JP2003098047A - Automobile automatic operation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automobile automatic operation system capable of shortening a time required for an exhaust gas test. SOLUTION: In this automobile automatic operation system for performing automatic operation of a test vehicle 1 by using a driving performance map, before the main test driving for performing the exhaust gas test, preconditioning driving of the test vehicle is performed based on a base map, and a corrected map is drawn by correcting the base map by using vehicle speed feedback information acquired by the preconditioning driving. In addition, a modified map is determined from the corrected map and the base map, and the driving performance map suitable for the test vehicle 1 is acquired.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic vehicle driving system.

[0002]

2. Description of the Related Art Conventionally, a chassis dynamometer is used to perform a simulated driving operation of an actual vehicle (hereinafter referred to as a test operation) for a dynamic driving performance test of an automobile.
In recent years, an automatic vehicle driving system using an automatic driving device (also called a driving robot) has been adopted for this test driving.

By the way, in the above-mentioned test driving, it is necessary to drive a vehicle to be tested (hereinafter referred to as a test vehicle) so as to follow a driving pattern using a driving performance map, but a test of an unregistered test vehicle is performed. When driving
It is necessary to create a driving performance map corresponding to the test vehicle. Here, the driving performance map is data capable of obtaining the accelerator operation amount from the speed and the acceleration,
Learning operation of the test vehicle is performed, speeds and accelerations at various different accelerator opening degrees are obtained, and data associating the accelerator opening degrees, speeds and accelerations is obtained. After that, based on the data obtained by the learning driving, the data is developed to the data (driving performance map) from which the accelerator operation amount can be obtained from the speed and the acceleration.

Conventionally, when a newly registered test vehicle is operated in a mode for exhaust gas, as shown in FIG. 6, the actuator section (mechanical unit) of the automatic driving device is installed in the driver's seat of the test vehicle. ) Is attached, position learning of the actuators that drive the accelerator, brake, clutch, and shift lever, learning operation, and preconditioning traveling are performed.

[0005]

The time required for mounting the actuator and learning the position is 2 to 4, respectively.
Although it is a short time such as about a minute, in the learning driving, it is necessary to change the accelerator pedal depression amount of the test vehicle variously and obtain the speed and acceleration at that time.
(5-speed manual transmission)
20-25 minutes by car, AT (Automatic Tra)
It took about 15 minutes by car.

As described above, with respect to the newly registered test vehicle, it is necessary to follow the above procedure before starting the exhaust gas test, and especially since a lot of time is spent for learning driving, the entire exhaust gas test is performed. It took that long to do.

The present invention has been made in view of the above matters, and an object thereof is to provide an automatic vehicle driving system capable of shortening the time required for an exhaust gas test.

[0008]

In order to achieve the above object, according to the present invention, in a vehicle automatic driving system in which a test vehicle is automatically driven using a driving performance map, an actual exhaust gas test is carried out for an exhaust gas test. Before, the test vehicle is pre-conditioning run based on the base map, using the vehicle speed feedback information obtained by this pre-conditioning run, the base map is corrected to create a correction map, and further, this correction map and A correction map is obtained from the base map and a travel performance map suitable for the test vehicle is obtained (claim 1).

In the automatic vehicle driving system according to the first aspect of the present invention, when the correction map is created, the correction acceleration obtained based on the deviation between the control target vehicle speed and the actual vehicle speed in the vehicle speed feedback system, and the accelerator operation amount , The actual vehicle speed is used as the vehicle speed feedback information (claim 2).

According to the automatic vehicle driving system having the above-mentioned structure, the learning driving can be performed during the preconditioning traveling, and the learning driving which is conventionally required is not required to be performed independently, and the time required before the main test can be reduced. The time required for the exhaust gas test can be shortened accordingly. By simplifying the procedure before the main test, the number of operations of the actuator unit is reduced, and at the same time the operability is improved.

Even in other exhaust gas test mode driving, when a human driver drives a new test vehicle, practice driving may be performed before the exhaust gas test.
When using the actuator section, preconditioning traveling is performed instead of the practice traveling. As a result, the actuator unit can be effectively used even in the exhaust gas test without significantly changing the exhaust gas test procedure conventionally performed by the human driver.

