JP2019018712A - Drive assistance device - Google Patents

Abstract

To make a drive assistance device inexpensive and smaller while improving safety performance by highly accurately estimating the direction of another vehicle when another vehicle joins a traffic lane in which the own vehicle is traveling.SOLUTION: A drive assistance device comprises: first and second determination means (122) that recognize at least part of a wheel of a second vehicle included in an image photographed by a front camera (102), determine on the basis of movement of a front wheel whether the second vehicle is starting to move or not and, in a case it is determined that the second vehicle is starting to move, estimate a direction of the second vehicle with respect to an advancing direction of the own vehicle (1) on the basis of a first angle of a front wheel of the second vehicle and a second angle of a rear wheel of the second vehicle when both the front wheel and the rear wheel of the second vehicle can be recognized from the image, (i), and estimate the direction of the second vehicle on the basis of the first angle when only a front wheel of the second vehicle can be recognized from the image, (ii); and drive assisting means (123) that performs predetermined drive assistance for a driver of the own vehicle on the basis of the estimated direction of the second vehicle.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a driving support device.

  There may be other vehicles trying to join ahead of the traveling direction of the host vehicle. The driver of the host vehicle estimates the course of another vehicle and determines whether the course of the host vehicle is affected. And when it affects, it will be necessary to perform avoidance actions, such as deceleration or a lane change.

  In such a situation, when there are a plurality of lanes, it is common for other vehicles to join the nearest lane (for example, a traveling lane). However, other vehicles may try to merge directly into an adjacent lane (for example, an overtaking lane) beyond the nearest lane in accordance with the convenience of subsequent driving, such as turning right after merging.

  In addition, other vehicles may join the opposite lane instead of the lane in the same traveling direction as the own vehicle. In this case, the other vehicle crosses the lane in which the host vehicle travels while taking a course in the opposite direction to the host vehicle.

  The driver of the own vehicle must drive the own vehicle running at high speed, guess the course of the other vehicle from the movement of the other vehicle seen in the distance, and determine whether the course of the own vehicle is affected. Don't be. There is a need for a driving support device that can reduce the burden on the driver and improve the preventive safety effect.

  Patent Document 1 infers the trend of other vehicles traveling in adjacent lanes and recognizes a line constituting the lower part of the side surface of the other vehicle in a vehicle recognition device that recognizes the vehicle before breaking into its own lane. The angle that the line makes with the white line that separates the vehicle lane and the adjacent lane is regarded as the inclination angle φ of the other vehicle, and if the other vehicle enters the own vehicle lane when the inclination angle φ is a large predetermined angle It is described to judge.

  Further, in Patent Document 2, when the host vehicle is controlled so as to secure the target inter-vehicle distance from the preceding vehicle, the oncoming vehicle is prevented from entering the own lane by making a right turn or a U-turn. When the front wheel angle of the oncoming vehicle is acquired and the front wheel angle is greater than or equal to a predetermined threshold for determining whether the oncoming vehicle enters the own lane or not with the aim of ensuring a smooth traffic flow It is described that it is determined that the oncoming vehicle enters the own lane.

  Further, in Patent Document 2, the front wheel side surface and the rear wheel side surface are extracted from the image of the oncoming vehicle, and the ratio of the major axis to the minor axis of the front wheel side surface shown in an ellipse and the ellipse are shown. It describes that the front wheel angle is obtained based on the ratio of the major axis to the minor axis of the side surface of the rear wheel.

JP-A-9-223235 JP 2009-96361 A

  However, in the driving support technology described in Patent Document 1, the course of the other vehicle is predicted based only on the inclination angle φ of the other vehicle, and the angle of the wheel is not taken into consideration. It is difficult to determine whether it will merge into the nearest lane, merge into an adjacent lane, or merge into an opposite lane.

  Further, in the driving support technique described in Patent Document 2, another vehicle is specified based on radar information input from the radar, and one that is stopped or slowed down is detected. Therefore, for example, in a situation where another vehicle starts or accelerates and accelerates while accelerating from the outside of a road such as an exit of a parking lot beside the road or a side road leading to the road, It is not assumed that the course will be predicted. Furthermore, since the technique is premised on the use of a radar and a vehicle-mounted camera, the driving support device becomes expensive. Considering not only the cost but also a small vehicle such as a saddle-ride type vehicle, if the driving support device becomes large, the mounting becomes difficult.

  Furthermore, in Patent Document 2, although the angle of the front wheel is taken into consideration, it is assumed that an image of the side surface of the rear wheel is obtained. For example, when another vehicle is trying to join the lane from a parking lot on the side of the road, the rear wheels are blocked by obstacles such as hedges and guardrails, and the image of the side of the rear wheel is obtained from the captured image. There can be nothing. In such a case, the driving assistance technology described in Patent Document 2 cannot be handled.

  The present invention has been made in view of such a point, and predicts the course of another vehicle when the other vehicle joins the lane on which the host vehicle travels while improving the preventive safety performance. Another object is to provide a driving support device that is inexpensive and can be miniaturized.

  One aspect of the driving support device of the present invention recognizes at least a part of wheels of an other vehicle included in an image captured by the imaging unit and an imaging unit that captures the front of the host vehicle, and detects the front wheels of the other vehicle. When it is determined whether or not the other vehicle has started to move based on the movement, and it is determined that the other vehicle has started to move, (i) When both the front and rear wheels of the other vehicle can be recognized from the image Predicts the course of the other vehicle based on the first angle formed by the front wheels of the other vehicle and the second angle formed by the rear wheels of the other vehicle with respect to the traveling direction of the host vehicle, and (ii) ) When only the front wheels of the other vehicle can be recognized from the image, based on the other vehicle course prediction means for predicting the course of the other vehicle based on the first angle, and based on the predicted course of the other vehicle. , Predetermined to the driver of the own vehicle Characterized by comprising a driving assistance means for performing rolling support, a.

  According to the present invention, it is possible to predict the course of another vehicle when the other vehicle joins the lane on which the host vehicle travels with higher accuracy and improve the preventive safety performance, while reducing the driving support device at a low price and a small size. Can be

It is a schematic diagram which shows the road where the own vehicle provided with the driving assistance device which concerns on this Embodiment travels. It is a functional block diagram of the driving assistance device concerning this embodiment. It is a figure which shows the control flow of the driving assistance device which concerns on this Embodiment. It is a schematic diagram which shows the picked-up image image | photographed with the front camera of the driving assistance device which concerns on this Embodiment. It is a schematic diagram which shows the front-wheel wheel and rear-wheel wheel of the other vehicle extracted from the picked-up image shown in FIG. It is a schematic diagram which shows the state by which the front wheel of the other vehicle was cut in the left turn direction of the other vehicle in this Embodiment. It is a schematic diagram which shows the relationship between the rear wheel of another vehicle and 2nd angle (theta) 2 in this Embodiment. It is a mimetic diagram showing the state where the front wheel of other vehicles was cut in the right turn direction of other vehicles in this embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following, an example in which the driving support device according to the present invention is applied to a four-wheeled vehicle will be described, but the application target can be changed without being limited thereto. For example, the driving support device according to the present invention may be applied to other types of vehicles (for example, straddle-type vehicles such as motorcycles and motorcycles). In the following description, the saddle riding type vehicle will be described using a motorcycle as a representative example.

