JP6674769B2 - Vehicle control device and vehicle control method - Google Patents

Description

  The present invention relates to a vehicle control device and a vehicle control method.

  Conventionally, for example, Patent Literature 1 below includes a turning assist control unit that generates a turning yaw moment by causing a time difference in the turning operation of the left and right wheels and generating a braking / driving force of the left and right wheels, If the road surface friction coefficient is determined to be small based on ABS (anti-lock brake) operation or information from various sensors, it is described that the assist gain is reduced.

JP 2013-39892 A

  By reducing the turning assist torque when a tire slip or a vehicle slip occurs on a low μ road surface, the behavior of the vehicle can be stabilized, but in an excessive spin state, the convergence of yaw is delayed and the vehicle enters a spin state. There's a problem.

  In the method described in Patent Document 1, since it is determined that the road surface friction coefficient is small based on information of various sensors and the like, the road surface condition is grasped unless slip occurs due to tire idling from the generation of driving force. Therefore, it is difficult to optimize the turning assist gain before the slip occurs. The state in which tire slip or slip actually occurs is a state in which the tire friction circle is exceeded, and it takes time to reduce the driving force or return to restore the vehicle to the control area. Therefore, the vehicle behavior becomes unstable, and the vehicle behavior becomes contrary to the driver's intention.

  Therefore, the present invention has been made in view of the above-described problems, and an object of the present invention is to estimate a low μ road surface in advance and reduce the vehicle body additional moment by turning assist control at an early stage. It is an object of the present invention to provide a new and improved vehicle control device and vehicle control method capable of stabilizing the behavior of the vehicle.

  In order to solve the above problems, according to an aspect of the present invention, a vehicle addition moment calculation unit that calculates a vehicle addition moment to be added to a vehicle body independently of a steering wheel steering system based on a yaw rate of the vehicle, Sometimes, a vehicle control device, comprising: a time difference information acquisition unit that acquires information about a time difference between a change in lateral acceleration with respect to a change in yaw rate, and a correction unit that corrects the vehicle body addition moment based on the information about the time difference. Provided.

  The time difference information acquisition unit may acquire information about the time difference when the vehicle starts turning.

  The time difference information acquisition unit calculates a first time until the model value of the lateral acceleration reaches the predetermined value as a starting point at which the model value of the yaw rate reaches the predetermined value; A second calculating unit that calculates a second time until the sensor value of the lateral acceleration reaches the predetermined value, as a starting point at which the model value of the predetermined value reaches the predetermined value. Information about the time difference may be obtained from a second time.

  In addition, the first calculation unit sets the model value of the lateral acceleration to a peak value after the start of the turn, with the time when the model value of the yaw rate reaches the peak value after the start of the turn at the start of the turn as the starting point. The first time up to may be calculated.

In addition, the second calculation unit sets the sensor value of the lateral acceleration to reach a peak value after the start of the turn, with the time when the model value of the yaw rate reaches a peak value after the start of the turn at the start of the turn as the starting point. The second time up to may be calculated.

  A correction unit configured to calculate a gain for correcting the vehicle-applied moment based on the information about the time difference, wherein the gain calculation unit determines the first time with respect to the second time; May be used as the gain.

  Further, the correction unit may correct the vehicle body additional moment by multiplying the vehicle body additional moment by the ratio.

  A target yaw rate calculator that calculates a target yaw rate based on a steering angle and a vehicle speed; a vehicle yaw rate calculator that calculates a yaw rate model value from a vehicle model; a yaw rate sensor that detects an actual yaw rate of the vehicle; A feedback yaw rate calculating unit that allocates the yaw rate model value and the actual yaw rate based on a difference between the model value and the actual yaw rate, and calculates a feedback yaw rate from the yaw rate model value and the actual yaw rate. The additional moment calculation unit may calculate the vehicle body additional moment based on a difference between the target yaw rate and the feedback yaw rate.

  Further, a motor required torque calculating unit for calculating a motor required torque for individually controlling motors for driving each of the rear left and right wheels of the vehicle based on the vehicle body additional moment may be provided.

  According to another embodiment of the present invention, there is provided a vehicle steering apparatus comprising: a step of calculating a vehicle body additional moment to be added to a vehicle body independently of a steering wheel steering system based on a yaw rate of the vehicle; A method of controlling a vehicle, comprising: acquiring information on a time difference of a change in lateral acceleration with respect to a change in a yaw rate; and correcting the vehicle body additional moment based on the information on the time difference.

  As described above, according to the present invention, it is possible to stabilize the behavior of the vehicle by preliminarily estimating the low μ road surface and reducing the vehicle body additional moment by the turning assist control at an early stage.

1 is a schematic diagram illustrating a vehicle according to an embodiment of the present invention. FIG. 4 is a schematic diagram for explaining general yaw rate feedback control. FIG. 4 is a diagram illustrating a relationship between a slip angle of a wheel and a lateral acceleration. FIG. 4 is a characteristic diagram illustrating a relationship between a tire slip ratio and a longitudinal force. FIG. 4 is a diagram illustrating a relationship between a longitudinal force and a lateral force of a rear wheel. FIG. 9 is a characteristic diagram showing a comparison between a model value obtained from a vehicle two-wheel model and a sensor value by inputting a vehicle speed V and a steering angle θh. . FIG. 2 is a schematic diagram showing in detail a control device 200 according to the embodiment and a configuration around the control device 200. FIG. 7 is a schematic diagram illustrating a gain map when a weighting gain calculator calculates a weighting gain a. It is a flowchart which shows the whole process of this embodiment. It is a flowchart which shows the process of step S122 of FIG. 9 is a flowchart illustrating a process in which a vehicle yaw rate calculation unit calculates a yaw rate model value γ_clc. It is a flowchart which shows the process which calculates additional torque Tvmot. 9 is a flowchart illustrating a process in which a low μ determination output gain calculation unit calculates a low μ determination output gain μG. 14 is a flowchart illustrating a process of calculating a target time constant Ttgt in step S246 of FIG. 14 is a flowchart illustrating a process of calculating a real time constant Treal in step S248 of FIG. FIG. 9 is a characteristic diagram for describing an effect when the control according to the present embodiment is performed. FIG. 9 is a characteristic diagram illustrating a state where parameters of a steering angle, a yaw rate, and a lateral acceleration change when a double lane change is performed on a low μ road. FIG. 9 is a characteristic diagram illustrating a state where parameters of a steering angle, a yaw rate, and a lateral acceleration change when a double lane change is performed on a low μ road.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

  First, the configuration of a vehicle 1000 according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a schematic diagram illustrating a vehicle 1000 according to the present embodiment. As shown in FIG. 1, a vehicle 1000 includes driving force generating devices (motors) 108, 110, 112 for driving front wheels 100, 102, rear wheels 104, 106, front wheels 100, 102, and rear wheels 104, 106, respectively. 114, gearboxes 116, 118, 120, 122 for transmitting the driving force of the motors 108, 110, 112, 114 to the front wheels 100, 102 and rear wheels 104, 106, respectively, and the motors 108, 110, 112, 114 respectively. Inverters 123, 124, 125, 126 to be controlled, wheel speed sensors 127, 128 for detecting respective wheel speeds (vehicle speed V) of the rear wheels 104, 106, a steering wheel 130 for steering the front wheels 100, 102, a longitudinal acceleration sensor 132, lateral acceleration sensor 134, battery 136, steering angle sensor 138, Steering mechanism 140, a yaw rate sensor 142, an inhibitor position sensor (IHN) 144, is configured to include an accelerator opening degree sensor 146, the control unit (controller) 200.

