JP2969175B1 - Main Line Traffic Flow Prediction Method for Merging Control System of Driving Support Road System - Google Patents

Abstract

Abstract: PROBLEM TO BE SOLVED: To provide a self-driving vehicle that can safely and smoothly control a self-driving vehicle even in a structure similar to a converging part of a conventional Japanese expressway having a short acceleration lane with respect to automatic driving of a vehicle in a driving support road system. A method for predicting main traffic flow in a system is provided. SOLUTION: A roadside detector 107 and a vehicle detector 113 for detecting a state of a running road 101, 102, a roadside control device 108,
In the driving support road system including the road-vehicle communication device and the road-vehicle / vehicle-to-vehicle communication device and the control device 112 of the vehicle, the processing device of the road-to-vehicle communication performs from the present time of the traveling vehicle in the confluence area until after a certain time. As a method of predicting the position and speed of the roadside control device, the prediction is performed in consideration of the fact that the driver recognizes the position and speed of other surrounding vehicles within the driver's visual recognition range and changes the lane to an adjacent lane. Use the prediction method in.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic driving system for a vehicle, and more particularly to a prediction method for predicting traffic flow on a main line at the time of merging.

[0002]

2. Description of the Related Art In recent years, research on automatic driving systems for vehicles has been advanced in the United States, Japan, and the like, and experiments using actual vehicles have been conducted. In order to automatically drive a vehicle, it is necessary for the system to detect the position and speed of surrounding vehicles, predict the state from the present time to the future, and create a travel plan based on the predicted state.

[0003] Regarding a merging control system by automatic driving of a vehicle, Bin Ran, Shawn Leight, Rnoda R. et al. Hu, Seth
Johnson, “Merging Process Analysis for an Automat
edHighway System ”, The 3rd ITS World Congress, Ju
ne 26, 1996, etc., describe the basic idea.
Further, Japanese Patent Application Laid-Open No. 3-118698 describes a case of merging in a mobile vehicle system that predicts the traveling of another vehicle and controls the traveling of the own vehicle. However, the prediction method here does not make predictions regarding lane changes of vehicles, but makes predictions of time and distance within the same lane.

[0004]

The basic concept of an automatic driving system for a vehicle has been studied so far.
Until its practical use, road conditions peculiar to Japanese highways, for example, the acceleration lane length in the merging area is often short, so the speed of the merging vehicle is adjusted in advance before the acceleration lane, and the acceleration is shortened. There is a problem that it is necessary to surely change lanes between vehicles traveling on the main lane in the lane, and specific means for these are required. For that purpose, it is necessary to predict the behavior of the traveling vehicle on the main line up to the near future and adjust the speed of the merging vehicle based on the prediction result, but with the conventional prediction method, it is traveling on the main line As a behavior that a vehicle can take by recognizing a merging vehicle that changes lanes to its own traveling lane, only acceleration / deceleration in the traveling direction is considered. However, in actual traffic, the driver often recognizes the lane change vehicle that is joining and changes the lane to avoid it, and the merging vehicle enters the vacant space. This prevents a decrease in the handling capacity at the junction. On the other hand, in the conventional method, since the lane change of the main lane was not taken into consideration, the merging vehicle actually escaped the space where the lane could be changed to the main lane, and a smooth Control is lost.

[0005] It is an object of the present invention to make the most of the space in which the merging vehicle can change lanes to the main line, and to smoothly control the automatic driving vehicle at the merging section.

[0006]

In order to predict the lane change of a manually driven vehicle, first, a roadside controller uses a roadside detector to ascertain the position and speed of all traveling vehicles. Next, in order to predict the behavior of the manually driven vehicle, all the vehicles recognized by the driver of the vehicle to be predicted are extracted, and from the information such as the position and speed of the vehicle, the vehicle to be predicted is extracted. It is determined whether or not to change lanes, and at the same time, a time and a position data sequence up to a certain time later are predicted. The above prediction is performed for all traveling vehicles. The roadside control device uses the road-vehicle communication based on these prediction results to instruct the self-driving vehicle merging with the main line on speed adjustment for smoothly changing lanes to the main line. The merged self-driving vehicle travels by creating a travel plan based on the roadside instruction and information from a vehicle-mounted detector. When changing lanes from a merging lane to a main lane, the on-board detector checks the space where the lane change is to be performed, and if the vehicle control device determines that it is safe, the lane change is performed. If it is determined that lane change is not possible, while adjusting the speed, the speed is adjusted in order to merge between vehicles that can merge behind and the lane is changed. As described above, by making predictions in consideration of lane changes of manually driven vehicles, it is possible to merge smoothly without missing a space where lanes can be changed, compared to prediction without considering lane changes become.

Next, a description will be given of the above processing.
Roadside detectors (cameras, etc.) can measure the passing time, position, speed, and vehicle type (large / normal) of an autonomous / manual driving vehicle traveling at a location on the road where the detector is located,
The roadside controller receives the information from the detector,
A prediction is made regarding the position and speed of a traveling vehicle (autonomous driving vehicle, manual driving vehicle) from the present to a certain time later.

Next, in order to predict the behavior of a manually driven vehicle on the main line, first, the passage time, position, speed, and type of vehicle (large / normal) are detected from a plurality of detectors installed on the roadside. Save the data. At this time, the detected vehicle is managed by attaching a vehicle ID, and matching with the past data is performed. In addition, the measurement data of the time and position for each vehicle may not be measured at the time of prediction (current time) depending on the installation interval of the detector, the measurable range in detection, and the measurement cycle. In this case, the running trajectory up to the present is complemented by linear approximation, etc., and for one vehicle,
It is stored in the form of a data string of vehicle ID, vehicle type, time and position up to the present, and lane number.

Next, for the vehicle to be predicted, a fan-shaped visual field in the forward direction of the driver having a radius R [m] centered on the driver's position and a center angle θ is set. As the values of R and θ, a driving simulator, experimental data using an actual vehicle, and the like are used. When setting the field of view, consider the increase or decrease of the field of view due to the height of the cab depending on the type of vehicle such as large / normal. Further, in addition to the front view, the left and right side views of the driver, the view by the rearview mirror, and the view by the left and right side mirrors are similarly set.

Next, the data (position / speed / vehicle type) of all the driver-recognizing vehicles in the visual field in front of the driver of the vehicle to be predicted set in consideration of the blind spot are set as described above. From the detector data. When a vehicle recognized by the driver is extracted from the set field of view, a blind spot of the driver due to another vehicle in front and a blind spot of the driver due to a road structure such as a side wall are taken into consideration. From these data, prediction is made regarding whether or not to change the lane of the vehicle to be predicted (whether or not lane change is to be performed) and prediction regarding the traveling direction (a sequence of time and position data until after a certain time).

The prediction regarding the lane change is performed as follows. First, a vehicle to be predicted (hereinafter, referred to as own vehicle).
Among the driver-recognized vehicles in the forward visual field of the driver at the position XO and the speed VO, a vehicle closest to the same lane as the own vehicle (hereinafter, referred to as a vehicle A. The position XA and the speed VA) is represented by F (XO). , XA, VO, VA), and a vehicle that always changes lanes ahead of the own vehicle among other vehicles adjacent to the own vehicle (for example, a vehicle running in an acceleration lane where the number of vehicle types decreases). Hereinafter, it is referred to as vehicle B. Position XB, speed VB)
When a certain condition is satisfied with respect to the output of the function of F ′ (XO, XB, VO, VB), the motive of the own vehicle to change lanes to the adjacent lane (slower preceding vehicle, forward interruption) It is determined that there is a car). Functions F, F '
If the conditions do not satisfy the conditions, and if the vehicles A and B do not exist, it is determined that the own vehicle does not change lanes, and a prediction (a data sequence of time and position until a certain time later) regarding the traveling direction is determined. Do.

Next, regarding the vehicle in which the motive to change lanes exists, the field of view of the left and right sides and the rear (back mirror, left and right side mirrors) of the driver of the own vehicle is set, and other vehicles and road structures are set. A driver-recognized vehicle within the range of the driver's field of view is extracted in consideration of the blind spot caused by the vehicle.

