JP5782708B2 - Driving support device - Google Patents

Description

  The present invention relates to a driving support device.

  Based on information from the CCD camera and laser radar, recognize the road shape and distance to obstacles, create a two-dimensional map of the surrounding environment of the vehicle, and check whether there is a narrow road and whether it can pass. In some cases, it is determined whether the vehicle travels by calculating the travel route of the host vehicle (see Patent Document 1).

JP 2002-352397 A

In the background art as described in Patent Document 1, the calculation cost increases because the travel route is constantly calculated based on the vehicle information such as the vehicle speed and the steering angle and the surrounding environment information of the host vehicle. .
An object of the present invention is to facilitate calculation of a travel route for travel support and to suppress calculation cost.

The driving support device according to the present invention recognizes the surrounding environment in the vehicle traveling direction, detects the vehicle position and the vehicle attitude with respect to the recognized surrounding environment, and associates the surrounding environment, the vehicle position, and the vehicle attitude. Stored in advance. Then, by referring to the accumulated information based on the vehicle position, a plurality of travel routes that the vehicle can travel are set as reference routes, and the vehicle position is determined based on the vehicle position and the vehicle attitude. The introduction route from each to the reference route is calculated. Then, according to the calculation result of the introduction route, one of the plurality of reference routes is selected, and the final vehicle is combined with the selected reference route and the introduction route up to this reference route. Set the travel route. Then, the shortest introduction route from the vehicle position to each travel route is calculated, and the reference route corresponding to the shortest introduction route among the calculated introduction routes is selected .

  According to the driving support device of the present invention, information that associates the surrounding environment, the vehicle position, and the vehicle attitude is stored, and the driving route of the vehicle is set by referring to the information as needed. Since it is a memory-based technology, it is possible to easily calculate a travel route for driving support and to reduce calculation costs.

It is a schematic block diagram of a driving assistance device. It is a calculation block diagram performed with a driving assistance device. It is a flowchart which shows a driving history preservation | save process. It is a flowchart which shows a driving assistance control process. It is a flowchart which shows an introduction route calculation process. It is a reference figure of introduction route calculation. It is an example of a driving scene. It is a calculation block diagram of 2nd embodiment performed with a driving assistance device. It is a calculation block diagram of 3rd embodiment performed with a driving assistance device. It is a figure which shows an example of a driving | running | working possible area | region. It is a figure which shows the reference route from which a narrow path degree differs. It is a calculation block diagram of 4th embodiment performed with a driving assistance device.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
<< first embodiment >>
"Constitution"
First, a schematic configuration of the travel support device 100 will be described.
FIG. 1 is a schematic configuration diagram of a driving support device.
FIG. 2 is a calculation block diagram executed by the driving support device.
The driving support device 100 includes a camera 2 installed forward with respect to the vehicle 1, a driving support controller 3, a vehicle information database 4, a driving history database 5, a display 6, a touch panel 7, and a vehicle controller 8. The vehicle sensor group 9 and the control unit group 10 are provided.

The camera 2 uses a solid-state imaging device such as a CCD, for example, and is installed on the back side of the in-car mirror portion of the vehicle 1 in a direction that allows imaging of the front of the vehicle. The captured image is sent to the driving support controller 3.
The driving support controller 3 is an electronic control unit that performs processing such as host vehicle position estimation using a captured image and driving support route selection. The driving support controller 3 may also be used as a controller used for other controls. The driving support controller 3 includes a CPU, a ROM, a RAM, and the like, like a general controller, and a program for realizing various processing units described later is stored in the ROM.

The vehicle information database 4 is a storage unit in which various parameters related to the vehicle 1 and the camera 2 are stored. In the vehicle information database 4, for example, information on a vehicle model required by an introduction route calculation unit, which will be described later, such as the minimum turning radius, vehicle width, tread length, and tire radius of the vehicle 1, the internal parameters of the camera 2 and the optical axis direction Information on camera design values such as the angle of view and the angle of view is stored.
The travel history database 5 is a storage unit that stores a captured image captured by the camera 2 together with additional information such as a camera position, a vehicle position, and a vehicle posture at the time of imaging. For example, a storage medium such as a hard disk or a memory card May be used.

The display 6 is installed, for example, in the center console in the vehicle 1 and displays the predicted course and the recommended route of the host vehicle superimposed on the camera image when driving is supported. The display 6 may also be used as a display generally used for a navigation device.
The touch panel 7 is a general touch panel that is arranged on the display unit of the display 6 and can specify the coordinate position of the part touched by the user and the movement locus of the contact part. In this embodiment, it is used when the user selects whether to start driving support or instructs to stop driving support. As an instruction input interface from the user, an input interface such as a button, a mouse, a keyboard, and a voice input may be used instead of the touch panel 7.

The vehicle controller 8 is an electronic control unit that controls the operation of an engine, a brake, and the like, and is connected to a control unit group 10 such as an engine control controller, a brake control controller, and an electric power steering control controller. The vehicle controller 8 calculates the accelerator opening, the brake control amount, and the steering angle based on the output from the travel support controller 3, and sends the drive command to the control unit group 10, thereby calculating the travel calculated by the travel support controller 3. It is possible to control the vehicle along the trajectory.
The vehicle sensor group 9 includes various on-vehicle sensors such as an accelerator sensor, a brake sensor, a steering sensor, a vehicle speed sensor, and an acceleration sensor, and is connected to the travel support controller 3. The driving support controller 3 is configured to calculate the vehicle speed, acceleration, steering angle, accelerator operation amount, brake operation amount, and the like of the vehicle 2 based on various output values from the vehicle sensor group 9.

Next, the calculation block executed by the driving support controller 3 will be described.
The travel support controller 3 includes a feature point detection unit 31, a data storage unit 32, a host vehicle position detection unit 33, a host vehicle posture detection unit 34, a reference route setting unit 35, an introduction route calculation unit 36, and a reference. A route selection unit 37 and a travel support unit 38 are provided.
The feature point detection unit 31 detects in-screen feature points from the captured image of the camera 2 and past captured images accumulated in the travel history database 5. A feature point refers to a point that can be followed when the screen moves, and generally refers to a point having a luminance difference in both the vertical and horizontal directions. Detection methods include SIFT, SURF, fast, and the like.