[0012]

DETAILED DESCRIPTION OF THE INVENTION The details of the present invention will be described below with reference to the drawings. 1 to 4 show one embodiment of the present invention. First, FIG. 1 schematically shows a configuration of a control system in an automatic vehicle driving system of the present invention. In FIG. 1, reference numeral 1 denotes a test vehicle, which is mounted on a chassis dynamometer (not shown). .
Reference numeral 2 is a control target value generator that outputs a control target vehicle speed V _nom to travel based on the reference vehicle speed V _tab . _{Reference numeral} 3 is a matching point at which the control target vehicle speed V _nom and the actual vehicle speed V _act of the test vehicle 1 are input and a difference (vehicle speed deviation) V _err between them is output. Numerals 4 and 5 are a proportional term (P; gain) and an integral term (T; time constant) for P control and I control of the vehicle speed deviation V _err , respectively. The proportional term correction acceleration α _p and the integral term correction acceleration α
Output _i respectively. _{Reference numeral} 6 is a differentiator that differentiates the control target vehicle speed V _nom, and a feedforward ratior 7 that outputs the feedforward acceleration α _ff is provided at the _subsequent stage.

Then, 8 is the proportional term correction acceleration α _p ,
Integral term corrected acceleration α _i (these are control outputs obtained by feedback control of speed deviation V _err ) and feedforward acceleration α _ff (this is control target vehicle speed V
It is the rate of change of _nom and is also referred to as the target acceleration) and outputs the control acceleration α _ctl at the matching point. Reference numeral 9 denotes an accelerator opening degree predicted value generator, which is stored in the vehicle data of the automatic vehicle driving system, and the actual vehicle speed V _act and the control acceleration α _ctl from the point 8 are input. Reference numeral 10 is a driving performance map 10, showing the relationship between speed and acceleration with the horizontal axis representing the speed and the vertical axis representing the acceleration, with the accelerator opening θ as a parameter. By inputting the acceleration and the acceleration, the accelerator opening to be controlled can be obtained. In this embodiment, a base map is stored. Here, the base map refers to a traveling performance map of a vehicle that has already been learned, and is a traveling performance map that is corrected during preconditioning traveling.

Reference numeral 11 is a correction map. here,
The correction map is a running performance map obtained for correcting the base map from the data obtained by measuring during the preconditioning running. This correction map 11 includes a corrected acceleration (integral term corrected acceleration α _{i in the} illustrated example) obtained based on the deviation V _err between the control target vehicle speed V _nom and the actual vehicle speed V _act , and the accelerator opening degree predicted value generator 9 Accelerator operation amount A _act and actual vehicle speed V
_{It is calculated} from _act .

The control arithmetic expression for _obtaining the control acceleration α _ctl when the current time of the control system shown in FIG. 1 is k is expressed by the following equation (1).

[0016]

[Equation 1]

Further, the accelerator operation amount A _act is obtained by inputting the control acceleration α _ctl and the actual vehicle speed V _act in the running performance map. A _act (k) = M _ap (α _ctl , V _act ) ... (2) M _ap : The operation amount obtained from the driving performance map, that is, the control acceleration α _ctl and the actual vehicle speed V _act , as described above (1), ( By calculating using the formula 2), the accelerator operation amount A _act
Can be asked.

Next, an example of a procedure for mode-driving the newly registered test vehicle 1 by the automatic vehicle driving system having the control system having the above configuration will be described.

(1) The actuator section is attached to the driver's seat of the test vehicle.

(2) The position of the actuator for driving the accelerator, the brake, the clutch and the shift lever is learned.

(3) Preconditioning run is performed. At this time, the preconditioning run is performed according to the following procedure. (1 of 3) The vehicle data that has already been learned is used. This vehicle data already has a driving performance map obtained by performing learning driving with another vehicle, and this is used as a base map. This base map may be provided in advance as a default value.

(3-2) Preconditioning traveling is performed based on the selected base map.

(3-3) During preconditioning traveling, the correction acceleration (in this case, the integral term correction acceleration α in the vehicle speed feedback information when the preconditioning traveling is performed using the traveling performance map learned by the other vehicle) is added.
_i ), the accelerator operation amount A _{ac t,} and the actual vehicle speed V _act are fetched.

(3-4) Preconditioning During the traveling, the corrected acceleration α _i , the accelerator operation amount A _act and the actual vehicle speed V _act are constantly monitored and used when the accelerator operation during traveling obtains the base map. The corrected acceleration α _i and the actual vehicle speed V _act in the vicinity of the accelerator depression amount at several places are measured. (4) After the completion of the preconditioning travel, a correction map that shows the relationship between the speed, the acceleration, and the accelerator opening by using the corrected acceleration α _i , the accelerator operation amount A _act, and the actual vehicle speed V _act obtained during the preconditioning travel. 11
To create. To create the correction map 11, a first-order approximation formula is obtained from the measured actual vehicle speed V _act and the corrected acceleration α _i data by the least square method.
How to obtain this correction map will be described later in detail with reference to FIG.

