JP2005004413A - Drive supporting device for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a drive supporting device for a vehicle with a sensing technology allowing a driver to be provided just enough information by sensing matching to a cognitive characteristics of the driver in consideration both of the cognitive characteristics and the sensing property of the driver. <P>SOLUTION: The supporting device comprises an environment detection means 13 having a function capable of detecting a body existing in an environment in which a vehicle travels with multi-step accuracy, and a function capable of arbitrarily setting a range conducing the detection of the body existing in the environment in which the vehicle travels; an information presenting means 14 for presenting information to the driver; a driving behavior detection means 10 for detecting the driving behavior of the driver; and a cognitive state determining means 11 for determining the cognitive state of the driver according to the detected driving behavior of the driver. The supporting system further has a detection property changing means 12 for changing a detection property indicated by at least either the detection accuracy or the detection area of the detection means 13 according to the determined cognitive state of the driver. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention belongs to the technical field of a vehicle driving support device having a function of recognizing the presence of a preceding vehicle, a pedestrian, or an obstacle existing on a traveling route of a vehicle and informing the driver.
[0002]
[Prior art]
As this type of prior art, there are known an object for speeding up recognition in obstacle recognition ahead of a vehicle and an object for improving the accuracy of recognition.
[0003]
For the purpose of speeding up, for example, in image processing for an image taken in front of a vehicle, first, a horizontal edge may be detected, and the horizontal edge may be an obstacle or a driving lane. There is known a vehicle obstacle detection device that detects a vertical edge only in an area where a horizontal edge is detected only when an edge is detected (see, for example, Patent Document 1). ).
[0004]
On the other hand, for the purpose of improving accuracy, for example, in the detection of a plurality of obstacles, a rule group that enables confirmation of continuity on a distance image based on the distance to the obstacle is constructed and recognized. In processing, there is known an obstacle detection apparatus that makes it possible to separate and identify a plurality of obstacles by making full use of these rules (see, for example, Patent Document 2).
[0005]
[Patent Document 1]
JP-A-10-97699
[Patent Document 2]
JP-A-7-129898
[0006]
[Problems to be solved by the invention]
However, when considering the situation of providing information such as the presence of obstacles to the actual driver, it is effective to provide information with high accuracy and detail, and conversely, even if accuracy and detail are low. In some cases, it is effective to provide information on the overall environmental condition in front of the vehicle. In other words, the level of accuracy and level of detail required varies depending on the driver's condition, so even if either speeding up of information provision or accuracy is improved, it will always be effective information for the driver. There was a problem that was not possible.
[0007]
The present invention has been made paying attention to the above-mentioned problem, and the purpose thereof is to perform sensing that matches the driver's cognitive characteristics in consideration of both the driver's cognitive characteristics and the sensing characteristics, An object of the present invention is to provide a driving support device for a vehicle equipped with a sensing technology that makes it possible to provide information to a driver who is not excessive or insufficient.
[0008]
[Means for Solving the Problems]
In order to achieve the above object, according to the present invention, a function capable of detecting an object existing in an environment in which the vehicle is traveling with multi-stage accuracy and a range for detecting an object present in the environment in which the vehicle is traveling are arbitrarily set. Environment detection means having at least one of settable functions, information presentation means for presenting information to the driver, driving behavior detection means for detecting the driving behavior of the driver, and detected driving behavior of the driver And a recognition state determination unit that determines a driver's recognition state according to the vehicle driving support device, the detection accuracy or detection range of the environment detection unit according to the determined driver recognition state A detection characteristic changing means for changing the detection characteristic represented by at least one of the above is provided.
[0009]
【The invention's effect】
Therefore, in the vehicle driving support apparatus of the present invention, the detection characteristic of the environment detection unit is changed according to the driver's recognition state determined by the recognition state determination unit, and therefore exists in the environment where the vehicle travels. When an object is detected, driving assistance capable of changing at least one of detection accuracy and detection range according to the driver's cognitive state is possible. In other words, in consideration of both the driver's cognitive characteristics and the sensing characteristics, by performing sensing that matches the driver's cognitive characteristics, it is possible to provide information to the driver without excess or deficiency.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment for realizing a vehicle driving support device of the present invention will be described with reference to the drawings.
FIG. 1 is a block configuration diagram illustrating a vehicle driving support apparatus according to an embodiment. The vehicle driving support apparatus includes a driving action detection unit 10, a recognition state determination unit 11, a detection characteristic change unit 12, an environment detection unit 13, and an information presentation unit 14.
[0011]
The driving behavior detection means 10 detects the driving behavior of the driver. The recognition state determination unit 11 determines the driver's recognition state according to the driving behavior of the driver detected by the driving behavior detection unit 10.