Next, the driver-recognizing vehicle in the field of view of the left and right sides, the left and right side mirrors, and the rearview mirror, and the driver-recognizing vehicle in the above-described front field of view, before and after the space where the lane change is to be performed. The vehicle is searched to determine whether or not the lane change is possible. When there is no vehicle before and after the space where the lane change is to be performed, it is determined that the lane change is to be performed and prediction is performed. If a vehicle exists before and after the space where the lane change is to be performed, the vehicle in front is moved to the vehicle C (position X).
G (XO, XC, XD, VO, VC, V) with the vehicle behind (vehicle D (position XD, speed VD))
When a certain condition is satisfied with respect to the output of the function D), it is determined that a lane change is to be performed, and prediction is performed. In other cases, it is determined that the lane change is not performed, and the prediction regarding the traveling direction is performed. When the lane change is possible for both the left and right lanes, it is determined that the lane change is to be performed for the right lane, and the prediction is performed.

The prediction regarding the traveling direction is performed as follows. When it is determined that the lane change is not to be performed, first, among the other vehicles in the visual field of the driver of the own vehicle (position XO, speed VO), the closest vehicle A (position XA, position XA, A speed VA) or a vehicle B (position XB, speed VB) that changes lanes ahead of the own vehicle is extracted. H (X
O, XA, VO, VA) or H (XO, XB, VO,
If the output of the function of VB) satisfies a certain criterion, it is determined that the own vehicle is in danger of hitting the vehicle A or B, and the speed of the own vehicle is changed to the speed VA of the vehicle A or the speed of the vehicle B. Assuming that the vehicle decelerates to the speed VB, a data sequence of time and distance until a certain time later (hereinafter referred to as a predicted time-distance curve) is created, and the prediction is made regarding the traveling direction. In other cases, a predicted time-distance curve is created assuming that the own vehicle will continue to travel at the same speed in the future, and the prediction is made regarding the traveling direction.

If it is determined that the lane change is to be performed, the output of the function of H (XO, XC, VO, VC) is given when the preceding vehicle (vehicle C) exists in the space to which the lane is changed. If a certain criterion is satisfied, prediction is made on the traveling direction on the assumption that the vehicle decelerates to the speed of the vehicle C. If vehicle C does not exist, or H (X
If the output of the function (O, XC, VO, VC) is other than the above, the prediction of the traveling direction is performed on the assumption that the vehicle travels at the same speed as before.

The roadside control device creates an instruction necessary for the automatic driving vehicle to adjust the speed based on the above prediction result, and transmits the instruction content to the roadside-to-vehicle wireless installed on both the roadside and the vehicle. Using a communication device, the information is transmitted from the roadside to the automatic driving vehicle by wireless communication. The instruction from the road side is given in the form of time and speed with a margin for the self-driving vehicle to pass the target point.

In the vehicle, a travel plan (position and speed from the present time to a certain time later) of the own vehicle is created in the vehicle control device based on the above-mentioned instruction from the road side for speed adjustment. At this time, since the instruction from the road side has a margin, the current acceleration performance of the own vehicle,
The traveling plan of the own vehicle is created in consideration of the vehicle condition, the traffic condition around the own vehicle measured by the on-board detector, the user's requirements, and the like. The self-driving vehicle travels by using a feedback function such as PID control or the like to determine various control amounts for realizing the determined travel plan.

Finally, a description will be given of a case where the prediction of the lane change of the vehicle traveling on the main line by the roadside control device is incorrect. Predicted when the merged autonomous vehicle will confirm that the vehicle is in the space where the roadside control device instructed the lane change, and that lane change is not possible, in a place where the main lane traveling vehicle can be grasped by the on-board detector Is removed.
The main line vehicle existing in the space where the lane of the merging vehicle is to be changed recognizes the merging vehicle merging ahead of the lane,
It is a vehicle whose roadside is predicted to change lanes to the adjacent lane so as to avoid the merged vehicle, and in fact, the vehicle has gone straight without changing lanes. In this case, the on-vehicle control device searches the inter-vehicles behind the space designated by the road-side control device using the on-vehicle detector, and adjusts the speed for changing lanes between the vehicles. According to this method, the lane can be changed between the vehicles on the main lane designated by the conventional prediction method in which the prediction of the lane change is not performed. Therefore, compared to the conventional prediction method of estimating the time and distance up to a certain time within the same lane without considering the lane change of the main line traveling vehicle, the safety is not reduced as compared with the conventional method at the same time, and The smoothness is improved over the conventional method.

According to the prediction method of the roadside control device, the instruction for the automatic driving vehicle on the roadside based on the prediction result, and the traveling plan of the automatic driving vehicle based on the instruction from the roadside and the information detected by the automatic driving vehicle, the automatic driving is performed. The vehicle can change lanes to the main line more smoothly than when the conventional prediction method is used, and the traffic flow at the junction is more smoothly controlled.

[0020]

Embodiments of the present invention will be described below. First, an outline of the present invention will be described with reference to FIG. In FIG. 1, reference numeral 101 denotes a main line portion of three lanes on one side of an expressway.
The self-driving vehicle 103 (hereinafter, the self-driving vehicle is referred to as an “AHS vehicle”), the self-driving vehicle 104 (hereinafter, the self-driving vehicle is referred to as
S vehicle). Reference numeral 102 denotes a one-lane merging channel that merges with the main line, and AH that is about to merge with the main line.
An S vehicle 105 (hereinafter, referred to as a merged AHS vehicle) is running. In order to smoothly change lanes to the main lane 101 in the acceleration lane 106, it is necessary to appropriately adjust the speed of the merging AHS vehicle 105 at the part of the acceleration lane 106 and at the part of the merging passage 102 just before the acceleration lane 106. . The merging AHS vehicle 105 can measure only the traveling vehicles around the own vehicle in the merging flow path 102, and the acceleration lane 106
Since vehicles traveling on the main line 101 upstream cannot be measured, the roadside processes information from the roadside measuring device and
In order for the HS vehicle to perform appropriate speed adjustment, it is necessary to give an instruction to the merging AHS vehicle.

The roadside running vehicle grasping device 107 detects the time when vehicles (AHS vehicles, non-AHS vehicles) traveling on the main line pass.
The position / speed and the vehicle type (large / normal) can be measured, and the information is sent to the roadside junction roadside controller 108 for processing. First, in the traveling vehicle prediction unit 109,
For each calculated vehicle, a time and a position from a present time to a certain time later are predicted. The prediction of the traveling direction of the vehicle is given by a data sequence of the time and the distance of the vehicle, which is hereinafter referred to as a “time distance curve”. The time distance curve is 110
As shown in FIG. 7, a graph can be represented with the horizontal axis representing time and the vertical axis representing distance. For more information about the time distance curve, see
It will be described later using 12. As for the prediction of lane change, a lane number is given on the time-distance curve as shown by 110.

On the basis of the prediction at 109, the vehicle-to-vehicle instruction creating unit 111 determines which of the vehicles running on the main line should be joined by the AHS vehicle, and merges between the vehicles on the main line. Of the instruction to be given to the merging AHS vehicle, which is necessary when performing speed adjustment for the vehicle, is created. The content of the instruction is given in the form of time and speed with a margin, passing through the target point (the end point of the acceleration lane 106).

The contents of the instruction for the merged AHS vehicle are transmitted to the merged AHS vehicle by road-vehicle communication, and processed by the merged AHS vehicle control device 112. In addition to the given instruction contents, the current position, speed, and acceleration performance of the own vehicle, measured by the own vehicle measuring device 114, such as the position and speed of the surrounding traveling vehicle, measured by the on-board measuring device 113 such as a camera and a radar. (Riding staff, tire wear, etc.
Taking into account the demands from the driver 115 (acceleration, maximum speed, ride comfort, etc.)
In the time distance curve creating section 116, a travel plan of the own vehicle is created. The travel plan of the own vehicle is represented in the form of a time-distance curve, and the merging vehicle travels using feedback functions such as PID control and the like to determine various control amounts for realizing the determined time curve.

The above processing is performed when the merged AHS vehicle is in the acceleration lane 10
Repeat until you reach 6. When the merged vehicle reaches the acceleration lane 106, the position and speed of the vehicle measured by the in-vehicle measuring device 113 of the vehicle are acquired while referring to the information on the traveling vehicle from the road side, and the vehicle safely moves to the main lane. Change lanes.

