JPH02231609A - Automatic vehicle operation controller - Google Patents

Abstract

PURPOSE:To attain the satisfactory automatic operations among different vehicles as well as an automatic vehicle operation approximate to a manual operation by correcting a vehicle driving error after estimating this error at a position ahead the vehicle by a prescribed distance. CONSTITUTION:In a teaching operation state, a teaching data storage means 14 stores the position and the driving state of a vehicle obtained by a position/ state detection means 10 and the information on the manipulated variable obtained via an actuator manipulated variable detection means 12. An estimated error arithmetic means 16 inputs the information received from the means 14 and also the information on the vehicle under operation given from the means 10 and operates an estimated error at a position ahead the vehicle by a prescribed distance. An actuator manipulated variable arithmetic means 18 inputs the manipulated variables of various actuators received from the means 14 and also the estimated error given from the means 16 and corrects the manipulated variable with the feedback applied to the teaching data to output the corrected manipulated variable. Thus an optimum course can be selected for each of different vehicles and an automatic vehicle operation similar to a manual operation can be satisfactorily applied.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an automatic pilot control system for a vehicle, and particularly to control contents of an automatic pilot control system for automatically driving a vehicle along a predetermined taxiway. be.

[Prior Art] Automated vehicles that automatically travel along a predetermined course are well known, and are used in unmanned guided vehicles in factories, endurance tests, and the like. This self-piloted vehicle, for example, installs a guidance cable on a predetermined course, detects the position of this guidance cable with a magnetic field sensor, and operates the steering wheel, accelerator, brakes, etc. using actuators, and automatically travels along the guidance cable. This makes it possible to drive the vehicle faithfully to the course.

Furthermore, in the past, as shown in Japanese Patent Laid-Open No. 63-3315, information necessary for setting a running course has been provided in advance by running the actual course with manned control and collecting course information. be. According to this,
To enable smooth running without deviating from a predetermined course even on relatively complex terrain.

[Problems to be Solved by the Invention] However, in conventional vehicle autopilots, the vehicle is operated based on the current course error, so it is not possible to perform smooth driving using predictions as would be the case when a human actually steers the vehicle. First, when transporting not only goods but also people, there is a problem of poor ride comfort.

In general, the most important thing for an autopilot vehicle is to follow the course faithfully, and because it uses high-speed control with a relatively high feedback gain, it is very different from the one operated by a human as described above, and the stability of the control is is getting worse. Note that even in the case of the above-mentioned Japanese Patent Application Laid-Open No. 63-3315, only the course is taught and the amount of maneuver is not taught.

Furthermore, when the types of vehicles are different, the maneuverability of the vehicle changes, and the same control may not be able to handle the change. However, in the past, without distinguishing between vehicle types,
Since steering control is performed using constant feedback control, there is a problem that some vehicles may go off course.

The present invention was made with the aim of solving the above-mentioned conventional problems, and its purpose is to provide a vehicle that can perform good autopilot even when the vehicle is different, and that is capable of autopiloting that is similar to that operated by a human. An object of the present invention is to provide an automatic pilot control device for use in automobiles.

[1,000 Steps to Solve the Problems] In order to achieve the above object, the present invention measures the current position and running condition of the vehicle with respect to the taxiway, and based on this position and running condition, performs various automatic steering operations. In a vehicle autopilot system that operates and controls actuators, a teaching data storage means stores a target course taught to a driver by driving along a predetermined taxiway and the amount of operation of each actuator when driving on this target course. and a prediction error calculation means for calculating a prediction error when the vehicle travels a predetermined distance from the current position according to the teaching contents of the teaching data storage means, and correcting the teaching data storage contents according to the prediction error, The present invention is characterized in that it includes actuator operation amount calculation means for calculating the operation amount of the actuator, and accurately controls travel on a taught target course while making corrections due to prediction errors.

[Operation] According to the above configuration, first, the driver actually runs on the taxiway and is taught the taxiway as a target course, and at the same time, the operation amounts of various actuators when automatically driving this target course are detected. , the target course and the operation amount of the actuator are stored in a teaching data storage means.