(5) The base map used for the preconditioning run and the same learning opening degrees of the correction map obtained based on various data collected by the preconditioning run are newly approximated using an approximation formula. A formula is calculated and used as a correction map (see FIG. 4).

(6) Test vehicle 1 using the modified map
This is the mode operation, and the main test run is performed.

As described above, according to the automatic vehicle driving system of the present invention, the learning operation can be performed during the preconditioning traveling, and it is not necessary to independently perform the learning operation which is conventionally required. That is, the predetermined mode operation can be performed by the procedure shown in FIG. Compared with FIG. 6 showing the flow of the conventional procedure, the flow of the procedure shown in this figure is reduced by the amount corresponding to the learning operation. That is, the time required before the main test can be shortened, and the time required for the exhaust gas test can be shortened accordingly. Further, by simplifying the procedure before the main test, the number of operations of the actuator section is reduced, and at the same time, the operability is improved.

Next, how to obtain the correction map in the automatic vehicle driving system of the present invention will be specifically described with reference to FIGS. 3 and 4.

(11) For the control in the preconditioning traveling, the control calculation formula according to the formula (1) is used. Therefore, the integral term correction acceleration α _i is measured at a constant cycle and those data are used.

(12) Since the map structure is the same as the driving performance map, the accelerator operation amount A used during the learning operation is set.
_The proportional term correction acceleration α _p , the integral term correction acceleration α _i, and the actual vehicle speed V _act in the accelerator operation amount near _act (within the opening error Δθ) are measured.

Θ ₀ −Δθ ₀ ≦ θ ₀ ^“ ≦ θ ₀ + Δθ ₀ (3) (θ ₀ ; learning opening, Δθ ₀ ; opening error at θ ₀ ) θ ₅ −Δθ ₅ ≦ θ ₅ ^” ≦ θ ₅ + Δθ ₅ (4) (θ ₅ ; learning opening, Δθ ₅ ; opening error at θ ₅ ) Note that the current opening error values Δθ _{0 to} θ ₅ are ± 0.5%. .

(13) θ_jComplement of integral term at (i = 0 to 5)
Positive acceleration α_iAnd actual vehicle speed V_actCalculate the correction map from
It FIG. 3 shows the actual vehicle speed V._actIs the horizontal axis and the acceleration of the correction of the integral term
Degree α _iIs a plot of a large number of points on the vertical axis,
Using these plots, the high-speed correction approximation formula (in the figure
(Indicated by reference numeral 12) α_FB(V) = A_FB+ B_FB× V Ask for. In this case, the data when the speed is near zero (in the figure,
The part indicated by reference numeral 13) is affected by the flapping at the time of starting.
Reliability to use as an indicator of map deviation
Is low, and data on the high speed side (indicated by reference numeral 14 in the figure)
(Part) has a small number of data, the speed range is V_lkm
/ H ~ V_hThe data of km / h is used as data.

(14) The method for obtaining the correction map approximation formula is as follows. Let the measured data be θ _i (V _act , α _act ). i = 0, ..., M-1 (m is the number of measured data).

A tentative first-order approximation formula is obtained by the method of least squares using the data in the range of V _l km / _{h to} V _h km / h. α _FB '(V) = A _FB ' + B _FB '× V ...... (4)

The measured data and the data of the approximate expression represented by the expression (4) are compared with each other at accelerations of the same speed to obtain an error from the approximate expression. α _err (i) = | α _i '(V _act ) −α _act (i) | (5)

The m pieces of data are deleted in descending order of the error calculated by the equation (5). The reason for adding this sequence is to ignore the influence of measurement noise or the like. However, when the number of measured data is p or more and q or less, the data to be ignored is r in descending order, and when it is p-1 or less, all the data are used.

A first-order approximation formula is obtained by the least-squares method using the data obtained again as described above. α _FB (V) = A _FB + B _FB × V (6)

Then, the map is corrected using the equation (6). In this case, cases are classified as shown in FIGS. 4A and 4B depending on the magnitude of the switching speed V _j between the low speed range and the high speed range. In this figure, reference numeral 1
Reference numerals 5 and 16 are temporary speed correction approximation formulas in the low speed region and the high speed region.