Next, referring to FIG. 2, the overall configuration of a system which is a premise for describing the present invention will be described. FIG. 2 illustrates the position of the merge control system in the driving support road system. Reference numeral 201 denotes a main lane portion of three lanes on one side of a highway, and reference numeral 202 denotes a merging passage merging with the main lane. The number of lanes of the merging passage is one. The self-driving vehicle (merging AHS vehicle) 203 that has traveled on the merging flow path merges with the lane on the leftmost side of the main lane 201 at the acceleration lane 204. Under a driving support road system central control device 205 (AHS center) for managing and controlling the entire driving support road system, a merging unit roadside control device 207 for managing and controlling vehicles traveling in the merging unit 206; Section, a short-path road-side control device 209 and the like for managing and controlling the vehicle traveling in the portion 208 other than the branching portion.Each of them manages the vehicle under the control of the central control device 205, and also communicates information with the adjacent control device. Interact.

Next, the configuration of the merging unit roadside control device will be described with reference to FIG. FIG. 3 is a diagram showing the merging unit roadside control device 20 of FIG.
7 is a detailed diagram of the configuration of FIG. Road-vehicle communication control unit 301
Performs bidirectional communication with the self-driving vehicle 303 using the road-vehicle communication device 302 using a leaky coaxial cable (LCX) or the like. The self-driving vehicle 303 detects the position of the lane to travel using the position detection support device 304 such as a magnetic nail embedded in the road. The traveling vehicle grasping unit 305 detects a passing time, a position, a speed, and the like of the traveling vehicle (automatic driving vehicle, manually driving vehicle) by using a traveling vehicle grasping device 306 such as a camera and a radar installed near the road. The environment grasping unit 311 performs measurement with the environment grasping device 310. Understand environmental information such as road surface conditions, weather conditions, and obstacle information. Road-to-road communication controller 309
Performs communication of instructions and information with the central control device 205 in FIG. 2, and exchanges information and the like with an adjacent device such as the single road unit roadside control device 209 in FIG.

The traveling vehicle prediction unit 307 includes a road-to-vehicle communication control unit 3
01 and traveling vehicle data from the traveling vehicle grasping unit 305, road surface condition, weather condition, obstacle information data of the environment grasping unit 311,
In addition, using road shape information and the like stored in advance,
The position and speed values of all vehicles traveling at the junction from the present to a certain time later are predicted (predicted time-distance curve creation). The vehicle-to-vehicle instruction creating unit 308 uses information of the traveling vehicle prediction unit 307, the road-to-road communication control unit 309, the environment grasping unit 311 and the like,
The instruction content given to the automatic driving vehicle 303 is created and transmitted to the road-vehicle communication control unit 301. In addition, the structure of the single road unit roadside controller 209 in FIG. 2 other than the merging unit roadside controller can be considered in the same manner as the structure of the merging unit roadside controller in FIG.

Next, the configuration of the self-driving vehicle will be described with reference to FIG. FIG. 4 is a detailed view of the self-driving vehicle 303 of FIG. The communication control unit 401 includes an on-vehicle road-to-vehicle communication device 402 that performs bidirectional data communication with a road-to-vehicle communication device 302 such as a leaky coaxial cable on the roadside, and an inter-vehicle communication that performs bidirectional data communication with another autonomous vehicle 403. Process information from the device 404. The sensor information processing unit 405 detects the magnetism of the radar 406 and the camera 407, which detect surrounding vehicles, obstacles, road surface, weather conditions, and the like, and the position detection support device 304 (such as a magnetic nail on a road) of FIG. The information from the magnetic sensor 408 is processed. Further, the sensor information processing unit 405 is a self-vehicle measuring device that measures a traveling distance of the self-vehicle, a current speed, an acceleration performance, a vehicle weight that changes according to a passenger, a tire wear degree, and the like.
It also processes information from 409. Further, requests (acceleration, maximum speed, ride comfort coefficient) from the driver are received and processed by the human interface unit 415.

The vehicle central processing unit 414 includes a communication control unit 401,
Human interface unit 415, sensor information processing unit 4
Exchange data with 05 and perform main processing. The own-vehicle time-distance curve creating unit 410 plans a time-distance curve from the present of the own vehicle based on the processing result of the vehicle central processing unit 414, and based on the plan, the vehicle behavior determining unit 411 performs acceleration / deceleration, steering. Determine the actual behavior, such as quantity. Then, the control amount determination unit 412 determines the brake / throttle opening / steering control amount in the vehicle drive unit for realizing the behavior, and transmits the control amount to the vehicle drive unit 413.

Next, the prerequisites of the merge control system and the flow of control will be described with reference to FIG. FIG. 5, on the left,
The whole process flow chart of the merge control is shown on the right side in the merge area of the self-driving vehicle 203 (merge AHS vehicle) corresponding to the flowchart. First, the prerequisites for the merge control system will be described. First, as the traveling vehicle grasping device 306 of FIG.
The traveling vehicle grasping devices 513 and 514 are installed at 512 traveling vehicle grasping devices and the B1 and B2 points of the merging channel 202, respectively.
Each of the traveling vehicle grasping devices 511, 512, 513, and 514 is capable of measuring the measurement time, position, and speed of a vehicle (autonomously driven vehicle or manually driven vehicle) traveling in each lane. Second
The junction 202 is one lane. Third, point A3 is a start point between lane change startable areas, point A4 is a lane change startable section end point, and point A5 is a lane change end end point. The positions of the points A3 and A5 depend on the road structure (the length of the acceleration lane) of the merging area, but the point A4 is a movable point that moves at the speed of each merging AHS vehicle (A4).
Will be described in the final part of the description for FIG. 11). Finally, the merging area 206 in FIG. 2 is from point A1 to point A5 on the main line 201 in FIG. 5, and from point B1 to point A5 in the merging channel 202 in FIG.

Next, the overall flow of the merge control process will be described below. When the merging AHS vehicle 203 passes the point B1 of the merging channel 202, merging processing is started (merging operation start processing).
501. Details will be described later with reference to FIG. ). Next, from the traveling vehicle prediction processing 502 (details will be described later with reference to FIG. 7), reference is made to the prediction data of the traffic in the merging area, and the on-vehicle instruction creation processing 503 (details will be referred to FIG. 9). At the time when the merging AHS vehicle passes the A5 point.
The speed (giving a margin) is determined and transmitted to the combined AHS vehicle 203 using the road-vehicle communication. Merging AHS car 203,
Based on the instruction from the vehicle-to-vehicle instruction creation process 503, in the own vehicle time distance curve creation / vehicle control process 504 (details will be described later with reference to FIG. 10), the current acceleration performance of the own vehicle, the vehicle condition, A travel plan (time-distance curve) of the own vehicle is created in consideration of a user's request and the like. These vehicle-to-vehicle instruction creation processing 503 and own-vehicle time-distance curve creation / vehicle control processing 504 include:
The process is repeated until the combined AHS vehicle 203 passes the point A3.
After the merged AHS vehicle 203 has passed, a lane change process 506 (details will be described later with reference to FIG. 11) is performed.

Next, the merging operation start processing will be described with reference to FIG. FIG. 6 is a detailed flowchart of the merging operation start process 501 in the flowchart on the left side of FIG. First, information such as the position and speed of the merging AHS vehicle sent from the merging AHS vehicle to the road side is always used by the road-to-vehicle communication, using roadside control adjacent to the merging unit roadside control device 207 shown in FIG. If the merging AHS vehicle passes the point B1 with reference to the device 209 (merging AHS vehicle position reference processing 601), then the merging A
A process 603 of transmitting the HS vehicle information to the junction roadside control device is performed. The process 603 of transmitting the merged AHS vehicle information to the merging unit roadside control device from the roadside control device adjacent to the merging unit to the merging unit roadside control device includes the merging AHS.
The vehicle information (vehicle ID, vehicle type, destination, acceleration performance, current position, speed, etc.) is transmitted to the junction roadside control device, and then the merged AHS vehicle information reading process (junction roadside control device) 60
At 4, the information of the merged AHS vehicle is saved. And 6
If you have not passed point B1 at branch 02,
Until passing the point B1, the merging AHS vehicle position reference processing 60
1 is repeated.