<Overview>
The present inventor has noted that when another vehicle enters a road including a travel lane on which the host vehicle travels, there are various possibilities in the course of the other vehicle.

  FIG. 1 is a schematic diagram showing a road on which the host vehicle is provided with the driving support apparatus according to the present embodiment. As shown in FIG. 1, the host vehicle 1 is traveling on the main road 2 from the right side to the left side shown in FIG. At this time, the lane in which the host vehicle 1 is traveling is referred to as a traveling lane 2a. The lane in which the other vehicle 3 traveling in the direction opposite to the traveling lane 2a is traveling is referred to as an oncoming lane 2b. The traveling lane 2a and the opposite lane 2b are separated by a white line 2c.

  On the main road 2, a sidewalk 4 is provided on the left side (the lower side shown in FIG. 1) of the traveling lane 2a. Curb stones 4 a and 4 b are respectively attached to both end portions of the sidewalk 4. The sidewalk 4 is provided with a cut 4c through which another vehicle 5 crosses the sidewalk 4 from a parking lot (not shown) and enters the main road 2, for example, and is provided with curbs 4a and 4b. It is not done. However, there are many steps (not shown) at the boundary between the main road 2 and the sidewalk 4.

  On the other hand, a curbstone 2d is attached to the main road 2 on the left side in the traveling direction of the opposite lane 2b (upper side shown in FIG. 1). Further, a side road 6 exists on the left side in the traveling direction of the oncoming lane 2 b with respect to the main road 2, and other vehicles 7 can enter the main road 2 from the side road 6.

  In the situation shown in FIG. 1, there are various possibilities for the path of the other vehicle 5. Specific examples thereof will be given below, but the invention is not limited thereto.

(1) When another vehicle 5 joins the travel lane 2a (part 1)
As indicated by an arrow α1 in FIG. 1, the other vehicle 5 may take a course (hereinafter referred to as “path α1”) so as to directly join the traveling lane 2a.

(2) When another vehicle 5 joins the travel lane 2a (part 2)
As indicated by an arrow α2 in FIG. 1, the other vehicle 5 crosses the travel lane 2a, enters the opposite lane 2b, and then joins the travel lane 2a (hereinafter referred to as “course α2”). May take.

(3) When the other vehicle 5 joins the opposite lane 2b As shown by the arrow β1 in FIG. 1, the other vehicle 5 crosses the traveling lane 2a, enters the opposite lane 2b, and joins the opposite lane 2b as it is. In some cases, a course (hereinafter referred to as “path β1”) may be taken. Similarly, as indicated by an arrow β2 in FIG. 1, the other vehicle 7 crosses the opposite lane 2b, enters the travel lane 2a, and joins the travel lane 2a as it is (hereinafter referred to as “course β2”). Sometimes).

  The present inventor predicts the course of the other vehicles 5 and 7 entering the main road 2 from the outside of the main road 2 such as the sidewalk 4 and the side road 6 having various possibilities, and If a driving support device capable of supporting the driving of the driver can be provided, it is considered that the burden on the driver of the host vehicle 1 can be reduced and the preventive safety performance can be improved, and the present invention has been completed.

  That is, the driving assistance apparatus according to the present embodiment recognizes at least a part of the wheels of the other vehicle included in the image captured by the imaging unit that captures the front of the host vehicle and the image captured by the imaging unit. It is determined whether or not the other vehicle has started to move based on the movement of the front wheel. When it is determined that the other vehicle has started to move, (i) both the front and rear wheels of the other vehicle are recognized from the image. When possible, predict the course of the other vehicle based on the first angle formed by the front wheels of the other vehicle and the second angle formed by the rear wheels of the other vehicle with respect to the traveling direction of the own vehicle; (Ii) When only the front wheels of the other vehicle can be recognized from the image, other vehicle course prediction means for predicting the course of the other vehicle based on the first angle, and the predicted course of the other vehicle Based on the driver of the vehicle Characterized in that it comprises a and a driving support means for performing a predetermined driving support by.

  An example of the driving support apparatus according to the present embodiment uses an imaging unit, but a radar is not essential. Thereby, a driving assistance device can be made inexpensive and can be easily downsized. As a result, it becomes easy to mount the driving support device on a small vehicle such as a saddle-ride type vehicle or a light vehicle.

  In addition, in the example of the driving support device according to the present embodiment, the case where the other vehicles 5 and 7 start to move is set as the monitoring target. The course is predicted, whether or not the other vehicles 5 and 7 obstruct the traveling direction of the own vehicle 1 (straight line X in FIG. 1) and whether or not avoidance is necessary. This is because the theme is to avoid contact.

  Further, in the example of the driving support device according to the present embodiment, the vehicle body of the other vehicle 5 is the host vehicle 1 in order to deal with various possibilities (1) to (3) of the course of the other vehicle 5 as described above. In addition to the angle formed with respect to the traveling direction X of the vehicle, the angle (first angle) formed by the front wheels of the other vehicle 5 with respect to the traveling direction X of the host vehicle 1 (θ1 ′ in FIGS. 6 and 8) is grasped. The course of the other vehicle 5 is predicted as accurately as possible. That is, the other vehicle course prediction means (i) can recognize both the front wheels and the rear wheels of the other vehicle 5 from the image taken by the imaging means with respect to the traveling direction X of the own vehicle 1. The course of the other vehicle 5 is predicted based on the first angle θ1 ′ formed by the front wheels and the second angle formed by the rear wheels of the other vehicle 5 (θ2 in FIGS. 1 and 7). As will be described later, the second angle θ2 formed by the rear wheel of the other vehicle 5 with respect to the traveling direction X of the host vehicle 1 is the angle formed by the vehicle body of the other vehicle 5 with respect to the traveling direction X of the host vehicle 1. It corresponds to.

