JP4735430B2 - Confluence support device - Google Patents

Description

  The present invention relates to a merging support device that supports a driving operation by a driver when a host vehicle traveling in a merging lane merges into a main lane.

Conventionally, when a host vehicle traveling in a merging lane joins the main lane, a merging support device has been proposed to assist the driving operation at the time of merging by presenting information such as the optimal merging timing to the driver of the host vehicle. (See, for example, Patent Document 1).
Japanese Patent Laid-Open No. 10-105884

  By the way, in order to obtain the optimum merging timing when the host vehicle joins the main lane from the merging lane, as a result of the host vehicle merging, the host vehicle and the surrounding vehicles traveling in the front and rear of the host vehicle are It is necessary to accurately predict the situation. Since the situation of such a merge result also varies greatly depending on the behavior characteristics at the time of merging of the own vehicle, in order to accurately predict the situation of the merge result, the merging behavior characteristics of the own vehicle are also considered. It is necessary to keep. Note that the merging behavior characteristics of the host vehicle are determined from characteristics of driving operation at the time of merging by the driver of the host vehicle and characteristics inherent to the vehicle such as acceleration / deceleration characteristics and steering characteristics of the host vehicle.

  However, since this type of conventional merging support device including the merging support device described in Patent Document 1 does not consider the merging behavior characteristics of the host vehicle, the prediction accuracy of the situation of the merging result is not sufficient. However, there is a problem that satisfactory information is not always presented.

  The present invention was devised to solve the problems of the prior art as described above, and predicts the situation of the merged result with high accuracy and presents accurate information to the driver of the own vehicle. An object of the present invention is to provide a merging support device that can appropriately support the driving operation by the driver at the time of merging.

The merging support apparatus according to the present invention predicts the situation of the host vehicle at the end of merging and the situation of surrounding vehicles when the host vehicle traveling in the merging lane immediately executes merging to the main lane, and the prediction Basically, the risk level is calculated according to the result, and information indicating the calculated risk level is presented to the driver of the host vehicle. In particular, in the merging support apparatus according to the present invention, when the merging is performed, based on the driver's driving characteristics that change according to the collision margin time of the surrounding vehicle traveling on the main lane with respect to the own vehicle. By calculating the data indicating the merging behavior characteristics as a learning value, when learning the merging behavior characteristics of the own vehicle and predicting the situation of the merging result, the situation of the own vehicle and surrounding vehicles, road conditions, etc. The situation of the merging result is predicted based on the current merging environment in accordance with the learning value of the merging behavior characteristic of the own vehicle.

  According to the merging support apparatus according to the present invention, the situation of the merging result is predicted with high accuracy in consideration of the merging behavior characteristic of the own vehicle, and information indicating the degree of risk corresponding to the predicted result is given to the driver of the own vehicle. Thus, the driver of the own vehicle can execute the merging at the optimum timing while accurately grasping the merging result according to his / her driving characteristics and the vehicle characteristics of the own vehicle.

[Outline of the present invention]
In the present invention, when the driver of the own vehicle takes an action and tries to change the driving behavior, the risk is calculated by predicting the resulting situation with high accuracy, and the risk is expressed. By presenting the information, the aim is to enable the driver of the vehicle to make a tactical decision as to whether or not to take action now. For that purpose, it is important to learn the behavior characteristics of the own vehicle according to the action of the driver that changes variously depending on the situation, and to predict the future situation based on the information, and the merge support device according to the present invention Is an application of this concept in a confluence scene.

  That is, the merging support device according to the present invention is the result when the merging (the lane change from the merging lane to the main lane) is executed in the merging scene where the own vehicle traveling in the merging lane merges with the main lane. That is, the situation of the host vehicle on the main lane at the end of merging and the situation of surrounding vehicles traveling around the host vehicle, the risk level is judged, and information representing the risk level is used to drive the host vehicle Present it to the person. At this time, the prediction of the situation of the merging result is based on the merging behavior characteristics of the host vehicle such as the time from the start to the end of merging (change of lane from the merging lane to the main lane), steering characteristics (vehicle trajectory), and speed change. And predict using the learned value. Since these behavior characteristics vary depending on the surrounding conditions, patterns are classified according to the surrounding conditions, and learning is performed for each pattern.

  In the present embodiment, the largest factor that affects the driving behavior of the driver of the host vehicle at the merging is the collision margin time (TTC: Time to Collision: relative distance) of the surrounding vehicle traveling on the main lane. (Divided by relative speed). As a general characteristic, when the TTC with the surrounding vehicles on the main lane is short, the driver of the own vehicle rushes to enter the main lane. On the other hand, when the TTC with the surrounding vehicles on the main lane is long, the driver of the own vehicle tends to enter the main lane slowly with his / her own rhythm. In view of such characteristics of the driver, the merging support device according to the present embodiment learns the merging behavior characteristic of the own vehicle according to the driver's driving characteristic change according to the TTC with the surrounding vehicle on the main lane. Based on. However, this is not the only factor that affects the merging behavior characteristics of the host vehicle in the merging scene, and various factors are involved. Therefore, in the merging support device of this embodiment, in addition to TTC with surrounding vehicles on the main lane, the merging execution position of the own vehicle in the merging lane, the congestion status of the main lane, the speed of the own vehicle, the lane of the merging lane Considering various situations such as shape, patterns are divided according to the merging environment identified in these situations, and the merging behavior characteristics of the host vehicle are learned. Then, using these learned learned values of the merging behavior characteristics of the own vehicle, what kind of situation when the driver of the own vehicle immediately takes action (merging behavior: lane change from merging lane to main lane) The degree of danger is calculated by predicting whether it will be, and information indicating the calculated degree of danger is presented to the driver of the host vehicle.

