JP2021062821A - Vehicle motion controller and program - Google Patents

Abstract

To provide a vehicle motion controller and a program that can achieve both suppression of oscillation of the heads of occupants, and follow-up of a target track.SOLUTION: A vehicle motion controller includes: deflection angle detection means for detecting a deflection angle between an advancing direction of a vehicle and a direction of a target arrival point after a forward gaze time on a target course along which the vehicle travels; target value arithmetic means for calculating a yaw angular velocity which is proportional to the detected deflection angle, as a first target value of a yaw angular velocity after a dead time of one third of the forward gaze time; target value correction means for correcting the first target value of the yaw angular velocity, in order to obtain a second target value of the yaw angular velocity in the case of performing feed forward control for controlling oscillation of heads of occupants; and vehicle motion control means for controlling vehicle motion so that the second target value of the yaw angular velocity is realized after the dead time. The dead time is a first dead time corresponding to a phase delay of the feed forward control, or a sum of the first dead time and a second dead time corresponding to a phase delay of a transfer function from an actual steering angle to a yaw angular velocity.SELECTED DRAWING: Figure 13

Description

The present invention relates to a vehicle motion control device and a program.

The techniques described in Patent Document 1 and Patent Document 2 are known as techniques for maintaining occupant comfort in vehicle motion control such as automatic driving. The technique described in Patent Document 1 provides a comfortable ride to the occupant by setting the lateral acceleration and lateral jerk of the vehicle (that is, "lateral jerk" which is the lateral jerk) to a predetermined value or less. Is. Further, the technique described in Patent Document 2 eliminates the discontinuity of the lateral jerk by continuously controlling the rate of change of the curvature of the traveling locus, which is the target of automatic driving, and maintains the comfort of the occupant. ..

Japanese Unexamined Patent Publication No. 7-179140 JP-A-2017-193189

In these conventional techniques, attention is paid to the movement of the vehicle, and the comfort of the occupant is maintained by imposing restrictions on the movement of the vehicle such as lateral acceleration and lateral acceleration. However, in order to maintain the comfort of the occupants, it is necessary to pay attention to the movements of the occupants moved in the vehicle interior, not the movements of the vehicle.

There is a dynamic characteristic between the vehicle movement and the occupant's movement, and even if the vehicle movement is restricted, the occupant's movement cannot always be restricted and the occupant's comfort cannot be maintained. For this reason, the above-mentioned prior art has problems such as the movement of the occupant in the vehicle interior cannot be appropriately suppressed, and excessive suppression is performed.

An object of the present invention is to provide a vehicle motion control device and a program capable of both suppressing the sway of the occupant's head and following the target trajectory.

The vehicle motion control device of the present invention includes a deviation angle detecting means for detecting an deviation angle between the traveling direction of the vehicle and the direction of the target arrival point after a predetermined forward gaze time on the target course on which the vehicle travels. A target value calculation means for calculating the detected yaw angle velocity proportional to the deviation angle as the first target value of the yaw angle velocity after a dead time of one-third of the forward gaze time, and the first target of the yaw angle velocity. The target value correction means for obtaining the second target value of the yaw angle velocity when the feed forward control for suppressing the head sway of the occupant is performed by correcting the value, and the second target value of the yaw angle velocity are wasteful. The waste time is the first waste time corresponding to the phase delay of the feed forward control, or the first waste time and the actual steering, including the vehicle motion control means for controlling the vehicle motion so as to be realized after the time. This is a vehicle motion control device that is the sum of the second dead time corresponding to the phase delay of the transmission function from the angle to the yaw angle velocity.

According to the present invention, it is possible to suppress the sway of the occupant's head and follow the target locus at the same time.

It is a figure which shows the human body behavior modeled as an equivalent double inverted pendulum. It is a figure for demonstrating the definition of a symbol. (A) and (B) are Bode plots of the head posture angle from the vehicle lateral acceleration. It is a graph which shows the comparison between the experimental result and the calculation result. (A) to (C) are graphs showing the time response of the posture angle and the head displacement at the time of inputting the vehicle lateral acceleration step (1 m / s ^2). (A) and (B) are Bode plots of the feedforward controller. (A) to (C) are graphs showing the effect of feedforward control at the time of vehicle lateral acceleration step input (1 m / s ^2). It is a graph which shows the deviation of the locus with feedforward control. (A) to (C) are graphs showing the dead time approximation of the phase lag. It is a graph which shows the target locus shifted by the waste time. It is a figure for demonstrating the travel locus control of a vehicle based on the forward gaze angle. It is a functional block diagram which shows an example of the structure of the vehicle steering control device which concerns on the prior application. It is a functional block diagram which shows an example of the structure of the vehicle steering control device which concerns on embodiment of this invention. It is a graph which shows the result of the traveling locus control of a vehicle. It is a graph which shows the time response of a vehicle lateral acceleration and a head displacement.

