JPH05280623A - Controller for automatic transmission - Google Patents

Abstract

PURPOSE:To contrive improvement of a feeling of running by performing a speed change of agreeing with a request of an idea based on the idea such as judging the driving request of deceleration, steady, acceleration, etc., by a driver from also information of a road picture before a vehicle, relating to a control device for an automatic transmission of utilizing the input information of the picture for controlling a speed change point of the automatic transmission. CONSTITUTION:A device has an outside picture detecting means 11 for detecting outside picture information before a vehicle and a deceleration factor judging means 12 for judging whether a deceleration factor is provided or not before the vehicle based on the outside picture information detected by the outside picture detecting means 11. The device comprises a step-in release detecting means 13 for detecting step-in release of an accelerator 16 and a shiftup inhibiting means 14 for inhibiting shiftup action in an automatic transmission 15 when the step-in release of the accelerator is detected by the step-in release detecting means 13 and further when the deceleration factor is judged to be provided by the deceleration factor judging means 12.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device for an automatic transmission, which uses input information of a road image to control a shift point of the automatic transmission.

[0002] An automatic transmission mounted on an automobile controls a shift point on the basis of operating parameters such as vehicle speed and throttle opening (TVO) as disclosed in Japanese Patent Laid-Open No. 63-6260. There is. FIG. 17 shows a basic shift diagram used for shift point control of a conventional automatic transmission. As shown in the figure, in the conventional shift point control, basically, the shift speed is uniquely determined by the vehicle speed and the throttle opening.

[0003]

The biggest problem with the conventional shift point control is that the accelerator is depressed despite the driver's demand for deceleration due to the presence of road corners or obstacles in front of the running vehicle. When is turned off (release of the pedal), the throttle opening is closed and the automatic transmission shifts up to the higher gear, which reduces the effectiveness of engine braking. When the driver makes an acceleration request at the next moment and depresses the accelerator to try to accelerate again, the automatic transmission shifts down to the original gear, which causes shock and unnecessary gear changes Comfort is compromised.

For example, as shown in FIG.
At the uphill corner, it was in 3rd speed before entering the corner, but it became 4th speed in the corner because the accelerator was released. When I exited the corner, I stepped down on the accelerator again, so I downshifted to 2nd or 3rd speed. There are many unnecessary shifts, which impairs driving comfort. Similarly, at the downhill corner, the vehicle was in the 3rd speed before entering the corner, but it was in the 4th speed at the corner because the accelerator was released.

As described above, it is always necessary to accurately determine whether the driver requests deceleration, steady operation, or acceleration only by the operating parameters such as the vehicle speed and the throttle opening which are used in the conventional shift point control. It was not possible to say that appropriate shift point control was not always performed.

The present invention has been made in view of the above circumstances, and an object of the present invention is to judge a driving request such as deceleration, steady state, acceleration, etc. of a driver from road image information in front of the vehicle. Based on the concept, it is intended to improve the driving feeling by shifting gears that match the demand.

[0007]

FIG. 1 is a diagram for explaining the principle of the present invention. In the control device for an automatic transmission according to the present invention, an external world image detecting means 11 for detecting external world image information in front of the vehicle, and a deceleration factor existing in front of the vehicle based on the external world image information detected by the external world image detecting means 11. Deceleration factor determining means 12 for determining whether or not to perform, and accelerator 16
The depression release detection means 13 for detecting the depression release of the accelerator pedal, and the depression release means 13 detect the depression of the accelerator pedal,
Further, the deceleration factor determining means 12 is provided with a shift-up inhibiting means 14 for inhibiting the shift-up operation in the automatic transmission 15 when it is determined that the deceleration factor exists.

[0008]

The outside world image detecting means 11 detects an image in the traveling direction of the vehicle. From the image information, the deceleration factor determination means 12 determines a deceleration factor ahead of the vehicle, such as a road corner, an overtaking vehicle, an obstacle on the road, braking or turning of the vehicle in front, approach of the vehicle in front, and the like. When the driver turns off the accelerator in response to the appearance of these deceleration factors, the depression release detecting means 13 detects it and the upshift inhibiting means 14 inhibits upshifting of the automatic transmission. As a result, the automatic transmission does not have to perform an unnecessary gear shifting operation, which improves the driving feeling.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 shows a control device for an automatic transmission as one embodiment of the present invention. In the figure, reference numeral 1 denotes a camera which is attached to a front position of a vehicle and captures an image of a road or the like extending in front of the vehicle. The camera 1 may be a black-and-white camera, but the accuracy of recognition is improved when the color image processing is performed by the road structure recognition device 2 in the subsequent stage using a color camera. Two
Is a road structure recognition device, and includes a road detection device 21 and a road change determination device 22. The road detection device 21 detects road information such as road edge lines on the left and right sides of the road and a road area surrounded by the left and right road edge lines based on the imaged image information from the camera 1. The road change judging device 22 judges the presence of factors such as the appearance of a corner of the road, the slope of the road, obstacles on the road, etc., which decelerate the vehicle, based on the road information detected by the road detecting device 21. ..