[Device configuration]
A schematic configuration of the merging support apparatus of the present embodiment is shown in FIG. As shown in FIG. 1, the merging support apparatus includes a controller 1 that forms the core of the apparatus, a GPS (Global Positioning System) receiver 2 connected to the input side of the controller 1, a vehicle speed sensor 3, a steering angle sensor. 4, a vehicle-mounted camera 5, a laser radar 6, a communication device 7, a map data storage device 8, and an information presentation device 9 connected to the output side of the controller 1.

  The GPS receiver 2 receives a signal from a GPS satellite every predetermined reception cycle, and detects latitude / longitude information indicating the absolute position of the host vehicle. The vehicle speed sensor 3 detects the traveling speed of the host vehicle, and the steering angle sensor 4 detects the steering steering angle of the host vehicle.

  The in-vehicle camera 5 captures an image of the surroundings of the own vehicle, in particular, an image of the surroundings of the own vehicle, such as a lane in the front-rear direction of the own vehicle (a start point, an end point, etc. of a road junction) Shoot. Further, the laser radar 6 scans a laser beam within a predetermined scan range around the host vehicle, and determines the presence or absence of the surrounding vehicle traveling around the host vehicle, the inter-vehicle distance and direction to the surrounding vehicle for each surrounding vehicle. To detect.

  The communication device 7 is for obtaining information from various communication infrastructures. Specifically, the communication device 7 receives traffic information by, for example, VICS (Vehicle Information and Communication System), and by road-to-vehicle communication, Detection information from a vehicle detection device installed at a junction or the like is received by DSRC (Dedicated Short Range Communications), etc., and by so-called inter-vehicle communication, position information of surrounding vehicles, blinker operation information, brake operation information, etc. Receive various information from surrounding vehicles.

  The map data storage device 8 includes, for example, a recording medium on which map data such as a CD-ROM, a DVD-ROM, and a hard disk is recorded, and a playback device that reads out the map data stored on the recording medium. Information from the GPS receiver 2, the vehicle speed sensor 3, the steering angle sensor 4, the in-vehicle camera 5, the laser radar 6, the communication device 7, and the map data storage device 8 is stored in the controller 1 at any time according to a command from the controller 1. Is taken in.

  The information presentation device 9 is for presenting information for confluence support created by the controller 1 to the driver of the host vehicle. For example, in the case of information presentation by sound, a speaker is used, In the case of information presentation, a display device is used.

  The controller 1 is mainly composed of a microcomputer with a built-in control program. When the control program is executed by the microcomputer, the own vehicle grasping unit 11 as a functional configuration characteristic of the merging support apparatus of the present embodiment, Each functional configuration of the surrounding vehicle grasping unit 12, the road condition grasping unit 13, and the merging support calculating unit 14 is realized.

  The own vehicle grasping unit 11 specifies the position of the own vehicle for each GPS reception cycle based on the latitude and longitude information detected by the GPS receiver 2, and determines the position of the own vehicle during the GPS reception cycle as a vehicle speed sensor 3. And the current position of the host vehicle is always grasped. In addition, in order to grasp | ascertain the present position of the own vehicle more correctly, the own vehicle grasping part 11 is an image of the surroundings of the own vehicle taken by the in-vehicle camera 5, road-to-vehicle communication using the communication device 7, and vehicle-to-vehicle communication. The detailed position on the road of the own vehicle may be specified using the information obtained in the above. The own vehicle grasping unit 11 grasps the current state of the own vehicle from the detection values of the vehicle speed sensor 3 and the rudder angle sensor 4 and also tracks the current position of the own vehicle that changes from time to time. , Grasp the behavior of the vehicle.

  The surrounding vehicle grasping unit 12 performs processing such as edge detection on the image of the surroundings of the host vehicle taken by the in-vehicle camera 5, and detects the positional relationship between the surrounding vehicle and the lane that travel on the main lane, In addition, the distance and direction of the vehicle to the surrounding vehicles on the main lane detected by the laser radar 6 as a reference, and information obtained by road-to-vehicle communication and vehicle-to-vehicle communication using the communication device 7 are also used. , Always know the current position of surrounding vehicles. Moreover, the surrounding vehicle grasping | ascertainment part 12 grasps | ascertains the behavior of a surrounding vehicle by tracking the present position of the surrounding vehicle on the main lane which changes every moment in time.