Hereinafter, an example of the embodiment of the present invention will be described in detail with reference to the drawings.

<Control to suppress the sway of the occupant's head>
The driver while driving a car utilizes these degrees of freedom to ensure forward visibility and controls muscles to maintain his / her posture while suppressing head sway. In the present embodiment, the posture control of a seated occupant when a lateral acceleration is applied to the vehicle is simulated by using a human body behavior model. In consideration of the dynamic characteristics between the vehicle and the occupant obtained as a result, feedforward control is performed with respect to the input of the lateral acceleration in order to suppress the sway of the occupant's head.

(Human body behavior model)
First, the human body behavior model will be described.
FIG. 1 is a diagram showing human body behavior modeled as an equivalent double inverted pendulum. As shown in FIG. 1, in the present embodiment, a two-mass system model in which the human body is simplified by three parts and two joints between them is used. The human body is approximated by a two-mass model, and the otolith feedback and somatic feedback described later are applied to the two-mass model.

As shown in FIG. 1, in the present embodiment, the human body seated on the automobile seat is held by the seat, which is the “cephalothorax” which is the non-contact part of the seat above the point of contact of the back with the seat back. It is assumed that the human body behavior model is divided into three parts, the "abdomen" including the pelvis and the "lower body" fixed to the vehicle below the pelvis.

The joint torque applied to the joint between the lower body and the pelvis plays a role of "physical feedback" for detecting the collapse of the posture from the joint angle and maintaining the posture. The joint torque applied to the joints between the head / chest and the abdomen, including the pelvis, acts as "otolith feedback" to invert the head from the acceleration detected by the otolith.

The main points of modeling this human body behavior model are as follows (1) to (3).

(1) The upper body is divided into a non-contact part (head / chest) above the seat from the point of contact of the back with the seat back and a part (pelvis / abdomen) held by the seat.

(2) Ear stone feedback (joint torque of the spine between the head / chest and pelvis) that keeps the posture of the head against the acceleration of the vehicle and physical feedback (of the pelvis and lower body) that tries to correct the posture collapse. The posture is stabilized by the joint torque between them).

(3) The joint position of the spine, which expresses the flexion of the spine with one joint, changes up and down with the force point as the body is supported by the seat and the driving position changes (when flexing to the center of the waist instead of the entire spine). Such).

When a lateral acceleration is applied, the binding force of the seat on the abdomen including the pelvis is smaller than in the anterior-posterior direction, and the center of gravity of the abdomen moves on the seat back. When this behavior is projected on the plane of the left-right-up-down axis of the vehicle, it is an equivalent inverted double pendulum as shown in FIG.

Next, the formulation is carried out based on the above points.
First, each physical quantity is defined for formulation. FIG. 2 is a diagram for explaining the definition of the symbol. As shown in FIG. 2, the joint (joint 1) angle between the lower body and the pelvis is θ, the joint (joint 2) angle between the head / chest and the abdomen is φ, the abdominal mass m ₁ , the head / chest mass. the m _2, the distance between the joint 1 and joint 2 L, l ₁ a distance from the joint 1 to the abdominal center of _gravity, the distance from the joint 2 to the center of gravity of the head, chest l, lower body (vehicle body) position (x , Y), the abdominal center of gravity position (x ₁ , y ₁ ), the head / chest center of gravity position (x ₂ , y ₂ ), the force applied to the joint 1 (u ₁ , v ₁ ), and the joint 2 applied. The force to be applied is described _{as (u 2} , v _2).

また、このときの頭部横加速度を下記記号で記述する。

In addition, the lateral acceleration of the lower body (vehicle body) is described by the following symbols.

In addition, the lateral acceleration of the head at this time is described by the following symbols.

The moment of inertia of the abdominal mass point with respect to the rotation of joint 1 is expressed by the following equation.

The moment of inertia of the cephalothorax mass point with respect to the rotation of the joint 2 is expressed by the following equation.

Furthermore, the equation of motion of the abdominal mass point and joint 1 is described as follows.