Therefore, if a target course is taught, the vehicle is controlled to travel according to this target course, but at this time, the prediction error when traveling a predetermined distance from the current position is calculated sequentially, and the prediction error is calculated according to the prediction error. The stored content of the taught data is corrected, and the operation amount of the actuator is calculated.

This prediction error is a value that also takes into account errors in the current state, and thereby enables accurate trajectory correction or correction of the operating state.

Also, since the target course is taught by the driver,
This increases the degree of freedom in choosing a course for a taxiway, allowing the optimal driving course for various types of vehicles to be taken.

Furthermore, since the operation amount of the actuator is also calculated based on data taught by the driver, driving that matches human senses is ensured.

[Embodiments] Hereinafter, preferred embodiments of the present invention will be described based on the drawings.

FIG. 1 shows a schematic configuration of an autopilot control device for a vehicle according to a first embodiment, and FIG. 2 shows a specific configuration of an autopilot vehicle.

In FIG. 1, the position/state detecting means 10 includes various sensors for detecting the position and running state of the vehicle with respect to the taxiway, and the actiniator operation amount detecting means 12 includes the steering amount, accelerator amount, and brake amount for driving the vehicle. It includes a sensor that detects various actuator operation amounts such as the amount of actuator operation.

A characteristic feature of the present invention is that automatic steering is performed using a taught target course as a guide, and that this automatic steering also compensates for and corrects errors in vehicle travel over a predetermined distance. , teaching data storage means 14, prediction error calculation means 16, and actuator operation amount calculation means 1
8 is provided.

The teaching data storage means 14 stores the position and running state of the vehicle obtained by the position/state detection means 10 and the operation amount information obtained by the actuator operation amount detection means 12 during teaching driving, and calculates a prediction error. The means 16 inputs the information from the storage means 14 and the current vehicle information being driven from the position/state detection means 10, and calculates a prediction error for a predetermined distance ahead. Further, the actuator operation amount calculating means 18 inputs the operation amounts of various actuators from the teaching data storage means 14, inputs the prediction error from the prediction error calculation means 16, and corrects the operation amount by applying feedback to the teaching data. and output it.

To explain more specifically with reference to FIG. 2, a front magnetic field sensor 110a, a front lateral deviation detection circuit 111a, a rear magnetic field sensor 11Qb, and a rear deviation detection circuit 11lb are provided to detect the lateral deviation of the vehicle. The magnetic field output from the induction cable 20 provided in the guideway is detected to determine the lateral deviation and also the yaw angle. That is, as shown in FIG. 3, if the front lateral deviation is Yr1 and the rear lateral deviation Yr, the lateral deviation Y at the center position O of the vehicle 32 is Yr.
- (Yf' 1 Yr) /2 - (1)
And if the length of the vehicle is a, then the yaw angle θ of the vehicle is
is θ1(Yr −Yr)/a−(2
).

Also, in order to find the vehicle speed, the magnetic pickup 11
4 is provided, and the vehicle speed is detected by detecting the number of rotations of the magnet rotating plate 113.

An antenna 115 and a reference position detection circuit 116 are provided to detect the position of the vehicle on the taxiway, and the antenna 115 receives radio waves output from position beacons provided at each predetermined point on the taxiway. By supplying the signal to the reference position detection circuit 116, the reference position on the course can be detected.

In addition, an arithmetic processing circuit (ECU) 24 is used to input detection signals etc. from the various sensors mentioned above and perform various arithmetic processing.
A feedback processing circuit 26 is provided which performs predetermined processing by feeding back the operating amounts of various actuators that operate the vehicle. An error calculation means 16 and an actuator operation amount calculation means are configured.

Various actuators provided in the vehicle 28 for automatic steering include the illustrated drive motor 29.
In addition to the steering wheel 30, an accelerator, a brake, etc. are provided, and the amounts of these operations are detected by sensors provided in various actuators, such as the steering amount sensor 32, and fed back.

Further, in the first embodiment, a manual switch 34 for switching between a teaching operation mode and an automatic operation mode, and a display lamp 36 for identifying the state of this mode are provided.