The data storage unit 32 is estimated by the captured image of the camera 2, the feature points obtained by the feature point detection unit, the host vehicle position estimated by the host vehicle position detection unit 33, and the host vehicle posture detection unit 34. The vehicle attitude is stored in the travel history database 5 in association with the vehicle attitude.
Here, the feature points are stored after the three-dimensional position measurement is performed by performing collation using the captured images of the camera 2 and the captured images stored in the travel history database 5, and further converted into the world coordinate system. The At this time, the position measurement result from the camera to the feature point is expressed in the camera coordinate system, and in order to convert this to the world coordinate system, the position of the camera in the world coordinate system, the three-dimensional configuration of the feature points, Need to be determined at the same time. As described above, SLAM has been proposed as a method for simultaneously generating a map and estimating the position of the vehicle, and various algorithms have been announced for this. Examples of SLAM using a camera include Mono-SLAM and PTAM.

The own vehicle position detection unit 33 estimates the own vehicle position using the captured image of the camera 2, the captured image and feature points stored in the travel history database 5, and the vehicle information stored in the vehicle information database 4. To do.
Specifically, first, the vehicle position is estimated in advance using prior knowledge (the vehicle position estimation result in the previous time step and the vehicle speed and steering angle information calculated from the vehicle sensor group 9). When the position and orientation of the host vehicle are obtained by pre-estimation, feature points that can be captured in the captured image of the camera 2 are selected from the feature points stored in the travel history database 5 based on the position and orientation. The position is estimated in the image, and the feature point is searched in the image with the estimated position in the image as the center. Thereby, a deviation between the position in the image assumed from the prior estimated value and the position in the image actually detected is detected. Then, the vehicle position is calculated from the difference between the feature point position in the image and the feature point position calculated from the pre-estimated value of the vehicle position. This is called posterior estimation. Here, post-estimation is performed by adjusting the result of the prior position estimation so that the position of the feature point matches the actual measurement position, and the final vehicle position estimation result is obtained.

The own vehicle posture detection unit 34 estimates the own vehicle posture using the picked-up image of the camera 2, picked-up images and feature points stored in the travel history database 5, and vehicle information stored in the vehicle information database 4. To do. Since the own vehicle posture is estimated together with the own vehicle position according to the same procedure as the calculation in the own vehicle position detection unit 33, the description thereof is omitted here.
The reference route setting unit 35 reads a plurality of past vehicle position loci accumulated in the travel history database 5 and sets them as reference routes. For example, if there are two driving histories in the past, two reference routes are calculated.

At this time, a three-dimensional feature point group obtained by collating a plurality of captured images stored in the travel history database 5 is also calculated at the same time. The trajectory of the past vehicle position, which is the reference route, is calculated from the camera position at the time of imaging of the captured image used for the feature point calculation, and is defined by the relative relationship with the three-dimensional feature point group.
The introduction route calculation unit 36 introduces, for each reference route set by the reference route setting unit 35, an excessive route (auxiliary route) from the current own vehicle position and own vehicle posture until the reference route is joined. Calculate as a route. Detailed procedures will be described later.

The reference route selection unit 37 calculates a final driving support route by selecting a reference route having the shortest route length of the introduction route and connecting the introduction route corresponding to the reference route.
The travel support unit 38 performs travel support using the travel support route uniquely determined by the reference route selection unit 37. Specifically, the generated driving support route information is converted into the camera coordinate system of the camera 2 and then displayed on the display 6 with the driving support route superimposed on the captured image. Furthermore, the driving support unit 38 can control the accelerator, the brake, and the steering, and can perform automatic driving by sending the deviation of the driving locus and the position / posture of the vehicle to the vehicle controller 8.

Next, the processing procedure of the driving support device 100 will be described in detail with reference to the flowcharts of FIGS. The processing of the driving support device 100 in the first embodiment is different between the driving history configuration time and the driving support time.
FIG. 3 is a flowchart showing the travel history storing process.
FIG. 4 is a flowchart showing the driving support control process.
3 and 4 are continuously performed in the driving support controller 3 at intervals of, for example, about 100 msec.

First, the travel history saving process will be described.
In step S <b> 110, a button is displayed on the display 6 as to whether or not to save the travel history, and it is determined whether or not there is a start instruction from the driver via the touch panel 7. Here, if there is a start instruction from the driver, the process proceeds to step S111. On the other hand, if there is no start instruction, it waits for an input from the driver.
In step S111, it is determined whether or not a past travel history exists in the travel history database. Here, if there is no travel history, the process proceeds to step S112. On the other hand, if a travel history exists, the process proceeds to step S113.

In step S112, a travel history initialization process is performed.
First, a feature point is calculated from the captured image of the camera 2 using the feature point detection unit 31.
Next, the captured image is captured again from the camera 2, the feature points are calculated in the same manner, and the feature points are tracked by comparing with the previously acquired feature points. Note that a generally known image processing method such as KLT-Tracker may be applied as the feature point tracking method.
Then, while performing feature point tracking continuously, when the vehicle travels a certain distance (for example, 0.5 m), the feature points of the first captured image and the current captured image are compared, and the generally known 8-point method is used. Using such a method, the feature points are three-dimensionalized, and the initial vehicle position and the initial vehicle attitude of the vehicle are defined as a relative relationship with the feature points.

Thus, the travel history is initialized.
The subsequent processing from step S113 to step S116 is a processing group that is repeatedly performed as the vehicle travels.
In step S113, from the captured image of the camera 2, using the above-described own vehicle position detecting unit 33 and own vehicle posture detecting unit 34, the own vehicle position and the own vehicle are determined in a relative relationship with the acquired three-dimensional feature point group. Detect (estimate) posture.
In subsequent step S114, a three-dimensional feature point serving as a map indicating the driving environment is configured based on the vehicle position / posture information obtained in step S113.
Specifically, the feature point of the captured image acquired in step S113 is compared with the feature point of the image captured immediately before or the image of the captured image stored in the travel history database 5, and the Using a known technique such as the 8-point method, the feature points are three-dimensionalized and added to a known feature point group.