However, g is the gravitational acceleration, c ₁ is the viscosity coefficient of the joint 1, c ₂ is the viscosity coefficient of the joint 2, T _θ is the control torque of the joint 1, and T _φ is the control torque of the joint 2. Similarly, the equations of motion for the head / chest mass points and joints 2 are described as follows.

From this relationship, the following equation can be obtained.

Substituting Eq. (9) into Eq. (3) gives the following equation.

Substituting Eq. (10) into Eq. (4) gives the following equation.

Substituting Eq. (11) into Eq. (6) gives the following equation.

Substituting Eq. (12) into Eq. (7) gives the following equation.

Further, by substituting Eq. (15) into Eq. (13), the following equation is obtained.

Substituting Eq. (16) into Eq. (14) gives the following equation.

Here, by substituting equations (15), (16), (17), and (18) into equation (5), the following equation is obtained.

Substituting equations (15) and (16) into equation (8) gives the following equation.

(Physical feedback)
In the joint 1, it is assumed that the somatic feedback that suppresses the collapse of the posture works, and the following control rule that feeds back the joint angle θ is set. However, k ₁ is the control gain of the somatic feedback.

(Otolith feedback)
Further, in the joint 2, it is assumed that the otolith feedback that tilts the head against the lateral acceleration works, and the following control rule is set. However, k ₂ is the control gain of the otolith feedback.

From the above equation of motion and control law, the equation of motion of the closed loop system is derived.
Substituting equations (21) and (22) into equation (19) gives the following equation.

Substituting equations (21) and (22) into equation (20) gives the following equation.

Here, the equations of motion of Eqs. (23) and (24) are expressed as equations of state.

However, J, A, and B are defined as follows.

The following equations of state are derived from these relational expressions.

However, A _h and B _h are defined as follows.

The equation of state shown in the above equation (29) is an equation for calculating the behavior of the occupant with respect to the lateral acceleration. The behavior of the occupant includes the joint angle θ of the lateral joint 1 and the joint angle φ of the lateral joint 2.

Next, the result of calculating the behavior of the occupant with respect to the lateral acceleration is shown.
Here, the simulation was performed using the following parameters.
m ₁ = 30 [kg]
m ₂ = 10 [kg]
L = 0.3 [m]
l ₁ = 0.25 [m]
l = 0.4 [m]
c ₁ = 12 [Nms / rad]
c ₂ = 12 [Nms / rad]
k ₁ = 60 [Ns ² / rad]
k ₂ = 5 [Ns ² / rad]

3 (A) and 3 (B) are Bode plots of the above equations (29) to (31). That is, it is a Bode diagram of the head posture angle from the vehicle lateral acceleration. The human body behavior is a non-minimum phase transition system in which the phase delay exceeds 180 deg.

FIG. 4 shows the experimental results (solid line) from the lateral acceleration of the floor to the lateral acceleration of the head when the seat with the occupant seated on the exciter was installed and vibrated in the lateral direction, and the entire spine was bent. It is the figure which compared the calculation result (broken line) of the case. In the experiment, the frequency is swept from 0.35 Hz to 15 Hz in 90 s under the condition that the lateral acceleration is 0.1 G and the amplitude is constant. In the model, it can be seen that the actual human body behavior can be expressed even though the upper body is described as a two-inertial system with an extremely low degree of freedom.

(Reverse response in human behavior)
5 (A) to 5 (C) are graphs showing the time response of the posture angle and the head displacement at the time of inputting the vehicle lateral acceleration step (1 m / s ^2). It can be seen that the human body behavior shows a reverse response at the initial stage of lateral acceleration application as surrounded by a circle. For example, when the vehicle turns, lateral acceleration occurs outward. The occupant's head is first swung outward by turning due to lateral acceleration, but then falls inward to balance the centrifugal force and gravity. This sway of the occupant's head is the reverse response.

(Reduction of reverse response)
In the present embodiment, feedforward control is performed with respect to the input of lateral acceleration in order to suppress the shaking of the occupant's head.

The output equation for outputting the head displacement y is expressed by the following equation.

The transfer function from the vehicle lateral acceleration to the head displacement can be obtained by expressing the state equation related to the human body behavior shown in the above equation (29) and the above output equation as a transfer function. The transfer function from vehicle lateral acceleration to head displacement is described by Eq. (32) below, where the numerator is a quadratic polynomial containing an unstable zero point and the denominator is a quaternary polynomial.