  The road condition grasping unit 13 grasps the road shape around the own vehicle based on the current position of the own vehicle grasped by the own vehicle grasping unit 11 and the map data stored in the map data storage device 8. The road condition grasping unit 13 also uses images taken by the in-vehicle camera 5 and information obtained by road-to-vehicle communication using the communication device 7 in order to grasp the shape of the road around the host vehicle more accurately. You may do it. Further, the road condition grasping unit 13 grasps the congestion situation of the main lane around the host vehicle based on the VICS traffic information received by the communication device 7. The traffic density on the main lane can be calculated from the section arrival time of the VICS traffic information.

  The merge support calculation unit 14 performs various calculation processes using information from the own vehicle grasping unit 11, the surrounding vehicle grasping unit 12, and the road condition grasping unit 13, and a situation that occurs as a result when the own vehicle immediately executes a merge. Is predicted to determine the risk level, and information indicating the risk level is output from the information presentation device 5. The main functions of the merging support calculation unit 14 are roughly divided into a function for identifying the current merging environment (merging environment identifying means), a function for learning the merging behavior characteristics of the host vehicle (merging behavior characteristic learning means), and , A function for determining the driving ability of the driver of the own vehicle (driving ability judging means), and a function for predicting the situation of the own vehicle and the situation of surrounding vehicles at the end of merging when the own vehicle immediately executes merging (merging Result prediction means) and a function (risk degree calculation means) for calculating the risk of merging according to the prediction result. Each of these functions is realized by, for example, the functional configuration illustrated in FIG. Hereinafter, the details of the merging support calculation unit 14 will be described with reference to FIG.

  In the functional configuration of the merge support calculation unit 14 shown in FIG. 2, the own vehicle state temporary storage unit 21 is configured to receive information from the own vehicle grasping means 11 when the own vehicle is traveling in the merge lane, specifically, Information such as the position, traveling speed, and steering angle of the host vehicle is temporarily stored as time-series information with a time stamp. In addition, the host vehicle state temporary storage unit 21 temporarily stores information from the host vehicle grasping unit 11, and at the same time, the host vehicle state calculation unit 29 at the end of the merging action described later and the case classification determination unit 24 of the merging environment. Information such as the current position, traveling speed, steering angle, etc. of the own vehicle is transmitted.

  The merging behavior characteristic extraction unit 22 extracts the merging behavior characteristic of the host vehicle based on the time series data of the host vehicle at the time of merging that is temporarily stored in the host vehicle state temporary storage unit 21. Specifically, the merging behavior characteristic extraction unit 22 determines, for example, the steering start time and the steering end time when performing merging (lane change from the merging lane to the main lane) from the time-series data of the steering angle. Extracting the time (Δt) it took for the vehicle to join, the steering characteristics from the steering start time to the steering end time, and the time series data of the traveling speed from the steering start time to the steering end time The acceleration / deceleration characteristics of the vehicle are extracted as the merging behavior characteristics of the host vehicle. Information on the merging behavior characteristic of the host vehicle extracted by the merging behavior characteristic extracting unit 22 is sent to a learning calculation unit 25 that is classified in the case described later.

  The ambient situation temporary storage unit 23 travels on the main lane when the host vehicle is traveling on the merge lane, information from the surrounding vehicle grasping unit 12 or information from the road condition grasping unit 13. Information such as the position and traveling speed of surrounding vehicles, the road shape around the host vehicle, and the congestion status of the main lane is temporarily stored as time-series information with a time stamp. The ambient situation temporary storage unit 23 temporarily stores information from the surrounding vehicle grasping unit 12 and information from the road situation grasping unit 13, and at the same time, with respect to the surrounding vehicle situation calculating unit 30 at the end of the merging action described later, Information from the surrounding vehicle grasping unit 12 is transmitted.

  Based on the time series data of surrounding vehicles and road conditions at the time of merging that is temporarily saved in the surrounding condition temporary storage unit 23, the case classification determination unit 24 in the merging environment tries to start merging immediately. Then, it is determined to which classification of the predetermined case the current merge environment belongs. Further, the case determination unit 24 in the merging environment performs steering at the time of merging execution extracted by the merging behavior characteristic extraction unit 22 after the merging actually ends in order to perform case categorization in performing the learning calculation. Based on the situation of surrounding vehicles and roads at the same time as the start time, and the state of the own vehicle, the case of the merging environment is classified, and the classification information of the case division is sent to the learning calculation unit 25 divided in the case described later.

  The case-by-case learning calculation unit 25 learns the merging behavior characteristics of the host vehicle extracted by the merging behavior characteristic extraction unit 22 under the conditions divided by the case classification determination unit 24 in the merging environment. Further, the case-by-case learning value storage unit 26 stores and saves the result (learning value) calculated by the case-by-case learning calculation unit 25 in association with the case-by-case conditions. Further, in this case, the divided learning value storage unit 26 stores the individual data used for learning together with the calculation result of the divided learning calculation unit 25.

  The driving ability determination unit 27 determines the driving ability of the driver of the host vehicle based on the individual data corresponding to the case classification condition stored in the storage unit 26 of the case-by-case learning value. Information on the driving ability of the own vehicle driver determined by the driving ability determination unit 27 is sent to the surrounding vehicle state calculation unit 30 at the end of the merging operation, which will be described later, and the surrounding vehicle state calculation at the end of the merging operation is performed. The unit 30 is used to change the calculation method according to the difference in driving ability.