The above processing is performed by the merging unit roadside controller (FIG. 2).
207) and the adjacent upstream roadside control device (209 in FIG. 2)
Is performed using road-to-road communication.

Next, the details of the traveling vehicle prediction processing will be described with reference to FIGS. FIG. 7 shows the traveling vehicle prediction process 50 of FIG.
FIG. 4 is a flowchart showing the details of the process 2; This process is performed in the traveling vehicle prediction unit 307 in FIG. FIG.
Is a graph showing the concept of a time-distance curve of a vehicle used when making a prediction. The horizontal axis in FIG. 12 indicates time, and the vertical axis indicates the position on the confluence, and A1 to A5, B1, B2
Correspond to the respective points on the road map on the right side of FIG. The lower part of the vertical axis, B1, B2, B3, corresponds to the junction, and the upper part of the vertical axis, A1, A2, A3, A4, A5
Corresponds to the main line, and A3 at the lower portion (merging channel) and A3 at the upper portion (main line) are connected. Symbol 1201 in the graph is
The measurement data (passing time and speed of each vehicle) at the traveling vehicle grasping devices 511 (point A1), 512 (point A2), 513 (point B1), and 514 (point B2) in FIG. 5 and complement these data The reference numeral 1202 indicates the A1, A2, B
A prediction of the future movement of each vehicle based on the data at the points 1 and B2 is shown (the prediction has a range as shown in the figure).
Symbol 1209 indicates time / position / velocity data of the merged AHS vehicle sent from the merged AHS vehicle to the roadside and a curve complementing those data, and symbol 1210 indicates the merged AHS vehicle created by the roadside. It is a driving plan. These, 1201, 12
Curve lines complementing the 09 data string, and 1202, 12
The curve representing the prediction / plan of the movement of each vehicle of 10 is referred to as a “time distance curve”.

Next, in the traveling vehicle grasping process (point A1) 701 in FIG. 7, data is extracted from the traveling vehicle grasping device 511 installed at the point A1 on the main line 201 in FIG. If there is an AHS vehicle or a non-AHS vehicle, go to the next step. If there is a vehicle that passed the A1 point (corresponding to the measurement data of 1203 in FIG. 12), change the lane.
In the time-distance curve prediction processing (points A1 to A5) 703, a prediction of a lane change from the point A1 to the point A5 (FIG. 5) of the vehicle and a prediction time-distance curve are created (the prediction of 1204 in FIG. 12). (Corresponds to the time-distance curve), saves it in a file 704, and proceeds to the next step (in the branch of 702, even if there is no vehicle passing through the A1 point,
Proceed to the next step 711).

In the traveling vehicle grasping process (point A2) 711, data from the traveling vehicle grasping device 512 installed at the point A2 on the main line 201 in FIG. 5 is extracted, and vehicles (AHS vehicles, non-AHS) passing through the point A2 are extracted. If there is a car, go to the next step. When there is a vehicle that has passed through point A2 (see FIG. 12)
In the lane change / time-distance curve prediction process (points A2 to A5) 713, the prediction regarding the lane change from the point A2 to the point A5 (FIG. 5) of the vehicle and the predicted time Create a distance curve (Fig. 12
The predicted time-distance curve of the vehicle and the predicted time-distance curve (corresponding to 1207 in FIG. 12) corresponding to the lane change of the vehicle created in the lane change / time-distance curve prediction processing (points A1 to A5) 703 described above. (Corresponding), save the file 714, and proceed to the next step (even if there is no vehicle passing the point A2 at the branch of 712,
Proceed to the next step 721. ). Further processing 721, 731
Performs the above-mentioned processes 701 to 704 and 711 to 714 in the same manner with respect to information from the two traveling vehicle grasping devices 513 and 514 installed at the points B1 and B2 of the merging channel 202 in FIG. Then, the processes of 701 to 704, 711 to 714, 721, and 731 are repeatedly performed.

Next, the lane change / time distance curve prediction processing will be described with reference to FIG. FIG. 8 is a detailed flowchart of the lane change / time distance curve prediction processing 703 in FIG. This process is performed in the traveling vehicle prediction unit 307 in FIG. First, in the traveling vehicle grasping process 801, traveling vehicle information (time and speed at which the traveling vehicle passed) measured by the traveling vehicle grasping device 306 in FIG. 3 is used.
The information (vehicle ID, position, speed, acceleration, etc.) sent from the AHS vehicle using the road-vehicle communication is extracted in the AHS vehicle information reading process 802 through the road-vehicle communication control 305 in FIG. It is extracted via the unit 301. Next, in the traveling vehicle grasp information / AHS vehicle information matching processing 803,
Matching is performed by comparing the position of the traveling vehicle by the traveling vehicle grasping process 801 with the position of the AHS vehicle by the AHS vehicle information reading process 802. Here, if the passing vehicle is an AHS vehicle, the time and distance curve information of the AHS vehicle is received and processed.
At 5, the vehicle receives information on the travel plan (time-distance curve) of the own vehicle created by the AHS vehicle, and the time-distance curve is
In the case of a curve that can reach the junction (point A5 in FIG. 5), the AHS vehicle time-distance curve prediction processing 807 predicts the time-distance curve based on the travel plan (time-distance curve) of the AHS vehicle. In the branch of 806, when the travel plan (time-distance curve) of the AHS vehicle is a curve that has not reached the junction, the prediction obtained by 807, 809, 810, 811 in the AHS vehicle time-distance curve interpolation processing 808 Based on the time distance curves, the shortage of the time distance curves is compensated so as not to be inconsistent with those time distance curves, and the time distance curves are predicted.

At the branch 804, if the passing vehicle is not an AHS vehicle, a point A5 where the non-AHS vehicle joins (FIG. 5)
Predict the time-distance curve up to and including lane change. First, in the non-AHS lane change prediction processing of 809, in consideration of a case where a vehicle changes lanes to an adjacent lane, classification is made into cases where there is a possibility of changing lanes and cases where there is no possibility of changing lanes. . Next, in 810 non-AHS vehicle traveling direction time distance curve prediction processing, a predicted time distance curve in the traveling direction of the vehicle is created, and the predictions of 809 and 810 are combined to form a predicted time distance curve with lane numbers.

Next, in the predicted time-distance curve correction processing 811 other than the target non-AHS vehicle, a new predicted time-distance curve is created in the processing of 809 and 810 described above, so that the current predicted target is obtained. If it is necessary to change the predicted time-distance curve (including the presence or absence of a lane change) other than the non-AHS vehicle, the vehicle is again subjected to the lane change / time-distance curve prediction process. 809,810
The details of this process will be described later with reference to FIGS. Then, all the data is stored in a file, and the process ends.

Next, the non-AHS lane change prediction process will be described with reference to FIG. FIG. 14 is a detailed flowchart of the non-AHS lane change prediction process 809 of FIG. First, in the vehicle data search processing around the non-AHS vehicle to be predicted in 1401, it is measured whether or not another vehicle exists within a range that the driver of the non-AHS vehicle to be predicted is considered to be aware of. Search within data that complements between data and measurement data. Details of the processing of 1401 will be described later with reference to FIG. In the branch of 1402, if the vehicle is within the driver's recognition range, the information (position, speed) of the vehicle recognized by the driver to be predicted is recognized in the information extraction processing of the driver recognition vehicle in 1410. Is extracted from the traffic flow measurement data and the data complementing the measurement data.

In the branch of 1403, the non-A
When the driver-recognized vehicle of the HS vehicle is a vehicle traveling in the same lane as the non-AHS vehicle to be predicted, the process proceeds to the branch of 1406 (case 1).
In the branch of 1403, if the driver recognition vehicle of the non-AHS vehicle to be predicted is not the same lane as the non-AHS vehicle to be predicted, the process proceeds to the branch of 1405, and the driver recognition vehicle of the non-AHS vehicle to be predicted is In the case of a vehicle that changes lanes to the own vehicle traveling lane (for example, a vehicle that is traveling on a merging lane), the process proceeds to the branch at 1406 (Case 2).