  In the example of the driving support device according to the present embodiment, in particular, the other vehicle 5 entering the main road 2 from the outside of the main road 2 (a parking lot, a side road, etc.) is behind the obstacles such as hedges and guardrails. It is assumed that the wheel cannot be recognized by the driving support device. Even in such a case, the other vehicle course prediction means (ii) can recognize only the front wheels of the other vehicle 5 from the image captured by the imaging means, based on the first angle θ1 ′, Can be predicted.

  Furthermore, in the example of the driving support apparatus according to the present embodiment, the other vehicle course prediction unit calculates the width of the travel lane 2a obtained based on the image captured by the imaging unit when the course of the other vehicle 5 is predicted. It is preferable to consider the ratio with the wheel base of the other vehicle 5. Thereby, since the course of other vehicles 5 can be predicted more correctly, it is excellent.

  Moreover, in an example of the driving support apparatus according to the present embodiment, the driving support means evaluates whether or not the other vehicle 5 affects the future traveling of the host vehicle 1 based on a predetermined criterion, and the evaluation Although driving assistance is performed based on the result, it is preferable that the predetermined reference includes the one based on the predicted course of the other vehicle 5.

  Here, it is preferable that the predetermined reference further includes a reference based on a distance at which the host vehicle 1 can safely stop.

  Further, in the example of the driving support device according to the present embodiment, the other vehicle course prediction means sets the first angle θ1 ′ and the second angle θ2 and the front wheel and the rear wheel of the other vehicle 5 are the own vehicle 1. It is preferable to obtain the angle formed with respect to the traveling direction X.

  Moreover, in an example of the driving support apparatus according to the present embodiment, it is preferable that the driving support means includes a warning generating means and / or vibration generating means for alerting the driver of the host vehicle 1 by vibration.

  In the example of the driving support apparatus according to the present embodiment, the driving support means is preferably a deceleration control means for reducing the speed of the host vehicle 1.

  In the example of the driving support apparatus according to the present embodiment, the driving support means is preferably a steering control means for changing the steering angle of the steering of the host vehicle 1.

  Furthermore, the example of the driving support device according to the present embodiment further includes raindrop detection means and / or road surface condition confirmation means, and the driving support means responds to the output from the raindrop detection means and / or road surface condition confirmation means. Thus, it is preferable to change the timing of the driving assistance.

  Hereinafter, embodiments of the present invention will be described in more detail with reference to the drawings.

<Drive assist device>
Hereinafter, the driving assistance apparatus according to the present embodiment will be described by taking driving assistance to the driver of the host vehicle 1 as an example.

  First, the configuration of the driving support apparatus according to the present embodiment will be described with reference to FIG. FIG. 2 is a functional block diagram of the driving support apparatus 100 according to the present embodiment. In FIG. 2, only the structure relevant to this invention is shown for convenience of explanation. In addition, in the own vehicle 1 to which the driving assistance apparatus 100 according to the present embodiment is applied, the configuration (engine, tire, etc.) that the four-wheel vehicle is normally provided is provided, and the description thereof is omitted. . The present invention can be applied to both a four-wheeled vehicle and a saddle riding type vehicle. In addition, the present invention can be applied to both autonomous driving vehicles and manned driving vehicles.

  The driving support device 100 according to the present embodiment is mounted on the host vehicle 1. As shown in FIG. 2, the driving support apparatus 100 includes a CPU (Central Processing Unit) 101 that controls the entire apparatus. The CPU 101 is configured by, for example, a processor that executes various processes. The CPU 101 is configured to be able to implement a first determination unit 121, a second determination unit 122, and a driving support unit 123, which will be described later, by executing a program stored in a storage unit (not shown).

  A front camera 102, a raindrop amount sensor 103, a road surface sensor 104, and a steering sensor 105 are connected to the CPU 101. A steering sensor 106 is connected to the steering sensor 105. The front camera 102, the raindrop amount sensor 103, the road surface sensor 104, and the steering sensor 105 are electrically connected to the CPU 101 so that signals can be transmitted to and received from the CPU 101, or signals can be transmitted or received in one direction. Has been.

  The front camera 102 constitutes an example of an imaging unit, and images a predetermined range in front of the host vehicle 1. More specifically, the front camera 102 images a predetermined range in front of the host vehicle 1 (see the alternate long and short dash line in FIG. 1). The front camera 102 outputs the captured image (captured image) to the CPU 101. The captured image is stored in a storage unit (not shown) disposed inside or outside the CPU 101 and used for determination processing by the first determination unit 121 as will be described in detail later.

  The raindrop amount sensor 103 constitutes an example of a raindrop detection unit, and is composed of, for example, an infrared sensor. The raindrop amount sensor 103 includes a light emitting unit and a light receiving unit, and detects the raindrop amount based on the amount of infrared reflection that varies depending on whether or not raindrops adhere to the windshield. The raindrop amount sensor 103 outputs the detected raindrop amount to the CPU 101.

  The road surface sensor 104 constitutes an example of a road surface state confirmation unit, and includes, for example, an ultrasonic detection type sensor. The road surface sensor 104 includes an ultrasonic oscillator that outputs ultrasonic waves to the road surface, and a detection unit that detects ultrasonic waves reflected from the road surface, and detects the road surface state based on the amount of reflected ultrasonic waves from the road surface. . For example, the road surface sensor 104 can detect the unevenness of the road surface and the presence or absence of foreign matter (such as gravel) on the road surface. The road surface sensor 104 outputs the detected road surface state to the CPU 101.

  The steering sensor 105 detects a change amount of the steering wheel 106 steered by the driver, and calculates a steering angle from the detected change amount. The steering sensor 105 outputs a signal (steering angle signal) corresponding to the calculated steering angle to the CPU 101.

  Further, the CPU 101 is connected with alarm generation means 107, steering vibration means 108, deceleration control means 109, and steering control means 111. A deceleration control unit 110 is connected to the deceleration control unit 109. A steering 106 is connected to the steering control means 111. The steering 106 includes an actuator (not shown) that steers the steering 106 in accordance with a control signal from the steering control unit 111. The alarm generation means 107, the steering vibration means 108, the deceleration control means 109, and the steering control means 111 constitute a part of the driving support means. The alarm generation means 107, the steering vibration means 108, the deceleration control means 109, and the steering control means 111 are electrically connected to the CPU 101 so that signals can be transmitted to and received from the CPU 101, or signals can be transmitted or received in one direction. Connected.

  The alarm generation means 107 constitutes an example of the alarm means, and issues an alarm to the driver of the host vehicle 1 under the control of the CPU 101. For example, the alarm generation unit 107 includes an alarm generator and a speaker. The alarm generation means 107 issues an alarm as an example of driving assistance for the driver when a predetermined condition is satisfied. The alarm is not limited to sound, but for example, an indicator lamp provided on an instrument panel (not shown) or the like of the host vehicle 1 is turned on, or a warning is displayed on a display unit of a navigation device (not shown). A message or the like may be displayed.