  The vehicle 1000 according to the present embodiment is provided with motors 108, 110, 112, 114 for driving the front wheels 100, 102 and the rear wheels 104, 106, respectively. Therefore, the drive torque can be controlled by each of the front wheels 100 and 102 and the rear wheels 104 and 106. Accordingly, the yaw rate can be generated by the torque vectoring control by driving each of the front wheels 100 and 102 and the rear wheels 104 and 106 independently of the generation of the yaw rate by the steering of the front wheels 100 and 102. The steering assist can be performed. That is, in the vehicle 1000 according to this embodiment, the turning moment (hereinafter, also referred to as the yaw moment) is controlled by the vehicle turning angle speed (hereinafter, the yaw rate), and the turning assist control for assisting the steering operation is performed. In particular, by providing inverters 123, 124, 125, and 126 having excellent responsiveness, a turning control for controlling a turning moment in a vehicle turning angle (yaw rate) in an electric vehicle capable of independently driving left and right to assist steering steering. Can be realized with high accuracy.

  The driving of each of the motors 108, 110, 112, and 114 is controlled by controlling the inverters 123, 124, 125, and 126 corresponding to the motors 108, 110, 112, and 114 based on a command from the control device 200. You. The driving force of each motor 108, 110, 112, 114 is transmitted to each of the front wheels 100, 102 and the rear wheels 104, 106 via each gear box 116, 118, 120, 122. The turning moment (yaw moment) is controlled by the body turning angular velocity (yaw rate) in a vehicle 1000 capable of independent left and right driving to which the motors 108, 110, 112, 114 having excellent responsiveness and the inverters 123, 124, 125, 126 are applied. The turning assist control for assisting the steering operation is performed.

  The power steering mechanism 140 controls the steering angles of the front wheels 100 and 102 by torque control or angle control according to the operation of the steering wheel 130 by the driver. The steering angle sensor 138 detects a steering angle θh input by the driver operating the steering wheel 130. Yaw rate sensor 142 detects actual yaw rate γ of vehicle 1000. Wheel speed sensors 127 and 128 detect vehicle speed V of vehicle 1000.

  Note that the present embodiment is not limited to this embodiment, and may be a vehicle in which only the rear wheels 104 and 106 independently generate a driving force. Further, the present embodiment is not limited to the torque vectoring by the driving force control, but can also be realized in a 4WS system or the like for controlling the steering angle of the rear wheels.

  FIG. 2 is a schematic diagram for explaining general yaw rate feedback control. The target yaw rate γ_tgt is obtained from the vehicle speed V and the steering angle θh. On the other hand, the actual yaw rate γ is detected by the yaw rate sensor 142. Then, the difference Δγ between the target yaw rate γ_tgt and the actual yaw rate γ is converted into a vehicle additional moment Mg based on the vehicle specifications, and the motor torque instruction values (Frl (left rear wheel), Frr ( Right rear wheel)) is calculated. In this way, by feeding back the actual yaw rate γ to the target yaw rate γ_tgt, the vehicle 1000 can turn in accordance with the target yaw rate γ_tgt.

  Next, the operation of the turning assist control will be described in detail with reference to FIGS. FIG. 3 is a diagram illustrating a relationship between a slip angle of a wheel (hereinafter, also referred to as a tire) and a lateral acceleration.

  As shown in FIG. 3, in the characteristics indicating the relationship between the slip angle (slip angle) of the tire and the lateral acceleration (hereinafter, also referred to as cornering characteristics or the lateral force characteristics of the tire), the lateral acceleration is linear with respect to the slip angle. In a linear region (a region where the slip angle is relatively small), the lateral acceleration increases as the slip angle increases. For example, in a plane two-wheel model, it is assumed that the cornering characteristics of the tire are linear, and in the linear region described above, the behavior of the model and the behavior of the actual vehicle approximately match.

  On the other hand, when the slip angle increases to some extent, the cornering characteristic of the tire becomes non-linear unlike the two-wheel model. That is, there is a nonlinear region where the lateral acceleration becomes nonlinear with respect to the slip angle, and in the nonlinear region, the rate of increase of the lateral acceleration with respect to the rate of increase of the slip angle decreases.

  As described above, when the slip angle increases to some extent, the obtained rate of increase in lateral acceleration decreases, so that the lateral acceleration tends to be saturated. Then, when the lateral acceleration of the front wheels is saturated, understeer occurs. Therefore, by applying the turning assist control for generating the yaw moment in the same direction to the rear wheel of the vehicle independently of the generation of the yaw moment due to the steering of the front wheels, the lateral acceleration can be additionally obtained, and the lateral acceleration can be obtained. Saturation is avoided. As a result, understeer is suppressed, and the vehicle can turn in accordance with steering.

FIG. 4 is a characteristic diagram showing a relationship between a tire slip ratio and a longitudinal force. In the characteristics shown in FIG. 4 (hereinafter, also referred to as tire longitudinal force characteristics), the longitudinal force increases while the slip ratio increases until the slip ratio increases to some extent. Then, for example, when the longitudinal force is increased to the upper limit of the friction circle characteristic of the tire, the longitudinal force is saturated. Generally, the slip ratio on the horizontal axis in FIG. 4 is calculated by the following equation (1).
Slip ratio = (vehicle speed−wheel speed) / vehicle speed (1)

  As shown in FIG. 4, when the slip ratio increases to some extent, the longitudinal force starts to decrease. This is because, when the slip ratio increases, the friction circle characteristic of the tire decreases, and the allowable amount of the longitudinal force decreases. In this state, when the gain of the turning assist control is increased, the longitudinal force characteristics of the tire change so as to approach a region surrounded by a broken line shown in FIG. 4, that is, the slip ratio increases and the longitudinal force decreases. . Conversely, when the gain of the turning assist control is decreased, the longitudinal force characteristics of the tire change so as to approach the region surrounded by the dashed line shown in FIG. 4, that is, the slip ratio decreases and the longitudinal force increases. . When the gain of the turning assist control is further reduced, the slip ratio decreases, and the longitudinal force also decreases.

  Here, the frictional circle characteristic of the tire is smaller on a road surface with a low μ than on a road surface with a high μ, so that the allowable amounts of the longitudinal force and the lateral force are reduced. In this state, when the gain of the turning assist control is increased, the slip ratio increases, and the friction circle characteristic of the rear tire is further reduced. Therefore, the lateral force tolerance is further reduced without obtaining the longitudinal force, and the lateral force is likely to be saturated. As a result, oversteer is likely to occur, and the vehicle may eventually spin.