3 is an electronic control unit for an automatic transmission,
Road structure information is input to the electronic control unit 3 from the road structure recognition unit 2 in addition to operating parameters (vehicle speed, engine speed, throttle opening, etc.). The electronic control unit 3 controls the automatic transmission 4 on the basis of these pieces of information, and when the road structure recognition unit 2 notifies the presence of a deceleration factor, the electronic transmission unit 3 is controlled so as to prohibit upshifting of the automatic transmission 4. Is done.

The operation of this embodiment device is described below with reference to the flow chart shown in FIG. Vehicle speed, throttle opening, etc. are input to the electronic control unit 3 (step S1),
Based on these operating parameters, the shift stage of the automatic transmission is determined by the basic conversion diagram as shown in FIG. 17 (step S2).

Further, in the road structure recognizing device 2, first, the road detecting device 21 detects road information such as road edge lines and road regions, and external environment image information such as forward vehicles on the road (step S3). The road change determination device 22 determines whether or not there is a factor for decelerating the vehicle ahead of the route based on the external image information (step S3). The deceleration factors include, for example, the appearance of road corners and downhills in front of the route, the entry of overtaking vehicles in the front of the route, the lighting of the brake lights and direction indicators of the vehicle ahead, the vehicle ahead decelerating or accelerating There is a rapid approach or the presence of obstacles on the road. Further, the distance between these deceleration factors is measured, and it is detected whether or not the distance has approached a predetermined dangerous distance (step S5).

When the electronic control unit 3 detects that the distance between the vehicle and the deceleration factor is within the dangerous distance and that the accelerator is released, the electronic control unit 3 controls the automatic transmission so as to prohibit upshifting. Perform (step S6). As a result, it is possible to prevent inappropriate shifting by the automatic transmission.

Various embodiments of the road structure recognition device 2 in the above-described embodiment device will be described below.

FIG. 4 shows an embodiment for detecting a corner appearing on the road by detecting a white line on the road (a road shoulder line, a lane marking, etc.) by using a window as the road structure recognition device 2. .. This road structure recognition device 2
The road detection device 21 in the window management device 211 generates and manages a window for white line fitting.
It is composed including.

As shown in FIG. 5, the window management device 211 sets a vertical detection line DL at, for example, a central position on an image taken by the camera 1, and selects a white line candidate (on the shoulder of the road) extracted from the taken image. When a running line drawn along the line, a white line-shaped portion including a traffic division line) intersects with the detection line DL, a rectangular window is generated around the intersection, and the window is rotated about the intersection. The white line candidates are tracked by performing white line fitting so that the white line candidates fit within the window, and information about the Y coordinate position and inclination of this window is sent to the road change determination device 22. Further, the white line width on the screen corresponding to the Y coordinate is calculated from the white line width given by the road structure model 23, and it is sent to each window and the window width is adjusted to that value to perform optimum white line tracking. It Therefore, each window moves independently so that the window moves up and down the detection line DL and rotates about the intersection with the detection line DL.

The operation of this embodiment apparatus will be described below. When the images captured by the camera are sequentially input, the road detection device 21 extracts white line candidates from the images.
The window management device 211, as shown in FIG.
When a white line candidate that intersects the detection line is captured, windows WD1, WD2, and WD3 are generated centering on the intersection. The generated window is rotated around the intersection so that the white line candidates in the image always fit within the window, and the white line candidates are tracked accordingly. Then, the inclination of the window and the Y coordinate value at that time are used as the road change determination device 22.
Send to.

The road change judging device 22 calculates the distance to the white line candidate (the distance from the lowermost point on the detection line in the screen to the intersection with the white line candidate) based on the inclination of this window and the Y coordinate value. At this time, the movement vector, the luminance distribution, the occupation frequency, etc. of the white line candidate are compared with the road structure model as judgment information, and the window WD3 of the white line candidate having a factor that cannot be recognized as the white line of the traveling line is deleted. The window management device 211 is notified as described above.