  The steering vibration means 108 constitutes an example of vibration generation means, and causes the steering 106 to generate vibration under the control of the CPU 101. The steering vibration means 108 is composed of, for example, a vibration device such as a vibration motor built in the steering 106 and an instruction device that gives a vibration instruction to the vibration device.

  The deceleration control means 109 constitutes an example of the deceleration control means, and is automatically controlled by the deceleration means 110 configured by mechanical brake, engine brake, electronic throttle opening control, fuel injection control, etc. under the control of the CPU 101. The speed of the vehicle 1 is reduced.

  The CPU 101 includes first determination means 121, second determination means 122, and driving support means 123. The first determination unit 121 can recognize the other vehicles 5 and 7 existing in front of the host vehicle 1 based on the captured image captured by the front camera 102. More specifically, the first determination unit 121 can perform pattern matching on the contours of the vehicle, wheels, wheels, tail lamps, and the like in the captured image to determine whether the other vehicles 5 and 7 are present. .

  Moreover, the 1st determination means 121 can determine whether the other vehicles 5 and 7 are moving based on a picked-up image, or has stopped. More specifically, for example, the front camera 102 continuously photographs the front of the host vehicle 1, compares the current captured image with the previous captured image in time series, If the position of the front wheel or the rear wheel or the front wheel or the rear wheel is changed, it can be determined that the other vehicles 5 and 7 have started to move.

  Further, the first determination unit 121 can determine whether the front wheels or the rear wheels of the other vehicles 5 and 7 or the front wheels or the rear wheels can be recognized based on the captured image. More specifically, the contour of the front wheel or rear wheel of the other vehicle 5, 7 or the wheel of the front wheel or rear wheel can be recognized by pattern matching, and the feasibility can be determined.

  In the present embodiment, it is assumed that a high-performance camera with high image resolution is used for the front camera 102. However, because the front wheels and the rear wheels are black, the front wheels and wheels are monitored at night. It can be difficult. On the other hand, the wheel is rarely black, and the recognition of the front wheel and the rear wheel is preferably replaced by the recognition of the wheel.

  Further, the first determination unit 121 sets the first angle (θ1 ′ in FIGS. 6 and 8) formed by the front wheels of the other vehicles 5 and 7 with respect to the traveling direction X of the host vehicle 1 based on the captured image. Can be calculated. Details of the calculation method of the first angle θ1 ′ will be described later.

  Further, the first determination unit 121 sets a second angle (θ2 in FIGS. 1 and 7) formed by the rear wheels of the other vehicles 5 and 7 with respect to the traveling direction X of the host vehicle 1 based on the captured image. Can be calculated. A method for calculating the second angle θ2 will be described later.

  Further, the first determination means 121 determines the ratio between the width W1 of the travel lane 2a (see FIG. 4) and the wheel base W2 of the other vehicles 5 and 7 (see FIG. 6) based on the captured image (hereinafter referred to as “W1 / W2 "can be calculated. Details of the method for calculating W1 / W2 will be described later.

  Further, the first determination unit 121 can recognize the height of the step generated at the boundary between the main road 2 and the sidewalk 4 based on the captured image.

  The second determination means 122 constitutes another vehicle course prediction means, and can predict the course of the other vehicles 5 and 7 based on the output of the first determination means 121. Details of the course prediction of the other vehicles 5 and 7 will be described later.

  In addition, the driving state of the host vehicle 1 can be determined. For example, the second determination unit 122 can recognize the traveling direction X (see FIG. 1) of the host vehicle 1 based on the steering angle signal from the steering sensor 105.

  The driving support means 123 constitutes a part of the driving support means, and provides predetermined driving support for the driver of the host vehicle 1 based on the course of the other vehicles 5 and 7 predicted by the second determination means 122. To implement. Details of the driving assistance will be described later.

  The driving support means 123 evaluates whether or not the other vehicles 5 and 7 have an influence on the future traveling of the host vehicle 1 based on a predetermined standard, and performs driving support based on the evaluation result. It is preferable that the reference includes one based on the predicted course of the other vehicles 5 and 7.

  More specifically, the driving support means 123 performs predetermined driving support. The driving support means 123 cooperates with the alarm generation means 107, the steering vibration means 108, the deceleration control means 109, and the steering control means 111 according to the function of each means 107, 108, 109, 111 for the driver. Implement driving assistance.

  Further, the driving support means 123 can change the timing of driving support for the driver based on the detection results from the raindrop amount sensor 103 and the road surface sensor 104.

  Next, a control flow of the driving support device 100 according to the present embodiment will be described. FIG. 3 is a diagram showing a control flow of the driving support apparatus 100 according to the present embodiment.

  First, the first determination unit 121 (see FIG. 2) determines whether or not there is another vehicle ahead based on the captured image in front of the host vehicle 1 captured by the front camera 102 (step (hereinafter referred to as “ST”). ) 101). FIG. 4 is a schematic diagram showing a photographed image photographed by the front camera 102 of the driving assistance apparatus 100 according to the present embodiment. As shown in FIG. 4, the captured image 10 includes images of other vehicles 5 and 7. In this case, the 1st determination means 121 can recognize the other vehicles 5 and 7 ahead by pattern matching as above-mentioned. Hereinafter, for the sake of convenience, processing for only the other vehicle 5 will be described.

  In FIG. 4, the monitored distance indicated by the broken line 11 varies depending on the vehicle speed of the host vehicle 1. For example, when the main road 2 is dry asphalt when traveling at a speed of 60 km, the host vehicle 1 is safe. Since the distance that can be stopped is 32 m, it is preferable that the distance to be monitored, for example, the range to be monitored by the front camera 102 (see FIG. 1 and FIG. 2) is up to 52 m, which is 20 m ahead.

  Here, the first determination means 121 determines whether or not there is another vehicle 5 ahead. However, it is determined whether or not the other vehicle 5 is outside the main road 2, that is, whether it is on the sidewalk 4. It is preferable to make it. In this case, the driving assistance apparatus 100 according to the present embodiment can particularly improve the safety prevention effect when the other vehicle 5 enters the main road 2 from the outside of the main road 2.

  When there is another vehicle 5 ahead based on the recognition result (ST101: YES), the first determination means 121 determines whether the other vehicle 5 is based on the change in the position of the front wheel 5c of the front wheel 5a of the other vehicle 5 in the captured image. It is determined whether 5 has started moving (ST102).

  When the other vehicle 5 has started to move (ST102: YES), the first determination means 121 determines whether or not the front wheel 5c (see FIG. 4) has been recognized (ST103).