  As described above, at the time of turning limit, the vehicle body slips due to saturation of the tire friction circle, and stability cannot be obtained. For this reason, it is conceivable to secure stability by lowering the turning assist gain when a slip or the like occurs beyond the limit of the tire. When turning assist control is performed, the amount of side slip of the vehicle body can be suppressed by reducing the turning assist torque during a tire slip and a vehicle slip state on a low μ road, and the stability of the vehicle is improved. In this state, the convergence of the yaw is delayed and the vehicle spins.

  If the turning performance improvement control is set on a high μ road condition, the tire may be saturated on a low μ road, causing the tire to spin or the vehicle to spin. On the other hand, if the control is set on a low μ road condition, turning performance on a high μ road cannot be obtained. If the control is set under a low μ road condition, since the friction circle of the tire is already saturated by the turning assist, the turning assist cannot be further performed by the driving force of the tire.

  Here, when turning, on a high μ road, a lateral force is generated by inputting a tire turning angle from steering. In particular, in the turning assist control, if a turning moment is applied to the rear wheels, a reaction force on a high μ road cannot be obtained on a low μ road. In particular, at the start of the steering operation, such as when a corner starts to turn, the occurrence of the lateral acceleration is delayed only by the occurrence of the yaw rate. In the present embodiment, by applying such characteristics, the time constant of the occurrence of the yaw rate and the lateral acceleration is defined as the ratio between the target value and the actual value, and the ratio of the time constant between the target value and the actual value is defined as the gain of the turning assist control. More specifically, a time constant difference between the turning yaw rate and the turning lateral acceleration generated by the steering angular velocity of the steering wheel is calculated, the state of the road surface μ is estimated from the ratio of the time constant difference, and the gain is used as the gain of the turning assist control. At this time, the turning time constant of the turning yaw rate and the turning lateral acceleration is calculated from the steering angular velocity of the vehicle, and it is estimated that the larger the time constant difference is, the lower the μ is. Vary the control gain. By estimating the road surface condition from the time constant difference, the gain of the turning assist control can be optimized before the occurrence of slip, so that the steering performance and the blackout performance can be achieved at the same time, and the yaw convergence can be greatly improved. . The turning assist control gain is changed between an estimated road surface μ high state and a low state, and the gain is set to 1 at a high μ and the gain is set to a value of 0 or more and less than 1 at a low μ, so that the turning assist is set at a low μ. Reduce torque. Thus, the road surface condition can be determined without the wheels and the vehicle body slipping. Then, since the state of the running vehicle can be grasped and the motor control can be performed with the braking / driving force (front / rear force) and the lateral force within the tire friction circle, the turning performance and the vehicle stability performance can be compatible, and the straight running stability and The steering stability is excellent, and the turning performance and the vehicle behavior stabilizing performance can be improved on both the high μ road and the low μ road.

  FIG. 5 is a diagram illustrating the relationship between the longitudinal force and the lateral force of the rear wheel. With reference to FIG. 5, the stabilization of the behavior of the vehicle 1000 will be described in detail.

  In the characteristics indicating the relationship between the front-rear force and the lateral force of the rear wheels 106 and 108 (hereinafter also referred to as tire friction circle characteristics), for example, when the road surface state is high μ and before the gain of the turning assist control is reduced. In FIG. 5, since the front-rear force is generated up to the front-rear axis arrow A51 shown in the left diagram of FIG. 5, the width of the left-right axis arrow A52 is the allowable lateral force. In the case of high μ, as shown by the solid line in the left diagram of FIG. 5, since the friction circle C1 is large, the lateral force does not saturate even if the longitudinal force is sufficiently used. If the road surface becomes low μ in this state, the friction circle of the tire becomes a state indicated by a broken line C2, and the allowable lateral force is saturated, so that oversteer occurs. However, in the present embodiment, when it is estimated that the road surface becomes low μ, the front-rear force decreases as indicated by arrow A53 in the left diagram of FIG. , The allowable amount of lateral force increases (arrow A54). Therefore, since the rear wheels 106 and 108 are prevented from slipping in the lateral direction, the behavior of the vehicle 1000 is stabilized without oversteering.

  Thereafter, when it is estimated that the road surface has returned to the high μ, the friction circle of the tire returns to C1 in the right diagram of FIG. Therefore, by increasing the gain of the turning assist control, the longitudinal force of the tire can be increased to the friction circle C1 indicated by the solid line. At this time, in the present embodiment, the gain of the turning assist control is increased according to the degree of slip of the vehicle 1000. Specifically, when the friction circle expands in the order of C2 → C3 → C1, the longitudinal force is increased in the order of arrows A55 → A56 → A57 according to the size of the friction circle. Thus, the gain increases as the friction circle of the tire increases, so that the longitudinal force can be increased in a short time without saturating the longitudinal force.

  FIG. 7 is a schematic diagram showing the control device 200 according to the present embodiment and the configuration around it in detail. The control device 200 includes an on-vehicle sensor 202, a target yaw rate calculator 204, a vehicle yaw rate calculator (vehicle model) 206, a yaw rate F / B calculator 208, subtractors 210 and 212, a weighting gain calculator 220, and a target lateral acceleration calculator. 224, target time constant calculation unit (time difference acquisition unit) 226, actual time constant calculation unit (time difference acquisition unit) 228, vehicle body added moment calculation unit 232, low μ determination output gain calculation unit (gain calculation unit) 234, multiplication unit ( (Correction unit) 236 and a required motor torque calculation unit 238.

  7, the on-vehicle sensor 202 includes the above-described wheel speed sensors 127 and 128, a longitudinal acceleration sensor 132, a lateral acceleration sensor 134, a steering angle sensor 138, a yaw rate sensor 142, and an accelerator opening sensor 146. The steering angle sensor 138 detects the steering angle θh of the steering wheel 130. The yaw rate sensor 142 detects the actual yaw rate γ of the vehicle 1000, and the wheel speed sensors 127 and 128 detect the vehicle speed (vehicle speed) V. The lateral acceleration sensor 134 detects a lateral acceleration Ay of the vehicle 1000.

  The target yaw rate calculation unit 202 calculates a target yaw rate γ_tgt based on the steering angle θh and the vehicle speed V. Specifically, the target yaw rate calculation unit 202 calculates the target yaw rate γ_tgt from the following equation (2) representing a general plane two-wheel model. The target yaw rate γ_tgt is calculated by substituting the values calculated from Expressions (3) and (4) into the right side of Expression (2). The calculated target yaw rate γ_tgt is input to subtraction section 210.

The variables, constants, and operators in Expressions (2) to (4) are as follows.
γ_tgt: target yaw rate θh: steering angle V: vehicle speed T: vehicle time constant S: Laplace operator N: steering gear ratio l: vehicle wheel base l _f : distance from vehicle center of gravity to front wheel center l _r : vehicle Distance from center of gravity to center of rear wheel m: Vehicle weight K _ftgt : Target cornering power (front wheel)
K _rtgt : target cornering power (rear wheel)

As described above, the target yaw rate γ_tgt is calculated from the equation (2) using the vehicle speed V and the tire steering angle δ as variables. The constant _Atgt in the equation (3) is a constant representing the characteristic of the vehicle, and is obtained from the equation (4).