Next, the teaching data storage means 14 and the prediction error calculation means 16
The operation of the actuator operation amount calculating means 18 will be specifically explained using steering amount control as an example.

FIG. 4 shows an arithmetic processing block circuit for determining the final actuator operation amount from the teaching data, and a target course map 141 and a steering amount map 142 are set in the teaching data storage means 14. This is what is being done.

The target course map 141 includes information on the position X detected from the position beacon 22 etc. during teaching driving and this position
Lateral deviation YO and yaw angle θ during teaching operation. are averaged by the averaging unit 241 and stored. Lateral deviation Y0 and yaw angle θ in this case. is a value measured with the guideway as a reference, and in the example, is a value measured from the guide cable 20. In the steering amount map 142, the amount by which the driver operated the steering wheel during the teaching driving is averaged by the averaging unit 242 and stored as the steering amount.

Further, as the prediction error calculation means 16, adder/subtractors 161,1
62, an error calculation section 163 and a gain multiplier 164 are provided, and an adder 181 is provided as the actuator operation amount calculation means 18.

For example, when the vehicle is running in the state shown in FIG. 5, the following calculations are performed.

In Figure 5, the target course 800 is automatically traveled.
Vehicle 28 is Δj from the target course! m (hl2 deviation) f/
The yaw angle of the vehicle at this time is Δθ. This yaw angle is the tangent line 801 (8
01-) is the direction of the vehicle. In this state, in normal feedback control, a steering amount is given to correct ΔJlm, but in the present invention, a prediction error amount Lm ahead is determined. Note that this distance L is 10 to 20
A distance of m is preferred.

That is, since the current yaw angle of the vehicle is Δθ, Lm
In the future, a new error of L・Δθ (Δθ is small (sinΔθ is assumed to be Δθ)) will occur, and the final prediction error ε0 for Lm ahead will be ε 0 − Δ ”e−L conflict Δ
θ...(3), which is performed by the error calculation unit 163.

In order to set the lateral deviation and yaw angle, which are the calculation information in this case, as values based on the guide cable (guideway), the current detected values are added to the output of the target course map 141 using the adder/subtractor 161.
, 162. That is, the above (3
) In the formula, Δl is Δl−YO−Y, and Δθ is Δθ−θ. It becomes ten. Then, the prediction error ε obtained by the error calculation unit 163. is multiplied by the gain Kp by the gain multiplier 164, and the final error correction amount ΔU is Δu−Kpε. ...(4), which is output to the adder 181 which is the actuator operation amount calculation means 18.

On the other hand, from the steering amount map 142, the steering amount programmed by teaching at the current position is supplied to the adder 181 as the actuator operation amount. That is, ε shown in FIG. 5 is the amount of deviation from the target course 800 when traveling with the current steering amount, and this ε The amount is added to the error correction amount ΔU by an adder 181.

Therefore, if the taught steering amount is all, the final steering amount u (t) is u (t)mu''+Δu - (5), and if the steering wheel 30 is operated by this steering jlu (t), the target The vehicle will be able to reach the course 800.

The first embodiment has the above configuration, and its operation will be explained below.

FIG. 6 shows the overall operation of the control contents, and when the power of the control system is turned on, the program starts. First, in step 61, a scheduled interrupt (IQms
It is determined whether or not there was a timing similar to ec), and if "YES", the process moves to step 62, where the sensor detecting the position and motion state of the vehicle and the sensor detecting the amount of actuator operation are detected. The sensor data is read (details are shown in FIG. 7).

In the next step 63, it is determined whether the teaching switch is ONL, and if "No", the process advances to step 66.
If "YES", the process moves to step 64. In this step 64, since the current mode is the teaching mode, the steering actuator control is turned OFF, and the teaching control in step 65 is performed (details are shown in FIG. 9), leaving the steering to the driver.

On the other hand, in step 66, it is determined whether the taught lamp is on or not. If "No", the autopilot is turned on even though no teaching has been performed yet. Proceed to step 67 and set the actuator $1g.
Turn a off. In this case, it is determined that the condition is abnormal, and the process proceeds to step 68, where an alarm is issued.