[0012]
The detection characteristic changing unit 12 changes the detection characteristic represented by at least one of the detection accuracy or the detection range of the environment detection unit 13 according to the driver's cognitive state determined by the cognitive state determination unit 11.
[0013]
The environment detection means 13 has a function capable of detecting an object existing in the environment in which the vehicle is traveling with multi-stage accuracy and a function capable of arbitrarily setting a range for detecting an object present in the environment in which the vehicle is traveling. And at least one of them. The information presenting means 14 displays the information detected by the environment detecting means 13 to the driver.
[0014]
Hereinafter, a vehicle driving support apparatus according to an embodiment will be described based on a first example and a second example.
[0015]
(First embodiment)
FIG. 2 is a block diagram showing the vehicle driving support apparatus of the first embodiment. The vehicle driving support device includes a driver recognition image processing device 20, a recognition state determination device 21, a detection characteristic changing device 22, an environment recognition image processing device 23, and an information presentation device 24.
[0016]
The driver recognition image processing device 20 corresponds to the driving behavior detection means 10 of the embodiment shown in FIG. The recognition state determination device 21 is the recognition state determination unit 11, the detection characteristic change device 22 is the detection characteristic change unit 12, the environment recognition image processing device 23 is the environment detection unit 13, and the information presentation device 24 is the information presentation unit. 14 respectively.
[0017]
The driver recognition image processing device 20 includes, for example, a camera device that captures an image of the driver, and a driver's image captured by the camera device, and the driver's line-of-sight direction and the horizontal rotation angle of the head. Is provided with an image processing circuit that calculates the above in real time. In particular, in this embodiment, the driver's line of sight is either in the driving environment or in the side mirror outside the passenger compartment, and in the interior of the passenger compartment, either in the room mirror or on the screen of the navigation device. Can be determined.
[0018]
The driver recognition image processing apparatus 20 uses an image processing circuit capable of calculating an angle at which the driver's head rotates in the left-right direction. Further, the image processing circuit is provided with line-of-sight direction holding means for holding the line-of-sight direction of the driver for a predetermined time, and an image processing circuit capable of detecting an area where the line of sight of the driver has been extended and an area where the line of sight has not been extended for a predetermined time Use the same device.
[0019]
The recognition state determination device 21 uses a device that determines the recognition state based on the driving behavior of the driver detected by the driver recognition image processing device 20. For example, the frequency at which the driver's line of sight detected by the driver recognition image processing device 20 is poured onto the screen of a room mirror, a side mirror, or a navigation device, or the driver's head detected by the driver recognition image processing device 20 Technology has been developed to determine whether the driver's cognitive state is a skill-based state or a knowledge-rule-based state based on the rotation angle in the horizontal direction of the head and the rotation frequency in the horizontal direction of the head Therefore, the one equipped with this technology is used.
[0020]
In the detection characteristic changing device 22, based on the driver's line-of-sight direction detected by the driver recognition image processing device 20 and the driver's recognition state determined by the recognition state determination device 21, environment recognition image processing described later. Parameters that determine how to recognize the environment in the device 23 are set. Note that there are three types of parameters: “detection range”, “scan interval”, and “similarity threshold”.
[0021]
In the environment recognition image processing device 23, based on the parameters “detection range”, “scan interval”, and “similarity threshold” set by the detection characteristic changing device 22, the vehicle travels safely in the environment where the vehicle travels. Detect objects that may interfere.
[0022]
The configuration of the environment recognition image processing apparatus 23 is shown in FIG. The environment recognition image processing device 23 includes a camera device 32 that captures a scene in front of the vehicle, and an image processing circuit 33 that extracts an edge by performing filter processing or the like on an image captured by the camera device 32. An image processed by the holding circuit 35 and the image processing circuit 33 holding the image features of an object such as an obstacle, a pedestrian, and a preceding vehicle that may interfere with the safe driving of the vehicle, and held in the holding circuit The image recognition circuit 34 selects a type that has a high possibility of matching by comparing various types of features such as an obstacle, a pedestrian, and a preceding vehicle. The parameters “detection range”, “scan interval”, and “similarity threshold” are input via the input circuit 31.
[0023]
The information presentation device 24 is a device that presents information to the driver. Specifically, a display device, a voice providing device, a vibration device provided in a seat, or the like can be used. In this embodiment, the driver recognizes the vehicle according to the type of the object present on the course of the vehicle recognized by the environment recognition image processing device 23 and the recognition state of the driver determined by the recognition state determination device 21. Provide information.
[0024]
Next, the operation will be described.