In subsequent step S115, the captured image, the three-dimensional feature point group, and the vehicle position / posture acquired in steps S112 to S114 are stored in the travel history database 5. The travel history database 5 may store various vehicle information such as the vehicle speed and the steering angle calculated from the vehicle sensor group 9 in association with the vehicle position / posture.
In the subsequent step S116, a button is displayed on the display 6 as to whether or not to save the travel history, and it is determined whether or not there is an end instruction from the driver through the touch panel 7. If there is no end instruction from the driver, the process returns to step S113. On the other hand, if there is an end instruction, the travel history storing process is ended and then the process returns to a predetermined main program.

Next, the driving support control process will be described.
Prior to the driving support control process, it is assumed that the driving history saving process has already been performed a plurality of times.
First, in step S120, a button indicating whether to start running support is displayed on the display 6, and it is determined whether there is a start instruction from the driver through the touch panel 7. If there is a start instruction from the driver, the process proceeds to step S121. On the other hand, if a start instruction is given, an input from the driver is awaited.

In step S121, a three-dimensional feature point group is read from the travel history database 5, and a trajectory of a past vehicle position defined by a relative relationship with the three-dimensional feature point group is calculated as a reference route. Since a plurality of travel histories are stored in the travel history database 5, a plurality of reference routes are calculated.
In the subsequent step S122, using the current captured image of the camera 2, using the vehicle position detection unit 33 and the vehicle posture detection unit 34 described above, the relative relationship with the read three-dimensional feature point group, The initial vehicle position and vehicle attitude are detected.

In subsequent step S123, an introduction route is calculated for the plurality of reference routes calculated in step S122. Detailed procedures will be described later.
In subsequent step S124, a reference route having the shortest introduction route length calculated in step S123 is selected from a plurality of reference routes, and the introduction route and references after the point where the introduction route is connected (joined). A route combined with the route is generated as a driving support route.
In a succeeding step S125, a button indicating whether or not to perform automatic driving is displayed on the display 6, and it is determined whether or not there is an automatic driving instruction from the driver through the touch panel 7. If there is an automatic driving instruction from the driver, the process proceeds to step S126. On the other hand, if there is no automatic driving instruction, the process proceeds to step S128.

In step S126, automatic driving is performed based on the driving support route calculated in step S124. At this time, the vehicle speed is, for example, a constant vehicle speed, and the steering angle is controlled such that the vehicle position and posture change according to the travel support route. The vehicle controller 8 calculates an accelerator opening, a brake control amount, and a steering angle for realizing the calculation result, and sends a command value to the control unit group 10 to automatically drive the vehicle along the driving support route. .
In a succeeding step S127, it is determined whether or not the vehicle position has reached a certain range at the end of the driving support route. Here, if not reached, the process proceeds to step S125. On the other hand, if it reaches within a certain range at the end, the driving support process is terminated and the process returns to the predetermined main program.

  On the other hand, in step S128, driving assistance is performed based on the driving assistance route calculated in step S124. For example, after the captured image of the camera 2 is displayed on the display 6, the driving support route converted into the coordinate system of the camera 2 and the predicted locus based on the current steering angle are displayed in a superimposed manner so that the driver can travel. Support driving along the support route. As another driving support method, for example, EPS control may be performed based on the steering angle calculated in step S126, and steer steering guidance may be performed using a reaction force.

Next, the introduction route calculation process will be described.
FIG. 5 is a flowchart showing the introduction route calculation process.
FIG. 6 is a reference diagram for calculating the introduction route.
Here, a process for calculating the introduction route for a certain reference route will be described.
In FIG. 6, P0 indicates the initial vehicle position / orientation calculated in step S122, X0, X1,..., Xe indicate the position coordinate group constituting the reference route, X0 is the start point, and Xe is the end point. Represent.
First, in step S130, X0 which is the route start point of the reference route is set as a locus point (route point) Xi to be processed.
In the subsequent step S131, the arrival route from the initial vehicle position / posture P0 is calculated for the locus point Xi.
First, a generally known two-wheel model of a vehicle is configured on a simulator based on various vehicle data such as the minimum turning radius of the vehicle stored in the vehicle information database 4. The vehicle speed is constant in the forward direction, the steering angle is determined from the angle difference with Xi, the vehicle model is run on the simulation, and reaches within a predetermined distance of Xi, or a constant multiple of the distance to Xi (for example, 1. The simulation is performed by moving forward by a distance of (5 times).

This simulation is performed along a route that reaches the trajectory point Xi so that it is smoothly connected (joined) to the reference route, not a straight route that connects the vehicle position and the trajectory point Xi. That is, the simulation is performed so that the vehicle posture is substantially tangent to the reference route at the locus point Xi.
The route at this time is set as a simulation route. As a method for calculating the simulation route, any other method may be used as long as it is a calculation method that can be realized by the vehicle 1 from the initial own vehicle position / posture P0 to the route point Xi. .

In a succeeding step S132, it is determined whether or not the simulation route calculated in the step S132 has reached within a predetermined distance of Xi. Here, if it has reached within the predetermined distance of Xi, this simulation route is set as a temporary introduction route, and the process proceeds to step S134. On the other hand, if it does not reach within the predetermined distance of Xi, the process proceeds to step S133.
In step S133, the route point Xi + 1 next to Xi is set as the route point Xi to be processed, and then the process proceeds to step S131.
In step S134, the route length of the introduction route calculated in step S132 is calculated. The path length may be, for example, the sum of distances between points through which the vehicle model has passed in the simulation, or may be simply calculated as a linear distance between P0 and Xi.
After the processes in steps S130 to S134 are executed for all reference paths, the introduction path calculation process is terminated.

<Action>
If the driving route is constantly calculated from the beginning (from the beginning) based on the vehicle information such as the vehicle speed and the steering angle and the environment information around the host vehicle, the calculation cost increases.
Therefore, first, a travel history is stored in advance. That is, the 3D feature point group extracted from the camera image is recognized as the surrounding environment, and the vehicle position and the vehicle posture relative to the recognized 3D feature point group are detected (S112 to S114). These are associated and stored in the travel history database 5 (S115).

When running support is performed, a plurality of reference routes are set by referring to (reading) past driving histories based on the vehicle position (S121). Furthermore, based on the own vehicle position and the own vehicle attitude, respective introduction routes from the own vehicle position to each reference route are calculated (S123). Then, based on the calculation result of the introduction route, one reference route is selected, and the final travel route is set by combining this reference route and the introduction route up to this reference route (S124).
As described above, the present embodiment accumulates information that associates the surrounding environment, the vehicle position, and the vehicle attitude, and sets the travel route of the vehicle while referring to the information as needed. Therefore, the calculation of the travel route for driving support can be facilitated, and the calculation cost can be suppressed.