  The learning value calling unit 28 that matches the merging environment is selected from the learning values stored in the case-by-case learning value storage unit 26, and the current merging case-specific determination unit 24 classifies the current merging. The learning value stored in the area corresponding to the environment is called up. The learning value information read out by the learning value calling unit 28 that matches the merging environment is sent to the host vehicle situation calculating unit 29 at the end of merging and the surrounding vehicle situation calculating unit 30 at the end of merging. .

  The own vehicle state calculation unit 29 at the end of the merging action is called by the calling unit 28 of the learning value 28 that matches the current state of the own vehicle transmitted from the own vehicle state temporary storage unit 21 and the merging environment. Based on the learning value of the merging behavior characteristics of the vehicle, if the own vehicle immediately starts the merging behavior at the present time, as a result, predict the situation that the own vehicle will be when the merging behavior ends Calculate.

  The surrounding vehicle situation calculation unit 30 at the end of the merging action is called by the calling unit 28 of the learning value 28 that matches the current surrounding vehicle situation and road situation transmitted from the ambient situation temporary storage unit 23 and the merging environment. Based on the learned value of the merging behavior characteristic of the own vehicle and the information on the driving ability of the own vehicle driver transmitted from the driving ability judging unit 27, when the own vehicle immediately starts the merging action, As a result, when the merging action is finished, a prediction calculation is performed on the situation of surrounding vehicles on the main lane.

  The risk calculation unit 31 at the end of the merging action is a predicted value of the situation of the host vehicle at the end of the merging calculated by the own vehicle situation calculating unit 29 at the end of the merging action, and a surrounding vehicle situation calculating unit at the end of the merging action Based on the predicted value of the situation of surrounding vehicles at the end of the merging calculated by 30, the degree of risk when the host vehicle starts the merging behavior at the present time is calculated. The risk level due to this merging behavior may be attributed to surrounding vehicles traveling behind the host vehicle on the main lane, or may be attributed to surrounding vehicles traveling in front of the host vehicle. In the predetermined range, the surrounding vehicle is present in both the front and the back of the own vehicle by the own vehicle situation calculating unit 29 at the end of the merging action and the surrounding vehicle situation calculating unit 30 at the end of the merging. When the merging action is finished, the risk level calculation unit 31 calculates the risk level due to the surrounding vehicle in front of the host vehicle and the risk level due to the surrounding vehicle behind the host vehicle, and the individual risk level. Calculate the overall risk level.

  The presentation information conversion unit 32 converts the risk information calculated by the risk calculation unit 31 at the end of the merging action into an information form that can be easily recognized by the driver of the host vehicle, and provides the information presentation device 9. introduce. Specifically, the presentation information conversion unit 32 converts, for example, risk information calculated by the risk calculation unit 31 at the end of the merging action into sound that represents the risk in a stepwise manner. 9 is output from the speaker. The presentation information conversion unit 32 converts the risk information calculated by the risk calculation unit 31 at the end of the merging action into an image and displays the image on the display device as the information presentation device 9. Good. Note that a specific example of such information presentation will be described later in detail.

[Specific examples of processing by the controller]
Next, a specific example of processing executed by the controller 1 in the merge support device of the present embodiment configured as described above will be described. The processing in the controller 1 mainly includes (1) processing related to calculation of the merging behavior characteristics of the own vehicle, (2) processing for judging the driving ability of the own vehicle driver, and (3) merging result prediction and risk. It is roughly divided into processing related to calculation. Hereinafter, specific examples of these processes will be described individually.

(1) Processing related to calculation of merging behavior characteristics of own vehicle First, processing related to calculation of merging behavior characteristics of the own vehicle will be described. This process includes the own vehicle state temporary storage unit 21, the merging behavior characteristic extraction unit 22, the surrounding situation temporary storage unit 23, and the merging environment case classification determination unit 24 among the functional configuration of the merging support calculation unit 14 illustrated in FIG. 2. This corresponds to the processing in the case-by-case learning calculation unit 25 and the case-by-case learning value storage unit 26.

  The flow of this process is shown in FIG. The processing flow of FIG. 3 is started when the joining support device of the present embodiment is activated by, for example, turning on an accessory switch of the host vehicle, and the operation of the joining support device is stopped at a predetermined processing cycle. It is executed repeatedly.

  When the processing flow of FIG. 3 is started, the controller 1 first determines in step S101 whether or not it is the first use of the joining support device of the present embodiment. In the case of the first use, since the data for learning the merging behavior characteristic of the own vehicle has not been obtained, the step is performed so that the merging behavior characteristic is learned using a predetermined initial value. Proceeding to S102, a predetermined initial value is stored in all data storage areas in which data for learning the merging behavior characteristic is stored, and then proceeding to Step S103. On the other hand, if the merge support apparatus of this embodiment has already been used in the past, the process directly proceeds to step S103.

  Next, in step S103, the controller 1 temporarily stores information on the own vehicle state grasped by the own vehicle grasping unit 11, for example, information such as the position, traveling speed, and steering angle of the own vehicle over a predetermined time width. Is stored with a time stamp to monitor the state of the host vehicle, and information grasped by the surrounding vehicle grasping unit 12 and the road condition grasping unit 13, for example, the position of the surrounding vehicle traveling on the main lane, Information such as travel speed, road shape around the host vehicle and congestion on the main lane is temporarily stored with a time stamp with a predetermined time width, and the status of surrounding vehicles and the road around the host vehicle are monitored. To do.