[Detection characteristic change control processing]
Hereinafter, the flow of the detection characteristic change control process performed by the detection characteristic change device 22 will be described with reference to the flowchart shown in FIG.
[0025]
In step S100, the recognition state of the driver determined by the recognition state determination device 21 is obtained. The driver's cognitive state is either a knowledge-rule-based state or a skill-based state. If the driver's recognition state is a knowledge-rule-based recognition state, the process proceeds to step S101. If the driver's recognition state is a skill-based recognition state, the process proceeds to step S105.
[0026]
In step S101, information on the driver's line-of-sight direction detected by the driver recognition image processing device 20 is acquired, and the process proceeds to step S102. Here, the information on the driver's line of sight is obtained by accumulating the direction in which the driver's line of sight has been poured for a predetermined time.
[0027]
In step S102, based on the information on the driver's line of sight obtained in step S101, on the contrary, a range where the line of sight is not poured is defined, and the process proceeds to step S103. The range where the line of sight is not poured is a “detection range” which is one of the parameters set by the detection characteristic changing device 22. The defined “detection range” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0028]
In step S103, “scan interval”, which is one of the parameters set by the detection characteristic changing device 22, is set to, for example, “per pixel”, and the process proceeds to step S104. The set “scan interval” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0029]
In step S104, “similarity threshold”, which is one of the parameters set by the detection characteristic changing device 22, is set to, for example, “0.7 or more”, and the process proceeds to step S108. The set “similarity threshold” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0030]
In step S105, “detection range”, which is one of the parameters set by the detection characteristic changing device 22, is set to “whole image”, for example, and the process proceeds to step S106. The set “detection range” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0031]
In step S106, “scan interval”, which is one of the parameters set by the detection characteristic changing device 22, is set to “every 5 pixels”, for example, and the process proceeds to step S107. The set “scan interval” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0032]
In step S107, “similarity threshold” that is one of the parameters set in the detection characteristic changing device 22 is set to “0 or more”, for example, and the process proceeds to step S108. The set “similarity threshold” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0033]
In step S108, the parameters “detection range”, “scan interval”, and “similarity threshold” held in the memory device provided inside the detection characteristic changing device 22 are output to the environment recognition image processing device 23, and This control is terminated.
[0034]
In this flowchart, when the driver's recognition state is a knowledge-rule-based state, the flow proceeds from step S100 to step S101 to step S102 to step S103 to step S104 to step S108. That is, a range in which the driver's line of sight is not poured is defined as a detection range in step S102, a scanning interval is set for each pixel in step S103, and a similarity threshold is set to 0.7 or more in step S104. .
[0035]
On the other hand, when the driver's recognition state is a skill veil state, the flow proceeds from step S100 to step S105 to step S106 to step S107 to step S108. That is, the entire image is defined as the detection range in step S105, the scanning interval is set for every five pixels in step S106, and 0 or more is set as the similarity threshold in step S107.
[0036]
[Environment recognition image control processing]
Next, the flow of the environment recognition image control process performed by the environment recognition image processing apparatus 23 will be described based on the flowchart shown in FIG.
[0037]
In step S110, the camera device 32 captures a scene in front of the vehicle, and the process proceeds to step S111.
[0038]
In step S111, the detection range a set by the detection characteristic changing device 22 is input via the input circuit 31, and the process proceeds to step S112.
[0039]
In step S112, the image processing circuit 33 sets, based on the detection range a input in step S111, a range in which the image captured in step S110 is tried to detect an object that may interfere with safe driving of the vehicle. Then, the process proceeds to step S113. That is, when the detection range a is “entire image”, the entire image captured in step S110 is a range where the detection of the object is attempted (see FIG. 6), and the driver's line of sight is the detection range a. If a range that cannot be poured is defined, it is converted into a coordinate value on the image photographed in step S110, and the coordinate value is deleted from the range where the object detection is attempted (see FIG. 7). .
[0040]
In step S113, the scanning interval b set by the detection characteristic changing device 22 is input via the input circuit 31, and the process proceeds to step S114.
[0041]
In step S114, the image processing circuit 33 performs image processing on the range on the image where the object set in step S112 is attempted to be detected, and the process proceeds to step S115. Note that the image processing in this embodiment performs a general edge extraction process for the purpose of extracting a boundary between an object and a background such as a road. Normally, edge extraction scans for each pixel in the horizontal direction and the vertical direction of an image, and a pixel having a large luminance difference between adjacent pixels is extracted as an edge. In this embodiment, the scanning for each pixel is determined by the scanning interval b input in step S113. For example, if the scanning interval b = 5, scanning is performed every 5 pixels.