FIG. 7 is an example of a travel scene.
For example, it is assumed that there is an obstacle in front of the traveling direction.
At this time, first, a travel route that the vehicle can travel is read from the travel history database 5. Here, a reference path 1 that bypasses to the left of the obstacle and a reference path 2 that bypasses to the right of the obstacle are set. These reference paths 1 and 2 are travel loci that have traveled in the past and exist in the vicinity of the vehicle position.
Next, an introduction route 1 leading to the reference route 1 and an introduction route 2 leading to the reference route 2 are calculated based on the vehicle position and the vehicle posture. These are the shortest routes from the vehicle position to the smooth joining to the reference paths 1 and 2.

Then, the shortest introduction route is selected from among the introduction routes 1 and 2. Here, since the introduction route 2 is shorter than the introduction route 1, the reference route 2 corresponding to the introduction route 2 is selected. Then, a combination of the introduction route 2 and the reference route 2 is set as a final travel route.
In this way, a reasonable driving route that can be passed in a short time is set by calculating the shortest introduction route and selecting a reference route corresponding to the shortest of a plurality of introduction routes. can do.

In addition, since the introduction route is calculated in consideration of not only the vehicle position but also the vehicle posture, it should smoothly merge with the reference route and provide driving support along the reference route accurately. Can do. And the change of a vehicle behavior can be suppressed to the minimum by the joint of an introduction path | route and a reference path | route.
Further, by extracting the feature point group from the image data captured by the camera and making it three-dimensional, the surrounding environment is recognized, so it is possible to cope with a step, a road surface gradient, and the like.
As described above, when saving the travel history, the vehicle position / orientation is estimated using the feature points extracted by the camera, and a plurality of travel trajectories are stored for the constructed feature point group. A plurality of reference routes can be calculated by reading past driving histories.

In driving support, not only the vehicle position but also the vehicle posture is estimated for the read feature points, and it is optimally introduced as an auxiliary route to the reference route based on highly accurate vehicle position and posture information. Since a route is calculated and a final driving support route is calculated based on the calculated introduction route length, a route optimal for the current vehicle state can be calculated at a low calculation cost. Thereby, in the situation as shown in FIG. 7 described above, it is possible to calculate the reference route 2 that is considered intuitively reasonable by a person.
Memory-based driving support and automatic driving technology is a very effective method for driving in parking lot environments where the white line cannot be easily recognized and where the road is not easily calculated, or on general roads where the road boundary is ambiguous. In the first embodiment, since the camera information is used, it can be applied to a three-dimensionally complicated driving environment.

"effect"
As described above, the processing of the camera 2 and steps S111, S112, and S114 corresponds to “peripheral environment recognition means”, the processing of steps S113 and S122 corresponds to “own vehicle position detection means”, and the processing of steps S113 and S122. Corresponds to “own vehicle attitude detection means”. Further, the travel history database 5 and the process of step S115 correspond to “information storage means”, the process of step S121 corresponds to “reference route setting means”, and the process of step S123 corresponds to “introduction route calculation means”. The processing in step S124 corresponds to “reference route selection means”. Further, the processing in step S124 corresponds to “travel route setting means”.

(1) The driving support device recognizes the surrounding environment in the vehicle traveling direction based on the image captured by the camera 2, detects the vehicle position and the vehicle posture with respect to the recognized surrounding environment, and the surrounding environment and the vehicle position. , And information in which the vehicle posture is associated. Then, by referring to the information accumulated based on the vehicle position, a plurality of travel routes that the vehicle can travel are set as reference routes, and the vehicle position is determined based on the vehicle position and the vehicle attitude. The introduction route from each to the reference route is calculated. In accordance with the calculation result, one of the plurality of reference routes is selected, and the traveling route of the host vehicle is set by combining the selected reference route and the introduction route leading to the reference route.
In this way, information for associating the surrounding environment, own vehicle position, and own vehicle posture is accumulated, and the travel route of the own vehicle is set by referring to the information as needed for driving support. The calculation of the travel route can be facilitated, and the calculation cost can be suppressed.

(2) The travel support device accumulates the travel route information of the own vehicle associating the surrounding environment, the own vehicle position, and the own vehicle posture as a history, and the accumulated travel route information history based on the own vehicle position The plurality of travel route information is set as a plurality of reference routes.
In this way, the history of past travel route information is referred to based on the vehicle position and used as it is as a reference route, so the calculation time is shortened.
(3) The travel support device calculates the shortest introduction route from the vehicle position to each travel route, and selects a reference route corresponding to the shortest introduction route among the calculated introduction routes.
Thus, by selecting the reference route corresponding to the shortest introduction route, it is possible to set a reasonable travel route that can be passed in a short time.

(4) The driving support device sets a reference route including a point sequence in which the vehicle position and the vehicle posture are associated with each other, and is determined in advance from the vehicle position and the vehicle posture in the point sequence. An introduction route to a point that can be reached with a movement amount within the range and the movement amount is the shortest is calculated.
Thus, by calculating the shortest introduction route that can be realized, it is possible to set a rational travel route that can be passed in a short time.
(5) The driving support device includes a camera that captures the vehicle traveling direction, and recognizes the surrounding environment by extracting a feature point group from the image data captured by the camera and making it three-dimensional.
Thus, by extracting the feature point group from the image data captured by the camera and making it three-dimensional, the surrounding environment is recognized, so it is possible to deal with steps, road gradients, and the like.

<< Second Embodiment >>
"Constitution"
In the second embodiment, the laser radar 11 is used to recognize the surrounding environment.
That is, the second embodiment is the same as the first embodiment described above except that the laser radar 11 is used instead of the camera 2, and detailed description of the same parts is omitted.
FIG. 8 is a calculation block diagram of the second embodiment executed by the driving support device.
The driving support device 200 includes a laser radar 11 instead of the camera 2.