Here, s is the Laplace operator, den (s) is a fourth-order polynomial whose 0th-order term with respect to s is 1, K is a steady-state gain, z ₁ is a stable zero point, and -z ₂ is an unstable zero point. is there.

In general, the response characteristics of the controlled object can be freely designed by exchanging the numerator and denominator of the transfer function and including an inverse model in which the pole zero is inverted in the feedforward control. However, when the controlled object contains an unstable zero point in the numerator, the inverse model diverges, the unstable zero point cannot be offset, and the design cannot be completely freely performed.

Therefore, the target transfer function is set by "waste time + second-order lag system".
That is, the time for counter-responding is suppressed to zero, the time for suppressing at zero is defined as "waste time", and the characteristic of rising from there with a response of "secondary delay" is set as the target characteristic. The target transfer function is expressed by the following equation.

The transfer function of wasted time is expressed by the following equation by Pade approximation.

Therefore, Eq. (33) is expressed by the following equation.

Here, assuming that L = 2z ₂ , the equation (35) is converted into the following equation (36).

That is, when Eq. (36) is set as the target characteristic, the same unstable zero point as Eq. (32) is included in the target characteristic. Assuming that the feedforward control for converting the _{transfer function Ph} (s) described by Eq. (32) into _{the target characteristic P 0} (s) of Eq. (36) _{is C (s), Ph} (s), P _{There is the following relationship between 0} (s) and C (s).

By solving Eq. (37), the feedforward control C (s) for realizing the target characteristic of Eq. (36) can be obtained as shown in the following equation.

Here, den ₀ (s) is defined by the following equation.

According to the feedforward control of the above equation (38), the transfer function from the vehicle lateral acceleration to the head displacement can be approximated to the target characteristic.

6 (A) and 6 (B) are Bode diagrams of the feedforward controller of the equation (38). As illustrated, feedforward control involves a phase lag. 7 (A) to 7 (C) are graphs showing the effect of feedforward control at the time of vehicle lateral acceleration step input (1 m / s ^2). In FIGS. 7A to 7C, the simulation result of the human body behavior via the feedforward control of the equation (38) is shown by a solid line, and the step response of FIG. 5 is shown by a broken line.

Although the simulation result is not a complete response of "waste time + second-order delay" due to the approximation, the step input of the lateral acceleration is performed via the feedforward control of Eq. (38). It can be seen that the head sway is reduced by 35%.

(Followability to the target trajectory)
FIG. 8 is a graph showing the deviation of the locus due to the feedforward control. In FIG. 8, the simulation result of the vehicle locus with feedforward control is shown by a solid line. In addition, the simulation result of the vehicle locus without feedforward control is shown by a broken line.

Here, the vehicle speed is set to 5 [m / s], the dynamic characteristics of the vehicle motion are ignored, and the vehicle body slip angle is always 0, that is, the yaw angular velocity is expressed by lateral acceleration ÷ vehicle speed. It is a simulation.

It can be seen that the feedforward control is accompanied by the phase delay shown in FIG. 6, and as a result, the turning is delayed and the locus is largely deviated. This means that if this feedforward control is simply incorporated into the control law of automatic driving, head sway is suppressed, but it becomes impossible to follow the target trajectory.

9 (A) to 9 (C) are graphs showing a phase lag waste time approximation. 9 (A) to 9 (C) show the results of approximating the phase delay of the feedforward controller with the "waste time L _f". Here, L _f = 0.8 [s]. The Bode plot of the dead time is shown by the solid line and compared with the Bode plot of the feedforward controller shown by the broken line. The dead time is set so as to match the low frequency components (0.1 Hz or less) of the phase diagram.

In FIG. 9C, the step response in which the time waveform of the lateral acceleration during feedforward control is delayed by the dead time is shown by a solid line and compared with the original time response shown by the broken line. Looking at the time response of the lateral acceleration shown by the solid line, it can be seen that the rise of the lateral acceleration is delayed until the timing when the lateral acceleration becomes ^{half of the steady value = 0.5 m / s 2.}

FIG. 10 is a graph showing a target locus shifted by a dead time. In FIG. 10, the simulation result of the vehicle locus with feedforward control is shown by a solid line. In addition, the vehicle locus (target locus) when the timing of stepwise start of the lateral acceleration is delayed by a wasteful time without feedforward control is shown by a broken line.