[0042]
In step S115, the similarity threshold value rt set by the detection characteristic changing device 22 is input via the input circuit 31, and the process proceeds to step S116.
[0043]
In step S116, in the image recognition circuit 34, the type of edge extracted in step S114 is based on which type, that is, whether it is a pedestrian, a preceding vehicle, or an obstacle. And the process proceeds to step S117.
[0044]
This identification process is performed by referring to the set of edges extracted in step S114 with reference to various features held in the holding circuit 35. Such a process includes a known neural network and a hidden Markov. A model can be used. In the image recognition processing using these neural networks and hidden Markov models, a set of edges is referred to various types of features, and which of the respective types best matches is represented stochastically. For example, when {pedestrian p1, preceding vehicle p2, obstacle p3} = {0.1, 0.7, 0.2} is calculated, the probability that the type is a pedestrian is 0.1. This means that a certain probability is 0.7, and the possibility of being an obstacle is 0.2. In this embodiment, the type showing the value with the highest probability is set as the type of identification. That is, the type of identification in the previous example is “preceding vehicle” indicating p2 = 0.7. Further, in this embodiment, the probability p2 is set as the similarity r (= 0.7).
[0045]
In step S117, the image recognition circuit 34 compares the similarity threshold rt input in step S115 with the similarity r of the object identified in step S116. If the similarity r is greater than or equal to the similarity threshold rt, the processing of the environment recognition image processing device 23 is terminated. On the other hand, when the similarity r is smaller than the similarity threshold rt, the process of step S116 is performed again. At this time, the identification of the type having the lowest probability is not performed in the previous processing of step S116. That is, in the example of step S116 described above, since the probability is “pedestrian”, the identification of “pedestrian” is omitted in the next processing of S116.
[0046]
[Change of sensing characteristics according to driver's cognitive characteristics]
Various vehicle driving support devices have been invented for the purpose of realizing safe vehicle driving for the driver. A so-called approach warning that informs the driver of the presence of a preceding vehicle, a pedestrian, or an obstacle existing on the traveling route of the vehicle is also one of the vehicle driving support devices. These vehicle driving support apparatuses are usually provided with sensors for detecting the state of the environment in which the vehicle travels. For example, a sensing device that includes a camera, captures a scene in front of the vehicle as an image, and performs image processing on the image to check whether there is an object that hinders safe driving of the vehicle, A sensing device, etc., has been invented and developed that confirms whether there is an object that hinders safe driving of a vehicle by transmitting laser light to the front and detecting reflected light from the object.
[0047]
There are two directions for the invention and development in this field. That is, detection of the presence of an object that hinders safe driving of the vehicle and identification of the type of the object, that is, whether the object is a preceding vehicle, a pedestrian, or an obstacle. It is “how fast” and “how accurately it is detected”.
[0048]
The background of the two directions in the present invention and development is that the amount of information contained in the state in front of the vehicle captured by the sensing device is enormous, and necessary information from the enormous information, that is, the existence confirmation of the object And the complexity of performing type recognition. Along with this complexity, the wider the region in which an object is attempted to be detected, the more accurate it is required to confirm the presence of an object, and the more detailed it is required in identifying the type of object, the cost, i.e. This is because the time and the capacity of the arithmetic processing unit required increase.
[0049]
However, when considering information provision situations such as the presence of obstacles to the actual driver, it is effective to provide information with high accuracy and detail, and conversely even if accuracy and detail are low. It is assumed that it is effective to provide information on the overall environmental condition in front of the vehicle.
[0050]
Driver operation can be broadly divided into the following two states.
(1) A skill-based driving state in which driving is performed while processing information mechanically and reflexively using skills acquired through experience for a given situation.
(2) Knowledge based on experience and learning. Driving knowledge while applying rules and rules-Rule-based driving state.
[0051]
Considering from the viewpoint of collecting driver information, in the case of a skill-based driving state, the driver achieves the operation of driving by grasping the overall state of the environment widely and shallowly. On the other hand, in the case of the knowledge-rule-based driving state, the operation of driving is realized by grasping the local state of the environment narrowly and deeply.
[0052]
Even in the case of providing information for preventing an accident, the information to be provided for each driving state is naturally different. For example, in the case of a skill-based driving state, there is a tendency that, for example, the presence confirmation of an obstacle is unconsciously removed from the routine of consciousness, or the countermeasure for the obstacle is mistakenly unconsciously. Prevention of accidents in such cases is accomplished by simply suggesting to the driver that something important is out of the routine of consciousness. In order to realize this, it is necessary for the driving support device to grasp the environment regarding the entire traveling environment in front of the vehicle.