  Next, in step S104, the controller 1 determines whether or not the host vehicle has passed the junction point based on the information from the host vehicle grasping unit 11. And when the own vehicle passes the junction, it progresses to step S105, and when not passing the junction, it progresses to step S110.

  When it is determined that the host vehicle has passed the joining point and the process proceeds to step S105, the controller 1 extracts the behavior characteristics of the host vehicle when passing the joining point in step S105. In extracting the merging behavior characteristic, information on the host vehicle state temporarily stored in step S103 is used. Specifically, the controller 1 calculates the steering start time and the steering end time at the time of execution of merging (lane change from the merging lane to the main lane) from the time-series data of the steering angle of the host vehicle at the time of the merging that is temporarily stored. And the steering characteristic from the start time to the steering end time is extracted, and the acceleration / deceleration characteristic is extracted from the time-series data of the traveling speed of the own vehicle during that time. And so on.

  Next, in step S106, the controller 1 is in the case of the merge environment at the merge point that has passed based on the information on the own vehicle state temporarily stored in step S103, the information on the surrounding vehicle state, the information on the road condition, and the like. Divide (category environment pattern classification). The case classification of the merging environment is performed according to the classification for each parameter indicated by the following (a) to (f), for example.

(A) TTC with surrounding vehicles behind the host vehicle on the main lane: (long, medium, short)
(B) TTC with surrounding vehicles in front of the vehicle on the main lane: (long, medium, short)
(C) Your vehicle's merging execution position on the merging lane: (front / middle / rear)
(D) Main line lane congestion: (High / Medium / Low)
(E) Travel speed of own vehicle: (High / Medium / Low)
(F) Road shape of merge lane: (parallel type, direct type)
In each of the above parameters, the specific value of each divided region may be set to an appropriate value in advance. In the present embodiment, for example, the surrounding vehicle behind the host vehicle on the main lane of (a) The TTC is set such that short = less than TTC 8 seconds, medium = TTC 8 seconds or more and less than 16 seconds, and long = TTC 16 seconds or more.

  Next, in step S107, the controller 1 defines an array variable for each data indicating the merging behavior characteristic of the host vehicle by each parameter described above, and each of the merging behavior characteristics extracted in step S105 is included in each array variable. Data (the time (Δt) taken for the own vehicle to join, the steering characteristics from the steering start time to the steering end time, and the acceleration / deceleration characteristics during that time) are stored. For example, the array variable DT for storing the time (Δt) taken for the own vehicle to join is defined as follows.

DT (R-TTC, F-TTC, POSITION, T-JAM, V, JUNCTION, N)
R-TTC (TTC with surrounding vehicles behind the host vehicle on the main lane): (Long: 1, Medium: 2, Short: 3)
F-TTC (TTC with surrounding vehicles in front of the host vehicle on the main lane): (Long: 1, Medium: 2, Short: 3)
POSITION (the merge execution position of the vehicle on the merge lane): (front: 1, medium: 2, rear: 3)
T-JAM (main line lane congestion): (High: 1, Medium: 2, Low: 3)
V (travel speed of own vehicle): (High: 1, Medium: 2, Low: 3)
Junction (road shape of merging lane): (Parallel type: 1, Direct type: 2)
N (Data serial number)
Next, in step S108, the controller 1 determines whether or not the number of data (N) in the data storage area in which the data is newly stored in step S107 exceeds a predetermined number, and the number of data reaches the predetermined number. If it exceeds, the process proceeds to step S109, the driving behavior characteristic of the host vehicle is learned from the data stored in the data storage area, the learning value is updated, and the process returns to step S103. In the present embodiment, an average value of a predetermined number of data is used as a learning value. On the other hand, if the number of data does not exceed the predetermined number, learning alone cannot be performed in the data storage area, and the process advances to step S114 to perform learning in the substitute area.

  The processing from step S110 to step S113 is processing for enabling learning by substituting data at the time of lane change when the number of data to be learned does not exceed a predetermined number in any data storage area. That is, when it is determined in step S104 that the host vehicle has not passed the junction, the controller 1 determines in step S110 whether or not the host vehicle has changed lanes. If the host vehicle has not changed the lane, the process returns to step S103, and monitoring of the host vehicle state, surrounding vehicle state, and road state is continued. On the other hand, if it is determined that the host vehicle has changed lanes, the controller 1 stores the host vehicle temporarily stored in step S103 in the next step S111, as in the merge behavior characteristic extraction process in step S105. Using the information on the state, the behavior characteristic of the host vehicle at the time of lane change is extracted. The data indicating the behavior characteristics at the time of the lane change extracted is an array variable similar to that at the time of merging described above using TTC (long: 1, medium: 2, short: 3) with the surrounding vehicle behind the host vehicle as a parameter. Is defined and stored in the lane change data storage area.

