JP2015089801A - Operation control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable an automatic operation in which an operator's feeling of discomfort is suppressed.SOLUTION: A plurality of items for specifying a vehicle operation environment (environment items) are set. During manual operation control, an operation environment is specified by a combination of values of the environment items detected on the environment items, and an operation of an operator is learned in association with the operation environment. During automatic operation control, values of the environment items are detected to specify the operation environment, and referring to the result of learning of the operation in the operation environment, automatic operation control is performed. In this way, the operation environment is appropriately specified even in a complicated operation environment, the operation of the operator can be learned for each operation environment so that the operator's feeling of discomfort during the automatic operation can be greatly suppressed.

Description

  The present invention relates to an operation control device mounted on a vehicle that can be operated by switching between manual operation and automatic operation.

  In recent years, automatic driving technology for automatically driving a vehicle has been developed in order to reduce the driving burden on the driver. In addition, a technique has been proposed in which a driver's driving habit is learned in accordance with the driving environment such as road width and weather, and the driving habit is reflected in automatic driving (for example, Patent Document 1). As a result, it is possible to perform automatic driving close to the driving habit of the driver, so that it is considered that the driver's uncomfortable feeling with respect to driving operation (for example, acceleration method and traveling speed) at the time of automatic driving can be suppressed. .

JP-A-5-58198

  However, the proposed technology still has a problem that it is difficult to suppress the driver's uncomfortable feeling during automatic driving. This means that even if you say “driver's driving habit”, you don't know how to evaluate the driving habit, and if you evaluate it wrongly, you will learn how to drive habit. This is because the driver's uncomfortable feeling during automatic driving cannot be suppressed.

  The present invention has been made in view of the above-described problems, and provides an operation control device capable of automatically driving in a state where a driver's uncomfortable feeling is suppressed.

In order to solve the above-described problems, the operation control device of the present invention learns the driving operation by the driver during manual operation control. Here, the driving operation is learned as follows. First, a plurality of items (environment items) for specifying the driving environment of the vehicle are set in advance. Then, the driving environment is specified by the combination of the environmental item values detected for these environmental items, and the driving operation of the driver is learned in association with the driving environment. Thereafter, at the time of automatic driving control, the ambient condition of the vehicle is detected and the environmental item value is also detected, and the driving environment is specified based on the environmental item value. Then, automatic driving control is executed with reference to the learning result of the driving operation in the driving environment.
In this way, it is possible to specify the driving environment by appropriately selecting the environmental items, no matter how complicated the driving environment is. In addition, a complicated driving environment can be efficiently identified by selecting a truly necessary environmental item. As a result, it becomes possible to learn the driving operation of the driver for each driving environment, so that it is possible to greatly suppress the driver's uncomfortable feeling during automatic driving.

In the above-described operation control device of the present invention, when detecting an environmental item value for each environmental item, the corresponding environmental item value is selected from a plurality of candidate values preset for each environmental item. You may do it.
In this way, every driving environment can be represented as a finite number of coordinate points set in an N-dimensional space corresponding to the number of environment items. As a result, similar driving environments are learned as the same driving environment, so that it is possible to efficiently learn driving operations.

In the operation control device of the present invention described above, the control mode of the automatic operation control may be switched to either the standard control mode or the specific control mode depending on the driver's selection. Then, when learning the driving operation during manual driving, the driving operation may be learned in association with the combination of the control mode selected by the driver and the driving environment.
Even for the same driver, it is considered that the driving operation without a sense of incongruity differs depending on the selected control mode. Therefore, by learning the driving operation in association with the combination of the control mode and the driving environment, even if the driver can select the control mode at the time of automatic driving, it is possible to perform automatic driving without giving a sense of incongruity. Become.

Further, in the above-described operation control device of the present invention, even when automatic operation control is being performed, when a driving operation by the driver is detected, the detected driving operation is learned in association with the driving environment at that time. It is good as well.
The driver performs the driving operation during the automatic driving control when the driver feels uncomfortable with the automatic driving, and the driving operation performed is considered to be a driving operation that the driver feels desirable. Therefore, if the driver's driving operation detected at the time of automatic driving control is learned in association with the driving environment at that time, it is possible to efficiently learn an appropriate driving operation.

In the above-described driving control device of the present invention, the environmental items for detecting the environmental item values are “driving area”, “brightness”, “weather”, “number of lanes”, “width of traveling lane”, “ `` Road shape '', `` Road gradient '', `` Road surface type '', `` Road surface condition '', `` Signal display contents '', `` Number of surrounding vehicles '', `` Front vehicle speed '', `` Front vehicle acceleration '', `` Near obstacle '' It may be at least two environmental items among “travel state of own vehicle”, “travel direction of own vehicle”, and “type of travel lane of own vehicle”.
By combining the environmental item values of such environmental items, various driving environments can be expressed, and driving operations can be learned in association with such various driving environments.