1) When V _{j ≤V l} α _ctl (V _l ) = α _ctl ′ (V _l ) −α _FB (V _l ) α _ctl (V _h ) = α _ctl ′ (V _h ) −α _FB (V _h ) From the acceleration and velocity obtained by the above equation 2, the following (7 l),
As shown in (7h), a speed correction approximate expression is obtained. α _ctl = A _l + B _l × V …… (7l) α _ctl = A _h + B _h × V …… (7 _h )

2) When V _j > V _l α _ctl (V _j ) = α _ctl ′ (V _j ) −α _FB (V _j ) α _ctl (V _j ) = α _ctl ′ (V _h ) −α _FB (V _h ) From the acceleration and velocity obtained by the above equation 2, the following (7 l),
As shown in (7h), a speed correction approximate expression is obtained. α _ctl = A _l + B _l × V …… (7l) α _ctl = A _h + B _h × V …… (7 _h )

The velocity correction approximation formula obtained as described above is represented by reference numeral 1 in FIGS. 4 (A) and 4 (B).
It is represented as 7,18.

In the above embodiment, when the correction map is created, the integral term correction acceleration α _i is used as feedback information together with the accelerator operation amount and the actual vehicle speed, but the proportional term correction acceleration α _p is also used. In this case, as shown in FIG. 5, these corrected accelerations α _i , α
You may make it multiply _p with a weighting coefficient. That is, in FIG. 5, w _I and w _P are circuits that multiply each with a weighting coefficient.

[0043]

As described above, according to the automatic vehicle driving system of the present invention, the test vehicle is preconditioned based on the base map before the actual test run for the exhaust gas test, and the preconditioned run is performed. Using the vehicle speed feedback information obtained by
A correction map is created by correcting the base map, and a correction map is obtained from the correction map and the base map to obtain a running performance map suitable for the test vehicle. The learning operation can be performed, the learning operation conventionally required is not necessary to be performed independently, the time required before the main test can be shortened, and the time required for the exhaust gas test can be shortened accordingly. By simplifying the procedure before the main test, the number of operations of the actuator unit is reduced, and at the same time the operability is improved.

In addition, even in the case of driving in another exhaust gas test mode, when a human driver drives a new test vehicle, practice driving is often performed before the exhaust gas test.
When using the actuator section, preconditioning traveling is performed instead of the practice traveling. As a result, the actuator unit can be effectively used even in the exhaust gas test without significantly changing the exhaust gas test procedure conventionally performed by the human driver.

[Brief description of drawings]

FIG. 1 is a diagram schematically showing an example of a configuration of a control system in an automatic vehicle driving system of the present invention.

FIG. 2 is a diagram showing a test procedure in the automatic vehicle driving system.

FIG. 3 is a diagram for explaining how to obtain a correction map.

FIG. 4 is a diagram showing an example of a correction map.

FIG. 5 is a diagram schematically showing another example of the configuration of the control system in the automatic vehicle driving system of the present invention.

FIG. 6 is a diagram showing a test procedure in a conventional automatic vehicle driving system.

[Explanation of symbols]

1 ... test vehicle, 10 ... driving performance map, V _nom ... control target vehicle speed, V _act ... the actual vehicle speed, V _err ... deviation, alpha _p ... correction acceleration, A _act ... accelerator operation amount.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Nobuhisa Mori             1 Toyota Town, Toyota City, Aichi Prefecture Toyota Auto             Car Co., Ltd. F-term (reference) 3D044 AA01 AB01 AC16 AC26 AC28                       AD12 AD14 AD17 AD21 AE01                       AE04 AE14 AE16 AE19 AE22                 3G093 AA04 BA23 CB10 DA06 DB05                       EA00 EB03 EB04 FA02 FA07                       FA09 FA10 FA11 FA12 FB01                       FB02

Claims (2)

[Claims] 1. An automatic vehicle driving system for automatically driving a test vehicle using a driving performance map,
Before the main test run for conducting an exhaust gas test, the test vehicle is preconditioned based on the base map, and the vehicle speed feedback information obtained by this preconditioning is used to correct the base map to obtain a correction map. An automatic driving system for a vehicle, characterized in that a correction map is created from the correction map and the base map to obtain a driving performance map suitable for the test vehicle. 2. When the correction map is created, the vehicle speed feedback system uses the corrected acceleration obtained based on the deviation between the control target speed and the actual vehicle speed, the accelerator operation amount, and the actual vehicle speed as the vehicle speed feedback information. The automatic vehicle driving system according to claim 1.