[0053]
On the other hand, in the case of a knowledge-rule-based driving state, there is a tendency that it is not known which knowledge or rule should be used for the situation or that wrong knowledge or rule is driven. Prevention of an accident in such a case is achieved by giving correct information to the driver, instructing the location of the information, and giving a priority order and a countermeasure method. In order to realize this, it is necessary for the driving line-of-sight device to accurately and accurately grasp the environment state.
[0054]
This suggests that even if the accuracy and level of detail are improved by the conventional technology, these technologies cannot be effectively used unless the driver is in a state of seeking accurate and detailed information. On the contrary, the detected information suggests that even if the accuracy and the level of detail are somewhat impaired, it can be effective information depending on the driver's condition.
[0055]
On the other hand, in the vehicle driving support device of the present embodiment, the environment recognition image processing device via the detection characteristic changing device 22 according to the determination or detection result of the recognition state determination device 21 and the driver recognition image processing device 20. The object recognition characteristic using the image recognition in 23 is changed.
[0056]
That is, when the driver's cognitive state is a skill-based state, it is possible to detect an object over a wide range even if it is rough and low-classification type identification for an image obtained by shooting a scene in front of the vehicle. Further, it becomes possible to provide information based on the result.
[0057]
On the other hand, when the driver's recognition state is a knowledge-rule-based state, the object detection range is limited to a range that does not reach the driver's line-of-sight direction, but is detailed and has high similarity. Identification can be performed, and further information can be provided based on the result.
[0058]
In this way, by changing the sensing characteristic according to the driver's cognitive characteristic, it is possible to provide information that matches the driver's cognitive characteristic and that is comfortable for the driver.
[0059]
Next, the effect will be described.
In the vehicle driving support device of the first embodiment, the following effects can be obtained.
[0060]
(1) The detection characteristic changing device 22 changes the detection range and the scanning interval of the environment recognition image processing device 23 according to the determined recognition state, and therefore detects an object present in the environment in which the vehicle travels. Can support driving in which the detection range and accuracy can be changed according to the driver's cognitive state (corresponding to claim 1).
[0061]
(2) When the driver's recognition state is a skill veil state, the detection characteristic changing device 22 defines the entire image as a detection range and sets the similarity threshold to 0 or more with a scanning interval of every 5 pixels. In the skill-based state in which it is effective to give the driver summary information in a wide area to set, it is possible to provide driving support that presents the results of object detection with rough accuracy for a wide range of environments. (Corresponding to claim 2).
[0062]
(3) When the driver's cognitive state is a knowledge-rule-based state, the detection characteristic changing device 22 defines a range in which the driver's line of sight cannot be poured as a detection range, and sets the scanning interval to 1 In order to set the similarity threshold to 0.7 or more for each pixel, it is effective to give the driver specific information in a narrow area. In a rule-based state, detailed accuracy for a narrow range environment The driving support that presents the result of the object detection is possible (corresponding to claim 3).
[0063]
(4) The driver recognition image processing device 20 includes a gaze direction detection unit that detects the driver's gaze direction, and the detection characteristic change device 22 detects the environment recognition image processing device 23 using the detection characteristic change device 22. The change to the narrow range of the characteristic is a range in which the presence of the object in the driver recognition image processing device 20 is detected based on the driver's line-of-sight direction detected by the line-of-sight direction detection unit and the driver's line of sight does not reach. On the other hand, the change of the detection characteristic to a wide range was detected by the gaze direction detection means for detecting the presence of the object in the driver recognition image processing device 20. Regardless of the driver's line-of-sight direction, the environment recognition image processing device 23 is changed so that it can be applied to the entire range in which the object can be detected. When the detection characteristic of the image processing device 23 is changed to detection of a narrow range, driving assistance can be performed by trying to detect an object around a range where the driver is not paying attention. When the detection is changed to a wide range, it is possible to assist driving by trying to detect an object in the entire traveling environment regardless of the driver's attention (corresponding to claim 4).
[0064]
(5) The environment recognition image processing device 23 is a camera device 32 that captures an environment in front of the vehicle, and an image that scans the captured image pixel by pixel and extracts an object that exists in the environment by extracting an edge. A change in the detection characteristics of the environment recognition image processing apparatus 23 to rough accuracy by the detection characteristic changing apparatus 22 is an image of the driving environment captured by the camera apparatus 32 by performing image processing in the image processing circuit 33. On the other hand, the change of the detection characteristics to detailed accuracy is achieved by setting the scanning interval for the image of the driving environment captured by the camera device 32 for each pixel. Because it was decided to realize, object detection in the environment in front of the vehicle using image processing is possible with different accuracy, based on different object detection accuracy Driving support is possible (corresponding to claim 5).