It is explanatory drawing which shows the vehicle carrying the driving control apparatus of a present Example. It is explanatory drawing which shows the internal structure of the operation control apparatus of a present Example. It is a flowchart of the main control process which an operation control apparatus implements. It is a flowchart of the driving operation learning process implemented at the time of manual driving. It is explanatory drawing which showed the method of specifying the driving environment of a vehicle. It is explanatory drawing which illustrated the environmental item which comprises the driving environment of a vehicle. It is explanatory drawing which illustrated the environmental item value about an operation area among the environmental items which comprise a driving environment. It is explanatory drawing which illustrated the environmental item value about a visibility state among the environmental items which comprise a driving environment. It is explanatory drawing which illustrated the environmental item value about a road among the environmental items which comprise driving environment. It is explanatory drawing which illustrated the environmental item value about a road among the environmental items which comprise driving environment. It is explanatory drawing which illustrated the environmental item value about a surrounding vehicle among the environmental items which comprise a driving environment. It is explanatory drawing which illustrated the environmental item value about a surrounding obstacle among the environmental items which comprise a driving environment. It is explanatory drawing which illustrated the environmental item value about the own vehicle among the environmental items which comprise a driving environment. It is explanatory drawing which illustrated the item detected as a driving operation by a driver | operator. It is explanatory drawing which showed having matched driving environment and learning driving operation. It is a flowchart of the automatic driving | operation process implemented at the time of automatic driving | operation.

Hereinafter, examples will be described in order to clarify the contents of the present invention described above.
A. Device configuration :
FIG. 1 shows a vehicle 1 on which an operation control device 100 of the present embodiment is mounted. The driving control device 100 is an electronic control unit (ECU) that includes a CPU, a ROM, a RAM, and the like, and can drive the vehicle 1 automatically.
As shown in the figure, various sensors, a car navigation device 220, a communication device 230, and the like are connected to the operation control device 100. Examples of sensors include various sensors that detect information related to the driving state and driving operation of the vehicle 1 (hereinafter collectively referred to as “vehicle information detection sensor 200”), traffic conditions around the vehicle 1, and other vehicles. Various sensors (hereinafter collectively referred to as “peripheral information detection sensor 210”) for detecting information on the connection are connected.

Among these, the vehicle information detection sensor 200 includes a vehicle speed sensor 202 that detects the speed of the vehicle 1, an accelerator sensor 204 that detects the operation state of the accelerator pedal, a brake sensor 206 that detects the operation state of the brake pedal, and a steering angle. A steering angle sensor 208 is detected.
The surrounding information detection sensor 210 includes a front camera 212a (or rear camera 212b) that acquires a front image (or rear image) of the vehicle 1, presence / absence of a front vehicle (or rear vehicle), and a front vehicle (or rear vehicle). The front laser radar 214a (or the rear laser radar 214b) that detects the speed of the rear is included.
The driving control device 100 uses these sensors (202 to 208, 214a, 214b) and the cameras (212a, 212b) to provide information on the driving state and driving operation of the vehicle 1, traffic conditions around the vehicle 1, and other vehicles. Get information about.

Further, the driving control device 100 is based on map information obtained from the car navigation device 220 (hereinafter referred to as the car navigation system 220), and information related to the area in which the vehicle 1 is traveling (for example, residential areas, urban areas, high speeds). Information about the shape of the road (for example, whether the road is a straight road or a curved road).
Furthermore, the driving control apparatus 100 acquires information related to the traveling state of the other vehicle by performing inter-vehicle communication using the communication device 230. Further, by performing road-to-vehicle communication using the communication device 230, information on the presence / absence of a traffic signal and the display content of the traffic signal is acquired.

Further, as shown in the figure, various actuators for driving the vehicle 1 (hereinafter collectively referred to as “traveling actuator 300”) are connected to the operation control apparatus 100. For example, the throttle actuator 302 that adjusts the engine output by adjusting the throttle opening, the brake actuator 304 that decelerates (or stops) the vehicle 1 by pressing the brake pad against the brake surface, and the traveling direction of the vehicle 1 is changed. A steering actuator 306 is connected.
The driving control apparatus 100 performs a process of automatically driving the vehicle 1 by controlling the operation of the travel actuator 300 based on various information obtained from the various sensors (200 to 230) described above.

  Further, an operation panel 400 and a display 500 are connected to the operation control apparatus 100. The operation panel 400 is provided with a manual / automatic switching SW 402 (see FIG. 2) for setting the “driving method” of the vehicle 1 to either “manual driving” or “automatic driving”. / Automatic switching SW 402 is operated by the driver of vehicle 1. When the operation control device 100 detects that the manual operation is switched to the automatic operation, the operation control device 100 starts a process of automatically driving the vehicle 1. In addition, the operation control apparatus 100 displays the control status of the driving operation during driving on the display 500.