The vehicle yaw rate calculation unit 206 calculates a yaw rate model value γ_clc from the following equation for calculating the vehicle yaw rate. Specifically, the vehicle speed V and the steering angle θh are substituted into the following equations (5) and (6), and the equations (5) and (6) are simultaneously solved to obtain the yaw rate model value γ_clc. (Γ in Expressions (5) and (6)) is calculated. Equation (5), in equation (6), _{K f} is the cornering power (front), the _{K r} represents a cornering power (rear). In Equation (4), the target yaw rate γ_tgt is calculated from the yaw rate model value γ_clc by using target cornering powers K _ftgt and K _rtgt different from the cornering powers K _f and K _r in Equations (5) and (6). To increase the turning performance. The yaw rate model value γ_clc is output to yaw rate F / B calculation section 208. The yaw rate model value γ_clc is input to the subtraction unit 212.

  On the other hand, the actual yaw rate γ of the vehicle 1000 detected by the yaw rate sensor 142 (hereinafter, referred to as the actual yaw rate γ_sens) is input to the subtraction unit 212. The subtraction unit 212 subtracts the yaw rate model value γ_clc from the actual yaw rate γ_sens to obtain a difference γ_diff between the actual yaw rate γ_sens and the yaw rate model value γ_clc. The difference γ_diff is input to the weighting gain calculator 220.

  The weighting gain calculator 220 calculates the weighting gain a based on the difference γ_diff between the actual yaw rate γ_sens and the yaw rate model value γ_clc.

The yaw rate F / B calculation unit 208 receives the yaw rate model value γ_clc, the actual yaw rate γ_sens, and the weighting gain a. The yaw rate F / B calculating unit 208 weights the yaw rate model value γ_clc and the actual yaw rate γ_sens with the weighting gain a based on the following equation (7), and calculates the feedback yaw rate γ_F / B. The calculated feedback yaw rate γ_F / B is output to subtraction section 210.
γ_F / B = a × γ_clc + (1−a) × γ_sens (7)

0082-0087
FIG. 8 is a schematic diagram illustrating a gain map when the weighting gain calculation unit 220 calculates the weighting gain a. As shown in FIG. 8, the value of the weighting gain a varies between 0 and 1 depending on the reliability of the vehicle model. As an index for measuring the reliability of the vehicle model, a difference (deviation) γ_diff between the yaw rate model value γ_clc and the actual yaw rate γ_sens is used. As shown in FIG. 8, the gain map is set so that the value of the weighting gain a increases as the absolute value of the difference γ_diff decreases. The weighting gain calculation unit 220 performs the map processing of FIG. 8 on the difference γ_diff, and calculates a weighting gain a according to the reliability of the vehicle model.

  In FIG. 8, the weighting gain a is a value of 0 to 1 (0 ≦ a <1). When −0.05 [rad / s] ≦ γ_diff ≦ 0.05 [rad / s], the weighting gain a is set to 1 (a = 1).

  When 0.05 <γ_diff or γ_diff <−0.05, the weighting gain a is set to 0 (a = 0).

When 0.05 [rad / s] <γ_diff <0.1 [rad / s], the weighting gain a is calculated by the following equation.
a = −20 × γ_diff + 2

When −0.1 [rad / s] ≦ γ_diff <−0.05 [rad / s], the weighting gain a is calculated by the following equation.
a = + 20 × γ_diff + 2

  The area A1 of the gain map shown in FIG. 8 is an area where the difference γ_diff approaches 0, an area where the S / N ratio of the actual yaw rate γ_sens is small, and an area where the tire characteristics are linear (dry road surface). The reliability of the yaw rate model value γ_clc calculated by the calculation unit 206 is high. Therefore, with the weighting gain a = 1, the feedback yaw rate γ_F / B is calculated from equation (7) with the distribution of the yaw rate model value γ_clc being 100%. As a result, the influence of noise of the yaw rate sensor 142 included in the yaw rate γ_sens can be suppressed, and sensor noise can be excluded from the feedback yaw rate γ_F / B. Therefore, the ride comfort can be improved by suppressing the vibration of the vehicle 1000.

  Here, as a factor that causes a difference between the actual yaw rate γ and the yaw rate model value γ_clc obtained from the vehicle model, there is a dynamic characteristic of the tire shown in FIG. The plane two-wheel model described above assumes a region where the relationship between the tire slip angle and the lateral acceleration (tire cornering characteristics) is linear. In this linear region, the actual yaw rate γ and the yaw rate model value γ_clc are substantially equal. Matches. In the characteristic showing the relationship between the slip angle and the lateral acceleration shown in FIG. 3, in a linear region where the lateral acceleration is linear with respect to the slip angle (a region where the steering speed is relatively slow), the influence of the sensor noise of the yaw rate sensor 142. Occurs. Therefore, in this region, the yaw rate model value γ_clc is used.

  On the other hand, in a region where the cornering characteristics of the tire are non-linear, the yaw rate and the lateral acceleration of the actual vehicle become non-linear with respect to the steering angle and the slip angle, and the two-wheel model and the yaw rate sensed by the actual vehicle deviate. In such a transient nonlinear region, since no noise is generated due to the sensor characteristics of the yaw rate sensor 142, the actual yaw rate γ can be used. The non-linear region corresponds to, for example, the timing of switching the steering. When the actual yaw rate γ exceeds the yaw rate model value γ_clc, it corresponds to a non-linear region and is not affected by sensor noise, so that control based on the true value is possible by using the actual yaw rate γ. When a model taking into account the nonlinearity of the tire is used, the control based on the yaw rate becomes complicated. However, according to the present embodiment, the reliability of the yaw rate model value γ_clc can be easily determined based on the difference γ_diff. In the non-linear region, the actual yaw rate γ can be increased and used. Further, a region that is hardly affected by the dynamic characteristics of the tire can be handled by the yaw rate model value γ_clc.

  The area A2 of the gain map shown in FIG. 8 is an area where the difference γ_diff is large, and corresponds to running on a wet road surface, running on a snowy road, or turning at a high G, and the like. Area. In this region, the reliability of the yaw rate model value γ_clc calculated by the vehicle yaw rate calculation unit 206 decreases, and the difference γ_diff increases. Therefore, the feedback yaw rate γ_F / B is calculated from the equation (7) with the weighting gain a = 0 and the distribution of the actual yaw rate γ_sens as 100%. Thereby, the accuracy of the feedback is secured based on the actual yaw rate γ_sens, and the feedback control of the yaw rate reflecting the behavior of the actual vehicle is performed. Therefore, the turning of the vehicle 1000 can be optimally controlled based on the actual yaw rate γ_sens. Further, since the tire is slipping, even if the signal of the yaw rate sensor 142 is affected by noise, the driver does not perceive the vibration of the vehicle 1000 as a vibration. Regarding the setting of the low μ area A2 shown in FIG. 8, the area where the weighting gain κ = 0 may be determined from design requirements, the steering stability performance when the vehicle 1000 actually runs on the low μ road surface, and the riding stability. It may be determined experimentally from comfort.