From FIG. 10, it can be seen that even with feedforward control, the target locus can be followed by delaying the target locus by a waste time. That is, if it is possible to advance the time relatively by utilizing the information of the target locus in advance by the amount of wasted time for matching the low frequency components of the phase of the feedforward controller, it is possible to follow the target locus. You can see that.

<Traveling trajectory control>
The travel locus control of the present embodiment is a case where the travel locus control is performed with the feedforward control of the equation (38) for the "waste time τ" included in the technology for performing the travel locus control devised earlier by the inventors. It is used for relative time advance (control based on the previous target information).

(Traveling trajectory control technology according to the prior application)
First, the traveling locus control according to the prior application will be described.
The inventors have applied for a patent as a "vehicle steering control device" for a technique for controlling a traveling locus (see JP-A-2010-170187, JP-A-2010-167817, etc.).

FIG. 11 is a diagram for explaining steering control of the vehicle based on the forward viewing angle.
Vehicle steering control apparatus according to the prior application detects declination of the expected arrival position and the target arrival point after a certain time in the travel direction of the vehicle (after forward fixed point time T _ahead) a (forward gaze angle theta _gaze) "polarized The "angle detecting means", the "target value calculating means" that calculates the yaw angular velocity proportional to the declination as the target value after a certain time (after the dead time τ), and the "steering control" that realizes the target value of the yaw angular velocity. It has "means".

Then, the following relationship exists between the _{dead time τ and the forward gaze time T ahead in} the calculation of the target value of the yaw angular velocity. That is, by inserting appropriate dead time corresponding to the forward fixed point time T _ahead tau a (= T _ahead / 3) to the steering control, can be controlled by anticipating the target value after the dead time tau, the target locus Can be followed.

FIG. 12 is a functional block diagram showing an example of the configuration of the vehicle steering control device according to the prior application. As shown in FIG. 12, the vehicle steering control device 10 according to the prior application includes an image pickup device 12, a vehicle speed sensor 14, a steering angle sensor 16, a steering device 20, and a computer 22. The computer 22 controls the steering device 20 based on the outputs of the vehicle speed sensor 14 and the steering angle sensor 16 and the forward image captured by the image pickup device 12.

The image pickup device 12 has an image pickup unit (not shown) composed of a monocular camera that captures an image of the front of the own vehicle and generates an image signal of the image, and an A / D conversion of the image signal generated by the image pickup unit. It includes a / D conversion unit (not shown) and an image memory (not shown) for temporarily storing an A / D converted image signal. The vehicle speed sensor 14 detects the vehicle speed of the own vehicle. The steering angle sensor 16 detects the actual steering angle of the front wheels of the own vehicle.

The steering device 20 is a steer-by-wire system in which the steering wheel and the front steering wheel are mechanically separated, includes an electric power steering device, and the assist torque of the electric power steering device is set by control by the computer 22.

The computer 22 includes a CPU, RAM, and ROM. The CPU reads the program stored in the ROM and executes the program using the RAM as a work area. The computer 22 is functionally composed of an image input means 24, a target arrival point declination detection means 26, a target yaw angular velocity calculation means 28, and a steering control means 30.

The image input means 24 acquires a front image output from the image pickup apparatus 12. The target arrival point declination detecting means 26 determines the predicted arrival position after the forward gaze time in the traveling direction of the own vehicle from the front image based on the vehicle speed from the vehicle speed sensor 14 and the actual steering angle of the front wheels from the steering angle sensor 16. Along with the detection, the target arrival point after the forward gaze time on the target course on which the own vehicle travels is detected from the front image, and the declination between the traveling direction of the own vehicle and the direction of the target arrival point after the forward gaze time. Is detected.

The target yaw angular velocity calculating means 28 calculates the target value of the yaw angular velocity after the dead time, which will be described later, based on the declination between the traveling direction of the own vehicle and the direction of the target reaching point after the forward gaze time. The steering control means 30 controls the steering device 20 so as to realize the calculated target value of the yaw angular velocity after a dead time based on the actual steering angle of the front wheels from the steering angle sensor 16.

The functions of each part will be described in detail below.

The target arrival point declination detecting means 26 detects _{the forward gaze angle θ gaze} as the declination between the direction of the vehicle speed vector and the target arrival point of the vehicle after the _{forward gaze time T ahead} . In order to detect the front viewing angle, the vehicle body slip angle β is obtained based on the _{front wheel actual steering angle δ f and the vehicle speed v.} The following relationship holds between the front wheel actual steering angle δ _f and the vehicle body slip angle β.