  FIG. 2 shows an internal configuration of the operation control apparatus 100 of the present embodiment. As illustrated, the driving control apparatus 100 includes a driving method determination unit 110, a driving operation detection unit 120, a driving environment identification unit 130, a driving operation learning unit 140, a database 150, an automatic driving execution unit 160, and the like. .

  Among these, the driving method determination unit 110 determines the “driving method” of the vehicle 1 based on the setting state of the manual / automatic switching SW 402 provided on the operation panel 400. As described above, the method of driving the vehicle 1 includes “manual driving” and “automatic driving”, and the driver sets which method to drive.

Further, the driving method determination unit 110 determines the “control mode” of the vehicle 1 during automatic driving. Here, the “control mode” represents the travel mode of the vehicle 1, and in the operation control device 100 of the present embodiment, “eco mode”, “sport mode”, and “normal mode” are control modes. It is prepared. In the eco mode, the vehicle travels while suppressing fuel consumption by slowly accelerating when starting. In the sport mode, the vehicle 1 travels taking advantage of the travel performance of the vehicle 1 by, for example, performing acceleration at the time of starting. In the normal mode, intermediate driving between the eco mode and the sport mode is performed.
A control mode switching SW 404 for switching these control modes is provided on the operation panel 400 and is operated by the driver. The driving method determination unit 110 determines the control mode of the vehicle 1 during automatic driving based on the set state of the control mode switching SW 404. The normal mode of the present embodiment corresponds to the “standard control mode” of the present invention, and the eco mode and the sport mode of the present embodiment correspond to the “specific control mode” of the present invention. Further, the control mode switching SW 404 of this embodiment corresponds to a “control mode switching unit” of the present invention.
In addition, the driving method determination unit 110 displays the setting states of “driving method” and “control mode” on the display 500.

  The driving operation detection unit 120 detects a driving operation performed by the driver based on various information obtained from the vehicle information detection sensor 200 described above with reference to FIG. Driving operations by the driver include operations such as an accelerator pedal, a brake pedal, and steering.

  Based on various information obtained from the various sensors and the like (200 to 230) described above with reference to FIG. ). As will be described in detail later, the driving environment is composed of a plurality of items (for example, the weather, the width of the driving lane, the presence / absence of surrounding vehicles, etc.), and the state of these items is specified. Thus, it can be considered that the driving environment is specified. Specific examples of items constituting the operating environment will be described later.

  Here, in the driving control apparatus 100 of the present embodiment, not only automatic driving of the vehicle 1 but also processing for learning driving operation by the driver during manual driving is performed. This is to perform automatic driving reflecting the driving operation by the driver. As will be described later, the process of learning the driving operation is performed by the driving operation learning unit 140, and the learning result is accumulated in the database 150. The database 150 is provided with three databases (eco mode DB 150a, sports mode DB 150b, and normal mode DB 150c) corresponding to the three control modes described above, and learning results of driving operations are accumulated for each control mode. (Details will be described later).

In the automatic driving | running | working execution part 160, the process which drives the vehicle 1 automatically is performed. As shown in the figure, the automatic operation execution unit 160 includes an actuator control module 161 and a control status display module 162.
When it is determined that the driving method of the vehicle 1 is “automatic driving”, the actuator control module 161 refers to the learning result of the driving operation stored in the database 150 and performs various actuators (traveling actuator 300). To control the operation.
The control status display module 162 displays the control status of the driving operation on the display 500.

B. Main control processing:
FIG. 3 shows a main control process for performing a process for automatically driving the vehicle 1 or a process for learning a driver's driving operation depending on whether the driving method of the vehicle 1 is set to automatic or manual. The flowchart of is shown. This process is started by the operation control device 100 when the engine of the vehicle 1 is started.
When the main control process is started, the driving method determination unit 110 of the driving control device 100 reads out whether the driving method of the vehicle 1 is set to automatic driving or manual driving (S100). Which operation method is used is read from the setting state of the manual / automatic switch SW 402 provided on the operation panel 400.
Subsequently, the driving method determination unit 110 reads out whether the control mode of the vehicle 1 is set to eco, sports, or normal (S102). The control mode is read from the setting state of the control mode switching SW 404 provided on the operation panel 400.

  In the operation control apparatus 100 of the present embodiment, the operation method of the vehicle 1 is set (initialized) to “manual operation” after the engine is started. On the other hand, in the control mode, the setting when the engine is stopped is stored in the storage unit (not shown), and the stored setting is read when the engine is started next time. To do.