When the distance to the white line intersecting the detection line DL is calculated by the road change determination device 22 in this way,
The appearance of a corner ahead of the route can be determined based on this distance. That is, when the road ahead of the route is a straight line, the traveling line drawn along the road shoulder does not intersect the detection line DL. On the other hand, since the traveling line on the shoulder crosses the detection line DL at the corner, the windows WD1 and WD2 are generated and their distances are calculated, so it can be estimated that the corner has appeared. ..

FIG. 6 shows another embodiment of the road structure recognition device 2. The road structure can be accurately recognized by precisely processing the image captured by the camera, but the amount of calculation becomes enormous, and it is difficult for the in-vehicle computer to process it in real time. In this embodiment, a road structure is roughly approximated by a first-order straight line, thereby making it possible to detect a corner while significantly reducing the calculation amount.

In FIG. 6, the road detecting device 21 includes a road edge detecting device 213 for detecting a road edge line and a linear approximation line (y = ax + b) for approximating a road edge on a straight road based on the detected road edge information. And a first-order approximation device 214 for calculating The first-order approximation straight line is calculated in a state where the steering angle is kept in a straight road condition for a certain period of time, and the first-order approximation straight line is constantly updated. In this linear approximation line y = ax + b, y = ax + b + σ and y = ax + b−σ in consideration of the variance σ of the movement of the road edge.
The road structure is approximated by.

In the road detecting device 21 whose operation will be described below with reference to FIG. 7, the road edge detecting device 213 detects the road edge based on the input image from the camera 1, and at the same time, the primary approximation is performed. The device 214 calculates the first-order approximation straight line y = ax + b of the road end of the straight road. This linear approximation line is constantly updated.

The steering angle and the vehicle speed are input to the road change determination device 22 every moment, and the road change determination device 22 considers the variance σ of the movement of the road edge in the calculated linear approximation line. , Determine the appearance of the corner. FIG. 7 shows the state of this determination. That is,
As shown in (a), a linear approximation road y = ax + b is calculated for a straight road, and a value obtained by adding the variance σ to it is set as a reference range (dotted line range in the figure), and the reference range is assumed to be a road edge structure. To do. When the road ahead approaches a corner [(b) → (c) → (d)], the linear approximation line of the road edge at that corner exceeds the reference range. It can be detected.

FIG. 8 shows another embodiment of the road structure recognition device 2. The apparatus of this embodiment can easily obtain the degree of curve of the road (roughly not the curvature) by detecting the center of gravity of the road area.

In FIG. 8, the road detecting device 21 includes a road edge detecting device 213 and a road area detecting device 215. The road edge detection device 213 detects the road edge and calculates an equation approximating it. This road edge may be approximated by a simple quadratic equation. The road area detection device 215 calculates the area of the road area surrounded by the left and right road edges detected by the road edge detection device 213, and calculates the center of gravity position thereof. These calculated information are sent to the road change determination device 22.

The road change judging device 22 is a road detecting device 21.
It is equipped with a buffer that stores several frames of information sent from, and is configured to store time-series road information.

The operation of this embodiment apparatus will be described below with reference to FIG. The road change determination device 22 determines the appearance of a road corner based on the center of gravity information of the road area detected by the road detection device 21. FIG. 9 shows how this judgment is made. That is, if the center of gravity of the road area on the straight road is detected as the initial position (the image at the left end of FIG. 9),
When the vehicle approaches a corner, the position of the center of gravity of the road area changes due to the change in the shape of the road area. Therefore, the appearance of the corner can be determined based on the magnitude of the change.

The road change determination device 22 sets the initial position of the center of gravity on the straight road from the center of gravity information stored in the buffer. Since the vehicle speed and the steering angle are input to the road change determination device 22, it is considered that the vehicle is traveling along a straight road when a certain time or more has elapsed while the steering is in a straight traveling state based on these information. The center of gravity position of is the initial position. The initial position of the center of gravity may be set in addition to the above, by adjusting a constant in advance as a simple method.

The road change judging device 22 judges the appearance of a corner based on the change of the center of gravity position of the road area as described above. At this center of gravity position, the photographing screen of the camera 1 is shaken by the vibration of the vehicle. It also depends on this, so it is necessary to eliminate this effect. Therefore, it can be expected that the center-of-gravity position will fluctuate significantly when the vehicle shakes significantly. Therefore, the smallest corner (the one with the smallest curvature) to be detected is assumed, and the center-of-gravity of the smallest corner is detected. The fluctuation range is calculated in advance as the variance σ of the center of gravity, and it is determined that the fluctuation of the center of gravity exceeding the fluctuation range is not due to the appearance of a corner but due to the vibration of the vehicle.