The laser radar 11 is generally called a laser range finder, and can calculate the distance to the target object directly by using a time-of-flight technique for measuring the arrival time of the laser beam. This is a two-dimensional distance acquisition device capable of acquiring distance information in the range of ° every 0.25 °. In addition, any apparatus may be used as long as it can acquire the distance to the target object two-dimensionally.
By using the laser radar 11 instead of the camera 2, the driving support controller 3 includes a map construction unit 39 instead of the feature point detection unit 31.

Hereinafter, elements different in configuration and processing contents from the first embodiment will be described.
The travel history database 5 is a storage unit that stores data acquired by the laser radar 11, information such as a vehicle position at the time of data acquisition, a vehicle posture, and a two-dimensional grid map formed by the map configuration unit 39 in association with each other. A storage medium such as a hard disk or a memory card may be used.
The data storage unit 32 includes data acquired by the laser radar 11, a host vehicle position estimated by the host vehicle position detection unit 33, a host vehicle posture estimated by the host vehicle posture detection unit 34, and a map construction unit 39. Are stored in the travel history database 5 in association with each other.
The own vehicle position detection unit 33, the own vehicle posture detection unit 34, and the map construction unit 39 are simultaneously executed using a generally known SLAM technique.

  More specifically, the data acquired by the laser radar 11 and the past two-dimensional grid map stored in the travel history database 5 are used to superimpose the data using a generally known ICP algorithm or the like. First, the vehicle position on the ground plane and the vehicle attitude indicating the direction, which are indicated by the relative relationship with the two-dimensional grid map, are calculated. Next, the map construction unit 39 reconstructs the two-dimensional grid map from the past two-dimensional grid map and the data acquired by the laser radar 11 and accumulates them in the travel history database 5.

The reference route setting unit 35 reads a plurality of past vehicle position loci accumulated in the travel history database 5 and sets them as reference routes. For example, if there are two driving histories in the past, two reference routes are calculated.
At this time, the two-dimensional grid map stored in the travel history database 5 is also read. The trajectory of the past vehicle position, which is the reference route, is defined by a relative relationship with the two-dimensional grid map.

The travel support unit 38 performs travel support using the travel support route uniquely determined by the reference route selection unit 37. Specifically, the generated travel support route information and the current vehicle position information are expressed in a bird's eye view and displayed on the display 6. Furthermore, the driving support unit 38 can control the accelerator, the brake, and the steering, and can perform automatic driving by sending the deviation of the driving locus and the position / posture of the vehicle to the vehicle controller 8.
The configuration other than the above is the same as that of the first embodiment.
Next, a processing procedure of the driving support device 200 will be described. Since the processing procedure in the second embodiment is substantially the same as that in the first embodiment, only the processing steps different from those in the first embodiment will be described using the flowcharts of FIGS.

First, the travel history storage process of FIG. 3 will be described.
The processing in steps S110 and S111 is the same as that in the first embodiment.
In step S112, as the travel history initialization process, the data acquired by the laser radar 11 is configured as the initial state of the two-dimensional grid map and stored in the travel history database 5.
In step S113, from the data acquired by the laser radar 11, using the above-described own vehicle position detecting unit 33 and the own vehicle posture detecting unit 34, the own vehicle position and the own vehicle are determined relative to the acquired two-dimensional grid map. Estimate the vehicle posture.

In step S114, based on the own vehicle position / posture information obtained in step S113, the map construction unit 39 is used to construct a two-dimensional grid map that is a map that constitutes the travel environment. Note that, unlike the first embodiment, the configured map is not three-dimensional but two-dimensional.
In step S115, the acquired data of the laser radar 11, the two-dimensional grid map, and the vehicle position / posture acquired in steps S112 to S114 are stored in the travel history database 5, and the process proceeds to step S116. The travel history database 5 may store various vehicle information such as the vehicle speed and the steering angle calculated from the vehicle sensor group 9 in association with the vehicle position / posture.
The process in step S116 is the same as that in the first embodiment.
This is the end of the travel history configuration process.

Next, the driving support control process of FIG. 4 will be described.
Prior to the driving support control process, it is assumed that the driving history saving process has already been performed a plurality of times.
The process of step S120 is the same as that of the first embodiment described above.
In step S121, the two-dimensional grid map is read from the travel history database 5, the past vehicle position trajectory defined by the relative relationship with the two-dimensional grid map is calculated as a reference route, and the process proceeds to step S122. Since a plurality of travel histories are stored in the travel history database 5, a plurality of reference routes are calculated.

In step S122, using the current acquisition data of the laser radar 11 and using the vehicle position detection unit 33 and the vehicle posture detection unit 34 described above, the initial relationship is determined based on the relative relationship with the read two-dimensional grid map. The vehicle position and vehicle attitude of the vehicle are estimated.
The processing from step S123 to step S127 is the same as in the first embodiment.
In step S128, driving assistance is performed based on the driving assistance route calculated in step S124. Specifically, the generated travel support route information and the current vehicle position information are expressed in a bird's eye view and displayed on the display 6. Further, the predicted trajectory based on the current steering angle is displayed in a superimposed manner to assist the driver in driving along the driving support route. As another driving support method, for example, EPS control may be performed based on the steering angle calculated in step S126, and steer steering guidance may be performed using a reaction force.
This is the end of the driving support process.

<Action>
In this embodiment, a two-dimensional grid map is constructed using a laser radar, a travel history is stored in a relative relationship with the two-dimensional grid map, and an optimal travel is selected from a plurality of route candidates based on the read travel history. Since the support route can be calculated, the amount of information to be stored as a travel history is smaller than when the camera is used with a camera. Also, the calculation cost can be kept low. In an environment where the runway is relatively flat and an obstacle exists at a position that can be recognized by a laser radar, a system that can be realized more simply than a camera can be constructed.
In the second embodiment, a two-dimensional grid map is constructed using a laser radar capable of obtaining two-dimensional data. However, a three-dimensional map is obtained using a three-dimensional laser scanner capable of directly obtaining three-dimensional distance information. It may be realized.