  A region A3 of the gain map shown in FIG. 8 is a region (non-linear region) where a transition is made from the linear region to the limit region, and the yaw rate model value γ_clc is also taken into consideration, if necessary, in consideration of the tire characteristics of the actual vehicle 1000. And the distribution of the actual yaw rate γ_sens (weighting gain a) is changed linearly. In the transition from the region A1 (high μ region) to the region A2 (low μ region), or in the region transitioning from the region A2 (low μ region) to the region A1 (high μ region), the torque accompanying the sudden change of the weighting gain a In order to suppress fluctuations and fluctuations in the yaw rate, a weighting gain a is calculated by linear interpolation.

  An area A4 of the gain map shown in FIG. 8 corresponds to a case where the actual yaw rate γ_sens is larger than the yaw rate model value γ_clc. For example, when an incorrect parameter is input to the vehicle yaw rate calculation unit 206 and the yaw rate model value γ_clc is incorrectly calculated, control can be performed using the actual yaw rate γ_sens based on the map of the area A4. It should be noted that the range of the weighting gain a is not limited to 0 to 1, and the configuration of the present invention may be modified so as to take an arbitrary value as long as the range is established as a vehicle control. Fall into a category that can be done.

The subtraction unit 210 subtracts the feedback yaw rate γ_F / B from the control target yaw rate γ_tgt input from the target yaw rate calculation unit 204 to obtain a difference Δγ between the control target yaw rate γ_tgt and the feedback yaw rate γ_F / B. That is, the difference Δγ is calculated from the following equation (8).
Δγ = γ_tgt−γ_F / B (8)
The difference Δγ is input to the vehicle body additional moment calculation unit 232 as a yaw rate correction amount.

  The vehicle body additional moment calculation unit 232 calculates the vehicle body addition moment Mg based on the input difference Δγ such that the difference Δγ becomes 0, that is, the control target yaw rate γ_tgt matches the feedback yaw rate γ_F / B. I do. Specifically, the vehicle body additional moment Mg is calculated from the following equation (9). As a result, at the center position of the vehicle 1000, the vehicle body additional moment Mg required for turning is obtained. A turning moment is applied to the vehicle 1000 based on the vehicle body addition moment Mg.

The target lateral acceleration calculation unit 224 calculates a target lateral acceleration Aylc from the steering angle θh and the vehicle speed V. The target lateral acceleration Ayclc is a model value calculated from the following equation (10) based on the steering angle θh and the vehicle speed V.
Ayclc = (dβ / dt + γ) × V (10)

  In equation (10), the sideslip angle β is calculated from a dynamic model. The sideslip angle β can be calculated from the following equations. In the following equation, ωn indicates a natural frequency, Δn indicates a damping ratio, and A indicates a stability factor.

  Further, the target lateral acceleration calculation unit 224 calculates the peak lateral acceleration during turning Aypeak from the following equations. The turning peak lateral acceleration Apeak is a calculated lateral acceleration in a steady circular turning state, and corresponds to Apeak shown in FIG. 6 described later. Since the peak lateral acceleration during turning Aypeak is a calculated value of lateral acceleration in a steady circular turning state, an immediate value can be obtained from the steering angle θh and the vehicle speed V without including a delay element of the vehicle body. The target lateral acceleration Aylcc and the turning-time peak lateral acceleration Aypeak are input to the target time constant calculator 226. The turning time peak yaw rate γpeak is input to the target time constant calculating unit 226. The peak yaw rate γpeak during turning is a calculated yaw rate value in a steady circular turning state, and is a yaw rate value corresponding to γpeak shown in FIG. The turning peak yaw rate γpeak is calculated from the following equation based on the steering angle θh and the vehicle speed V. Since the peak yaw rate γpeak during turning is a calculated value of the yaw rate in the steady circular turning state, an immediate value can be obtained from the steering angle θh and the vehicle speed V without including a delay element of the vehicle body. The turning peak yaw rate γpeak may be calculated by the target yaw rate calculation unit 204 or may be calculated by the target time constant calculation unit 226.

  The target time constant calculation unit 226 calculates the time Ttgt from the time when γ_clc reaches γpeak to the time when Aylcc reaches Apeak. γ_clc is calculated from a dynamic model and can be calculated from the following equations. Note that γ_clc can also be obtained from the above-described equations (5) and (6).

  The real time constant calculator 228 receives the vehicle lateral acceleration Aysens detected by the lateral acceleration sensor 134. The actual time constant calculation unit 228 receives the actual yaw rate γ_sens detected by the yaw rate sensor. The turning time peak yaw rate γpeak is input to the actual time constant calculation unit 228. The real time constant calculator 228 calculates a time Treal from the time when γ_clc reaches γpeak to the time when Aysens reaches Apeak.

  The low μ judgment output gain calculation section 234 calculates a low μ judgment output gain μG from the target time constant Ttgt and the real time constant Treal. Details will be described later with reference to the flowchart shown in FIG.

  The low μ determination output gain μG calculated by the low μ determination output gain calculation unit 234 is input to the multiplication unit 236. To the multiplying unit 236, the vehicle body additional moment Mg calculated by the vehicle body additional moment calculation unit 232 is also input. The multiplication unit 236 calculates a correction value Mg ′ of the vehicle body additional moment Mg by multiplying the vehicle body additional moment Mg by the low μ determination output gain μG.

  The correction value Mg ′ is input to the motor required torque calculator 238. The motor required torque calculation unit 238 calculates ΔTv from the following equation (11) in order to convert the moment into torque using the correction value Mg ′ obtained by multiplying the vehicle body additional moment Mg by the turning assist gain βG. Then, the required motor torque calculation unit 238 calculates the additional torque Tvmot from the following equation (12).

  In the equation (11), TrdR is the tread width of the rear wheels 104 and 106. TireR is the tire radius of the front wheels 100 and 102 and the rear wheels 104 and 106, and Gratio is the gear ratio of the gearboxes 120 and 122 of the rear wheels 104 and 106. According to equation (11), the correction value Mg 'of the vehicle body additional moment Mg at the center position of the vehicle 1000 is converted into the motor torque? Tv of the rear wheels 104 and 106. Then, the motor torque of each of the rear wheels 104 and 106 necessary to generate the correction value Mg 'is obtained from the equation (12).

By the way, the driving force of the front wheels 100 and 102 and the rear wheels 104 and 106 is determined by a motor torque instruction value reqTq determined from a required driving force (opening degree of an accelerator pedal) of the driver when the vehicle 1000 goes straight. Here, the motor torque instruction value reqTq is calculated from the following equation (13).
reqTq = reqF * TireR * Gratio (13)
In the equation (13), reqF is a required driving force determined from the opening degree of the accelerator pedal. The opening of the accelerator pedal is detected by an accelerator opening sensor 146.

