 1988, p. 392) and (122) to (124) and (126) to (128), the rotational torque derived from inertia can be calculated:

[0219]

[0220] Regarding aerodynamic traction and lift, it is possible to incorporate the traction lever arm height h, which describes the height of the generated ground traction force. d To represent the aerodynamic traction percentage of any forces and moments acting at the center of gravity of vehicle 1 (M. Mitschke, Motor Vehicle Dynamics, Springer, Berlin, 5th edition, 2014):

[0221]

[0222] Here, C d A represents the aerodynamic traction coefficient. a Represents an aerodynamic surface, ρ a This indicates air density.

[0223] Similarly, aerodynamic lift (M. Mitschke, Motor Vehicle Dynamics, Springer, Berlin, 5th edition, 2014) is described by the following equation:

[0224]

[0225] In this case, C l,f and C l,r These are the front and rear aerodynamic lift coefficients.

[0226] The quasi-static steering system model applies as follows: Steering force limitations are another important aspect in assessing the physical feasibility (=drivability) of a motion trajectory. Considering the tire forces discussed above with respect to equations (70) to (95), the limitations of the steering system and power will be discussed below. Model assumptions are made as discussed above.

[0227] Assessing the drivability of a driving trajectory is particularly relevant at higher speeds. In this case, limits on steering force and steering speed should be considered to ensure the selection of a drivable driving trajectory. If a discrepancy arises between the expected and actual steering actuator movements, slow cornering and stopping maneuvers can be easily addressed through replanning.

[0228] For higher speeds and smaller steering angles, W. Matschinsky's "Wheel Control of Road Vehicles" (Springer, Berlin, 3rd edition, 2007) states that the steering torque is determined by the lateral tire force F. Y T,1L F Y T,1R Motion trail n k and tire trail n p Main control. The trail distance, defined by the steering mechanism, generally varies with the vertical wheel travel. Tire trail varies with longitudinal tire slip and tire sideslip angle. The closer the tire is to slipping, the smaller the tire trail.

[0229] Under the same left and right side drive / brake force conditions as described above, F XT,1L and F XT,1R The effect on steering force disappears. Therefore, the required steering force is generated by the lateral force F. Y T,1L and F Y T,1R The determination is based on the proportion of the resetting torque. Therefore:

[0230]

[0231] Because of the motion drag n k and tire trail n p Depending on the vertical wheel travel and slippage, the "quasi-static model that does not consider chassis motion, excitation and slippage caused by the road surface" cannot provide a solution for n. k and n p The exact value. But it is feasible to calculate the upper limit P. s,max ≥P s The maximum motion drag distance n is defined as follows.k Tire trail n p and average steering speed :

[0232]

[0233] Thus, the upper limit is given:

[0234]

[0235] Therefore, another of the three necessary conditions for drivability should be checked: P s,max Is it below the current available power P of the electronic servo steering (EPS)? EPS If the inequality is satisfied... If so, then this necessary condition for drivability is met; otherwise, the trajectory is not drivable.

[0236] The quasi-static dynamic system model is applicable to:

[0237] The core idea for assessing drivability limits within a powertrain is to compare the required traction and wheel power with the available force and power at the wheels (supply characteristic curves). Details regarding the definitions of these terms are provided in M. Mitschke's *Dynamics of Motor Vehicles* (Springer, Berlin, 5th edition, 2014).

[0238] Traction demand F traction,demand As specifically used in (145) to distinguish the following cases:

[0239] 1. Driving: F traction,demand ≥0,

[0240] 2. Engine brake: F traction,hyd brake threshold < F traction,demand <0,

[0241] 3. Engine brakes and hydraulic brakes: F traction,demand ≤F traction,hyd brake threshold .

[0242] To simulate the required and available traction and wheel power, the following model assumptions are made.