"effect"
From the above, the processing of the laser radar 11 and steps S111, S112, and S114 corresponds to “peripheral environment recognition means”. Other correspondence is the same as that of the first embodiment described above.
(1) The driving support device recognizes the surrounding environment in the vehicle traveling direction based on the two-dimensional grid map using the laser radar 11, detects the own vehicle position and the own vehicle posture with respect to the recognized surrounding environment, and these surrounding environments The information which matched the own vehicle position and the own vehicle attitude | position is accumulate | stored. Then, by referring to the information accumulated based on the vehicle position, a plurality of travel routes that the vehicle can travel are set as reference routes, and the vehicle position is determined based on the vehicle position and the vehicle attitude. The introduction route from each to the reference route is calculated. In accordance with the calculation result, one of the plurality of reference routes is selected, and the traveling route of the host vehicle is set by combining the selected reference route and the introduction route leading to the reference route.
In this way, information for associating the surrounding environment, own vehicle position, and own vehicle posture is accumulated, and the travel route of the own vehicle is set by referring to the information as needed for driving support. The calculation of the travel route can be facilitated, and the calculation cost can be suppressed.
Other functions and effects are the same as those of the first embodiment described above.

<< Third embodiment >>
"Constitution"
In the third embodiment, the travelable area of the host vehicle is calculated, and the narrow road degree of the travelable area is calculated.
In other words, the third embodiment is the same as the first embodiment described above except that a travelable area calculation unit 301, a narrow road degree calculation unit 302, and an introduction route availability determination unit 303 are newly added. Therefore, detailed description of the same part is omitted.
FIG. 9 is a calculation block diagram of the third embodiment executed by the driving support device.
The travel support apparatus 300 includes a travelable area calculation unit 301, a narrow road degree calculation unit 302, and an introduction route availability determination unit 303.

  The travelable area calculation unit 301 calculates a travelable area in which the vehicle can travel based on information such as the three-dimensional feature point group accumulated in the travel history database 5 and the vehicle width of the vehicle 1 accumulated in the vehicle information database 4. Calculate on the dimensional plane. First, a three-dimensional feature point group is projected onto a two-dimensional plane on the ground surface where the vehicle travels, and a two-dimensional grid map including information on whether or not an obstacle exists in each grid is constructed. Next, based on the vehicle width in the vehicle information database 4, the obstacle occupation area of the two-dimensional grid map is expanded by half the vehicle width. Then, on the two-dimensional grid map, an area that is not occupied by an obstacle is calculated as a travelable area.

FIG. 10 is a diagram illustrating an example of the travelable area.
The narrow road degree calculation unit 302 calculates the narrow road degree by comparing each of the plurality of reference routes set by the reference route setting unit 35 with the travelable region calculated by the travelable region calculation unit 301. First, for each route point on the reference route, the width within the travelable area is calculated. Here, the width indicates the length of the travelable area in a direction perpendicular to the route direction. Of all the route points, the width at the route point with the shortest width is calculated as the narrowness degree in the reference route.

The introduction route availability determination unit 303 compares the introduction route calculated by the introduction route calculation unit 36 with the travelable region calculated by the travelable region calculation unit 301 so that the introduction route exists in the travelable region. Determine whether or not.
The reference route selection unit 304 selects one reference route from a plurality of reference routes, and connects the introduction route corresponding to the reference route to calculate a final driving support route.

Specifically, first, based on the determination of whether or not the introduction route is travelable, which is calculated by the introduction route availability determination unit 303, a reference route on which the introduction route cannot travel is excluded. Next, a reference route having a narrow road degree larger than a certain threshold calculated by the narrow road degree calculation unit 302 is excluded. Then, among the remaining reference routes, the route having the shortest introduction route is calculated as the final reference route.
In addition to the above, as a reference route selection method, a method of performing a comprehensive evaluation by weighting based on the narrowness degree of the reference route and the route length of the introduction route is also conceivable.
The configuration other than the above is the same as that of the first embodiment.

Next, a processing procedure of the driving support device 300 will be described. The processing procedure in the third embodiment is substantially the same as that in the first embodiment, and only step S124 in the driving support control process is different. Therefore, only the processing in step S124 will be described.
In step S124, first, the travelable area calculation unit 301 calculates a travelable area on a two-dimensional plane. Next, based on the calculated travelable area, the narrow road degree calculation unit 302 calculates the narrow road degree of each reference route, and the introduction route availability determination unit 303 determines whether the introduction route can travel. Then, based on the narrow road degree and the result of whether or not the introduction route can be traveled, the reference route selection unit 304 calculates a final travel support route. The details of each processing unit are as described above.

<Action>
In the present embodiment, the travelable area is calculated from the three-dimensional feature points calculated by the camera, the narrowness degree of the reference route and the travelability determination of the introduction route are performed, and the travel support route is selected based on the result, so Robust and highly feasible route selection is possible than in the first embodiment.
FIG. 11 is a diagram illustrating reference paths having different degrees of narrow paths.
For example, it is assumed that the reference path 1 that detours to the left side of the obstacle is not a narrow path, whereas the reference path 2 that detours to the right side of the obstacle is a narrow path. Comparing the introduction path 1 leading to the reference path 1 and the introduction path 2 leading to the reference path 2, the introduction path 2 is shorter, so in the first embodiment described above, the reference becomes a narrow path. Although the route 2 is selected, in this embodiment, the reference route 1 can be selected while avoiding a narrow road.

Since the driving support route is composed entirely of past driving histories, the car should not be able to pass through.However, for example, when the driving support route is used in the reverse direction, the driving support route is opposite due to the vehicle characteristics such as the inner wheel difference. There may be a state where the vehicle cannot travel in the direction. In addition, there is an error in the route trace at the time of driving support, so it is desirable to avoid going too narrow.
Therefore, under the environment as described above, the route selection according to the present embodiment in consideration of the narrow road degree is very effective.

"effect"
As described above, the travelable area calculation unit 301 corresponds to the “runnable area calculation unit”, the narrow road degree calculation unit 302 corresponds to the “narrow road degree calculation unit”, and the introduction route availability determination unit 303 sets the “introduction route availability”. Corresponds to “determination means”. Other correspondence is the same as that of the first embodiment described above.
(1) The travel support device calculates the travelable area of the host vehicle based on the information accumulated in the travel history database 5, and determines the narrowness degree of the travelable area based on the travelable area and the reference route. calculate. Then, one of a plurality of reference routes is selected according to the calculation result of the introduction route and the calculation result of the narrow road degree.
As described above, by calculating the travelable area, it is possible to recognize the surrounding environment on the travel route including the place that cannot be seen from the own vehicle position, although it is in the past. Further, by calculating the degree of narrow path, it is possible to select a reference path having a large space and a high margin.