  When the vehicle 1000 is traveling straight, the driving force of each of the four motors 108, 110, 112, 114 for driving the front wheels 100, 102 and the rear wheels 104, 106 is a motor torque instruction value reqTq based on the driver's required driving force reqF. It becomes a value obtained by dividing into four (= reqTq / 4). On the other hand, when the vehicle 1000 turns, an additional torque Tvmot based on the vehicle body additional moment Mg ′ calculated from the equation (12) is added to the motor torque instruction value reqTq for the rear wheels 104 and 106 by the torque vectoring control. Since the additional torque Tvmot based on the vehicle body additional moment Mg 'is a couple, in the case of a right turn, the motor torque instruction value of the left rear wheel 104 is obtained by adding the additional torque Tvmot to the motor torque instruction value reqTq / 4 for straight ahead. The motor torque command value for the right rear wheel 106 is a value obtained by subtracting the additional torque Tvmot from the motor torque command value reqTq / 4 for straight traveling. Similarly, in the case of a left turn, the motor torque instruction value of the right rear wheel 106 is a value obtained by adding the additional torque Tvmot to the motor torque instruction value reqTq / 4 for straight traveling, and the motor torque instruction value of the left rear wheel 104 is obtained. Is a value obtained by subtracting the additional torque Tvmot from the motor torque instruction value reqTq / 4 when traveling straight.

Therefore, the motor torque instruction value of each of the motors 108, 110, 112, 114 at the time of turning can be expressed by the following equations (14) to (17). The motor required torque calculating unit 228 calculates the motor torque instruction values TqmotFl, TqmotFr, TqmotRl, TqmotRr of each of the motors 108, 110, 112, 114 based on the equations (14) to (17).
TqmotFl (instruction value of motor torque for front left wheel) = reqTq / 4 (14)
TqmotFr (instruction value of motor torque for front right wheel) = reqTq / 4 (15)
TqmotRl (motor torque indication value for left rear wheel)
= ReqTq / 4-(± Tvmot) (16)
TqmotRr (right rear wheel motor torque indication value)
= ReqTq / 4 + (± Tvmot) (17)
The sign of the additional torque Tvmot is set according to the turning direction.

  FIG. 6 is a characteristic diagram showing a comparison between a sensor value and a model value obtained from a two-wheel vehicle model using the vehicle speed V and the steering angle θh as inputs. FIG. 6 shows how the model value γclc of the yaw rate and the lateral acceleration of the vehicle change over time when the steering angle θh is changed stepwise at time t0. As shown in FIG. 16, when the steering angle θh is changed stepwise, the model value γclc of the yaw rate decreases after reaching the peak value γpeak. In addition, the lateral acceleration also decreases after reaching the peak value Apeak. FIG. 6 shows the calculated values of the yaw rate in two road surface states of high μ (solid line characteristic) and low μ (dashed line characteristic). The yaw rate γclc does not include a factor of the road surface condition. Further, regarding the yaw rate γclc, since the braking / driving force control of the left and right wheels is performed by the turning assist control, the yaw rate does not greatly differ between the high μ road and the low μ road. Therefore, as shown in FIG. 6, with respect to the yaw rate, the temporal transitions of the model value and the measured value coincide with each other at both the high μ and the low μ.

  On the other hand, since the lateral acceleration (Ay) is affected by the force-reaction force relationship between the tire and the road surface, its generation is delayed with respect to the yaw rate according to the road surface μ. As shown in FIG. 6, when the time t1 at which the yaw rate reaches the peak value γpeak is used as a reference, the time t2 at which the model value Aylcc of the lateral acceleration reaches Apeak is delayed by Ttgt, and the sensor value Aysens of the lateral acceleration reaches Apeak. The time t3 is delayed by Real.

  As for the lateral acceleration, the model value does not include a factor of the road surface condition, and therefore, the model value matches at high μ and low μ. On the other hand, as the sensor value becomes lower μ, the reaction force from the road surface to the tire becomes more difficult to be transmitted, so that the rise of the lateral acceleration becomes slower than when the sensor value becomes high μ. Therefore, the road surface condition can be estimated according to the ratio between Ttgt and Real.

  Here, if Treal matches Ttgt, a reaction force equivalent to the model value is generated on the tire from the road surface, and it can be estimated that the tire has a high μ. On the other hand, as Treal is larger than Ttgt, the occurrence of the lateral acceleration of the sensor value is delayed, and the reaction force applied to the tire from the road surface is delayed, so that it can be estimated that the μ is low. In the present embodiment, the road surface μ is estimated according to the ratio between Real and Ttgt, and the gain of the turning assist control is changed according to this ratio.

  In the present embodiment, Ttgt / Treal is calculated as this ratio and used as a gain for turning assist control. If Ttgt and Treal are equal, Ttgt / Treal = 1, and since the μ is high, the turning assist torque is not reduced. On the other hand, as the value of Ttgt / Treal becomes smaller, the lateral acceleration is delayed with respect to the yaw rate and becomes lower μ. Therefore, the turning assist torque is reduced by multiplying the turning assist torque by the value of Ttgt / Treal. As a result, the front-rear force of the tire is reduced and the lateral force can be increased, so that the occurrence of slip, spin, and the like of the vehicle can be reliably suppressed.

  Here, the turning assist torque is controlled based on Ttgt / Treal, but since the value of Treal changes according to the road surface condition, the turning assist torque may be controlled based only on the value of Real. . Thus, control can be performed with simpler processing.

  Next, processing performed by the control device 200 according to the present embodiment will be described. FIG. 9 is a flowchart showing the overall processing of the present embodiment. First, in step S100, it is determined whether or not an ignition key (ignition SW) is on. If the ignition key has been turned on, the process proceeds to step S102, and if the ignition key has not been turned on, the process waits in step S100.

  In step S102, it is determined whether or not the inhibitor position sensor (IHN) 144 indicates the position of P (parking) or N (neutral). If the position is P (parking) or N (neutral), step S104 is performed. Proceed to. If it is not the P (parking) or N (neutral) position in step S102, the process proceeds to step S106, where it is determined whether or not the ignition key is turned on. If the ignition key is turned on, the process returns to step S102. . If the ignition key is turned off in step S106, the process proceeds to step S108, ends the vehicle start-up process, and returns to step S100.

  In step S104, the vehicle 1000 is started, and in the next step S110, it is determined whether or not the inhibitor position sensor (IHN) 144 indicates the position of D (drive) or R (reverse). When the inhibitor position sensor (IHN) 144 indicates the position of D (drive) or R (reverse), the process proceeds to step S112, and the processing of the traveling control is started. On the other hand, if the inhibitor position sensor (IHN) 144 does not indicate the position of D (drive) or R (reverse) at step S110, the process proceeds to step S113, where it is determined whether the ignition key is turned on, and the ignition is performed. If the key has been turned on, the process returns to step S110. If the ignition key is turned off in step S113, the process proceeds to step S108, ends the vehicle start-up process, and returns to step S100.