[0065]
(6) The environment recognition image processing device 23 holds the characteristics of a plurality of objects that may exist in the environment in which the vehicle is traveling for each type of object, and a plurality of types held in the holding circuit 35. And the image processed by the image processing circuit 33, the higher the degree of similarity, the higher the degree of similarity, and the type with the highest degree of similarity is assigned to the environment recognition image processing device 23. An image recognition circuit 34 that performs an image recognition process that identifies the type of the object detected by the image recognition circuit 34, and the detection characteristic changing device 22 changes the detection characteristics of the environment recognition image processing device 23 to rough accuracy. If the image recognition processing in 34 is equal to or greater than the similarity threshold 0, it is realized by specifying the type of the object detected by the environment recognition image processing device 23. On the other hand, the change of the detection characteristics to detailed accuracy is as follows. Image recognition In the image recognition processing in the recognition circuit 34, it is realized by repeatedly performing the similarity threshold until the similarity threshold becomes 0.7 or more. Therefore, the detection of an object in the environment in front of the vehicle using image processing is different. The object can be detected with high accuracy, and driving assistance based on different object detection accuracy is possible (corresponding to claim 7).
[0066]
(Second embodiment)
FIG. 8 is a block diagram showing the vehicle driving support apparatus of the second embodiment. Since the basic configuration of the second embodiment is substantially the same as the block configuration diagram of the first embodiment shown in FIG. 2, the same components are denoted by the same reference numerals, and redundant description is omitted.
[0067]
The vehicle driving support device includes a driver recognition image processing device 20, a recognition state determination device 21, a detection characteristic change device 40, a radar device 41, and an information presentation device 24. The radar apparatus 41 corresponds to the environment detection means 13.
[0068]
In the detection characteristic changing device 40, the environment recognition in the radar device 41 is performed based on the driver's line-of-sight direction detected by the driver recognition image processing device 20 and the driver's recognition state determined by the recognition state determination device 21. Set the parameters that determine the method. There are two types of parameters, “detection direction” and “resolution”.
[0069]
The radar device 41 detects an object that may hinder the safe traveling of the vehicle in the environment where the vehicle travels based on the parameters “detection direction” and “resolution” set by the detection characteristic changing device 40. . The configuration of the radar device 41 is shown in FIG. The radar device 41 is of a type that emits laser light, and includes a transmitter 52 that transmits laser light, a receiver 53 that receives reflection of the transmitted laser light, a timer 54, and a motor that rotates the transmitter in the left-right direction. 55, a motor control circuit 56 for controlling the motor 55.
[0070]
Since the transmitting unit 52 can transmit laser light only in one direction, it is necessary to rotate the transmitting unit 52 in the left-right direction by the motor 55 in order to detect an object existing in front of the traveling direction of the vehicle. is there. The parameters “detection direction” and “resolution” are input via the input circuit 51.
[0071]
Next, the operation will be described.
[Detection characteristic change control processing]
Hereinafter, the flow of the detection characteristic change control process performed by the detection characteristic change device 40 will be described based on the flowchart shown in FIG.
[0072]
In step S200, the recognition state of the driver determined by the recognition state determination device 21 is obtained. The driver's cognitive state is either a knowledge-rule-based state or a skill-based state. If it is a knowledge-rule-based recognition state, the process proceeds to step S201, and if it is a skill-based recognition state, the process proceeds to step S204.
[0073]
In step S201, information on the driver's line-of-sight direction detected by the driver recognition image processing apparatus 20 is acquired, and the process proceeds to step S202. Here, the information on the driver's line of sight is obtained by accumulating the direction in which the driver's line of sight has been poured for a predetermined time.
[0074]
In step S202, the direction in which the line of sight is not poured is defined on the basis of the information on the driver's line of sight obtained in step S201, and the process proceeds to step S203. The direction in which the line of sight is not poured is a “detection direction” which is one of the parameters set by the detection characteristic changing device 22. The defined “detection direction” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0075]
In step S203, “resolution” which is one of the parameters set by the detection characteristic changing device 22 is set to 0.1 °, for example, and the process proceeds to step S206. The set “resolution” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0076]
In step S204, “detection direction”, which is one of the parameters set by the detection characteristic changing device 22, is set to “all directions”, for example, and the process proceeds to step S205. The set “detection direction” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0077]
In step S205, “resolution” which is one of the parameters set by the detection characteristic changing device 22 is set to 0.3 °, for example, and the process proceeds to step S206. The set “resolution” is held in, for example, a memory device provided inside the detection characteristic changing device 22.