(2) The driving support device determines whether or not the calculated introduction route is in the travelable region, and a plurality of values are determined according to the introduction route calculation result, the narrow road degree calculation result, and the introduction route availability determination result. One of the reference paths is selected.
As described above, by determining whether or not the introduction route is in the travelable region, it is possible to exclude the introduction route outside the travelable region and set a travel route that is easy to pass.

<< 4th embodiment >>
"Constitution"
In the fourth embodiment, various types of information are exchanged by communication with the outside of the vehicle.
That is, in the fourth embodiment, the first information described above is used except that the route information management device 401 installed outside the vehicle is used instead of the travel history database 5 and the own vehicle position positioning unit 404 is added. Since it is the same as that of embodiment, detailed description about the same part is abbreviate | omitted.

FIG. 12 is a calculation block diagram of the fourth embodiment executed by the driving support device.
The travel support device 400 includes a route information management device 401 having a received data storage unit 402 and a stored data selection unit 403, and a host vehicle position measurement unit 404, instead of the travel history database 5.
The data storage unit 32 is estimated by the captured image of the camera 2, the feature points obtained by the feature point detection unit, the host vehicle position estimated by the host vehicle position detection unit 33, and the host vehicle posture detection unit 34. The vehicle posture is associated and sent to the route information management device 401. Further, as necessary, information such as a captured image, a trajectory of the vehicle position, and a three-dimensional feature point group is received from the route information management device 401 and sent to each processing unit.

The route information management device 401 is a so-called server that is installed outside the vehicle and includes a communication unit that can wirelessly communicate with a plurality of vehicles, and includes a received data storage unit 402 and a stored data selection unit 403. Note that the route information management device 401 may be mounted on a single vehicle as long as it can communicate with a plurality of vehicles.
The reception data storage unit 402 is a storage unit that receives and stores information, such as a captured image captured by the camera, a camera position at the time of imaging, a vehicle position, and a vehicle posture, sent from the data storage unit 32 from a plurality of vehicles. For example, a storage medium such as a hard disk or a memory card may be used.

The accumulated data selection unit 403 receives the vehicle position calculated by the vehicle position positioning unit 404, selects a data group suitable for the vehicle position of the vehicle from the accumulated data group based on the vehicle position, It is sent to the driving support controller 3.
The own vehicle position measuring unit 404 is information indicating an approximate own vehicle position output from, for example, a navigation system mounted on a general vehicle, and combines the GPS information and the road information into the own vehicle position.
The configuration other than the above is the same as that of the first embodiment.
Next, a processing procedure of the driving support device 400 will be described. Since the processing procedure in the fourth embodiment is almost the same as that in the first embodiment, only the steps different from those in the first embodiment will be described using the notations in FIGS. 3 and 4.

First, the travel history storage process of FIG. 3 will be described.
In step S111, it is confirmed whether or not a past travel history exists in the route information management device 401 through the communication unit. When there is no travel history, the process proceeds to step S112, and when there is a travel history, the process proceeds to step S113.
In step S115, the captured image, the three-dimensional feature point group, and the vehicle position / posture acquired in steps S112 to S114 are sent to the route information management device 401 through the communication unit, and the process proceeds to step S116. In the route information management device 401, the received data storage unit 402 stores the received data.

Next, the driving support control process of FIG. 4 will be described.
In step S121, first, the own vehicle position is estimated by the own vehicle position measurement unit 404, and the estimation result is sent to the route information management device 401 through the communication unit. In the route information management device 401, the accumulated data selection unit 403 passes through the three-dimensional feature points around the vehicle position and within the threshold range of the vehicle position from all the data based on the sent vehicle position. The trajectory of the past vehicle position to be selected is selected and sent to the travel support controller 3. Finally, the reference route setting unit 35 calculates a plurality of sent travel tracks as reference routes, and the process proceeds to step S122.
Other processing steps are substantially the same as those in the first embodiment, and thus are omitted.

<Action>
In the present embodiment, only data with a close vehicle position is selected from all travel history data, so even if the number of data increases enormously, only rationally relevant routes can be selected, and calculation processing costs can be reduced. Can be suppressed.
In addition, since information is exchanged with the route information management device 401, it is possible to collect travel history information from a plurality of vehicles, so even if the route has never passed the vehicle, If another vehicle has already passed, driving support is possible, and even a route on which the vehicle has never traveled can be traveled.

"effect"
From the above, the own vehicle position positioning unit 404 corresponds to the “own vehicle position positioning unit”, and the communication unit of the route information management device 401 corresponds to the “communication unit”. Other correspondence is the same as that of the first embodiment described above.
(1) The driving support device is included in the navigation system, and includes a host vehicle position positioning unit 404 that measures the host vehicle position. The host vehicle position measured by the host vehicle position positioning unit 404 and the host vehicle position detection unit 33 Based on the detected own vehicle position, a plurality of travel routes that can be traveled by the host vehicle are set as reference routes by referring to past travel histories.
In this way, candidates are extracted based on the vehicle position obtained from the navigation system, and necessary information can be narrowed down. Therefore, even if the accumulated data increases, the calculation cost can be stably suppressed.

(2) The route information management device 401 includes a communication unit that is provided outside the host vehicle and transmits / receives information to / from a plurality of vehicles including the host vehicle.
As described above, since reception data from a plurality of vehicles can be accumulated, it is possible to refer to traveling histories of other vehicles. That is, driving support can be implemented even on a route on which the host vehicle has never traveled.