  After step S112, the process proceeds to step S114, where the amount of operation of the accelerator pedal (accelerator opening) by the driver is detected from the detection value of the accelerator opening sensor 146. In the next step S115, it is determined whether or not the operation amount of the accelerator pedal is 0.1 or more. If the operation amount is 0.1 or more, the process proceeds to step S116. In step S116, the required driving force reqF is calculated based on the operation amount of the accelerator pedal. The required driving force reqF can be calculated based on, for example, a map that defines the relationship between the accelerator opening and the required driving force reqF. On the other hand, when the operation amount of the accelerator pedal is less than 0.1, the process proceeds to step S118, and regenerative braking control of each of the motors 108, 110, 112, 114 is performed.

  After steps S116 and S118, the process proceeds to step S120. In step S120, it is determined whether or not the absolute value of the steering angle θh detected by the steering angle sensor 138 is 1 [deg] or more, and if the absolute value of the steering angle θh is 1 [deg] or more, Proceed to step S122. In step S122, the additional torque Tvmot is calculated by the above-described method, and feedback control to the target moment γ_Tgt is performed based on the additional torque Tvmot. Therefore, in the next step S124, the motor torque instruction values of the motors 108, 110, 112, 114 are calculated from the equations (14) to (15) based on the additional torque Tvmot, and the respective motors 108, 110, 112 are calculated. , 114 for output. In the next step S126, the acceleration of the vehicle 1000 is detected by the longitudinal acceleration sensor 132 and the lateral acceleration sensor 134. After step S126, the process returns to step S114.

  Next, main processing of the processing in FIG. 9 will be described in detail. FIG. 10 is a flowchart showing the process of step S122 in FIG. Here, FIG. 10 is a flowchart illustrating a process in which the weighting gain calculator 220 calculates the weighting gain a. The process of FIG. 10 functions as a process of removing the noise of the yaw rate sensor 142 by calculating the feedback yaw rate γ_F / B by allocating the actual yaw rate γ_sens and the yaw rate model value γ_clc based on the weighting gain a. First, in step S200, an actual yaw rate γ_sens and a yaw rate model value γ_clc are acquired. In the next step S201, a difference γ_diff between the actual yaw rate γ_sens and the yaw rate model value γ_clc is calculated. In the next step S202, a weighting coefficient a is calculated based on the gain map of FIG. In the next step S204, a feedback yaw rate γ_F / B is calculated based on equation (7) described above. The calculated feedback yaw rate γ_F / B is used for calculating the difference Δγ in step S224 of FIG.

  FIG. 11 is a flowchart illustrating a process in which the vehicle yaw rate calculation unit 206 calculates the yaw rate model value γ_clc. First, in step S210, the steering angle θh and the vehicle speed V are acquired. In the next step S212, the yaw rate model value γ_clc is calculated by solving the equations (5) and (6) simultaneously. The calculated yaw rate model value γ_clc is used for calculating the feedback yaw rate γ_F / B in step S204 of FIG.

  FIG. 12 is a flowchart showing a process for calculating the additional torque Tvmot. First, in step S220, the target yaw rate calculation unit 204 acquires the steering angle θh and the vehicle speed V. In the next step S222, a target yaw rate γ_tgt is calculated from the equations (2) to (4) based on the steering angle θh and the vehicle speed V. In the next step S224, a difference Δγ between the control target yaw rate γ_tgt and the feedback yaw rate γ_F / B is calculated based on equation (7). In the next step S226, a vehicle body additional moment Mg is calculated from equation (9).

  In the next step S228, the low μ judgment output gain calculation section 234 calculates the low μ judgment output gain μG. In the next step S230, ΔTv is calculated based on equation (11), and additional torque Tvmot is calculated based on equation (12). Based on the calculated additional torque Tvmot, a motor torque instruction value for each wheel is calculated in step S124 in FIG.

  FIG. 13 is a flowchart showing a process for calculating the low μ determination output gain μG. First, in step S240, various parameters (sensor values) such as the lateral acceleration Aysens detected by the lateral acceleration sensor 134, the actual yaw rate γ_sens detected by the yaw rate sensor 142, the steering angle θh, and the vehicle speed V are acquired. In the next step S242, model values γ_clc and Aylc of the yaw rate and the lateral acceleration are calculated.

In the next step S244, a peak yaw rate γ _peak during turning and a peak lateral acceleration Ay _peak during turning are calculated. The turning peak yaw rate γ _peak and the turning peak lateral acceleration Ay _peak are both calculated values in a steady circular turning state, and are model values (theoretical values) calculated from the steering angle θh and the vehicle speed V.

  In the next step S246, a target time constant Ttgt is calculated. In the next step S248, a real time constant Treal is calculated. The specific processing of steps S246 and S248 will be described later with reference to FIGS. In the next step S250, an occurrence time difference Tdec_clc is calculated. Here, Tdec_clc = Ttgt / Treal.

  In the next step S252, it is determined whether or not Tdec_clc ≧ 1, and if Tdec_clc ≧ 1, the process proceeds to step S254. In this case, since the value of Ttgt is equal to or greater than the value of Real, there is no delay in the sensor value with respect to the model value of the lateral acceleration, and it can be estimated that the road surface is high μ. Therefore, in step S254, the low μ determination output gain μG is set to 1 (μG = 1).

  On the other hand, if Tdec_clc <1 in step S252, the process proceeds to step S256. In this case, since the value of Ttgt is less than the value of Real, a delay occurs in the sensor value with respect to the model value of the lateral acceleration, and it can be estimated that the road surface is low μ. Therefore, in step S256, the low μ determination output gain μG is set to Ttgt / Treal (μG = Ttgt / Treal). Therefore, ΔTv calculated from the equation (11) is reduced by the low μ determination output gain μG, and the turning assist torque is reduced. Thereby, the behavior of the vehicle 1000 can be stabilized at the time of low μ.

  FIG. 14 is a flowchart showing a process for calculating the target time constant Ttgt in step S246 in FIG. First, in step S300, it is determined whether or not γpeak = γ_clc. If γpeak = γclc, the process proceeds to step S302. In this case, since γclc has reached the yaw rate peak value γpeak obtained from the steering angle θh and the vehicle speed V, in step S302, from when the turning peak yaw rate γpeak is generated until when the turning peak lateral acceleration Atypepeak (model value) is generated. The count-up for calculating the time difference Ttgt is started. Note that γpeak is a calculated value of the slip angle in a steady circular turning state, and γ_clc is calculated from a dynamic model. Therefore, γ_clc has a longer time delay than γpeak.

  On the other hand, if γpeak ≠ γclc in step S300, the process waits in step S300.

  In the next step S304, it is determined whether or not Aypeak = Aylcc. If Aypeak = Aylcc, the process proceeds to step S306. In this case, since the vehicle lateral acceleration Aylcc has reached the peak lateral acceleration Aypeak during turning obtained from the steering angle θh and the vehicle speed V, the count-up is ended in step S306, and the target time constant Ttgt = I (I is the count-up end. Hour count value). On the other hand, if AypeakＡAyclc in step S304, the process returns to step S302 and continues counting up. Note that Aypeak is a calculated value of lateral acceleration in a steady circular turning state, and Aylc is calculated from a dynamic model, so that Ayclc has a longer time delay than Aypeak.