Subsequently, the driving method determination unit 110 determines whether or not the read driving method is set to automatic driving (S104). As a result, when “manual operation” is set in an initialized state after the engine is started (S104: no), it is set to “manual operation” and the setting state of “control mode” Is displayed on the display 500 (S108). As a result, the driver can check whether or not the current driving method and control mode of the vehicle 1 are the desired settings.
Thereafter, the operation control apparatus 100 performs a process (driving operation learning process) for learning the driving operation of the driver during manual driving (S200). The driving operation learning process will be described later.

On the other hand, when it is determined in the determination process of S104 described above that the driving method is set to “automatic driving”, that is, the manual / automatic switching SW 402 is operated by the driver to start from “manual driving”. If it has been switched to “automatic operation” (S104: yes), the fact that it is set to “automatic operation” and the setting state of “control mode” are displayed on the display 500 (S106). Accordingly, the driver can confirm that the driving method has been switched to “automatic driving” and the control mode setting state.
Thereafter, the operation control apparatus 100 performs a process of automatically driving the vehicle 1 (automatic driving process) (S300). The automatic operation process will be described later.

  Thus, after the driving operation learning process or the automatic driving process is performed, the process returns to the first process (S100) and the above-described series of processes is repeated. That is, when the operation method and the control mode are read out (S100, S102), the setting method of the operation method and the control mode is displayed on the display 500 (S104, S106, S108). When the driving method is set to manual driving, driving operation learning processing is performed (S200). When the driving method is set to automatic driving, automatic driving processing is performed (S300).

C. Driving learning process:
FIG. 4 shows a flowchart of a driving operation learning process that is performed when the driving method of the vehicle 1 is set to “manual driving”. In the driving operation learning process, first, the driving environment specifying unit 130 of the driving control device 100 specifies the driving environment of the vehicle 1 (S202).

  Here, as a preparation for explaining a method of specifying the driving environment by the driving environment specifying unit 130 of the present embodiment, first, a basic concept will be described. Generally, when learning the driving operation by the driver, it is necessary to specify the driving environment of the vehicle 1. The driving environment of the vehicle 1 is composed of various items (hereinafter referred to as “environment items”) such as the weather, the width of the driving lane, and the presence or absence of surrounding vehicles. As a result, it is considered that one driving environment is specified.

For the sake of simplicity, it is assumed that the driving environment is composed of three environmental items (environment items A, B, and C). In addition, the value of each environmental item (hereinafter referred to as “environment item value”) does not change continuously, but usually takes several values. For example, if the environmental item A is “number of lanes”, it is assumed that the environmental item value has four values such as “1 lane”, “2 lanes”, “3 lanes”, and “4 lanes”. Can do. The same applies to the environmental items B and C.
If the three environmental items A, B, and C are made to correspond to the three axes of the three-dimensional space, as shown in FIG. 5, all the driving environments can be represented by coordinate points set in the three-dimensional space. it can. For example, it is possible to assume an operating environment 1 in which the value of environmental item A is “a4”, the value of environmental item B is “b3”, and the value of environmental item C is “c2”. Alternatively, it is possible to assume an operating environment 2 in which the value of the environmental item A is “a2”, the value of the environmental item B is “b4”, and the value of the environmental item C is “c4”.

  In this way, a plurality of environment items constituting the driving environment are set in advance, and the environment item value that is the value of the item is specified for each environment item. Thereby, it becomes possible to identify no matter how complicated the driving environment is by combining the environmental item values of each environmental item. Further, if the environment item value of each environment item is selected from a plurality of preset candidate values, all driving environments are limited to an N-dimensional space set in an N-dimensional space corresponding to the number of environment items. It can be expressed using individual coordinate points. For this reason, since driving environments that are very similar can be handled as the same operating environment, innumerable operating environments can be appropriately classified.

FIG. 6 illustrates seventeen environmental items constituting the driving environment in the driving control apparatus 100 of the present embodiment. These environmental items are roughly classified into “driving area”, “visibility state”, “road”, “peripheral vehicle”, “peripheral obstacle”, and “own vehicle”. Among these, the “visibility state” is classified into “brightness” and “weather”. “Road” is classified into “number of lanes”, “width of driving lane”, “road shape”, “road gradient”, “road surface type”, “road surface condition”, and “signal display contents”. The “Nearby vehicles” are classified into “number of surrounding vehicles”, “front vehicle speed”, and “front vehicle acceleration”, and “own vehicle” is classified as “travel state”, “travel direction”, “travel lane”. Classified into "types". By using such environmental items, it is possible to represent various operating environments by combining the environmental item values of the respective environmental items. In this embodiment, in order to specify the driving environment, the above-described 17 environmental items are used. However, all of these environmental items may not be used, and at least two are used. Good.
Below, the environmental item value which these environmental items can take is demonstrated. Note that these environmental items and environmental item values are stored in advance in the databases 150a to 150c.