As described above, the road change judging device 22 measures the variation of the center of gravity after determining the variance σ, and if the steering angle is almost constant and the variation of the center of gravity is less than the variance σ. , It is determined that a meaningful corner appears in front of the vehicle.

In this embodiment, it is also possible to detect the intrusion of an overtaking vehicle in the front of the route or the approach of a forward vehicle or an obstacle in the front of the route based on the area information of the road region.
That is, since these forward vehicles and obstacles have two-dimensional areas as image information, if the road area area is detected by excluding the areas of these objects, these objects will appear in front. In this case, the area of the road region detected momentarily changes with the ratio of the quadratic equation (for example, y = Ax + B). In this way, when the change of the road area becomes the ratio of the quadratic equation, it is considered that a vehicle or the like exists ahead. In the case of an intruding vehicle, the intruding direction can be determined based on the changing direction of the center of gravity of the road area.

FIG. 10 shows still another embodiment of the road structure recognition device 2. The apparatus of this embodiment detects the appearance of a corner based on the difference information between frames of time-series image information.

In FIG. 10, the road detecting device 21 includes a road edge detecting device 213 for detecting the road edge.
The road change determination device 22 includes a buffer that accumulates road edge information in time series in order to calculate the time course of the road edge information. The accumulated road edge information is, for example, image information in which the road edge portion is "1" and the other portions are "0". The road change determination device 22 sequentially obtains the difference between frames of the road edge information, accumulates the differences calculated over a certain time, and detects the road change based on the result.

Generally, when observing a temporal change of a road image, the rate of change D of the image is the approach speed V to the corner of the vehicle.
And the curvature R of the corner. That is, D = αVR. Here, α is a coefficient at that time. From this, when the change rate D of the image becomes large, it means that the vehicle has entered the corner at a high speed, the corner of the entry destination is steep, or both. It can be considered, and in any case, it is considered to be a deceleration factor of the vehicle, so it is desirable to prohibit the shift-up of the automatic transmission.

Therefore, as shown in FIG. 11, the road change determination device 22 obtains the inter-frame difference of the road edge information [(a) in FIG. 11] input momentarily in a time series [FIG.
1 (b)], a value obtained by accumulating the differences over a fixed time is obtained [(c) of FIG. 11]. Since the road edge does not change with respect to a straight road, a difference between frames does not occur, and this cumulative value is small. On the other hand, at the corner, a difference between frames occurs and the cumulative value increases. Therefore, when the cumulative value exceeds the predetermined value determined in advance, it is determined that a deceleration factor such as a corner in which the shift-up should not be performed appears.

Still another embodiment of the road structure recognition device 2 will be described below with reference to FIGS. In this embodiment, not only the appearance of a corner but also the appearance of an uphill or a downhill can be easily detected.

It is considered that it is important to know the gradient of the road shape for an autonomous vehicle, for example, but it is considered that sufficient prediction control can be performed if the information on the gradient required by the autonomous vehicle is rough gradient information. Be done. If a two-dimensional image is used, the road gradient cannot be estimated without some hypothesis. For example, a method may be considered in which the road bank is assumed to be zero, a hypothesis that the roads are parallel is provided, and the gradient curvature of the road is calculated by matching with the road model. However, since these methods require a huge amount of calculation, it is difficult to perform real-time calculation with a computer that can be mounted on a vehicle. In this embodiment, the empirical feature is used to detect the slope of the road so that the rough determination of the corner and the slope of the road can be made.

The outline of the processing will be described. As shown in FIG. 12, when an input image is given, first, the road edge is detected from the image. The method of extracting the road edge at this time may be a method of searching for the road edge from the lower part of the screen to the upper part, assuming that the road always exists at the nearest position in front of the camera. Next, a quadratic curve representing the road edge required for the determination is calculated from the detected road edge. As the calculation method, the least square method or the like can be used.

Next, the unevenness is judged from the obtained quadratic curve on the image plane. For the determination of unevenness, the quadratic curve is y = ax +
This can be done depending on whether the line segment given by b has a positive slope or a negative slope.

If the unevenness of the left road edge and the right road edge on the image can be determined, the road condition can be determined using the road shape determination table shown in FIG. That is, FIG.
As shown in FIG. 4, if both the right road edge and the left road edge are convex, the slope is downhill, if both are concave, the slope is uphill, and if the right roadside is concave and the left roadside is convex, the right corner and the right road are If the edge is convex and the left road edge is concave, it is the left corner. If the road shape is approximated by such a method, the determination can be performed with a simple quadratic curve on the image, and thus high-speed processing can be performed.