If the answer is “YES” in step 66,
The process moves to step 69 and the autopilot control described above is performed.

Note that when the process moves to step 69 in which this automatic pilot control is performed, the automatic pilot lamp of the display lamp 36 is turned on.

FIG. 7 shows the operation of reading the sensor data. First, in step 621, the front and rear lateral deviations yr,
yr is read, and in step 622, the lateral deviation Yl'
, Yr, calculate the yaw angle θ-(Yl'-Yr)/a,
In step 623, the lateral deviation Y- CYr +Yr)/
Calculate 2. Further, in step 624, the vehicle speed is read, in step 625, the vehicle position X is evaluated (details are shown in FIG. 8), and in step 626, the current steering Qu
Load.

FIG. 8 shows the operation of evaluating the vehicle position. In step 62a, the reference position signal obtained from the position beacon 22 is taken in, and in step 62b, it is determined whether the vehicle position is the true reference position or not. If 'YES', the process moves to step 62c to identify the reference position (point A,
Point B, point C, etc.) are performed. And step 62
In d, set the reference distance, for example, if it is the beacon position of point B, then XB^5. SW X. is set, and the process moves to step 62e, where the traveling distance ΔX between beacons is set to 0.
Reset to .

On the other hand, if the result in step 62b is "No", the process moves to step 62f, in which the detected vehicle speed VR is captured (V=VR), and the process moves to step 62g, at predetermined intervals τ. The traveling distance of The position X, X = X BASB + ΔX ・(
Calculate in 7).

In this way, the traveling distance HX is calculated from the vehicle speed at predetermined intervals from this reference position using the beacon position as a reference position.

FIG. 9 shows the operation of the teaching control. First, in step 651, it is determined whether or not the data storage position is reached. If "NO", the process moves to the end, and if "YES", the process moves to the end. The process moves to step 652, where it is determined whether n>a. In the embodiment, information for automatic piloting is stored for each predetermined position on the course, for example, for each position set at 1 m intervals, and in step 651, this position is searched. Furthermore, the stored data is an average value of several times, and in step 652, the number of times is counted.

For example, when taking the average value of three times, set it to a-3, and nm3
If so (“YES”), the process moves to step 653 and the sensor data v2- (X, θr Y*'
J) is read, and the process proceeds to step 654, where the data average value of the following equation is calculated.

a υ2'-(1/a)Σ hi21 (8) Next, the process moves to step 655, where writing is performed on the teaching map, and the target course map 141 and the steering amount map 14
Each piece of information is stored in 2.

On the other hand, in the case of "NO1" in step 652, the data up to the previous time is averaged as shown in the following equation, and step 657
Move to.

In step 657, it is determined whether or not the current data is appropriate as recruitment data using the following formula, I V2 - Vz
l < b ++ (1 0) lVz
If V21 is larger than b, the process moves to the end without storing it, and if it is smaller than b, the process moves to step 658.

In this step 658, data v2 is stored and set as nmn-'+t, and the process proceeds to step 659, where it is determined whether all courses have been stored. Then, when all the courses have been memorized, the taught lamp is turned on in step 660, and if the course has not been memorized yet, the taught lamp remains OFF (
Step 661) Move to end.

The present invention is characterized by automatically piloting a target course taught in this manner. In automatic piloting control, as shown in the above-mentioned FIG. is detected, and the equations (1) and (2
), the prediction error εo from equation (3) and the correction amount ΔU from equation (4) are calculated and output, and finally the steering amount u from equation (5) is calculated. (t) is calculated.

This steering amount u(t) is then given to the steering wheel 28, and this steering amount corrects the prediction error and achieves good vehicle steering.

Next, a second embodiment of vehicle speed control according to the present invention will be described in a tenth embodiment.
This will be explained based on the diagram.