[0078]
In step S206, the parameters “detection direction” and “resolution” held in the memory device provided inside the detection characteristic changing device 22 are output to the radar device 41, and this control is terminated.
[0079]
In this flowchart, when the driver's recognition state is a knowledge-rule-based state, the flow proceeds from step S200 to step S201 to step S202 to step S203 to step S206. That is, a range in which the driver's line of sight is not poured is defined as the detection direction in step S202, and the resolution is set to 0.1 ° in step S203.
[0080]
On the other hand, when the driver's recognition state is a skill veil state, the process proceeds from step S200 to step S204 to step S205 to step S206. That is, all directions are defined as detection directions in step S204, and the resolution is set to 0.3 ° in step S205.
[0081]
[Object detection control processing]
Next, the flow of the object detection control process performed by the radar apparatus 41 will be described based on the flowchart shown in FIG.
[0082]
In step S210, the “detection direction” and “resolution” set by the detection characteristic changing device 40 are input via the input circuit 51, and the process proceeds to step S211.
[0083]
In step S211, the motor control circuit 56 sets a direction in which the radar device 41 tries to detect an object that may interfere with safe driving of the vehicle based on the “detection direction” input in step S210. The process proceeds to S212.
[0084]
That is, when the detection direction is “omnidirectional”, all directions that can be operated in design are set as the detection direction in the transmission unit 52 and the motor 55 that rotates the transmission unit 52 (FIG. 12). reference). Further, when a direction in which the driver's line of sight is not poured is defined as the detection direction, the direction within a predetermined angle (± α) is set as the detection direction with the direction as the center ( (See FIG. 13).
[0085]
In step S212, the motor control circuit 56 determines a unit for controlling the rotation of the transmitter 52 in the left-right direction, and the process proceeds to step S213. This unit determination is based on the “resolution” input in step S210. In other words, when the “resolution” is 0.1 °, the motor control circuit 56 is set so that the rotation of the motor 55 is controlled in units of 0.1 °, and the “resolution” is 0.3. If it is, the motor control circuit 56 is set so that the rotation of the motor 55 is controlled in units of 0.3 °.
[0086]
In step S213, the motor control circuit 56 controls the motor 55 based on the “detection direction” set in step S211 and the “resolution” set in step S212. During that time, the transmitter 52 transmits laser light. Then, the laser beam reflected and returned by the object or the like is received by the receiving unit 53, so that the object is detected, and this control is finished. Further, the distance from the transmission of the laser beam to the reception thereof is measured by the timer 54, thereby measuring the distance to the object.
[0087]
Next, the effect will be described.
In the vehicle driving support device of the second embodiment, the following effects can be obtained.
[0088]
(7) The change of the detection characteristic of the radar device 41 to the rough accuracy by the detection characteristic changing device 40 is realized by setting the resolution of the radar emitted by the radar device 41 to 0.3 °. The change to the detailed accuracy of the detection characteristics is realized by setting the resolution of the radar device 41 to 0.1 °. Therefore, different accuracy is detected in the detection of an object in the environment in front of the vehicle using the radar device 41. Thus, it becomes possible to detect an object, and driving assistance based on different object detection accuracy is possible (corresponding to claim 6).
[Brief description of the drawings]
FIG. 1 is a block configuration diagram showing a vehicle driving support apparatus according to an embodiment.
FIG. 2 is a block configuration diagram showing the vehicle driving support device of the first embodiment.
FIG. 3 is a block diagram illustrating an environment recognition image processing apparatus.
FIG. 4 is a flowchart showing a flow of detection characteristic change control processing performed by the detection characteristic change device.
FIG. 5 is a flowchart showing a flow of environment recognition image control processing performed by the environment recognition image processing apparatus.
FIG. 6 is an explanatory diagram of an object detection range in the environment recognition image processing apparatus.
FIG. 7 is an explanatory diagram of an object detection range in the environment recognition image processing apparatus.
FIG. 8 is a block diagram showing a vehicle driving support apparatus according to a second embodiment.
FIG. 9 is a block diagram showing a radar apparatus.
FIG. 10 is a flowchart showing a flow of detection characteristic change control processing performed by the detection characteristic change device.
FIG. 11 is a flowchart showing a flow of object detection control processing performed by the radar apparatus.
FIG. 12 is an explanatory diagram of an object detection range in a radar apparatus.
FIG. 13 is an explanatory diagram of an object detection range in a radar apparatus.