In the above embodiment, the automatic transmission is prohibited from shifting up when there is a corner or the like mainly by recognizing the shape of the road. When the indicator is turned on, this may be used as a deceleration factor to prohibit upshifting of the automatic transmission. For example, as shown in FIG. 15, when a vehicle running in the same direction is detected in the road image ahead, when the brake lamp of the vehicle is lit or when the turn indicator can be confirmed to be lit, It judges that there is something that causes a deceleration in front of the car, and controls so that the automatic transmission does not shift up even if the driver turns off the accelerator. The lighting of the brake lamp or the turn signal can be determined by detecting that the current screen has undergone an extreme brightness change with respect to the previous screen in the time-series screen.

As a method for measuring the distance to the deceleration factor (corner, vehicle ahead, etc.) in the above embodiments, a simple method as shown in FIG. 16 can be used. That is, for example, a window WD is provided along the bottom edge of the screen to detect the brightness of that portion. Then, the white line drawn on the road (running line on the shoulder and traffic line)
The brightness becomes higher in the area, and becomes lower in other areas. Here, since the traffic division line has line segments drawn in a broken line shape at a predetermined distance, the luminance of the traffic division line portion periodically increases or decreases. By observing this change in brightness, the traffic division line is detected.
Then, by detecting the number of line segment of the traffic division line up to the deceleration factor, the distance to the deceleration factor can be easily calculated.

[0043]

As described above, according to the present invention, driving demands such as deceleration, steady state, acceleration, etc. of the driver can be judged from the road image information in front of the vehicle, and the automatic driving that meets the demand can be determined. By shifting the transmission, it is possible to reduce the number of unnecessary shifting operations and improve the driving feeling.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the present invention in principle.

FIG. 2 is a diagram showing a control device for an automatic transmission as one embodiment of the present invention.

FIG. 3 is a flowchart for explaining the operation of the apparatus of the embodiment.

FIG. 4 is a diagram showing a configuration example of a road structure recognition device in the embodiment device.

5 is a diagram for explaining the operation of the road structure recognition device in FIG.

FIG. 6 is a diagram showing another configuration example of the road structure recognition device in the embodiment device.

FIG. 7 is a diagram for explaining the operation of the road structure recognition device in FIG.

FIG. 8 is a diagram showing still another configuration example of the road structure recognition device in the embodiment device.

FIG. 9 is a diagram for explaining the operation of the road structure recognition device in FIG. 8.

FIG. 10 is a diagram showing still another configuration example of the road structure recognition device in the embodiment device.

FIG. 11 is a diagram for explaining the operation of the road structure recognition device in FIG.

FIG. 12 is a diagram illustrating a processing flow of still another road structure recognition device in the embodiment device.

FIG. 13 is a diagram showing a road shape determination table of the road structure recognition device.

FIG. 14 is a diagram showing a determination example by the road structure recognition device described above.

FIG. 15 is an explanatory diagram of a method of detecting the lighting of a brake lamp of a vehicle ahead.

FIG. 16 is a diagram illustrating a method of measuring a distance to a deceleration factor.

FIG. 17 is a basic conversion diagram of an automatic transmission.

FIG. 18 is a diagram illustrating a problem caused by an upshift of the automatic transmission.

[Explanation of symbols]

 1 Camera 2 Road Structure Recognition Device 3 Electronic Control Device for Automatic Transmission 4 Automatic Transmission 21 Road Detection Device 22 Road Change Judgment Device 23 Road Structure Model 211 Window Management Device 213 Road Edge Detection Device 214 First Approximation Device 215 Road Area Detection apparatus

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Jun Shinnobu, 3-1, Shinchi Fuchu-cho, Aki-gun, Hiroshima Prefecture Mazda Co., Ltd. Within the corporation

Claims (1)

[Claims] 1. An outside world image detecting means (11) for detecting outside world image information in front of the vehicle, and whether or not a deceleration factor exists in front of the vehicle based on the outside world image information detected by the outside world image detecting means. Deceleration factor judging means (12) for judging, depression release detecting means (13) for detecting depression release of the accelerator (16), and depression release of the accelerator is detected by the depression release means,
Further, control of the automatic transmission provided with shift-up prohibiting means (14) for prohibiting shift-up operation in the automatic transmission (15) when the presence of the deceleration factor is judged by the deceleration factor judging means. apparatus.