  Next, in step S112, the controller 1 determines whether or not the number of data in the lane change data storage area exceeds a predetermined number necessary for learning, and if the number of data exceeds the predetermined number. Proceeding to step S113, the driving behavior characteristic at the time of lane change of the host vehicle is learned from the data stored in the data storage area of the lane change and the learning value is updated as in the learning of the merging behavior characteristic at step S109. Then, the process returns to step S103. The learning value of the driving behavior characteristic at the time of changing the lane is used when the number of data for learning at merging is not enough. On the other hand, if the number of data in the lane change data storage area does not exceed the predetermined number, the driving behavior characteristic at the time of lane change of the host vehicle cannot be learned, so the process returns to step S103 and the host vehicle state and surrounding vehicle state Continue to monitor road conditions.

  The processing from step S114 to step S126 is processing for performing learning by substituting the data stored in the other data area when the number of data stored in the individual data area is less than the predetermined number. . This is intended to provide assistance in accordance with the characteristics of the driver even for the driver who rarely passes through the junction. In addition, for a driver who passes a lot of merging points, the driver gradually shifts to a learning value that is classified in detail, thereby enabling driving assistance with higher accuracy.

  Specifically, in step S114, the controller 1 first matches the conditions of the merge environment (rear vehicle condition) classified in step S106 in the data storage area divided by the parameter (a) described above. It is determined whether or not the number of data in the data storage area to be exceeded exceeds a predetermined number. If the number of data in the data storage area that matches the rear vehicle condition of the merging environment does not exceed the predetermined number, in step S115, the merging behavior characteristic is substituted with the data indicating the behavior characteristic when the lane of the host vehicle is changed. The learned value is calculated, and the calculated learned value is stored in the corresponding data storage area.

  On the other hand, when the number of data in the data storage area that matches the rear vehicle condition exceeds a predetermined number, the controller 1 is divided into the parameter (a) and the parameter (b) described above in the next step S116. In the data storage area, it is determined whether or not the number of data in the data storage area that matches the conditions of the confluence environment (rear vehicle condition and forward vehicle condition) classified in step S106 exceeds a predetermined number. If the number of data in the data storage area that matches the conditions of the merge environment including the preceding vehicle conditions does not exceed the predetermined number, in step S117, the merge behavior characteristics are calculated from all the data that match the rear vehicle conditions. A learning value is calculated, and the calculated learning value is stored in the corresponding data storage area.

  On the other hand, if the number of data in the data storage area that matches the conditions of the merging environment including the preceding vehicle condition exceeds a predetermined number, the controller 1 determines that the parameter (a) and parameter described above are the next step S118. In the data storage area divided by (b) and parameter (c), the data storage area that matches the conditions of the merge environment (rear vehicle condition, forward vehicle condition, and merge execution position condition) divided in step S106. It is determined whether or not the number of data exceeds a predetermined number. If the number of data in the data storage area that matches the conditions of the merge environment including the merge execution position condition does not exceed the predetermined number, in step S119, all data that matches the rear vehicle condition and the front vehicle condition. The learning value of the merging behavior characteristic is calculated from, and the calculated learning value is stored in the corresponding data storage area.

  When the number of data in the data storage area that matches the conditions of the merge environment including the merge execution position condition exceeds a predetermined number, the controller 1 determines that the parameter (a) and In the data storage area divided by the parameter (b), the parameter (c), and the parameter (d), the conditions of the merge environment (rear vehicle condition, forward vehicle condition, merge execution position condition, It is determined whether or not the number of data in the data storage area meeting the congestion condition) exceeds a predetermined number. If the number of data in the data storage area that matches the conditions of the merge environment including the congestion condition does not exceed the predetermined number, in step S121, the rear vehicle condition, the forward vehicle condition, and the merge execution position condition are combined. The learning value of the merging behavior characteristic is calculated from all the data, and the calculated learning value is stored in the corresponding data storage area.

  On the other hand, if the number of data in the data storage area that matches the conditions of the confluence environment including the congestion condition exceeds the predetermined number, the controller 1 determines in step S122 that the parameter (a) and the parameter described above are used. In the data storage area divided by (b), parameter (c), parameter (d), and parameter (e), the conditions of the merge environment (rear vehicle condition, forward vehicle condition and merger) classified in step S106. It is determined whether or not the number of data in the data storage area that matches the execution position condition, the congestion degree condition, and the own vehicle speed condition) exceeds a predetermined number. If the number of data in the data storage area that matches the conditions of the merge environment including the vehicle speed condition does not exceed the predetermined number, in step S123, the rear vehicle condition, the forward vehicle condition, the merge execution position condition, and the congestion degree A learning value of the merging behavior characteristic is calculated from all data that matches the condition, and the calculated learning value is stored in the corresponding data storage area.