Here, l is the wheelbase, l _f is the distance from the front shaft to the center of gravity, l _r is the distance from the rear shaft to the center of gravity, m is the vehicle mass, I _z is the yaw moment of inertia, c _f is the front wheel cornering power, and c. _r is the rear wheel cornering power. Therefore, the vehicle body slip angle can be calculated from the vehicle speed and the front wheel actual steering angle signal based on the above equations (41)-(45).

Further, the target arrival point declination detecting means 26 calculates the traveling direction of the own vehicle on the front image based on the vehicle front-rear direction obtained in advance on the front image and the calculated vehicle body slip angle, and the own vehicle A point in front of the forward gaze distance obtained by the product of the preset forward gaze time and the vehicle speed in the traveling direction of is detected as an estimated arrival position after the forward gaze time.

Further, the target arrival point declination detecting means 26 recognizes the target course from the front image by image recognition processing, and is on the target course and is obtained by the product of the preset forward gaze time and the vehicle speed. A point in front of the gaze distance is detected as a target arrival point after the forward gaze time.

Further, the target arrival point declination detecting means 26 determines the direction of travel of the own vehicle and the direction of the target arrival point after the forward gaze time based on the detected arrival predicted position after the forward gaze time and the position of the target arrival point. Detects the declination of.

Regarding the target arrival point, the target course of the own vehicle may be acquired from a navigation system or the like, and the position of the target arrival point may be detected. Further, the case where the vehicle body slip angle is calculated using the outputs of the vehicle speed sensor and the steering angle sensor has been described as an example, but the present invention is not limited to this, and the vehicle body slip angle is calculated by using another conventionally known method. You may ask.

Next, in the target yaw angular velocity calculation means 28, the target value r _{ref of the yaw angular velocity is calculated from the forward gaze angle θ gaze,} which is the declination between the expected arrival position and the target arrival point, based on the following equation.

Next, in the steering control means 30, the _{steering angle required to obtain the target yaw angular velocity rref according} to the above equation (46) is calculated and steering is controlled. The following relational expression holds between the front wheel actual steering angle δ _f and the yaw angular velocity r.

Here, equations (47)-(49) are approximated as follows.

However, G is constant gain (6), L _v is the dead time when approximated by a dead time delay due to the dynamic characteristics of the expression (47). Therefore, the front wheel actual steering angle δ _f for realizing the yaw angular velocity command value r _ref can be derived as follows.

The steering control means 30 calculates the actual steering angle of the front wheels after the dead time τ according to the above equation (52) based on the calculated target value of the yaw angular velocity after the dead time τ, and stores it in a memory (not shown). I will do it. In this as dead time τ, taking into account the delay of the vehicle movement, the (50) is a value obtained by only subtracting the dead time L _v of equation used.

When the dead time τ elapses, the steering control means 30 reads the corresponding front wheel actual steering angle from the memory, detects the front wheel actual steering angle from the steering angle sensor 16, and reads the front wheel actual steering angle and the detected front wheel. The steering assist torque required to obtain the read front wheel actual steering angle is calculated based on the actual steering angle, and the calculated steering assist torque is added to the assist torque in the electric power steering device of the steering device 20. Set. In this way, the steering device 20 realizes the target value of the yaw angular velocity.

(Traveling locus control according to this embodiment)
Next, the traveling locus control according to the present embodiment will be described.
Travel trajectory control according to this embodiment, the "dead time τ (= T _{ahead / 3)"} included in the traveling locus control according to the above prior application, the relative required feedforward control equation (38) It is a technology that integrates feedforward control for the purpose of suppressing head sway based on the human body behavior model and control based on the forward gaze angle that accurately follows the target trajectory, focusing on the fact that it can be used for a rapid time advance.

FIG. 13 is a functional block diagram showing an example of the configuration of the vehicle steering control device according to the present embodiment. The vehicle steering control device 10A is different from the computer 22 of the vehicle steering control device 10 shown in FIG. 12 in that the computer 22A includes a forward gaze time calculating means 23 and a feedforward control means 29.

The forward gaze time calculation means 23 is inserted in front of the image input means 24. The feedforward control means 29 is inserted between the target yaw angular velocity calculation means 28 and the steering control means 30. The same components as those of the vehicle steering control device 10 shown in FIG. 12 are designated by the same reference numerals to simplify the description.