[0243] The following aspects were considered:

[0244] •Acceleration resistance of chassis and body

[0245] • Acceleration resistance of the powertrain

[0246] • Uphill resistance

[0247] • Two-wheel drive, four-wheel drive, independent engine four-wheel drive

[0248] • Power loss caused by wheel slippage

[0249] • Power loss in the dynamic system

[0250] • Electric drive unit or internal combustion engine

[0251] • Shifting transmission under static conditions (power and force envelope curves for all gears)

[0252] • Roll drag (no speed correlation)

[0253] • Degradation of electric drive units (starting power, hourly power, continuous power)

[0254] • Aerodynamic traction

[0255] The following aspects are disregarded:

[0256] • Shifting strategy and shifting process. For example, assuming the transmission controller selects an appropriate gear to provide the required wheel power and traction, and there is no significant loss of traction during the shift,

[0257] • Friction potential, i.e., the available traction force only simulates the engine limit, while the friction limit has already taken into account the longitudinal tire forces.

[0258] • Electric enhancement of internal combustion engines,

[0259] • Hybrid braking between engine / electric drive and hydraulic brakes is a prerequisite, but it has not yet been specifically simulated.

[0260] Situation 1: Driving

[0261] The following considerations assume a single internal combustion engine or a single electric drive unit. However, this approach can be easily extended to include situations where drive torque is distributed to multiple axles using a drive torque distribution factor.

[0262] According to M. Mitschke's *Dynamics of Motor Vehicles* (Springer, Berlin, 5th edition, 2014), the total traction force requirement is obtained from the following formula:

[0263]

[0264] Here,

[0265] • Roll drag,

[0266] •F X R,cg The sum of slope angle (96), chassis and body inertia (111), and aerodynamic traction (135),

[0267] • According to the mass coefficient λ used for rotating mass m The acceleration resistance of the power system.

[0268] The wheel power requirement for zero tire slippage loss is given below in a simplified manner:

[0269]

[0270] The wheel power under tire slippage conditions is obtained as follows:

[0271]

[0272] And adopt the following definition:

[0273] •η T Tire efficiency, which describes the loss caused by longitudinal slippage.

[0274] •s: The average slip of the driven shaft, where it is assumed, for example, that s≈s 2L ≈s 2R ,

[0275] •r d : Dynamic wheel radius, where, U represents the distance traveled by a freely rolling wheel on the ground.

[0276] •r s : Static wheel radius, that is, the distance from the center of the wheel to the road surface.

[0277] The following distinctions should be taken into consideration:

[0278] • The distance traveled on the ground by a tire that does not slip.

[0279] • The distance traveled on the ground by a slippery tire.

[0280] • On the ground, there is a radius r s The distance traveled by a point on a fixed wheel.

[0281] For the quasi-static vehicle model described above, longitudinal slip is unknown. With the anti-slip controller engaged, longitudinal slip is, in the worst-case scenario, almost the typical severe slip of around 10%. c (M. Mitschke, *Dynamics of Motor Vehicles*, Springer, Berlin, Germany, 5th edition, 2014). Therefore, tire efficiency during severe slippage can be calculated as follows:

[0282]

[0283] This leads to the use of P traction,demandThe upper limit estimate. A more accurate estimate can be obtained by using values ​​from 1 to η. T,c Interpolation η T This can be obtained, for example, based on the use of friction limit:

[0284]

[0285] Figure 9 This diagram illustrates a lookup table (LUT, family of supply characteristic curves).

[0286] F traction,supply and P traction,supply The lookup tables (supply characteristic curve family) are given for electric motors with fixed transmission ratios (solid lines) and for internal combustion engines with shift transmissions (dashed lines). The supply lookup tables describe the traction and power available at the axle. That is, they have already accounted for powertrain losses. In the case of shift transmissions, the outer envelope curve is used for all gears. The lookup tables can be measured based on the degraded DEG.

[0287] To determine drivability, F traction,demand and P traction,demand Compared to Figure 9 The F shown is presented in the form of a lookup table (family of supply characteristic curves). traction,supply and P traction,supply The envelope curves are compared. The lookup table already accounts for powertrain losses, assuming they are simulated as constant efficiency coefficients for each gear. Degradation can be considered by using, for example, a reduction factor based on engine temperature, supply voltage, etc.