  When the front wheel 5c of the other vehicle 5 is recognized (ST103: YES), the first determination means 121 calculates the first angle θ1 ′ based on the captured image (ST104). In the present embodiment, the first angle θ <b> 1 ′ is calculated as an angle formed by the traveling direction X (see FIG. 1) of the host vehicle 1 and the front wheel 5 c of the other vehicle 5.

  First, a case where the other vehicle 5 makes a left turn to join the traveling lane 2a of the host vehicle 1 will be described.

  FIG. 5 is a schematic diagram showing the front wheel 5c and the rear wheel 5d of the other vehicle 5 extracted from the captured image 10 shown in FIG. FIG. 6 is a schematic diagram showing a state in which the front wheel 5a of the other vehicle 5 is cut in the left turning direction (left side shown in FIG. 6) of the other vehicle 5 in the present embodiment. In FIG. 6, a two-dot chain line X is parallel to the traveling direction X of the host vehicle 1, and a one-dot chain line I indicates a straight line passing through the side surface of the front wheel 5 c of the other vehicle 5. The first angle θ1 ′ corresponds to the angle formed by the two-dot chain line X and the one-dot chain line I in FIG.

  The longitudinal dimension A ′ of the front wheel 5c shown in FIGS. 5 and 6 corresponds to the diameter of the front wheel 5c, and is constant without depending on the first angle θ1 ′ (see FIG. 6) of the front wheel 5c. The lateral dimension B 'of the front wheel 5c changes depending on the first angle? 1' of the front wheel 5c. Specifically, the lateral dimension B 'of the front wheel 5c becomes smaller as the steering of the other vehicle 5 is largely turned and the first angle? 1' of the front wheel 5c is increased. For convenience of explanation, it is assumed in FIG. 6 that the longitudinal dimension A 'and the lateral dimension B' of the front wheel 5c are substantially the same as the respective dimensions of the front wheel 5a.

  Therefore, since the front wheel 5c is circular, the first angle θ1 ′ of the front wheel 5c can be calculated from the lateral dimension B ′ / vertical dimension A ′ of the front wheel 5c using an inverse trigonometric function. .

Specifically, as shown in FIG. 6, a right triangle abc is assumed in which the longitudinal dimension A ′ of the front wheel 5c is the hypotenuse (ab) and the lateral dimension B ′ is the base (ac). Therefore, the angle θ1 formed by the hypotenuse (ab) and the remaining one side (bc) can be obtained using the following equation (1) using an inverse trigonometric function.
θ1 = 90 ° −arccos (B ′ / A ′) (1)

Furthermore, since the angle θ1 corresponds to the complementary angle of the first angle θ1 ′, the first angle θ1 ′ can be obtained from the angle θ1 using the following equation (2).
θ1 ′ = 180 ° −θ1 (2)

  Next, the first determination means 121 determines whether or not the rear wheel 5d (see FIG. 4) of the other vehicle 5 has been recognized (ST105).

  When the rear wheel 5d of the other vehicle 5 is recognized (ST105: YES), the first determination unit 121 calculates the second angle θ2 based on the photographed image (ST106). In the present embodiment, the second angle θ is calculated as an angle formed by the traveling direction X of the host vehicle 1 (see FIG. 1) and the rear wheel 5d of the other vehicle 5.

  FIG. 7 is a schematic diagram showing the relationship between the rear wheel 5b of the other vehicle 5 and the second angle θ2 in the present embodiment. In FIG. 7, a two-dot chain line X is parallel to the traveling direction X of the host vehicle 1, and a one-dot chain line II indicates the direction of the rear wheel 5 b of the other vehicle 5. For convenience of explanation, it is assumed in FIG. 7 that the vertical dimension A and the horizontal dimension B of the rear wheel 5d are substantially the same as the respective dimensions of the rear wheel 5b.

  The vertical dimension A of the rear wheel 5d shown in FIGS. 5 and 7 corresponds to the diameter of the rear wheel 5d and is constant regardless of the angle of the rear wheel 5d, that is, the direction of the body of the other vehicle 5. is there. The lateral dimension B of the rear wheel 5d varies depending on the angle of the rear wheel 5d, that is, the direction of the vehicle body of the other vehicle 5. Specifically, the lateral dimension B of the rear wheel 5d becomes smaller as the steering of the other vehicle 5 is largely turned and the second angle θ2 becomes larger.

Therefore, since the rear wheel 5d is circular, the angle of the second angle θ2 is calculated from the lateral dimension B / vertical dimension A of the rear wheel 5d according to the following expression (3) using an inverse trigonometric function. Can do.
θ2 = 90 ° −arccos (B / A) (3)

  Next, a case where the other vehicle 5 makes a right turn while trying to join the opposite lane 2b will be described.

  FIG. 8 is a schematic diagram showing a state in which the front wheel 5a of the other vehicle 5 is cut in the right turning direction (left side shown in FIG. 8) of the other vehicle 5 in the present embodiment. In FIG. 8, a two-dot chain line X is parallel to the traveling direction X of the host vehicle 1, and a one-dot chain line I indicates a straight line passing through the side surface of the front wheel 5 c of the other vehicle 5. The first angle θ1 ′ corresponds to the angle formed by the two-dot chain line X and the one-dot chain line I in FIG.

  The vertical dimension A ′ of the front wheel 5c shown in FIGS. 5 and 8 corresponds to the diameter of the front wheel 5c, and is constant without depending on the first angle θ1 ′ (see FIG. 8) of the front wheel 5c. The lateral dimension B 'of the front wheel 5c changes depending on the first angle? 1' of the front wheel 5c. Specifically, the lateral dimension B 'of the front wheel 5c becomes smaller as the steering of the other vehicle 5 is largely turned and the first angle θ1' of the front wheel 5c is reduced. For convenience of explanation, it is assumed in FIG. 8 that the longitudinal dimension A 'and the lateral dimension B' of the front wheel 5c are substantially the same as those of the front wheel 5a.

  Therefore, since the front wheel 5c is circular, the first angle θ1 ′ of the front wheel 5c can be calculated from the lateral dimension B ′ / vertical dimension A ′ of the front wheel 5c using an inverse trigonometric function. .