The forward gaze time calculation means 23 is the front gaze time T from the sum of _{the waste time L f} corresponding to the phase delay of the feed forward controller _{and the waste time L v} corresponding to the dynamic characteristics from the front wheel actual steering angle to the yaw angular velocity. _{Calculate the phase} based on the following equation. Incidentally, without taking into consideration the dead time _{L v,} three times the value of the dead time _{L f} may forward fixed point time _{T ahead.}

The image input means 24 acquires a front image output from the image pickup apparatus 12.

Next, target arrival points deviation angle detecting means 26 detects the forward gaze angle theta _gaze as the argument of the target arrival points of the vehicle rear direction and forward fixed point time T _ahead of the velocity vector of the vehicle.

Next, the target yaw angular velocity calculation means 28 sets the target value r of the yaw angular velocity after the dead time τ (= L _f + L _{v ) from the forward gaze angle θ gaze} , which is the deviation angle between the predicted arrival position and the target arrival point. _{Calculate ref} based on the following equation.

Next, the feedforward control means 29 converts the _{yaw angular velocity target value r ref into a yaw angular velocity target value r ref FF} modified based on the following equation.

The above equation (38) is an equation expressing the transfer function of feedforward control for obtaining the target value of the lateral acceleration for suppressing the head sway of the occupant using the human body behavior model. As will be described later, the yaw angular velocity can be approximated to be proportional to the lateral acceleration when the vehicle speed is constant. Therefore, using the transfer function expressed by the above equation (38), the yaw angular velocity target value r _ref when feedforward control is not performed is corrected to the yaw angular velocity target value r _{refFF when feedforward control is performed.} be able to.

Next, in the steering control means 30, the _{steering angle required to obtain the modified target yaw angular velocity r refFF} of the above equation (55) is calculated based on the following equation and steering control is performed.

FIG. 14 is a graph showing the result of traveling locus control of the vehicle. FIG. 14 shows the result of superimposing the target locus and the traveling locus of the vehicle controlled by the present invention. Here, as in the simulation of FIG. 8, the vehicle speed is set to 5 [m / s], the dynamic characteristics of the vehicle motion are ignored, the vehicle body slip angle is always 0, that is, the yaw angular velocity is expressed by lateral acceleration ÷ vehicle speed. The simulation is carried out under the assumption that It can be seen that the vehicle after the travel locus control according to the present embodiment can accurately follow the target locus indicated by the broken line.

FIG. 15 is a graph showing the time response of vehicle lateral acceleration and head displacement. The simulation results of lateral acceleration and head sway when the vehicle after controlling the traveling locus according to the present embodiment follows the target locus of FIG. 14 are shown by solid lines. The broken line shows the lateral acceleration and the head sway at that time when the lateral acceleration is stepwise increased at time 3 [s] by completely following the target trajectory without performing feedforward control.

Control vehicle according to the present embodiment, starts the steering at the target locus becomes circular turn (time 3 [s]) from the forward fixed point time _T ahead = 2.4 [s] before time 0.6 [s] However, the lateral acceleration is gradually increasing. That is, even if the target value of the yaw angular velocity calculated from the forward gaze angle changes stepwise, the target value of the yaw angular velocity corrected by the Ford forward control changes gradually, and the reverse response does not occur. ..

As a result, the reverse response of the head sway (the response of the broken line near 3 to 4 [s]) is suppressed, and the amplitude is reduced by about 40% as compared with the case where the lateral acceleration is changed stepwise. You can see that. Such suppression of head sway maintains the comfort of the occupants and also has the effect of reducing the load when working in the passenger compartment of the autonomous driving vehicle.

The present invention is not limited to the above-described embodiment, and various modifications and applications are possible without departing from the gist of the present invention.

For example, in deriving the control law, a human body behavior model divided into three is assumed, but when utilizing it in vehicle design, the two joint torques derived here are converted into a plurality of joint torques. It can also be applied to an articulated musculoskeletal human body behavior model by conversion to equivalent muscle strength. In this case, the joint angle between the lower body and the pelvis used for feedback is replaced by the sum of a plurality of joint angles. In addition, when converting the derived joint torque into multiple muscle strengths, for example, based on optimal control, the distribution that minimizes the sum of squares of muscle activity corresponding to each muscle and the maximum value of muscle activity The allocation that minimizes is conceivable.