In the branch at 1406, regarding the forward running vehicle (hereinafter referred to as vehicle A. position XA, speed VA) recognized by the driver of the non-AHS vehicle to be predicted, VA-VO <V
1 (V1 <0) and | XA-XX | / | VA-VO | ≦
In the case of T1, or a vehicle whose lane is to be changed lane ahead (hereinafter referred to as vehicle B. Position XB, speed VB), X1 ≦ XB−XO ≦ X2 (VO = VB) and T2 ≦
| XB-XO | / | VB-VO | ≦ T3 (VO <VB)
In the case of, it is determined that there is a motive to change lanes to the adjacent lane, and the process proceeds to the branch of 1407. Here, V1, X1, X2, T
2, The value of T3 is determined by the following two methods.

As a first method, there is a method in which statistical data measured in a driving simulator or an actual vehicle using a subject in advance or statistical data obtained by traffic flow measurement is used. The second method is to further improve the accuracy of the prediction based on the data of the first method, and updates the statistical data from the integration of the actual traffic flow measurement data of the target place, This is to correct the value by the method of 1. Specifically, a vehicle whose lane has been changed is extracted from the traffic flow measurement data and the data complemented between the data, and the vehicle within the driver's recognition range when the vehicle starts changing lanes. The position and speed of the vehicle (considering the height of the driver's seat, the blind spot of the driver due to other vehicles, road side walls, etc.) are extracted, and from the extracted data, the vehicle that has been motivated to change the lane of the vehicle (the relevant vehicle) (The vehicle that has changed lanes ahead of the vehicle, the vehicle that is slow ahead, etc.). Further, the type, position, and speed of the vehicle at the time when these extracted data are generated, and the surrounding environment (traffic volume of the lane in which the vehicle is traveling, traffic volume of the lane to which the vehicle changes lanes) , Weather, road conditions, road structure)
The information is stored with the information attached, and is used as a data source for matching in a similar situation. With this method, it is possible to incorporate not only road conditions specific to the section, but also changes in parameters due to real-time environmental changes such as current traffic volume, traffic flow trends (large, ordinary passenger cars, etc.), weather, brightness, and the like. , Leading to improved prediction accuracy.

Next, in the branch at 1406, the non-AH
If the S vehicle has a motive to change lanes in the adjacent lane, the process proceeds to branch 1407 to determine whether the lane can be changed to the adjacent lane. In 1407, first, the field of view of the left and right sides of the driver of the own vehicle and the rear (rear mirror, left and right side mirrors) are set. In case 1, the left and right side mirrors and left and right side mirrors are not used (if there is no adjacent lane),
In the case of the rearview mirror, the case 2, the field of view is set on the side opposite to the vehicle B, the side mirror on the opposite side (do not perform when there is no adjacent lane), the rearview mirror, and the other vehicle / road structure A driver-recognized vehicle within the blind spot of the driver is extracted in consideration of the blind spot caused by an object. The processing up to this point is the same as the processing of 1401 described above, and the details will be described later with reference to FIG. From these driver-recognized vehicles and the driver-recognized vehicle in the front field of view, vehicles before and after the space where the lane change is performed are searched to determine whether the lane change is possible. Assuming that a lane change will be made,
Go to lane change processing of 09. If there are vehicles before and after the lane change, the preceding vehicle is referred to as vehicle C (position XC,
Speed VC), the vehicle behind (vehicle D (position XD, speed V)
D) as | XC-XO | / | VC-VO | ≧ T4
(VO> VC), | XC-XO | ≧ X3 (VO ≦ VC)
And | XD-XO | / | VD-VO | ≧ T5 (VO <V
D), | XD-XO | ≧ X4 (VO ≧ VD)
The prediction is performed assuming that the lane change is performed, and the process proceeds to the lane change process of 1409 (if the vehicle is only one of the front and rear, only one condition is used). When the lane change of both the left and right lanes is possible, the prediction is performed on the assumption that the lane change is performed on the right lane. In these predictions, parameters such as X3, X4, T4, and T5 are determined in the same manner as in the first and second methods.

On the other hand, if no vehicle is present within the recognition range of the driver of the non-AHS vehicle to be predicted at the branch of 1402, the vehicle recognized by the driver of the non-AHS vehicle to be predicted travels at the branch of 1405. If the vehicle is not a lane change vehicle to a lane, the speed difference from the preceding vehicle is not within a predetermined range at the branch in 1406, and if the inter-vehicle is not within the predetermined range, the non-AHS vehicle to be predicted cannot change lanes to the adjacent lane at the branch in 1407 In each of the cases, it is determined that there is no possibility of changing lanes, and the non-AHS lane change prediction process ends, and the lane changes in the 810 non-AHS vehicle traveling direction time-distance curve prediction process (FIG. 8). The prediction process is performed assuming that no change is made. Last
In the lane change processing of 1409, the number of the current lane and the number of the lane to which the lane is changed are given for the non-AHS vehicle to be predicted.

Next, the non-AHS vehicle traveling direction time distance curve prediction processing will be described with reference to FIG. FIG. 15 is a detailed flowchart of the non-AHS vehicle traveling direction time distance curve prediction process 810 of FIG. First, in the surrounding vehicle data search processing 1501 of the non-AHS vehicle to be predicted, the surrounding vehicle information 1401 (FIG.
14) Extract (position / velocity), 1505 prediction target non-AHS
In the vehicle information extraction process, the current position, speed, and vehicle type of the non-AHS vehicle to be predicted are extracted. In the branch 1507, if the non-AHS vehicle to be predicted is the vehicle predicted not to change lanes in the non-AHS lane change prediction processing (809 in FIG. 8), the flow proceeds to the branch 1502. At branch 1502, the nearest vehicle A ahead of the same lane (position XA, speed V
A), and a preceding vehicle B (position XB, speed VB) in an adjacent lane that changes lanes ahead of the same lane is extracted. Assuming that the position of the own vehicle is XO and speed VO, VA-VO ≧ V
O and VA-VO <O and | XA-XO | / | VA-
If VO |> T1, or VB−VO ≧ V for vehicle B
O and VB-VO <O and | XB-XO | / | VB-
In the case of VO |> T1, the process proceeds to the single travel predicted time distance curve creation processing of 1506. As the value of T1, an experiment using a driving simulator and an actual vehicle, and a value of traffic flow observation data are used (the above-described method 1). In addition, in order to further improve the accuracy of the prediction based on the data of the first method, the statistical data is updated from the integration of the actual traffic flow measurement data of the target place, and the value obtained by the first method is used. (Method 2 described above). In 1506, assuming that the own vehicle will continue to travel at the same speed in the future, a predicted time-distance curve up to a certain time later is created to be a prediction regarding the traveling direction. At branch 1502, VA-VO <O and | XA-X for vehicle A
O | / | VA-VO | ≦ T (when there is a danger that the vehicle will hit the vehicle A) or VB-VO for the vehicle B
In the case of <O and | XB-XO | / | VB-VO | ≦ T (there is a danger that the vehicle will hit the vehicle B), the process proceeds to 1504, a follow-up running predicted time distance curve creation process. In the follow-up running predicted time-distance curve creation processing of 1504, a predicted time-distance curve is created assuming that the speed of the vehicle decelerates to the speed of the vehicle A or the vehicle B, and the prediction is made regarding the traveling direction.

At the branch 1507, the non-AHS vehicle to be predicted changes the non-AHS lane change prediction process (809 in FIG. 8).
If the vehicle is predicted to change lanes, the flow proceeds to branch 1508. In the branch 1508, regarding the space where the lane is to be changed, if there is a preceding vehicle (vehicle C, position XC, speed VC), and VC-VO <O and | XC-XO | / | VC-VO | ≦ In the case of T1 (when there is a risk that the own vehicle collides with the vehicle C), the process proceeds to the following traveling predicted time distance curve creation processing of 1504 (that is, the prediction is performed on the assumption that the own vehicle speed is reduced to the speed of the vehicle C). Create a time-distance curve.). In the branch of 1508, if the vehicle C does not exist, or if the vehicle C is not VC-VO <O and | XC-XO | / | VC-VO | ≦ T1, the independent traveling predicted time distance curve creation processing of 1506 is performed. Proceed to.