Specifically, as shown in FIG. 8, a right-angled triangle a′b′c having a longitudinal dimension A ′ of the front wheel 5c as a hypotenuse (a′b ′) and a lateral dimension B ′ as a base (a′c ′). Assume '. In the right triangle a′b′c ′, the angle formed by the hypotenuse (ab) and the remaining one side (bc) is equal to the first angle θ1 ′. Therefore, the first angle θ1 ′ can be obtained using the following equation (4) using an inverse trigonometric function.
θ1 ′ = 90 ° −arccos (B ′ / A ′) (4)

  Thereafter, the first determination means 121 calculates a ratio W1 / W2 between the width W1 of the traveling lane 2a and the wheel base W2 of the other vehicle based on the captured image (ST107). More specifically, as shown in FIG. 4, the distance between the curb 4a and the white line 2c of the sidewalk 4 attached to the travel lane 2a is measured from the photographed image to obtain the width W1. Further, the distance D1 between the line Y passing through the center of the front wheel 5c of the other vehicle 5 and the line Y 'passing through the center of the rear wheel 5d is obtained from the captured image. Then, based on the width W1, the interval D1, and the second angle θ2, the wheel base W2 (see FIG. 6) on the captured image of the other vehicle 5 can be calculated using a trigonometric function. Then, W2 / W1 can be calculated based on the calculated width W1 and the wheel base W2 of the other vehicle 5.

(Course prediction process A)
Next, the second determination unit 122 performs the course prediction process A of the other vehicle 5 based on the first angle θ1 ′, the second angle θ2 and W1 / W2 calculated by the first determination unit 121 ( ST108).

  First, the course prediction process A when the other vehicle 5 described with reference to FIG. 6 makes a left turn will be described. For example, when the first angle θ1 ′ is larger than a predetermined angle a (for example, 135 °), the second determination unit 122 has a large steering angle by the steering and the other vehicle 5 turns small. It can be predicted that the bulge when entering the traveling lane 2a is small and the possibility of taking the course α1 (see FIG. 1) is high.

  On the other hand, when the first angle θ1 ′ is smaller than the predetermined angle a, the steering angle by the steering is small, and the other vehicle 5 turns greatly, so that the bulge when the other vehicle 5 enters the traveling lane 2a is large. It can be predicted that there is a high possibility of taking the course α2 (see FIG. 1).

  In this way, it is possible to roughly predict whether the other vehicle 5 is the course α1 or the course α2 from the first angle θ1 ′ formed by the front wheel 5c with respect to the traveling direction X of the host vehicle 1.

  Furthermore, the second determination means 122 can realize the course prediction with higher accuracy by taking the second angle θ2 (see FIGS. 1 and 7) into consideration. For example, when the second angle θ2 is smaller than a predetermined angle b (for example, 45 °), the other vehicle 5 travels in an acute direction with respect to the traveling direction X of the host vehicle 1, and therefore the other vehicle 5 It can be predicted that the bulge when entering the traveling lane 2a becomes smaller.

  Therefore, when the first angle θ1 ′ is larger than the predetermined angle a and the second angle θ2 is smaller than the predetermined angle b (θ1 ′> a and θ2 <b), the other vehicle 5 travels. The bulge when entering the lane 2a is further reduced, and the possibility of the course α1 is increased.

  When the first angle θ1 ′ is smaller than the predetermined angle a and the second angle θ2 is smaller than the predetermined angle b (θ1 ′ <a and θ2 <b), the other vehicle 5 When the vehicle enters the driving lane 2a, the bulge becomes slightly large, and the possibility of the path α2 increases.

  On the other hand, when the second angle θ2 is larger than the predetermined angle b, the other vehicle 5 travels in a direction close to a right angle with respect to the traveling direction X of the host vehicle 1, so that the other vehicle 5 is in the direction of the traveling lane 2a. It can be predicted that the bulge when entering will be larger.

  Therefore, when the first angle θ1 ′ is smaller than the predetermined angle a and the second angle θ2 is larger than the predetermined angle b (θ1 ′ <a and θ2> b), the other vehicle 5 travels. The bulge when entering the lane 2a is further increased, and the possibility of the course α2 is increased.

  When the first angle θ1 ′ is larger than the predetermined angle a and the second angle θ2 is larger than the predetermined angle b (θ1 ′> a and θ2> b), the other vehicle 5 When the vehicle enters the traveling lane 2a, the bulge becomes slightly large, and the possibility that the other vehicle 5 protrudes into the oncoming lane 2b and takes the course α2 is increased.

  In this way, by taking both the first angle θ1 ′ and the second angle θ2 into consideration, it is possible to predict the course of the other vehicle 5 with higher accuracy.

  Furthermore, the second determination unit 122 can realize the course prediction with higher accuracy by performing the course prediction taking W1 / W2 into consideration. For example, when W1 / W2 is close to 1, since the wheel base W2 of the other vehicle 5 is long and the minimum turning radius of the other vehicle 5 is large, the bulge when the other vehicle 5 enters the traveling lane 2a increases. Cheap. As a result, the possibility that the other vehicle 5 becomes the course α2 is higher than the result of the course prediction based on both the first angle θ1 ′ and the second angle θ2.

  Note that it is not essential in the present invention to predict the course of the other vehicle 5 in consideration of W1 / W2.

  As described above, by taking W1 / W2 into consideration in addition to both the first angle θ1 ′ and the second angle θ2, it is possible to predict the course of the other vehicle 5 with higher accuracy.

  Next, the course prediction process A when the other vehicle 5 described with reference to FIG. 8 turns to the right will be described. For example, when the first angle θ1 ′ is smaller than a predetermined angle c (for example, 45 °), the second determination unit 122 makes a turn by the other vehicle 5 with a large steering angle by the steering. Although the bulge when entering the traveling lane 2a is small and the path β1 (see FIG. 1) is taken, it can be predicted that there is a high possibility that the vehicle will quickly join the oncoming lane 2b.

  On the other hand, when the first angle θ1 ′ is larger than the predetermined angle c, the steering angle by the steering is small, and the other vehicle 5 turns greatly, so that the bulge when the other vehicle 5 enters the traveling lane 2a is a course. It is larger than β1, and it can be predicted that there is a high possibility of merging in the opposite lane 2b relatively slowly.

  Thus, even when the other vehicle 5 turns to the right, the opposite lane 2b on which route the other vehicle 5 takes from the first angle θ1 ′ formed by the front wheel 5c with respect to the traveling direction X of the host vehicle 1 is determined. It is possible to roughly predict whether or not to join.

  Furthermore, the second determination means 122 can realize the course prediction with higher accuracy by taking the second angle θ2 (see FIGS. 1 and 7) into consideration. For example, when the second angle θ2 is larger than a predetermined angle d (for example, 90 °), the other vehicle 5 travels in the direction opposite to the traveling direction X of the host vehicle 1, and thus the other vehicle 5 It can be predicted that the bulge when entering the traveling lane 2a becomes smaller. On the other hand, when the second angle θ2 is smaller than the predetermined angle d, the other vehicle 5 travels in the same direction as the traveling direction X of the host vehicle 1, and therefore the other vehicle 5 moves with respect to the travel lane 2a. It can be predicted that the bulge when entering will be larger.