FIG. 7 shows environmental item values for the “driving area”. The environmental values of the “driving area” include “residential area”, “urban area”, “highway”, “mountainous part”, “countryside”, “parking lot”, “entry into the site of the building (or "Exit) point" can be illustrated. These environmental item values may be further classified, for example, urban areas may be classified into “main roads” and “other than main roads”. Alternatively, the highway may be classified into “highway main line”, “merge point to main line”, and “branch point from main line”.
Which environmental item value the “driving area” corresponds to is determined based on map information that the car navigation system 220 has and a front image acquired by the front camera 212a. Alternatively, the driver may set the environmental item value of “driving area”. For example, the driver sets on the operation screen displayed on the display 500.

  FIG. 8 shows the environmental item values for the “visibility state”. Among these, FIG. 8A shows environment item values for “brightness”. Examples of the “brightness” environment item value include “bright”, “dim”, and “dark”. Which environmental item value “brightness” corresponds to is determined based on a detection result of an illuminance sensor (not shown). Alternatively, the brightness may be estimated from the time by detecting the time with a clock built in the operation control apparatus 100.

  FIG. 8B shows environmental item values for “weather”. Examples of the environmental item value of “weather” include “sunny”, “cloudy”, “rain”, “snow”, and “fog”. Which environmental item value “weather” corresponds to is determined based on, for example, a setting state of a wiper switch (not shown).

  FIG. 9 shows environmental item values for “road”. Among these, FIG. 9A shows environmental item values for “number of lanes”. Examples of the environmental item value of “number of lanes” include “one-way one lane”, “one-way two lanes”, “one-way three lanes or more”, and “one lane in which an oncoming vehicle also travels”. Which environmental item value the “number of lanes” corresponds to is determined based on, for example, map information of the car navigation system 220 or a front image acquired by the front camera 212a.

  FIG. 9B shows the environmental item values for “traveling lane width”. Examples of the environmental item value of “width of travel lane” include “narrow”, “normal”, and “wide”. These environmental item values may be further classified by combining with “width of shoulder (narrow or wide)”, “existence of sidewalk”, and “existence of bicycle area”. Which environmental item value corresponds to “the width of the traveling lane” is determined based on, for example, map information of the car navigation system 220 or a front image acquired by the front camera 212a.

  FIG. 9C shows environmental item values for “road shape”. Environmental item values of “Road shape” include “Straight road”, “Corner with a large R”, “Corner with a small R”, “Hairpin corner”, “Cross road”, “T-shaped road”, “Trail crossing” It can be illustrated. Which environmental item value the “road shape” corresponds to is determined based on, for example, map information of the car navigation system 220 or a front image acquired by the front camera 212a.

  FIG. 9D shows the environmental item values for “road gradient”. Examples of the environmental item value of “road gradient” include “no gradient (flat)”, “steep climb”, “slow climb”, “steep descent”, and “slow descent”. Which environmental item value corresponds to “road gradient” is determined based on, for example, map information of the car navigation system 220 or an inclination angle sensor (not shown).

  FIG. 10 shows the environmental item values for “road” following FIG. 9. Among these, FIG. 10A shows environmental item values for “type of road surface”. Examples of the environmental item value of “type of road surface” include “paved road”, “step road”, “gravel road”, and “elevated joint”. Which environmental item value the “road surface type” corresponds to is determined based on, for example, map information of the car navigation system 220 or information on vibration of the vehicle 1 obtained from an acceleration sensor (not shown).

  FIG. 10B shows environmental item values for “road surface condition”. Examples of the environmental item value of “road surface condition” include “state wet by rain”, “snow cover”, and “freezing”. Which environmental item value the “road surface condition” corresponds to is determined based on, for example, an operating state of a so-called ABS (not shown) provided in the vehicle 1 or an outside air temperature detected by a thermometer (not shown). The

  FIG. 10C shows environment item values for “display contents of traffic light”. The environmental item values for “Signal display contents” include “Permitted to pass (display is blue)”, “Stop. However, if safe stop is not possible, pass is permitted (display is yellow)”, “Unable to pass (display is red) ) ”,“ No traffic light ”. Which environmental item value corresponds to “display content of traffic light” is determined based on information regarding the display content of traffic light obtained by road-to-vehicle communication.

  FIG. 11 shows environmental item values for “peripheral vehicles”. Among these, FIG. 11A shows environmental item values for “the number of surrounding vehicles”. As the environmental item value of “the number of surrounding vehicles”, “0”, “less than N (free)”, “N or more (congested)” can be exemplified. Which environmental item value the “number of surrounding vehicles” corresponds to is determined based on, for example, images around the vehicle 1 acquired by the front camera 212a and the rear camera 212b.