  On the other hand, if the number of data in the data storage area that matches the conditions of the merging environment including the vehicle speed condition exceeds a predetermined number, the controller 1 performs the above-described parameter (a) and parameter in the next step S124. In the data storage area divided by (b), parameter (c), parameter (d), parameter (e), and parameter (f), the conditions of the merged environment (rear vehicle conditions and It is determined whether the number of data in the data storage area that matches the preceding vehicle condition, the merge execution position condition, the congestion degree condition, the own vehicle speed condition, and the road shape condition) exceeds a predetermined number. If the number of data in the data storage area that matches the conditions of the merge environment including the road shape condition does not exceed the predetermined number, in step S123, the rear vehicle condition, the forward vehicle condition, the merge execution position condition, and the congestion degree A learning value of the merging behavior characteristic is calculated from all data in which the condition and the vehicle speed condition match, and the calculated learning value is stored in the corresponding data storage area. On the other hand, when the number of data in the data storage area that matches the conditions of the merge environment including the road shape condition exceeds a predetermined number, in step S126, the rear vehicle condition, the forward vehicle condition, the merge execution position condition, and the congestion are determined. The learning value of the merging behavior characteristic is calculated from all data in which the degree condition, the vehicle speed condition, and the road shape condition are matched, and the calculated learning value is stored in the corresponding data storage area.

  As described above, in the present embodiment, the conditions for identifying the merging environment are subdivided, and the learning value is calculated by gradually narrowing down the subdivided conditions according to the storage state of the data. Learning the characteristics of the merging behavior of the That is, when learning cannot be performed in individual data storage areas that match all the conditions of the current merge environment, the learning values for each area include the initial values in the following order (a) to (h). Whether or not the vehicle can be used is sequentially checked, and the merging behavior characteristic of the host vehicle is learned using the learning value for each region under the most subdivided available conditions.

(B) Predetermined initial value (b) Learning value according to the area divided by the TTC condition with the surrounding vehicle behind the host vehicle at the time of lane change (c) TTC of the surrounding vehicle ahead of the host vehicle on the main lane Region-specific learning values divided by conditions (rear vehicle conditions) (d) (c) by region added with TTC conditions (front vehicle conditions) with surrounding vehicles ahead of the vehicle on the main lane Region-specific learning values (f) (e) and congestion of the main lanes added to the learning value (e) (d) by adding the conditions of the vehicle's merging execution position (merge execution position condition) on the merging lane Region-by-region learning value (g) (f) and (f) added with the condition of own vehicle traveling speed (vehicle speed condition) ) (G) Learning value for each region divided by adding the condition of the road shape of the merged lane (road shape condition) (2) Processing for Determining Driving Capability of Own Vehicle Driver Next, processing for determining the driving capability of the own vehicle driver will be described. This process corresponds to the process in the driving ability determination unit 27 in the functional configuration of the merge support calculation unit 14 shown in FIG.

  The flow of this process is shown in FIG. The processing flow in FIG. 4 is executed at predetermined time intervals, and information on the driving ability of the driver of the vehicle is updated each time the processing flow in FIG. 4 is executed.

  When the processing flow of FIG. 4 is started, the controller 1 first, in step S201, stores data indicating behavior characteristics of the own vehicle at the time of merging stored for each pattern of the merging environment (step S107 in the processing flow of FIG. 3). Call the data stored in. At this time, the controller 1 sets a data storage area in advance for determining the driving ability, stores the data for determining the driving ability in this area, and calls only the data for determining the driving ability. Good. In this embodiment, the data of the merging execution time for each TTC with the surrounding vehicle traveling behind the host vehicle on the main lane (the time (Δt) taken for the host vehicle to join) is used as data for judging driving ability. Used.

  Next, in step S202, the controller 1 calculates a deviation amount of the merging execution time (Δt) for each TTC division (long / medium / short) with surrounding vehicles traveling behind the host vehicle on the main lane. In step S203, the average value is calculated. The deviation amount (dispersity) of the merging execution time in the same section becomes smaller as the technical driving ability becomes stable, and it can be judged that the ability is excellent. Therefore, in step S204, the controller 1 uses, for example, a map that represents a relationship between the technical driving ability and the merging execution time deviation that has been created in advance, and in the same category calculated in step S203 on this map. The technical driving ability of the driver of the vehicle is determined by a method of checking the average deviation amount of the merge execution time.

  Next, in step S205, the controller 1 calls the learning value of the merging execution time (Δt) for each TTC division (long / medium / short) with surrounding vehicles traveling behind the host vehicle on the main lane. In S206, the amount of change (dispersity) between the TTC sections of the learning value of the merge execution time (Δt) called in step S205 is calculated. From this amount of change, it can be determined whether or not the driver of the host vehicle is changing the driving behavior according to the difference in TTC with the surrounding vehicle behind the host vehicle that has the greatest influence on the surrounding environment. This reflects tactical driving ability. That is, a driver with a high tactical driving ability has the ability to change the driving method flexibly according to changes in the surrounding environment, while a driver with a low tactical driving ability changes the driving method according to the surrounding environment. There is no ability. Therefore, it is possible to determine the tactical driving ability of the own vehicle driver from the amount of change in the learned value of the merge execution time (Δt) between the TTC sections. Therefore, in step S207, the controller 1 uses, for example, a map representing the relationship between the tactical driving ability and the amount of change in the merging execution time learning value between the TTC sections, which has been created in advance, and step S206 on this map. The tactical driving ability of the host vehicle driver is determined by a method of collating the amount of change in the merge execution time learning value between the TTC sections calculated in (1).