In the case of vehicle speed control, a mode map 143 that stores the taught speed instead of the target course map, and an accelerator/brake map 144 that stores the taught accelerator and brake operation amounts instead of the steering amount map are provided. /The brake map 144 is provided with an averaging unit 38 that averages the amount taught by the driver. In addition, a primary prediction unit 40 that estimates the speed after a predetermined time (tp) from the acceleration and speed calculates the difference between the taught speed Vm outputted from the mode map 143 and the estimated speed V with an error ε - Vm - V
a subtracter 42 that calculates the amount of brake correction, a switch 44 that switches between calculating the brake correction amount and accelerator correction amount, a brake gain multiplier 48 that outputs the brake correction amount according to the error ε, and an accelerator correction amount that corresponds to the error ε. An accelerator gain multiplier 48 is provided as a preliminary error calculation means. Note that 50a and 50b are adders.

In the case of vehicle speed control, switching is performed using the accelerator and brake, and according to the above configuration, ε〉0 and u,〉0 →
Brake control ε〉O and u, lllIO → Accelerator control ε〉O and u. 〉0 → Accelerator control εku0
and u @ = 0 → brake control.

In the case of vehicle speed control, first, the speed Vm at a predetermined position 52x is entered in the mode map 143, the accelerator amount W a taught at the position X, and the brake Jlw are taught by the driver.
b is stored in the accelerator/brake map 144. On the other hand, the primary prediction unit 40 estimates the speed after a predetermined time tp from the longitudinal acceleration V- (t) and the speed V (t) using the following equation.

△ V (t+tp)-V (t)+1)” V- (t
)...(11) Then, in the subtracter 42, Vm (t+tp)-v(t+
tp)-ε is calculated, and either the accelerator or the brake is controlled by this prediction error ε, and the brake gain multiplier 46 outputs the brake correction amount Δub, and the accelerator gain multiplier 48 outputs the accelerator correction amount ΔU. is output. Therefore, from the adder 50a, the program accelerator amount W
The accelerator control value with a corrected value and the brake control value with the programmed brake amount Wb corrected are output from the adder 50b, thereby enabling good predictive control.

[Effects of the Invention] As explained above, according to the present invention, since the target course is first taught based on the taxiway, it is possible to freely set the actual driving course of the vehicle with respect to the taxiway. This makes it possible to select the most suitable course for different vehicles, and since the prediction error is corrected, the vehicle can be driven accurately on the target course and will not run off course.

Furthermore, by teaching and controlling the amount of operation of various actuators and performing prediction error correction control, it is possible to obtain the same riding comfort as when actually operated by a human, making it suitable for transporting not only objects but also people. Autopilot can be successfully applied.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing a schematic configuration of a first embodiment of the present invention, Fig. 2 is a diagram showing a configuration when the present invention is applied to an actual autopilot vehicle, and Fig. 3 is a lateral view of the taxiway. 4 is a block diagram showing the configuration for controlling the steering amount in the first embodiment; FIG. 5 is a diagram illustrating the prediction error calculation performed by the embodiment of FIG. 4. Fig. 6 is a flowchart showing the overall operation of the first embodiment, Fig. 7 is a flowchart showing the sensor data reading operation, Fig. 8 is a flowchart showing the vehicle position evaluation operation, and Fig. 9 is the teaching control. 10th flowchart showing the operation of
The figure is a block diagram showing the configuration of a second embodiment in which the present invention is applied to vehicle speed control. 10 ... Position/state detection means 12 ... Actuator operation amount detection means 14 ・
... Teaching data storage means 16 ... Prediction error calculation means 18 ... Actuator operation amount calculation means 20
... Induction cable 22 ... Position beacon 24 ... Arithmetic processing circuit (ECU) 26 ...
Feedback processing circuit 28...Vehicle 30...Steering 32...Steering amount sensor 36...Display lamp 141...Target course map 142...Steering amount map 181...Adder

Claims (1)

[Claims] (1) In a vehicle autopilot system that measures the current position and running condition of the vehicle with respect to the taxiway and operates and controls each actuator for autopilot based on this position and running condition, a predetermined taxiway is A teaching data storage means for storing a target course taught by the driver while driving and the amount of operation of each actuator when driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course, and a teaching data storage means for storing a target course taught by the driver while driving the target course. a prediction error calculation means for calculating a prediction error when the vehicle travels to An autopilot control device for a vehicle is characterized in that it accurately controls traveling on a taught target course while making corrections based on the following.