  FIG. 11B shows environmental item values for “front vehicle speed”. Examples of the environmental item value of “front vehicle speed” include “faster than own vehicle”, “substantially the same as own vehicle”, and “slower than own vehicle”. Which environmental item value the “front vehicle speed” corresponds to is determined based on, for example, the vehicle speed sensor 202 and the front laser radar 214a.

  FIG. 11C shows the environmental item values for “front vehicle acceleration”. Examples of the environmental item value of “front vehicle acceleration” include “sudden acceleration”, “slow acceleration”, “sudden deceleration”, and “slow deceleration”. Which environmental item value the “front vehicle acceleration” corresponds to is determined based on, for example, the vehicle speed sensor 202 and the front laser radar 214a.

  FIG. 12 shows environmental item values for “surrounding obstacles”. Examples of environmental item values for “Nearby obstacles” include “normal vehicles”, “light vehicles”, “medium-sized vehicles”, “large-sized vehicles”, “two-wheeled vehicles”, “pedestrians / bicycles”, and “parked vehicles”. can do. Which environmental item value the “peripheral obstacle” corresponds to is determined based on, for example, a front image acquired by the front camera 212a or information obtained from the front laser radar 214a.

  FIG. 13 shows environmental item values for “own vehicle”. Among these, FIG. 13A shows the environmental item values for the “running state”. The environmental item values for “running state” are “accelerate from stop state”, “run at constant speed”, “accelerate from constant speed run state”, “decelerate from constant speed run state”, “decelerate and stop” Can be illustrated. Which environmental item value the “running state” corresponds to is determined based on, for example, the vehicle speed sensor 202.

  FIG. 13B shows environmental item values for “traveling direction”. Examples of the environmental item value of “travel direction” include “straight direction”, “right direction”, and “left direction”. Which environmental item value the “traveling direction” corresponds to is determined based on the steering angle sensor 208.

  FIG. 13C shows environmental item values for “type of travel lane”. Examples of the environmental item value of “type of driving lane” include “right turn lane”, “left turn lane”, “straight forward / left turn lane”, “passing lane”, and “uphill lane”. Which environmental item value the “type of driving lane” corresponds to is determined based on, for example, a front image acquired by the front camera 212a or position information of the vehicle 1 obtained from a GPS receiver (not shown). Is done.

  In the above, the environmental item value of 17 environmental items which comprise a driving environment was demonstrated. In the process of S202 of FIG. 4, the driving environment specifying unit 130 detects the environmental item values using the various sensors and the like (200 to 230) described above with reference to FIG. Identify the driving environment. The operating environment specifying unit 130 of this embodiment corresponds to the “environment item value detecting unit” of the present invention.

After specifying the driving environment in this way, the driving operation detection unit 120 detects the driving operation by the driver using the vehicle information detection sensor 200 (S204). FIG. 14 illustrates a driving operation detected by the driving operation detection unit 120 of the present embodiment. As shown in the figure, the detected driving operations are roughly “accelerator / brake”, “steering”, “shift lever”, “speed / acceleration / deceleration”, “3-axis G”, “control mode”. are categorized.
Among these, “accelerator / brake”, “accelerator (throttle) opening”, “brake depression amount”, “accelerator and brake operation speed”, and “brake assist amount” are detected. For "steering", "steering angle", "steering angular velocity", "power steering assist amount", "winker switch timing", "lateral position in lane", and "type of driving lane" are detected Is done. Further, for “shift lever”, “shift-up (or shift-down) timing” and “mission gear setting status” are detected.
In the process of S204 in FIG. 4, the driving operation detection unit 120 detects these driving operations using the vehicle information detection sensor 200.

  When the driving operation by the driver is detected in this manner, the driving method determination unit 110 determines which of the control modes is set (eco, sports, or normal) (S206). Whether the control mode is set is determined based on the setting state of the control mode switching SW 404. Here, as described above, the “control mode” represents the driving mode of the vehicle 1 and is set to “at the time of automatic driving” so as to be the driving mode desired by the driver. In the operation control apparatus 100, it is possible to set this control mode also during “manual operation”. This is because the driving operation by the driver is learned in association with the travel mode when the driving operation is performed. That is, as a matter of course, if the driving mode of the vehicle 1 when the driver is driving changes, the content of the driving operation also changes. Therefore, when the control mode corresponding to (or close to) the driving mode during manual driving is set by the driver, the driving operation by the driver is learned in association with the set control mode, as will be described later.