[0288] When the following inequality is satisfied and At that time, the third necessary condition for drivability is met; otherwise, the trajectory is not drivable.

[0289] Scenario 2: Engine brake

[0290] Small negative traction force requirement F traction,demand This is achieved using only engine braking. For a two-wheel drive vehicle 1, this means that braking tire force only applies to the front or rear tires. Without additional measures, this could cause tire lock-up, especially on surfaces with low friction. Therefore, the control described is required by the electronic stability program to prevent tire lock-up. In this case, the hydraulic brakes are activated, and situation 2 transitions to situation 3. This means there are no limitations on traction and traction supply (except for friction limitations), which have already been taken into account in the longitudinal tire forces (see above).

[0291] Scenario 3: Engine brakes and hydraulic brakes

[0292] In this case, there are no limitations on traction force and traction supply (except for friction limitations), which have already been taken into account in the longitudinal force of the tire (see above).

[0293] Table 1 below provides an overview of the vehicle parameters.

[0294] parameter unit illustrate m kg Total vehicle mass l m Wheelbase <![CDATA[l f ]]> m Distance between center of gravity and front axle <![CDATA[l r ]]> m <![CDATA[Distance between the center of gravity and the rear axle (l r = l – l f )]]> <![CDATA[h cg ]]> m Center of gravity height <![CDATA[h rf ]]> m Forward roll center height <![CDATA[h rr ]]> m rear roll center height <![CDATA[h rcg ]]> m <![CDATA[Roll center height (h at the center of gravity rcg = h rr + (h rf − h rr )(1 − l f / l))]]> <![CDATA[b f ]]> m Front wheel track <![CDATA[b r ]]> m Rear wheel track κ - Roll moment distribution between the front and rear axles (1: front only, 0: rear only) <![CDATA[γ d ]]> - Drive torque distribution between the front and rear axles (1: front only, 0: rear only) <![CDATA[γ b ]]> - Braking torque distribution between the front and rear axles (1: front only, 0: rear only) <![CDATA[k FZ ]]> - <![CDATA[Wheel load - regression coefficient (0 ≤ k FZ ≪ 1)]]> <![CDATA[F ZT,N,1R ,...,F ZT,N,2L ]]> N Nominal wheel load under static load conditions <![CDATA[μ max ]]> - Friction potential <![CDATA[μ ql ]]> - <![CDATA[Ratio of lateral grip to longitudinal grip (concept: μ 2 F z 2 =F x 2 +F y 2 / μ 2 ql l)]]> <![CDATA[n k,max ]]> m Maximum travel distance <![CDATA[n p,max ]]> m Maximum tire trail <![CDATA[J xx , J yy , J zz ]]> <![CDATA[kg m 2 ]]> Moment of inertia of vehicle 1 about its axle at its center of gravity <![CDATA[C d ]]> - Aerodynamic traction coefficient <![CDATA[A a ]]> <![CDATA[m 2 ]]> Aerodynamic surfaces <![CDATA[h d ]]> m Traction lever arm height <![CDATA[C l,f ]]> - Front aerodynamic lift coefficient <![CDATA[C l,r ]]> - Rear aerodynamic lift coefficient <![CDATA[ρ a ]]> <![CDATA[kg / m 3 ]]> air density <![CDATA[λ m ]]> - Mass coefficient for rotating masses within the dynamic system (based on gear). <![CDATA[f R ]]> - Roll drag coefficient <![CDATA[r s ]]> m Static wheel radius (distance from the wheel center to the road surface) <![CDATA[r d ]]> m <![CDATA[Dynamic wheel radius U = 2πr d , U: The distance traveled by the freely rolling wheel]]> <![CDATA[η T,c ]]> - Tire power efficiency during severe slippage

[0295] Figure 11 The functional principle of the drivability check FP for autonomous vehicle 1 is shown.

[0296] The input parameters are:

[0297] - The vehicle's movement trajectory should be checked to ensure drivability.

[0298] - The slope λ and lateral inclination η of the road ahead of vehicle 1, for example, from sensor data or map data.