  Thus, even when the other vehicle 5 makes a right turn, it is possible to predict the course of the other vehicle 5 with higher accuracy by taking into consideration both the first angle θ1 ′ and the second angle θ2. Become.

  Similarly, when the other vehicle 5 described with reference to FIG. 8 turns to the right, the other vehicle 5 can travel to the travel lane 2a by taking into account both the first angle θ1 ′ and the second angle θ2. It is possible to more accurately predict the course until the vehicle crosses the road and merges with the opposite lane 2b.

  Whether the other vehicle 5 turns left or right is determined based on, for example, the direction of the front end of the front wheel 5a of the other vehicle 5 based on the photographed image taken by the front camera 102. Can be identified by pattern matching, and it can be determined whether the vehicle is turning left or turning right.

(Course prediction process B)
When the rear wheel 5d of the other vehicle 5 is not recognized (ST105: NO), the second determination unit 122 determines that the other vehicle 5 is based on only the first angle θ1 ′ calculated by the first determination unit 121. The route prediction process B is performed (ST109).

  The course prediction process B will be described by taking as an example the case where the other vehicle 5 described with reference to FIG. 6 makes a left turn. For example, the second determination unit 122 has a first angle θ1 ′ (see FIG. 6). When the angle is larger than the predetermined angle a, the steering angle by the steering is large and the other vehicle 5 turns small, so that the bulge when the other vehicle 5 enters the traveling lane 2a is small, and the course α1 (see FIG. 1) It can be predicted that there is a high possibility of taking.

  On the other hand, when the first angle θ1 ′ is smaller than the predetermined angle a, the steering angle by the steering is small, and the other vehicle 5 turns greatly, so that the bulge when the other vehicle 5 enters the traveling lane 2a is large. It can be predicted that there is a high possibility of taking the course α2 (see FIG. 1).

  Thus, according to the course prediction process B, whether the other vehicle 5 is the course α1 or the course α2 from the first angle θ1 ′ formed by the front wheel 5c with respect to the traveling direction X of the host vehicle 1 is roughly determined. Can be predicted. The course prediction process B is less accurate than the course prediction process A, but there is an advantage that the course of the other vehicle 5 can be predicted even when the rear wheel 5d of the other vehicle 5 cannot be recognized by an obstacle. is there.

(Necessity determination of avoidance action)
When the process of ST108 or ST109 is completed, the driving support means 123 determines a predetermined amount for the driver of the host vehicle 1 based on the course of the other vehicle 5 predicted by the course prediction process A or the course prediction process B. Driving assistance is implemented (ST110 to ST112).

  First, the driving support means 123 evaluates, based on the predicted course of the other vehicle 5, whether or not the other vehicle 5 affects future travel of the host vehicle 1 based on a predetermined criterion, and the evaluation Based on the result, it is determined whether an avoidance action is necessary (ST110).

  Here, the avoidance action includes deceleration control by the deceleration control means 109 (see FIG. 2) and / or course change control by the steering control means 111.

  The predetermined reference includes at least the course predicted by the course prediction process A (ST108) and the course prediction process B (ST109). For example, if the course of the other vehicle 5 is the course α1 (see FIG. 1), the other vehicle 5 has a small bulge when the other vehicle 5 enters the traveling lane 2a and smoothly joins. Therefore, it can be determined that avoidance control is unnecessary.

  On the other hand, if the course of the other vehicle 5 is the course α2 (see FIG. 1), the other vehicle 5 has a large bulge when the other vehicle 5 enters the traveling lane 2a and cannot smoothly join. Therefore, it is possible to determine that avoidance control is necessary.

  The predetermined standard described above can include a distance (described above) that the host vehicle 1 can safely stop. More specifically, when the position of the other vehicle 5 is in front of the distance (described above) at which the host vehicle 1 can safely stop, even if the course of the other vehicle 5 is the course α1 (see FIG. 1). Since there is a high possibility of affecting future travel of the host vehicle 1, it can be determined that avoidance control is necessary. On the other hand, if the path of the other vehicle 5 is the path α1 and the position of the other vehicle 5 is beyond the distance (described above) at which the host vehicle 1 can be safely stopped, the path of the other vehicle 5 is the host vehicle 1. Therefore, it is possible to determine that the avoidance control is unnecessary.

  Further, according to the above-mentioned predetermined standard, the height of the step generated at the boundary between the main road 2 and the sidewalk 4 recognized by the first determination means 121, or the step and the front wheel 5 a or the front wheel of the other vehicle 5. A ratio with a longitudinal dimension A ′ of 5c (see FIG. 5) can be included. For example, when the height of the step is large or the ratio between the step and the vertical dimension A ′ is large, it is inferred that the other vehicle 5 slowly enters the main road 2 in order to exceed the step. Is likely to affect future travel of the host vehicle 1, it can be determined that avoidance control is necessary.

  Further, the above-mentioned predetermined reference can include the above W1 / W2. For example, when W1 / W2 is less than 1, the wheel base W2 of the other vehicle 5 is long enough to exceed the width W1, and therefore the other vehicle 5 is assumed to be a large truck, a trailer, or the like. In this case, since it is presumed that the approach speed of the other vehicle 5 is slow, it is highly likely that the course of the other vehicle 5 will affect future travel of the host vehicle 1, and therefore it is determined that avoidance control is necessary. be able to.

  Furthermore, the driving support means 123 can change the timing of driving support for the driver based on the detection results from the raindrop amount sensor 103 and the road surface sensor 104. Specifically, when it is raining heavily or there is gravel on the road surface of the main road 2, the timing of driving support is advanced to prevent the own vehicle 1 from colliding with the other vehicle 5 and preventive safety. The performance can be improved.

  When the avoidance action is not necessary (ST110: NO), the driving support means 123 performs control for alerting the driver (hereinafter referred to as “attention control”) (ST111). In this alerting control, the driving support means 123 causes the alarm generation means 107 to generate an alarm, or the steering vibration means 108 causes the steering 106 to vibrate. Note that both alarm and vibration may be performed simultaneously. By performing the alerting control in this way, it is possible to alert the driver of the own vehicle 1 and prevent the own vehicle 1 from contacting the other vehicle 5.

  In addition, about the content performed by alerting control, it is not limited to these, It can change suitably.

  On the other hand, when the avoidance action is necessary (ST110: YES), the driving support means 123 performs avoidance action control for the host vehicle 1 (ST112). As this avoidance action control, for example, the driving support means 123 causes the deceleration control means 109 to operate the deceleration means 110 to reduce the speed of the host vehicle 1.