Further, the case where the reaction force generated by the contact between the human body and the vehicle and the influence from the arm gripping the steering wheel are ignored has been described as an example, but the present invention is not limited to this. The calculation may be made in consideration of the reaction force generated by the contact between the human body and the vehicle and the influence from the arm gripping the steering wheel. In this case, the reaction force generated by the contact between the human body and the vehicle and the joint torque according to the influence from the arm gripping the steering wheel may be added and calculated.

Further, although an example of performing feedforward control with respect to the input of lateral acceleration has been described, the present invention is not limited to this. Feedforward control may be performed for at least one input of front-back acceleration and lateral acceleration.

That is, at least one target value of the front-rear acceleration and the lateral acceleration when the feedforward control is performed is obtained, and the vehicle motion is controlled so as to realize the obtained target value (controlling the steering device, etc.).

When performing feed-forward control for the input of the anteroposterior acceleration and the lateral acceleration, the joint angle of the joint 2 in the anteroposterior direction and the joint angle of the joint 1 in the lateral direction are the same as in the state equation shown in the above equation (30). And the state equation for calculating the behavior of the occupant including the joint angle of the joint 2 in the lateral direction is obtained.

This state equation and the output equation for outputting the head displacement y from the joint angle of the joint 2 in the anterior-posterior direction, the joint angle of the joint 1 in the lateral direction, and the joint angle of the joint 2 in the lateral direction are expressed as a transfer function. Therefore, the transfer function from the front-rear acceleration and the lateral acceleration of the vehicle to the head displacement is obtained in the same manner as the transfer function shown in the above equation (32).

Similar to the equation of state shown in Eq. (36) above, the target transfer function (target characteristic) is obtained. Then, similarly to the transfer function shown in the above equation (38), the transfer function of feedforward control for converting the transfer function from the vehicle's longitudinal acceleration and lateral acceleration to the head displacement into the target characteristic is obtained.

Further, the program of the present invention may be stored in a storage medium and provided.

10 Vehicle steering control device 10A Vehicle steering control device 12 Imaging device 14 Vehicle speed sensor 16 Steering angle sensor 20 Steering device 22 Computer 22A Computer 23 Forward gaze time calculation means 24 Image input means 26 Target arrival point deviation angle detection means 28 Target yaw angle speed calculation Means 29 Feed forward control means 30 Steering control means

Claims (5)

Declination detecting means for detecting the declination between the traveling direction of the vehicle and the direction of the target reaching point after a predetermined forward gaze time on the target course on which the vehicle travels.
A target value calculation means for calculating the detected yaw angular velocity proportional to the declination as the first target value of the yaw angular velocity after a dead time of one-third of the forward gaze time.
A target value correcting means for obtaining a second target value of the yaw angular velocity when feedforward control is performed to correct the first target value of the yaw angular velocity and suppress the head sway of the occupant.
A vehicle motion control means for controlling the vehicle motion so that the second target value of the yaw angular velocity is realized after the dead time.
Including
The waste time is the sum of the first waste time corresponding to the phase delay of the feed forward control or the second waste time corresponding to the phase delay of the transfer function from the actual steering angle to the yaw angle velocity. To
Vehicle motion control device. The transfer function of the feedforward control
The sign of the unstable zero point existing in the numerator of the first transfer function, which obtains the displacement of the occupant's head from the lateral acceleration acting on the vehicle defined using the human body behavior model, is inverted, and the numerator and denominator are exchanged. Let it be the second transfer function including the model,
The vehicle motion control device according to claim 1. When the head displacement of the occupant obtained by the first transfer function includes an inverse response, a third transfer function having a characteristic of canceling the inverse response is set as a target characteristic.
Obtain the second transfer function from the first transfer function and the third transfer function.
The vehicle motion control device according to claim 2. The human body behavior model is
A human body behavior model that represents the behavior of an occupant seated on a vehicle seat, wherein the occupant's body is divided into an upper part including the head and chest, a middle part including the pelvis, and a lower part including the lower body below the pelvis. A human body behavior model having a joint between the middle part and the lower part and a joint between the upper part and the middle part.
To calculate the joint torque between the middle part and the lower part to prevent the posture from collapsing from the joint angle between the middle part and the lower part, and to invert the head against the acceleration applied to the vehicle. By calculating the joint torque between the upper part and the middle part according to the acceleration of the head, the joint angle between the middle part and the lower part and the joint angle between the upper part and the middle part are calculated. Calculate the occupant's head displacement,
The vehicle motion control device according to claim 2 or 3. Computer,
A program for functioning as each means of the vehicle motion control device according to any one of claims 1 to 4.

Images