[0299] - For example, the potential friction coefficient μ of a road from sensor data or map data. max ,

[0300] - Information on the degradation of the steering and drive actuator systems (available power / nominal power, available torque / nominal torque).

[0301] The output parameters are:

[0302] - Drivability assessment (yes / no)

[0303] - Selectively and quantitatively assess the utilization rate of the friction coefficient potential.

[0304] Technical signal processing is performed as follows:

[0305] In the first step S1, the requirements for the target motion trajectory (target position, target acceleration, target velocity, target curvature, target curvature change, etc.) as well as the slope and lateral tilt are converted into requirements for vehicle parameters (target traction force, target traction power, target steering power, target tire force). This conversion is performed using a reverse vehicle model, such as the quasi-static model described above.

[0306] In the second step S2, the requirements for vehicle parameters are compared with the limits preset by vehicle 1 and the road, that is:

[0307] - The powertrain limit check (DTLC) checks whether the target traction force and target traction power fall below the limits set by the traction force characteristic curve and traction power characteristic curve at the target speed.

[0308] - Steering system limit check (SLC): This involves determining whether the target steering power is below the power limit of the steering system.

[0309] - The friction coefficient limit check (FLC) is to check whether the horizontal target tire force is within the friction coefficient limit (Cam circle).

[0310] The limits in the second step can be adapted to the currently available power with the help of information about degradation.

[0311] The results of the drivability check are ultimately integrated and displayed in binary format as "drivable / indestructible" for each evaluated trajectory. It is conceivable to assess drivability more accurately by calculating whether limits are exceeded or by calculating the remaining percentage difference from those limits. One possible exemplary feedback could be: left front tire force is too high, 120% of the friction coefficient potential is being utilized. This extended approach offers the advantage of providing more precise auxiliary information for trajectory planning.

[0312] List of reference numerals

[0313]

[0314]

Claims

1. A method for checking whether a target motion trajectory is suitable for motion trajectory control of a vehicle (1), wherein, The target's trajectory includes route information about the direction of the route to be traversed and dynamic information about the dynamics that can be used to traverse that route. The slope (λ), lateral inclination (η), and coefficient of friction of the road along the target trajectory are determined and / or estimated using sensors and / or from map data. Based on the model formula of the vehicle (1), the target values ​​required for traversing the target trajectory are calculated from the route and dynamic information of the target trajectory, as well as from the determined slope (λ) and lateral inclination (η). The target trajectory is evaluated as suitable for motion trajectory control under the following conditions: - The target traction force and target traction power are lower than the predetermined traction force characteristic curve or traction power characteristic curve. - The target steering power is lower than the predetermined steering system power limit. - The horizontal target tire force on each wheel is within the friction coefficient limit determined by this friction coefficient. Otherwise, the target's trajectory is evaluated as inappropriate.

2. The method according to claim 1, characterized in that, This method is used on each target motion trajectory from a predetermined set of target motion trajectories.

3. The method according to claim 1 or 2, characterized in that, Each target trajectory deemed inappropriate is considered invalid and discarded.

4. The method according to claim 1 or 2, characterized in that, The traction force characteristic curve and the traction power characteristic curve are obtained from the corresponding lookup table (LUT).

5. The method according to claim 4, characterized in that, The traction force characteristic curve and traction power characteristic curve take into account the degradation phenomenon of the power system.

6. The method according to claim 1 or 2, characterized in that, The utilization rate of the potential value of this friction coefficient was quantitatively evaluated.

7. The method according to claim 1 or 2, characterized in that, Consider the degradation phenomenon of the steering actuator system.

8. The method according to claim 1 or 2, characterized in that, The model formula for the vehicle (1) is based on the quasi-static model method.

9. The method according to claim 1 or 2, characterized in that, In the first step, the parameters of the target's trajectory, as well as the slope (λ) and the lateral tilt (η), are converted into vehicle parameters using the model formula.

10. An apparatus configured to perform the method according to any one of the preceding claims.

Images