  As described above, the target time constant Ttgt in step S246 in FIG. 13 is calculated by the processing in FIG.

  FIG. 15 is a flowchart showing a process of calculating the real time constant Treal in step S248 of FIG. First, in step S310, it is determined whether or not γpeak = γsens. If γpeak = γsens, the process proceeds to step S312. In this case, since γsens has reached the yaw rate peak value γpeak obtained from the steering angle θh and the vehicle speed V, in step S312, the time difference Real from the generation of the turning peak yaw rate γpeak to the generation of the turning peak lateral acceleration Apeak is calculated. To start counting up.

  On the other hand, if γpeak ≠ γsens in step S310, the process waits in step S310.

  In the next step S314, it is determined whether or not Atype = Aysens, and if Atype = Aysens, the process proceeds to step S316. In this case, since the sensor value Aypeak of the lateral acceleration has reached the peak lateral acceleration Aypeak during turning obtained from the steering angle θh and the vehicle speed V, the count-up is terminated in step S316, and the target time constant Treal = J (J is the count Count value at the end of up). On the other hand, if AypeakＡAysens in step S314, the process returns to step S312 to continue counting up.

  As described above, the actual target time constant Treal in step S248 in FIG. 13 is calculated by the processing in FIG.

  FIG. 16 is a characteristic diagram for explaining an effect when the control according to the present embodiment is performed. When the control according to the present embodiment is not performed, when the torque control is performed only on the rear wheels 104 and 106 and the yaw of the vehicle body is to be made to converge, the grip of the rear wheels 104 and 106 is already saturated, and the occurrence of the yaw A phase difference occurs in the occurrence of steering (St_ang). In FIG. 16, the upper characteristic group shows the steering angle and the yaw rate, and the lower characteristic group shows the lateral acceleration. As shown in FIG. 16, it can be seen that the delay of the lateral acceleration with respect to the yaw rate changes between the time of high μ (hi) and the time of low μ (low). According to the present embodiment, the turning assist control is performed on a low μ road, and when the slip of the rear wheels 104 and 106 and the vehicle body slip occur, the convergence of the yaw rate (γ_low) at the time of the low μ is achieved in addition to the reduction of the skid amount. Improved.

  FIGS. 17 and 18 show how the parameters of the steering angle, the yaw rate, and the lateral acceleration change when a double lane change is performed on a low μ road when the control according to the present embodiment is performed. FIG. 9 is a characteristic diagram showing a comparison between (with control) and not (without control). When the control according to the present embodiment is performed, it can be seen that the steering amount and the yaw rate converge because the driving force control is suppressed by the road surface determination. In addition, it can be seen that the lateral force (lateral acceleration) is obtained because the slip can be prevented in the lane change in the latter half of the traveling.

  As described above, according to the present embodiment, when turning, the lateral acceleration occurs later than the yaw rate of the vehicle, so that the road surface μ can be estimated according to the time difference between the yaw rate and the lateral acceleration. Therefore, by controlling the assist torque of the turning assist control based on this time difference, it becomes possible to greatly increase the behavior stability of the vehicle when the vehicle body slips.

  As described above, the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to such examples. It is apparent that those skilled in the art to which the present invention pertains can conceive various changes or modifications within the scope of the technical idea described in the claims. It is understood that these also belong to the technical scope of the present invention.

142 Yaw rate sensor 200 Control device 204 Target yaw rate calculation unit 206 Vehicle yaw rate calculation unit 208 Yaw rate F / B calculation unit 220 Weighting gain calculation unit 224 Target lateral acceleration calculation unit 226 Target time constant calculation unit 228 Real time constant calculation unit 232 Vehicle body added moment Calculation unit 234 Low μ judgment output gain calculation unit 236 Multiplication unit 1000 Vehicle

Claims (10)

A vehicle body addition moment calculation unit that calculates a vehicle body addition moment to be added to the vehicle body independently of the steering wheel steering system based on the yaw rate of the vehicle;
At the time of turning the vehicle, a time difference information acquisition unit that acquires information about a time difference between a change in the lateral acceleration with respect to a change in the yaw rate,
Based on the information about the time difference, a correction unit that corrects the vehicle body additional moment,
A control device for a vehicle, comprising:   The vehicle control device according to claim 1, wherein the time difference information acquisition unit acquires information on the time difference when the vehicle starts turning. The time difference information acquisition unit,
A first calculation unit that calculates a first time until the model value of the lateral acceleration reaches the predetermined value as a starting point at which the model value of the yaw rate has reached the predetermined value;
A second calculation unit that calculates a second time until the sensor value of the lateral acceleration reaches the predetermined value, as a starting point at which the model value of the yaw rate has reached the predetermined value; The control device for a vehicle according to claim 1, wherein information about the time difference is obtained from the second time.   The first calculation unit sets the time when the model value of the yaw rate reaches a peak value after the start of the turn at the start of the turn, as the starting point, until the model value of the lateral acceleration reaches the peak value after the start of the turn. The control device for a vehicle according to claim 3, wherein the first time is calculated. The second calculator is configured such that the time when the model value of the yaw rate reaches a peak value after the start of the turn at the start of the turn is the starting point, and the time until the sensor value of the lateral acceleration reaches the peak value after the start of the turn. The vehicle control device according to claim 3, wherein the second time is calculated. Based on the information about the time difference, the correction unit includes a gain calculation unit that calculates a gain for correcting the vehicle body additional moment,
The vehicle control device according to claim 3, wherein the gain calculation unit sets a ratio of the first time to the second time as the gain.   The vehicle control device according to claim 6, wherein the correction unit corrects the vehicle body additional moment by multiplying the vehicle body additional moment by the ratio. A target yaw rate calculation unit that calculates a target yaw rate based on the steering angle and the vehicle speed;
A vehicle yaw rate calculation unit that calculates a yaw rate model value from the vehicle model;
A yaw rate sensor for detecting an actual yaw rate of the vehicle,
A feedback yaw rate calculator that distributes the yaw rate model value and the actual yaw rate based on a difference between the yaw rate model value and the actual yaw rate, and calculates a feedback yaw rate from the yaw rate model value and the actual yaw rate;
With
The vehicle body additional moment calculation unit,
The vehicle control device according to any one of claims 1 to 7, wherein the vehicle body additional moment is calculated based on a difference between the target yaw rate and the feedback yaw rate.   9. A motor required torque calculating unit for calculating a required motor torque for individually controlling motors driving each of the rear left and right wheels of the vehicle based on the vehicle body additional moment. The control device for a vehicle according to any one of the above. Calculating a body addition moment to be added to the body independently of the steering wheel system based on the yaw rate of the vehicle;
Acquiring information about a time difference between a change in the lateral acceleration and a change in the yaw rate when the vehicle is turning;
Correcting the vehicle body additional moment based on the information about the time difference;
A control method for a vehicle, comprising:

Images