  Further, the driving support means 123 changes the steering angle of the steering 106 by the steering control means 111 and changes the course of the host vehicle 1. More specifically, the driving support means 123 causes the steering control means 111 to drive an actuator for steering the steering 106 based on the steering angle signal from the steering sensor 105 so that the steering angle of the steering 106 becomes a desired value. The course of the host vehicle 1 can be changed.

  The course of the own vehicle 1 is changed, for example, when the width of the travel lane 2a is sufficiently wide, or there is no oncoming vehicle in the oncoming lane 2b, or there is another running lane adjacent to the running lane 2a. It is preferable to carry out while ensuring safety when the distance between the vehicle and the vehicle is sufficiently secured. By performing the avoidance control in this way, it is possible to prevent the host vehicle 1 from contacting the other vehicle 5.

  Thereafter, the driving support means 123 determines whether or not to end the driving support (ST113). When driving support is ended (ST113: YES), the process is ended.

  Note that, in all the determinations of ST101 to ST103 and ST113, if the determination result is NO, the process returns to ST101, and the processes after ST101 are repeated.

  As described above, in the driving support device 100 according to the present embodiment, there are various possibilities, the other vehicle 5 entering the main road 2 from the outside of the main road 2 such as the sidewalk 4 and the side road 6, 7 is predicted, the driving of the driver of the host vehicle 1 is supported, the burden on the driver of the host vehicle 1 is reduced, and the preventive safety performance can be improved. Such an effect is extremely important when considering an aging society.

  In particular, in the driving support apparatus 100, since the front camera 102 is used in each of the steps ST101 to ST103 and ST105 to ST109, no special detection means (particularly radar) is added, and a man-powered small-sized four-wheeled automobile In addition, since it can be easily realized in a straddle-type vehicle such as a motorcycle, the preventive safety performance can be improved economically.

  Further, in the driving support apparatus 100, the driving support means 123 changes the timing of the driving support (for example, alerting control and deceleration control) according to the detection result from the raindrop amount sensor 103 and / or the road surface sensor 104. This is preferable as an embodiment. In this case, since the driving support execution timing is changed according to the amount of raindrops and the road surface condition, appropriate driving support can be performed according to the environment in which the host vehicle 1 travels.

  In addition, this invention is not limited to the said embodiment, It can change and implement variously. In the above-described embodiment, the size, shape, and the like illustrated in the accompanying drawings are not limited to this, and can be appropriately changed within a range in which the effect of the present invention is exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention.

  For example, in the above embodiment, the other vehicle 5 that directly joins the traveling lane 2a of the main road 2 (see FIG. 1) has been described as an example. However, the vehicle travels across the opposite lane 2b of the main road 2. The other vehicle 7 that joins the lane 2a can also be handled by the driving support device 100 of the present embodiment.

  In the above embodiment, the case where the main road 2 (see FIG. 1) includes the traveling lane 2a and the opposite lane 2b has been described as an example. However, the main road includes two or more lanes in the same direction, Even when another vehicle joins the lane adjacent to the lane in which the host vehicle travels, the driving support device 100 of the present embodiment can cope with it.

  In the above embodiment, the alerting control (ST111) is performed only when the avoidance action is not necessary (ST110: NO in FIG. 3), but also when the avoidance action is necessary (ST110: YES). You may also perform alerting control.

  As described above, the present invention predicts the course of another vehicle when the other vehicle joins the lane on which the host vehicle travels with higher accuracy and improves the preventive safety performance, while driving the driving support device. It is advantageous in that it can be made inexpensive and downsized, and is particularly useful for driving assistance devices such as automobiles and saddle-type vehicles.

DESCRIPTION OF SYMBOLS 1 Own vehicle 2 Main road 4 Sidewalk 6 Side road 5, 7 Other vehicle 100 Driving support apparatus 101 CPU
102 Front camera (imaging means)
103 Raindrop amount sensor (raindrop detection means)
104 Road surface sensor (road surface condition confirmation means)
105 Steering sensor 106 Steering 107 Alarm generating means (alarm means)
108 Steering vibration means (vibration generating means)
109 Deceleration control means 110 Deceleration means 111 Steering control means 121 First determination means 122 Second determination means (other vehicle course prediction means)
123 Driving support means

Claims (10)

Imaging means for photographing the front of the vehicle;
Recognizing at least some of the wheels of the other vehicle included in the image captured by the imaging means, and determining whether the other vehicle is moving based on the movement of the front wheel of the other vehicle;
When it is determined that the other vehicle is moving,
(I) When both the front wheels and the rear wheels of the other vehicle can be recognized from the image, the first angle formed by the front wheels of the other vehicle and the rear wheels of the other vehicle with respect to the traveling direction of the host vehicle Predicting the course of the other vehicle based on the second angle formed by
(Ii) When only the front wheels of the other vehicle can be recognized from the image, other vehicle course prediction means for predicting the course of the other vehicle based on the first angle;
A driving support device, comprising: driving support means for performing predetermined driving support for a driver of the host vehicle based on the predicted course of the other vehicle.   The other vehicle course prediction means considers the ratio of the width of the lane in which the host vehicle travels obtained based on the image and the wheel base of the other vehicle when predicting the course of the other vehicle. The driving support apparatus according to claim 1, wherein   The driving support means evaluates whether or not the other vehicle affects future driving of the host vehicle based on a predetermined criterion, performs the driving support based on the evaluation result, and the predetermined vehicle The driving support apparatus according to claim 1, wherein the reference includes a basis based on the predicted path of the other vehicle.   The driving support apparatus according to claim 3, wherein the predetermined reference further includes a value based on a distance at which the host vehicle can be safely stopped.   The said other vehicle course prediction means calculates | requires the said 1st angle and the said 2nd angle as an angle which the wheel of the said front wheel and the said rear wheel makes with respect to the advancing direction of the said own vehicle. The driving assistance device according to any one of claims 1 to 4.   6. The driving support apparatus according to claim 1, wherein the driving support means includes a warning means.   The driving support apparatus according to any one of claims 1 to 6, wherein the driving support means includes a vibration generating means for alerting the driver by vibration.   The driving support device according to any one of claims 1 to 7, wherein the driving support means is a deceleration control means for reducing the speed of the host vehicle.   The driving support device according to any one of claims 1 to 8, wherein the driving support means is a steering control means for changing a steering angle of the steering of the host vehicle.   It further comprises raindrop detection means and / or road surface condition confirmation means, and the driving support means changes the timing of execution of the driving assistance according to the output from the raindrop detection means and / or road surface condition confirmation means. The driving support device according to any one of claims 1 to 9, wherein

Images