Next, with reference to FIG. 16, the process of extracting the surrounding vehicle data of the non-AHS vehicle to be predicted of 1401 in FIG. 14 will be described in detail. In FIG. 16, reference numeral 1601 denotes a vehicle to be predicted, and 1602 denotes the position of the driver. 1603, 1604, 1605, 1606, 1607, 16
08 is the driver's front view, right side view, left side view, right side mirror view, left side mirror view, respectively.
It is a rearview mirror field of view. The field of view is represented by a sector with a specified radius and center angle. For the values of the radius and the central angle, values such as a driving simulator and experimental data from an actual vehicle are used. When a road structure such as a side wall is present as in 1069, a blind spot is formed beyond the side wall, and it is determined that the vehicle existing in the blind spot range is not recognized by the driver. Also, considering the blind spot by other vehicles, the vehicle of 1610,
In this case, it cannot be seen by the 1611 vehicle. In this way, the setting of the driver's field of view and the vehicles recognizable by the driver are extracted.

Next, referring to FIG. 9 and FIG. FIG. 9 is a detailed flowchart of the on-vehicle instruction creation process 503 in FIG. This processing is performed in the vehicle instruction creating unit 308 in FIG. First, in the merged AHS vehicle information reading process 901, information (position, speed, acceleration performance, etc.) of the merged AHS vehicle transmitted using road-to-vehicle communication is read, and based on the information, the merged vehicle time at the maximum acceleration is read. Distance curve (1) creation processing
In 902, a temporary time-distance curve (1) of the merged AHS vehicle is created (corresponding to 1208 in FIG. 12) when it is assumed that the merged AHS vehicle has accelerated at the maximum acceleration. 5, the traveling vehicle prediction process 5 in FIG.
Of the traveling vehicle prediction processing (predicted time-distance curve) created in 02, only the vehicles existing in the leftmost lane between the points A3 and A5 are read (1204 and 1206 in FIG. 12).
Equivalent).

Next, the information of these 901, 902 and 903 will be compared. A of FIG. 5 calculated based on the time distance curve of 902
The position and speed of the merging AHS vehicle at the five points (corresponding to 1211 in FIG. 12) and the position and speed of the main line traveling vehicle based on 703 and 713 in FIG. 7 at the expected time when the merging AHS vehicle reaches the A5 point (1212, 1213 in FIG. 12) to extract the inter-vehicle distance between main-vehicles (between 1212 and 1213 in FIG. 12) and the speed difference from the main-vehicles, where merging AHS vehicles can merge. . As a result, if the inter-vehicle distance and the speed difference between the merging AHS vehicle and the main line vehicle are within a predetermined range for the merging AHS vehicle, an instruction content creation processing 905 for merging between the main line vehicles is performed.
In step (3), instruction contents (time and speed with a margin to pass point A5 in FIG. 5) for merging between vehicles on the main line are created.

In the branch of 904, if the inter-vehicle distance and the speed of the main line vehicle at which the merging AHS vehicle is to join at the point A5 are not within the predetermined range for the merging AHS vehicle, the main line inter-vehicle search processing 907 performs The merged AHS vehicle can reach the intervening vehicle at the specified minimum speed in the merging path with reference to the inter-vehicle in the leftmost lane immediately behind the processed main line. When the distance between the vehicles reaches the point A5, when the distance between the vehicles on the main line and the speed of the vehicles on the main line are within a predetermined range for the merging AHS vehicle, in the instruction content creation processing 905 for merging between the vehicles on the main line, Instructions for merging with other vehicles (time and speed with a margin when passing point A5 in FIG. 5)
And transmits the instruction to the merging AHS vehicle.

At the branch of 909, when the distance between the main lanes searched in the processing of 907 reaches the point A5, if the distance between the main lanes and the speed of the main lane are not within the predetermined ranges for the merged AHS vehicle, the main lane search is performed. In the process 907, the next inter-vehicle is referred to.

At the branch of 908, the AHS at the minimum speed specified in the merging path is entered between the vehicles searched at 907.
If the vehicle is unreachable, search for an AHS vehicle in the leftmost lane of the main line. If an AHS vehicle exists in the leftmost lane of the main line, in the process of 911, the AHS vehicle Confluence AH calculated based on the time distance curve (1)
At the time when the S vehicle reaches the point A5 (corresponding to 1211 in FIG. 12), the AHS vehicle on the leftmost lane of the nearest main line, which is upstream of the main line from the point A5, is referred to. The main line AHS
If the vehicle can change lanes to the acceleration, deceleration, and right lanes, the main lane AHS vehicle instruction processing 913 moves to the acceleration, deceleration, and right lanes so that the merging AHS vehicles can merge safely and smoothly. In the lane change, one or a plurality of instructions are performed, and in the instruction content creation processing 905 for merging between the main lanes, the instruction content for merging between the main lanes generated by the 913 main lane AHS vehicle instruction (FIG. 5) (The time and speed with a margin that passes through the A5 point) is created, and the instruction content is transmitted to the merged AHS vehicle using the road-vehicle communication.

At the branch 912, the main line AHS searched
If the vehicle cannot accelerate, decelerate, or change lanes to the right lane, the AHS vehicle in the leftmost lane, which is one line behind, is referred to (914), and the merged AHS vehicle is defined in the merging flow path. When traveling at the lowest speed, main line A searched at 914 at A5
If the vehicle can reach the HS vehicle, the process returns to 912 to check whether the AHS vehicle can accelerate, decelerate, and change lanes to the right lane.

At the branch of 915, when the merging AHS vehicle runs at the minimum speed specified in the merging channel, when it is not possible to reach the vicinity of the main AHS vehicle searched at 914 at the point A5, and at the branch of 910, the main line If there is no AHS vehicle in the leftmost lane, the process proceeds to the step of emergency processing 916 to return to manual operation, and an instruction to safely stop etc.
Do it for cars.

Next, the own vehicle time-distance curve creation / vehicle control process will be described with reference to FIG. FIG. 10 is a detailed flowchart of the own vehicle time distance curve creation / vehicle control process 504 of FIG. In the following processing, the processing of 1001 to 1004 is performed by the vehicle central processing unit 414 of FIG. 4 and the processing of 1005 to 1006 is performed by the own vehicle time distance curve creation unit 410 of FIG.
The process of FIG. 7 is performed by the vehicle behavior determining unit 411 of FIG. 4, the process of 1008 is performed by the control amount determining unit 412 of FIG. 4, the process of 1009 is performed by the communication control unit 401 of FIG. This is performed in the vehicle instruction creation unit 308, respectively.

First, in the instruction content reading process 1001, the instruction content (time and speed at which the merging vehicle passes point A5 in FIG. 5) from the roadside is read from the communication control unit 401 in FIG. Next, in the own vehicle position / speed acquisition processing 1002,
The current vehicle position, current vehicle speed, and current vehicle acceleration performance (engine performance, vehicle-specific , And the number of passengers, the wear state of tires, and other factors that change every moment) are read, and in the peripheral information acquisition processing (from the onboard sensor) 1003, the radar 406 of FIG.
From the camera 407 via the sensor information processing unit 405, surrounding information (the number of other vehicles in the vicinity, their respective positions / speeds / vehicle types, obstacles (presence / absence, if an obstacle exists, the number and shape),
Road surface conditions (friction coefficient, presence or absence of snow), weather conditions (wind speed))
Read. Further, in the driver request receiving process 1004, the human interface unit 415 of FIG.
Accepts driver requirements (acceleration, maximum speed, ride comfort factor).

On the basis of the information obtained in the processes 1001 to 1004, a time-distance curve creation process (vehicle) 1005 creates a time-distance curve of the own vehicle. Of the processing from 1001 to 1004,
Among the information obtained by the processing of 1003 (surrounding information acquisition processing), for example, when there is a vehicle traveling at a low speed ahead of the merging flow path, a time-distance curve that does not accelerate at first but accelerates immediately before changing lanes is created. For example, when the coefficient of friction is small and the wind speed is large in a road surface condition, a time-distance curve is created on the assumption that too rapid acceleration is not possible. 1004
Out of the information from the process (driver request acceptance process)
For example, if the driver issues a request that places importance on riding comfort rather than quick acceleration, a time-distance curve is created at an acceleration that does not feel sharp acceleration.