DESCRIPTION OF SYMBOLS 1 Vehicle 2 Camera 3 Driving support controller 4 Vehicle information database 5 Driving history database 6 Display 7 Touch panel 8 Vehicle controller 9 Vehicle sensor group 10 Control unit group 11 Laser radar 31 Feature point detection unit 32 Data storage unit 33 Own vehicle position detection unit 34 Self-vehicle posture detection unit 35 Reference route setting unit 36 Introduction route calculation unit 37 Reference route selection unit 38 Travel support unit 39 Map construction unit 100 Travel support device 200 Travel support device 300 Travel support device 301 Travelable region calculation unit 302 Narrow road degree Calculation unit 303 Introduced route availability determination unit 304 Reference route selection unit 400 Driving support device 401 Route information management device 402 Received data storage unit 403 Stored data selection unit 404 Own vehicle position measurement unit

Claims (10)

A surrounding environment recognition means for recognizing the surrounding environment in the vehicle traveling direction;
Vehicle position detection means for detecting the vehicle position relative to the surrounding environment recognized by the recognition means;
Own vehicle posture detection means for detecting the own vehicle posture relative to the surrounding environment recognized by the recognition means;
Information accumulating means for accumulating information associated with the surrounding environment recognized by the surrounding environment recognition means, the own vehicle position detected by the own vehicle position detection means, and the own vehicle attitude detected by the own vehicle attitude detection means;
Reference route setting for setting, as a reference route, a plurality of travel routes that can be traveled for the host vehicle by referring to information stored in the information storage unit based on the host vehicle position detected by the host vehicle position detection unit Means,
Introduction from the own vehicle position to each reference route set by the reference route setting unit based on the own vehicle position detected by the own vehicle position detection unit and the own vehicle posture detected by the own vehicle posture detection unit An introductory route calculating means for calculating a route;
A reference route selection unit that selects one of a plurality of reference routes set by the reference route setting unit according to a calculation result by the introduction route calculation unit;
A travel route setting means for setting the travel route of the host vehicle by combining the one reference route selected by the reference route selection means and the introduction route leading to the reference route ;
The introduction route calculation means calculates the shortest introduction route from the vehicle position to each travel route,
The reference route selection unit selects a reference route corresponding to the shortest introduction route among the introduction routes calculated by the introduction route calculation unit. A surrounding environment recognition means for recognizing the surrounding environment in the vehicle traveling direction;
Vehicle position detection means for detecting the vehicle position relative to the surrounding environment recognized by the recognition means;
Own vehicle posture detection means for detecting the own vehicle posture relative to the surrounding environment recognized by the recognition means;
Information accumulating means for accumulating information associated with the surrounding environment recognized by the surrounding environment recognition means, the own vehicle position detected by the own vehicle position detection means, and the own vehicle attitude detected by the own vehicle attitude detection means;
Reference route setting for setting, as a reference route, a plurality of travel routes that can be traveled for the host vehicle by referring to information stored in the information storage unit based on the host vehicle position detected by the host vehicle position detection unit Means,
Introduction from the own vehicle position to each reference route set by the reference route setting unit based on the own vehicle position detected by the own vehicle position detection unit and the own vehicle posture detected by the own vehicle posture detection unit An introductory route calculating means for calculating a route;
A reference route selection unit that selects one of a plurality of reference routes set by the reference route setting unit according to a calculation result by the introduction route calculation unit;
A travel route setting means for setting the travel route of the host vehicle by combining the one reference route selected by the reference route selection means and the introduction route leading to the reference route;
Based on the information stored in the information storage means, a travelable area calculation means for calculating a travelable area of the host vehicle;
A narrow road degree calculating means for calculating a narrow road degree of the travelable area based on the travel possible area calculated in the travelable area and the reference route set by the reference route setting means ,
The reference route selection unit selects one of a plurality of reference routes set by the reference route setting unit according to a calculation result by the introduction route calculation unit and a calculation result by the narrow path degree calculation unit. A driving support device characterized by that. The information accumulating unit is configured to associate the surrounding environment recognized by the surrounding environment recognizing unit, the own vehicle position detected by the own vehicle position detecting unit, and the own vehicle posture detected by the own vehicle posture detecting unit. Accumulate travel route information as history,
The reference route setting means refers to a history of travel route information stored in the information storage means based on the vehicle position detected by the vehicle position detection means, and refers to the plurality of travel route information. drive assist system according to claim 1 or 2, characterized in that to set as a reference path. The introduction route calculation means calculates the shortest introduction route from the vehicle position to each travel route,
The driving support apparatus according to claim 2 or 3 , wherein the reference route selection unit selects a reference route corresponding to the shortest introduction route among the introduction routes calculated by the introduction route calculation unit. The reference route setting means sets a reference route composed of a point sequence in which the own vehicle position detected by the own vehicle position detection means and the own vehicle posture detected by the own vehicle posture detection means are associated with each other,
The introduction route calculation means calculates an introduction route from the point sequence to a point that is reachable with a movement amount within a predetermined range from the vehicle position and the vehicle posture, and the movement amount is the shortest. The travel support device according to any one of claims 1 to 4 , wherein the travel support device is calculated. Based on the information stored in the information storage means, a travelable area calculation means for calculating a travelable area of the host vehicle;
A narrow road degree calculating means for calculating a narrow road degree of the travelable area based on the travel possible area calculated in the travelable area and the reference route set by the reference route setting means,
The reference route selection unit selects one of a plurality of reference routes set by the reference route setting unit according to a calculation result by the introduction route calculation unit and a calculation result by the narrow path degree calculation unit. The travel support apparatus according to any one of claims 1 to 3 to 5 , wherein An introduction route availability determination unit that determines whether the introduction route calculated by the introduction route calculation unit is in the travelable region calculated by the travelable region calculation unit;
The reference route selection means has a plurality of references set by the reference route setting means according to a calculation result by the introduction route calculation means, a calculation result by the narrow road degree calculation means, and a determination result by the introduction route availability determination means. The driving support device according to claim 2 or 6 , wherein one of the routes is selected. It is included in the navigation system and has own vehicle position measuring means for measuring the own vehicle position.
The reference route setting means refers to the information accumulated in the information accumulation means based on the own vehicle position measured by the own vehicle position positioning means and the own vehicle position detected by the own vehicle position detection means. drive assist device according to any one of claim 1 to 7, characterized in that to set as a reference path a plurality of travel route runnable for the vehicle. The said information storage means is provided in the exterior of the own vehicle, and is provided with the communication means which transmits / receives the said information between several vehicles including the own vehicle, The any one of Claims 1-8 characterized by the above-mentioned. The driving support device according to item. The surrounding environment recognition unit includes a camera that captures a vehicle traveling direction, and recognizes the surrounding environment by extracting a feature point group from the image data captured by the camera and making it three-dimensional. The travel assistance apparatus according to any one of 1 to 9 .

Images