  When the operation control device 100 determines which control mode is set using the operation method determination unit 110 (S206), the operation control device 100 selects the databases 150a to 150c corresponding to the set control mode (S208, S210, S212). Then, the driving operation learning unit 140 stores the driving operations detected in the process of S204 in the selected databases 150a to 150c in association with the driving environment specified in the process of S202 (S214).

FIG. 15 shows a state in which driving operations are accumulated in the databases 150a to 150c. As illustrated, various driving operations can be stored in the databases 150a to 150c in association with the driving environment. Thereby, the driving operation by the driver is learned. As described above, in the driving control device 100 of the present embodiment, by combining the environmental item values of a plurality of environmental items constituting the driving environment, It can be identified even in a complicated driving environment. For this reason, the driving operation performed under any driving environment can be learned by storing the driving operation by the driver in association with the driving environment thus specified.
Further, the process of storing the driving operation in association with the driving environment is performed on the databases 150a to 150c corresponding to the control mode (control mode set by the driver) when the driving operation is performed. For this reason, the driving operation by the driver can be learned in association with the driving environment and the control mode. Thereby, although the driving environment is the same, if the content of the driving operation is different due to the different driving modes, each driving operation can be learned separately. The learning result of the driving operation thus obtained is used for automatic driving of the vehicle 1.

D. Automatic operation processing:
FIG. 16 shows a flowchart of an automatic driving process for automatically driving the vehicle 1. This automatic driving process is a process performed when it is determined that the driving method is set to “automatic driving” in the main control process described above with reference to FIG. 3 (see S104 in FIG. 3). . Whether or not the automatic driving is set is determined by the driving method determination unit 110 based on the setting state of the manual / automatic switching SW 402.

  When the automatic driving process is started, the driving environment specifying unit 130 specifies the driving environment of the vehicle 1 (S302). The driving environment is specified by the same method as in the driving operation learning process. That is, the driving environment specifying unit 130 detects the environmental item value of each environmental item constituting the driving environment based on various information obtained from the various sensors (200 to 230) described above with reference to FIG. The driving environment is specified by combining the environmental item values. As a result, it is possible to specify even a complicated operating environment.

  Subsequently, the driving method determination unit 110 determines which of the eco, sports, and normal control modes is set (S304). Which control mode is set is determined based on the setting state of the control mode switching SW 404. Then, the databases 150a to 150c corresponding to the set control mode are selected (S306, S308, S310).

Subsequently, the actuator control module 161 reads the learning result of the driving operation corresponding to the driving environment identified in the above-described processing of S302 from the selected databases 150a to 150c (S312). Then, the vehicle 1 is automatically driven by controlling the operation of various actuators (traveling actuator 300) with reference to the learning result of the driving operation (S314). As described above, since the operation control apparatus 100 according to the present embodiment can specify any complicated operation environment, it is possible to learn the operation performed under any operation environment. For this reason, it is possible to suppress the driver's uncomfortable feeling with respect to the driving operation (acceleration method, traveling speed, etc.) during automatic driving.
Further, the driving operation by the driver is learned in association with the combination of the driving environment and the control mode (control mode set by the driver). For this reason, regardless of which control mode is selected, automatic driving can be performed without causing the driver to feel uncomfortable.

  Subsequently, the driving operation detection unit 120 detects the control state of the driving operation using the vehicle information detection sensor 200 (S316). Then, the control status display module 162 displays the control status of the driving operation on the display 500 (S318). For example, a state in which the speed of the vehicle 1 is kept constant or a state in which the steering angle of the steering wheel automatically changes when traveling on a curved road is displayed. As a result, the driver can grasp the control status of the driving operation by the automatic driving.

  Subsequently, the operation control apparatus 100 determines whether or not there has been an intervention operation by the driver (S320). Whether or not there has been an intervention operation by the driver is, for example, comparing various driving operations estimated by controlling the operation of the various actuators 300 with various driving operations actually detected by the driving operation detection unit 120. Judging by doing. As a result, when a difference equal to or greater than a predetermined threshold is detected between these driving operations, it is determined that there has been an intervention operation by the driver (S320: yes). In this case (when there is an intervention operation), it is considered that the driver felt uncomfortable with the driving operation by automatic driving, and the content of the intervention operation is a driving operation that the driver feels desirable. Conceivable. Therefore, the driving control apparatus 100 performs a driving operation learning process for the intervention operation (S200). In this way, since the contents of the intervention operation can be reflected in the next automatic driving, the appropriate driving operation can be efficiently learned while the vehicle 1 is driven in the automatic driving. If it is determined that there is no intervention operation by the driver (S320: no), the automatic driving process is terminated as it is.