[Explanation of symbols]
10 Driving behavior detection means
11 Cognitive state determination means
12 Detection characteristic changing means
13 Environment detection means
14 Information presentation means
20 Image processing device for driver recognition
21 Cognitive state determination device
22 Detection characteristic changing device
23 Image processing device for environment recognition
24 Information presentation device
31 Input circuit
32 Camera device
33 Image processing circuit
34 Image recognition circuit
35 Holding circuit
40 Detection characteristic changing device
41 Radar equipment
51 Input circuit
52 Transmitter
53 Receiver
54 Timer
55 motor
56 Motor control circuit

Claims (7)

Provided with at least one of a function capable of detecting an object existing in the environment where the vehicle travels with multi-stage accuracy and a function capable of arbitrarily setting a range for detecting an object present in the environment where the vehicle travels Environmental detection means;
Information presentation means for presenting information to the driver;
Driving behavior detecting means for detecting the driving behavior of the driver;
According to the detected driving behavior of the driver, a recognition state determination means for determining the driver's recognition state;
In a vehicle driving support device comprising:
Vehicle driving support characterized by comprising detection characteristic changing means for changing a detection characteristic represented by at least one of detection accuracy or detection range of the environment detection means in accordance with the determined recognition state of the driver apparatus. The vehicle driving support device according to claim 1,
The cognitive state determination means is a means for determining whether the driver's cognitive state is a skill-based state or a knowledge-rule-based state,
When the detection characteristic changing means determines that the driver's cognitive state is a skill-based state, the detection characteristic changing unit is configured to detect the presence of an object with rough accuracy over a wide range of environments. A driving support apparatus for a vehicle, wherein the detection characteristic of the detection means is changed. The vehicle driving support device according to claim 2,
The detection characteristic changing means detects the presence of an object with a detailed accuracy in a narrow range of environment when it is determined that the driver's cognitive state is a knowledge-rule-based state. In addition, the vehicle driving support device is characterized in that the detection characteristic of the environment detection means is changed. In the vehicle driving support device according to claim 2 or 3,
The driving behavior detection means includes a gaze direction detection means for detecting the gaze direction of the driver,
The detection characteristic changing means includes
When the detection characteristics of the environment detection means are set to detailed accuracy, the detection range of the object by the environment detection means is limited to a range where the driver's line of sight does not reach based on the detected driver's line-of-sight direction,
On the other hand, when the detection characteristic of the environment detection unit is set to rough accuracy, the detection range of the object by the environment detection unit is set to all the detectable ranges regardless of the detected driver's line-of-sight direction. A vehicle driving support device. The vehicle driving support device according to any one of claims 2 to 4,
The environment detection means includes a camera that captures an environment in front of the vehicle,
Image processing means for scanning an imaged image in units of pixels and extracting an object existing in the environment by extracting an edge;
With
The detection characteristic changing means includes
When the detection characteristics of the environment detection means are rough accuracy, the image processing in the image processing means is changed so that the scanning interval for the image of the driving environment imaged by the camera is rough,
On the other hand, when the detection characteristics of the environment detection means are set to detailed accuracy, the image processing in the image processing means is changed so as to be performed with a scan interval for the image of the traveling environment imaged by the camera. A vehicle driving support device. The vehicle driving support device according to any one of claims 2 to 5,
The environment detection means is a radar device that transmits radar at a predetermined angular interval in front of the vehicle and detects the presence or absence of an object present in the environment from the received radar.
The detection characteristic changing means includes
When making the detection characteristic of the environment detection means rough accuracy, the radar device emits a change in the angular interval of the radar.
On the other hand, when the detection characteristics of the environment detection means are set to detailed accuracy, the vehicle driving support device is changed so as to be implemented by narrowing the angular interval of the radar emitted by the radar device. In the vehicle driving support device according to claim 5 or 6,
The environment detection means includes
Type feature holding means for holding, for each type of object, characteristics of a plurality of objects that may exist in the environment in which the vehicle travels;
The feature of the plurality of types held is compared with the image processed by the image processing means, and the higher the similarity is given as the features are matched, the type having the highest similarity is determined as the environment. Image recognition means for performing an image recognition process that identifies the type of object detected by the detection means;
With
The detection characteristic changing means includes
When the detection characteristics of the environment detection means are rough accuracy, the image recognition process in the image recognition means is not performed, or the image recognition process in the image recognition means is detected by the environment detection means even if the similarity is low. Specify the type of object,
On the other hand, when the detection characteristic of the environment detection unit is set to detailed accuracy, it is specified as the type of the object detected by the environment detection unit only when high similarity is obtained in the image recognition processing in the image recognition unit. Alternatively, the vehicle driving support device is characterized in that the image recognition processing in the image recognition means is repeatedly performed until a high similarity is obtained.

Images