As a result of the time-distance curve creation processing, an instruction from the road side (time and speed at which the merging vehicle passes point A5 in FIG. 5)
However, in the case of a feasible time-distance curve, the acceleration / steering amount for realizing the time-distance curve created in 1005 is determined in the acceleration speed / steering amount determination processing 1007, and the brake / steering amount is determined.
In the throttle opening / steering control amount determination process 1008, 10
A brake / throttle opening / steering control amount for realizing the acceleration / steering amount determined in 07 is determined, and the control amount is transmitted to the vehicle drive device 413 in FIG. Here,
It is assumed that the position and speed of the vehicle are brought closer to the determined time-distance curve using a feedback function such as PID control.

If the time-distance curve cannot be created in the branch of 1006, the time-distance curve cannot be created.
The time and speed that allow the merging AHS vehicle to pass the target point (point A5) are transmitted to the roadside by road-to-vehicle communication (1009), and the instructions for the merging vehicle are corrected on the roadside. (1010), and returns to the process of 1001.

Next, the lane change processing will be described with reference to FIG. FIG. 11 is a detailed flowchart of the lane change processing 506 in FIG. In the following processing, 1101 to 1103 and 1109 correspond to FIG.
In the vehicle central processing unit 414, reference numerals 1104 to 1106 denote the own vehicle time distance curve creation unit 410 in FIG. 4, 1107 denotes a vehicle behavior determination unit 411 in FIG. 4, and 1108 denotes a control amount determination unit 4 in FIG.
At 12, each is performed.

First, peripheral vehicle information acquisition processing (from the roadside)
At 1101, from the roadside, using the road-to-vehicle communication,
The position and speed of the surrounding vehicles of the HS vehicle are determined by the communication control unit 40 shown in FIG.
Conveyed through one. The vehicle information on the road side is the information from the traveling vehicle grasping device 306 of FIG.
There are two types of information from the on-vehicle sensors of HS vehicles, and information provided to the roadside using road-to-vehicle communication. Next, in a surrounding vehicle information acquisition process (from an onboard sensor) 1102, information obtained from the radar 406 and the camera 407 in FIG. Acquire the position, speed, and type of surrounding vehicles.
Further, in the vehicle position / speed acquisition processing 1103, the current vehicle position / current vehicle speed / current vehicle acceleration performance (engine performance, body weight, etc.) measured by the vehicle measurement device 409 in FIG. Of the vehicle, and the factors that change every moment, such as the number of occupants and the wear of tires, are acquired via the sensor information processing unit 405 in FIG.

In the lane change start point determination processing 1105, a point at which the own vehicle starts lane change is determined based on the information obtained in the processing of steps 1101 to 1103. As a result of the processing of 1101 to 1105, the distance between two vehicles (1302 and 1303 in FIG. 13) between the merging vehicles (1301 in FIG. 13) and 1304 in FIG.
If the positions and speeds of the vehicle ahead and the vehicle behind the merged vehicle are within a predetermined range with respect to the position and speed of the merged vehicle, it is determined that the lane can be changed. If it is determined that the merging AHS vehicle can change lanes, in the acceleration speed / steering amount determination processing 1107, the acceleration for realizing the time-distance curve created at 1005 in FIG.・
In the brake / throttle opening / steering control amount determination processing 1108, the brake / throttle opening / steering control amount for realizing the acceleration / steering amount determined in 1107 is determined, and the control amount is determined. This is transmitted to the vehicle drive device 413 in FIG. Here, it is assumed that the position and speed of the vehicle are brought closer to the determined time-distance curve (including the movement in the lane changing direction) using a feedback function such as PID control. Then, the above-described processes of 1101-1109 are repeated until the lane change is completed.

If the lane change is not possible at the branch at 1106 and the merging AHS vehicle does not pass through the point A4 (the end point of the section where lane change can be started) in FIG. According to the information obtained in step (1), speed adjustment such as acceleration start, deceleration start, acceleration end, deceleration end, and the like is performed, and the process returns to 1101.

At the branch of 1110, the merged AHS vehicle
If the vehicle is passing through a point (the end point of the section where lane change can be started), the vehicle stops at a safe deceleration in the acceleration lane without changing lanes to the main line (emergency processing 1112). Here, the A4 point is the last point where the merging AHS vehicle can start changing lanes to the main lane, that is, the merging AHS vehicle starts decelerating at a safe deceleration from the A4 point, and safely in the acceleration lane. Can be stopped. The position of the A4 point is not a fixed point on the physical road, but is unique to each merging AHS vehicle, and the location moves moment by moment according to the state (position and speed) of the merging AHS vehicle. Is a point. Therefore, when the speed of the merged AHS vehicle is high, the point A4 moves upstream, and when the speed of the merged AHS vehicle is low, the point A4 is
Move downstream.

[0066]

As described above, the present invention relates to a structure similar to the confluence of a conventional Japanese highway, for example, when the acceleration lane is relatively short in the confluence area of the driving support road system. (In the case of inter-city highways, it is often 250 m or less.) Compared to the conventional method of predicting traveling vehicles that does not consider lane changes, there are more opportunities for merging vehicles to change lanes to the main line, and smooth This has the effect that the vehicle at the junction can be controlled.

[Brief description of the drawings]

FIG. 1 is a drawing showing an outline of the present invention.

FIG. 2 is a diagram showing a position of a merging control system according to an embodiment of the present invention in a driving support road system.

FIG. 3 is a diagram showing a configuration of a merging unit control device 207 of FIG. 2 in detail.

FIG. 4 is a drawing showing the automatic driving vehicle 303 of FIG. 3 in detail.

FIG. 5 is a drawing showing a control flow in a merging control system in the driving support road system, which is one embodiment of the present invention.

6 is a merging operation start process 501 in the flowchart on the left side of FIG. 5;
It is a flowchart which shows in detail.

FIG. 7 is a flowchart showing details of a traveling vehicle prediction process 502 in FIG. 5;

FIG. 8 is a flowchart showing in detail the time distance curve prediction processing 703 in FIG. 7;

FIG. 9 is a flowchart showing the details of a vehicle-to-vehicle instruction creation process 503 in FIG. 5;

10 is a diagram illustrating a vehicle time distance curve creation / vehicle control process 5 in FIG. 5;
It is a flowchart which shows the process of 04 in detail.

FIG. 11 is a flowchart showing the lane change processing 506 in FIG. 5 in detail.

FIG. 12 is a graph showing a time-distance curve of a traveling vehicle.

FIG. 13 is a drawing showing an example of an arrangement of vehicles traveling on a main line immediately before a merging AHS vehicle changes lanes to a main line.

FIG. 14 is a flowchart showing the non-AHS lane change prediction processing 809 in FIG. 8 in detail.

FIG. 15 is a flowchart showing the non-AHS vehicle traveling direction time distance curve prediction processing 810 in FIG. 8 in detail.

FIG. 16 is a view showing a method of searching for surrounding vehicle data of a non-AHS vehicle at 1401 in FIG. 14;

[Explanation of symbols]