  As described above, in the driving control apparatus 100 of the present embodiment, the driving operation by the driver is learned as follows, and the vehicle 1 is automatically driven using the learning result. First, a plurality of environmental items that specify the driving environment are set. Then, the driving environment is specified by the combination of the environmental item values detected for each environmental item, and the driving operation by the driver is learned in association with the specified driving environment. During automatic driving, the driving environment is identified by detecting the environmental item value, and the vehicle 1 is automatically driven with reference to the learning result of the driving operation in the identified driving environment.

  In this way, it is possible to identify even the most complex driving environment by appropriately selecting environmental items. Further, by selecting the really necessary environmental items, it is possible to efficiently identify a complicated driving environment. As a result, since the driver's driving operation can be learned for each driving environment, the driver's uncomfortable feeling with respect to the driving operation during automatic driving can be suppressed.

  In addition, when detecting an environmental item value for each environmental item, a selection is made from among preset candidate values for each environmental item. In this way, every driving environment can be represented as coordinate points set in an N-dimensional space corresponding to the number of environment items. For this reason, driving environments that are similar to each other are handled as the same driving environment, so that the driving environments can be appropriately classified, and the driving operation by the driver can be efficiently learned.

In the above description, the driving operation by the driver of “Vehicle 1” is described as being learned. That is, many of the environmental item values that specify the driving environment of the surrounding vehicle are common to the vehicle 1 such as the driving area, the visibility state, and the item value relating to the road. Therefore, based on such common environmental item values, environmental item values (traveling state, etc.) related to the surrounding vehicle obtained by inter-vehicle communication, and the driving operation of the driver of the surrounding vehicle, the surrounding vehicle The driving environment by the driver of the surrounding vehicle (peripheral vehicle driving operation) can be learned by specifying the driving environment.
The learning result of the surrounding vehicle driving operation obtained in this way can be used for driving support of the vehicle 1. For example, when it is desired to automatically drive the vehicle 1, a peripheral vehicle that performs a driving operation similar to the driving operation by the driver of the vehicle 1 is selected from the surrounding vehicles. As a result, automatic follow-up driving can be performed without causing the driver of the vehicle 1 to feel uncomfortable.

DESCRIPTION OF SYMBOLS 1 ... Vehicle, 100 ... Operation control apparatus,
110: Driving method determination unit, 120 ... Driving operation detection unit,
130 ... Driving environment identification unit, 140 ... Driving operation learning unit,
150 ... database, 160 ... automatic operation execution part,
161 ... Actuator control module,
162 ... Control status display module, 200 ... Vehicle information detection sensor,
210 ... Surrounding information detection sensor, 220 ... Car navigation system,
230 ... communication device, 300 ... traveling actuator,
400 ... operation panel, 500 ... display.

Claims (5)

Manual driving control for detecting driving operation by the driver to control the driving of the vehicle, and automatic driving control for detecting the surrounding situation of the vehicle and controlling the driving of the vehicle instead of the driving operation of the driver An operation control device (100) that can be switched and executed,
A driving operation detector (120) for detecting the driving operation by the driver;
An environment item value detection unit (130) that detects an environment item value that is a value of the item for a plurality of environment items that are set in advance as items for specifying a driving environment including the surrounding environment of the vehicle;
When the driving operation is detected by the driving operation detection unit during the manual driving control, the driving operation learning unit (140) learns the driving operation in association with the driving environment specified by the combination of the environmental item values. )When,
When the environmental item value is detected together with the ambient condition, the driving environment is identified based on the environmental item value, and the automatic driving control is performed by referring to the learning result of the driving operation associated with the driving environment. An operation control device comprising: an automatic operation execution unit (160) to execute. The operation control device according to claim 1,
The said environmental item value detection part is a detection part which selects the said environmental item value applicable from the some candidate value preset for every said environmental item. Operation control apparatus. The operation control device according to claim 1 or 2, wherein
The automatic operation execution unit is an execution unit that executes the automatic operation control in a control mode of a predetermined standard control mode or a specific control mode different from the standard control mode,
A control mode switching unit (404) for switching between the standard control mode and the specific control mode based on the driver's selection;
The driving operation learning unit is a learning unit that learns the driving operation in association with a combination of the switched control mode and the driving environment. The operation control device according to claim 3,
The driving operation learning unit is associated with the driving environment specified by the combination of the environmental item values when the driving operation is detected by the driving operation detection unit even during the automatic driving control. A driving control device that is a learning unit that learns the driving operation. The operation control device according to any one of claims 1 to 4, wherein
The environmental item value detection unit is configured to detect the environmental item values such as driving area, brightness, weather, number of lanes, driving lane width, road shape, road gradient, road surface type, road surface state, traffic light Display item, number of surrounding vehicles, forward vehicle speed, forward vehicle acceleration, peripheral obstacle, traveling state of own vehicle, traveling direction of own vehicle, and at least two environmental items of the type of traveling lane of own vehicle An operation control device that is a detection unit.

Images