101 Expressway main line (one lane on one side) 102 Expressway merging path (one lane on one side) 103 Own vehicle driving vehicle 104 Manual driving vehicle 105 Merging automatic driving vehicle 106 Acceleration lane 107 Running vehicle grasping device 108 Merging section roadside control device 109 Running vehicle Prediction unit (in the roadside control unit at the junction) 110 Time distance curve (created by the traveling vehicle prediction unit) 111 Vehicle-to-vehicle instruction creation unit (inside the roadside control unit at the junction) 112 Vehicle control device for the automatic merging vehicle 113 Onboard measuring instrument (In the automatic merging vehicle) 114 Self-vehicle measuring device (in the automatic merging vehicle) 115 Driver of the automatic merging vehicle 116 Time distance curve creation unit (in the vehicle control unit) 201 Expressway main line (3 lanes on one side) 202 Expressway merger Road (one lane on one side) 203 Autonomous driving vehicle (AHS vehicle) 204 Acceleration lane 205 Central control unit (AHS center) 206 Highway merging section 207 Merging section roadside controller 208 Single road section 209 Single road section roadside controller 301 Road to vehicle Communication control unit 302 Road-to-vehicle communication device 303 Automatic driving vehicle 304 Position detection support device 305 Traveling vehicle grasping unit 306 Traveling vehicle grasping device 307 Traveling vehicle prediction unit 308 Anti-vehicle instruction creation unit 309 Road-to-road communication control unit 310 Environment grasping device 311 Environment grasping unit 401 Communication control unit 402 Road-to-vehicle communication device 403 Other autonomous driving vehicle 404 Vehicle-to-vehicle communication device 405 Sensor information processing 406 Radar 407 Camera 408 Magnetic sensor 409 Self-vehicle measuring device 410 Self-vehicle time-distance curve creation unit 411 Vehicle behavior determination unit 412 Control Quantity determination unit 413 Vehicle drive unit 414 Vehicle central processing unit 415 Human interface unit 501 Merging operation start processing 502 Traveling vehicle prediction processing 503 Vehicle instruction creation processing 504 Self-vehicle time-distance curve creation / vehicle control processing 505 Branch processing for passing through A3 point 506 Lane change processing 511, 512, 513, 514 Traveling vehicle grasping device 601 Merging AHS vehicle position reference processing 602 Branch processing for passing B1 point 603 Processing for transmitting AHS vehicle information to the merging unit roadside controller 604 Merging AHS vehicle information reading processing 701, 711 Running vehicle grasping processing 702 Branching processing for vehicle passing at A1 point 703, 713 Lane change / time distance curve prediction processing 704, 714 files Save process 712 Branch process for passing A2 point vehicle 721,731 Traveling vehicle prediction process for merged channel 801 Traveling vehicle grasping device 802 AHS vehicle information reading process 803 Traveling vehicle grasp information / AHS vehicle information matching process 804 Is passing vehicle an AHS vehicle? 805 AHS vehicle time distance curve information receiving processing 806 AHS vehicle time distance curve prediction processing 808 AHS vehicle time distance curve prediction processing 808 AHS vehicle time distance curve interpolation processing 809 Non-AHS lane lane change prediction processing 810 Non-AHS vehicle traveling direction time distance curve prediction processing 811 Predicted time distance other than target non-AHS vehicle Line correction processing 901 Merging AHS vehicle information reading processing 902 Merging AHS vehicle time-distance curve creation processing at maximum acceleration 903 Driving vehicle prediction information reading processing 904 Branching processing for inter-vehicle / speed difference determination with vehicles ahead and behind the main line 905 Merging between main line vehicles 906 Instructions by road-to-vehicle communication 907 Inter-vehicle inter-vehicle search processing 908 Branch processing to determine whether the vehicle can be reached at the minimum speed in the combined flow path 909 Branch processing to determine the inter-vehicle / speed difference on the main line 910 An AHS vehicle exists on the main line 911 Branch processing for the 911 Main AHS vehicle reference processing 912 Whether the AHS vehicle on the main line can accelerate, decelerate, and change lanes
913 Branch processing for main line AHS vehicle for ensuring inter-vehicle connection 914 Reference processing for AHS vehicle one immediately behind 915 Branch processing for whether or not the vehicle can be reached at the minimum speed in the merge flow path 916 Emergency processing 1001 Instruction reading processing 1002 Own vehicle position / speed value acquisition processing 1003 Peripheral acquisition processing from in-vehicle sensors 1004 Driver request acceptance processing 1005 Time distance curve creation processing 1006 Branch processing for whether a time distance curve is possible 1007 Acceleration / deceleration and steering amount determination processing 1008 Brake / throttle Opening / steering control amount determination processing 1009 Time distance curve creation disabled transmission processing 1010 Roadside instruction content correction processing 1101 Peripheral vehicle information acquisition processing from roadside 1102 Peripheral vehicle information acquisition processing from in-vehicle sensors 1103 Own vehicle position / speed acquisition processing 1104 Lane change start point determination processing 1105 Lane change start point determination processing 1106 Branch for whether lane change is possible Processing 1107 Acceleration / deceleration / deceleration / steering amount determination processing 1108 Brake / throttle opening / steering control amount determination processing 1109 Branch processing for whether lane change is completed 1110 Branch processing for passing A4 point 1111 Speed adjustment processing 1112 Emergency processing 1201 Time-distance curves supplemented with measured data 1202, 1204, 1206, 1207 Predicted time-distance curves 1203, 1205 Measurement data 1208 1210 Merging AHS vehicle traveling plan created by roadside 1209 Merging AHS vehicle time / position / speed data 1211 The vehicle in front of the merging space of the merging AHS vehicle 1213 The vehicle behind the merging space of the merging AHS vehicle 1301 The vehicle in front of the merging space of the merging AHS vehicle 1301 Vehicle 1303 Vehicle behind merging space of merging AHS vehicle 1304 Merging space of merging AHS vehicle 1401 Vehicle around non-AHS vehicle to be predicted Data search processing 1402 Divergence processing for whether a vehicle is within the driver's recognition range 1403 Branch processing for whether the recognized vehicle is the same lane vehicle 1405 Whether the recognized vehicle is a lane change vehicle to the own vehicle lane 1406 Branch processing for whether the speed difference with the preceding vehicle is within a predetermined range and whether the distance between vehicles is within a predetermined range 1407 Branch processing for whether the lane can be changed to the adjacent lane 1409 Lane changing processing 1501 Vehicles around the non-AHS vehicle to be predicted Data search processing 1502 Within the same lane and recognition range as the non-AHS vehicle to be predicted,
Bifurcation processing to determine whether there is a vehicle within a predetermined range 1503 Information extraction processing of the preceding vehicle of the non-AHS vehicle to be predicted 1504 Following travel prediction time distance curve creation processing 1505 Information extraction processing of the non-AHS vehicle to be predicted 1506 alone Driving predicted time / distance curve creation processing 1507 Divergence processing for whether the vehicle is predicted to change lanes 1508 Bifurcation processing for whether vehicles within a predetermined range exist ahead of the lane change destination 1601 Vehicle to be predicted 1602 Driver's position 1603 Driver's forward view 1604 Driver's right view 1605 Driver's left view 1606 Driver's right side mirror view 1607 Driver's left side view 1608 Driver's rear view 1609 Road structures such as side walls 1610 Vehicles that cannot be seen by the driver's blind spot 1611 Vehicles that create the driver's blind spot

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-10-105880 (JP, A) JP-A-7-334790 (JP, A) JP-A 8-194882 (JP, A) JP-A 9-1998 147285 (JP, A) (58) Field surveyed (Int. Cl. ⁶ , DB name) G08G 1/00-1/16

Claims (3)

(57) [Claims] 1. A roadside detector for detecting a situation on a traveling road, a roadside controller for receiving information from the detector and processing the information, and a roadside-to-vehicle wireless communication device installed near the traveling road. And a wireless communication device of the vehicle that performs wireless communication with the roadside and other vehicles, a detector of the vehicle that detects a situation on the traveling road, and a control device of the vehicle that processes information of the roadside, other vehicles, and the own vehicle detector. In a driving support road system that includes a vehicle, as a method of predicting the position and speed of the traveling vehicle from the present to a certain time later, consider that the traveling vehicle changes lanes to the adjacent lane at the junction A traffic flow prediction method in a merging control system of a driving support road system, wherein 2. The main road traffic flow prediction method in the merging control system of the driving support road system according to claim 1, wherein the roadside control device creates an instruction content for the automatic driving vehicle for the purpose of running the vehicle smoothly. In order to predict the position and speed of the running vehicle from the present to a certain time later, the driver needs to recognize the position and speed of the surrounding vehicles within the driver's visual recognition range. A main traffic flow prediction method in a merging control system of a driving support road system, wherein prediction is performed in consideration of lane change to a lane. 3. The main line traffic flow prediction method in the merging control system of the driving support road system according to claim 1, wherein the on-vehicle control device is required when creating a traveling plan for the purpose of smooth traveling of the vehicle. consisting, as a method of predicting the position and speed of until after a certain time from the present around the vehicle, the position and surrounding the other vehicles in the visual perception range of the driver of the other vehicle
A prediction method in a vehicle control device, wherein a prediction is performed in consideration of a driver of another vehicle recognizing a speed and changing a lane to an adjacent lane.