  Next, in step S208, the controller 1 determines that the driver of the host vehicle from the technical driving capability of the host vehicle driver determined in step S204 and the tactical driving capability of the host vehicle driver determined in step S207. To determine the driving ability. The overall driving ability of the own vehicle driver may be determined in detail in multiple stages, but in this embodiment, for example, according to the criteria shown in FIG. Judging by stage.

(3) Process Related to Merge Result Prediction and Risk Calculation Next, a process related to merge result prediction and risk calculation will be described. This processing includes the learning value calling unit 28 that matches the merging environment, the own vehicle situation calculation unit 29 at the end of the merging action, the end of the merging action, among the functional configurations of the merging support calculating unit 14 illustrated in FIG. This corresponds to processing in the surrounding vehicle situation calculation unit 30, the risk calculation unit 31 at the end of the merging action, and the presentation information conversion unit 32.

  The flow of this processing is shown in FIG. The processing flow of FIG. 6 is a point at a predetermined distance (for example, 50 m) from the own vehicle traveling on the merge lane based on information from the own vehicle grasping unit 11 or the road condition grasping unit 13, for example. The process is started when it is determined that the vehicle has arrived, and is repeatedly executed at a predetermined processing cycle (for example, 100 msec) until it is determined that the host vehicle has passed the junction end point.

  When the processing flow of FIG. 6 is started, the controller 1 first, in step S301, information related to the own vehicle state grasped by the own vehicle grasping unit 11, such as the position, traveling speed, steering angle, etc. of the own vehicle. Information is temporarily stored with a time stamp with a predetermined time width to monitor the state of the host vehicle, and information grasped by the surrounding vehicle grasping unit 12 and the road condition grasping unit 13, for example, on the main lane Information such as the position and speed of surrounding vehicles traveling around the vehicle, the shape of the road around the host vehicle and the congestion of the main lane is temporarily saved with a time stamp with a predetermined time width, and the status of surrounding vehicles And monitoring the road conditions around the vehicle.

  Next, in step S302, the controller 1 determines that the current merging environment is based on the information on the host vehicle state temporarily stored in step S301, the information on the surrounding vehicle state, the information on the road state, and the like as shown in FIG. It is determined to which classification of the merging environment classified in step S107 in the processing flow is applicable. The conditions for dividing the case of the merging environment are the parameters indicated by (a) to (f) described above, and it is determined for each parameter which condition the current merging environment applies according to the classification.

  Next, in step S303, the controller 1 calls the learned value of the merging behavior characteristic of the host vehicle under the conditions that apply to the current merging environment determined in step S302. Then, the controller 1 in step S304, based on the state of the host vehicle monitored in step S301 and the learned value of the merging behavior characteristic of the host vehicle suitable for the current merging environment called in step S303. When the own vehicle immediately performs the merging action (lane change from the merging lane to the main lane), the position of the own vehicle at the end of merging is calculated. In step S305, the own vehicle immediately executes the merging action. In this case, the traveling speed of the host vehicle at the end of merging is calculated. Note that the host vehicle position at the end of merging can be predicted from, for example, current host vehicle position and travel speed information, and learned steering characteristics and acceleration / deceleration characteristics. Further, the traveling speed of the host vehicle at the end of merging can be predicted from, for example, information on the current traveling speed of the host vehicle and the learned acceleration / deceleration characteristics.

  Next, in step S306, the controller 1 calls the driving skill (high, medium, low) of the host vehicle driver determined and stored by the above-described process of determining the driving ability of the host vehicle driver, and step S307. The vehicle driver determines whether or not the driving skill of the own vehicle driver is high, that is, whether or not the own vehicle driver is an advanced driver. If the driver of the host vehicle is an advanced driver, in order to calculate the situation of the surrounding vehicle in consideration of the allowable deceleration in accordance with the degree of congestion on the main lane, the controller 1 in step S308, in step S301. Recalls temporarily stored information indicating the current main lane congestion. Information indicating the congestion status of the main lane is determined by the road condition grasping unit 13 based on VICS traffic information and the like. On the other hand, if the driver of the own vehicle is a beginner driver or an intermediate driver, the controller 1 in step S309, in order to simplify the information without considering the deceleration of surrounding vehicles according to the congestion situation of the main lane, A variable representing the congestion degree of the main lane is set to 0.

  Next, in step S310, the controller 1 calls up information on the allowable deceleration of the surrounding vehicle corresponding to the congestion degree of the main lane. When the congestion on the main lane is high, the surrounding vehicles generally have a high acceleration / deceleration frequency, and the driver is paying attention to the surroundings, and there is little psychological resistance to decelerate at the junction. . For this reason, there is a high possibility that the surrounding vehicles will decelerate at the meeting point on a highly congested road, and if such a tendency is taken into consideration, more detailed behavior prediction can be performed. Of particular relevance are surrounding vehicles (hereinafter referred to as rear vehicles) traveling behind the host vehicle on the main lane. In the present embodiment, the allowable deceleration of the rear vehicle corresponding to the congestion degree of the main lane is obtained using a map indicating the correspondence relationship between the congestion degree and the allowable deceleration. It should be noted that since the variable representing the degree of congestion is set to 0 in step S309, the allowable deceleration of the rear vehicle is not taken into consideration for the beginner driver and the intermediate driver. In consideration of the fact that the level of information that can be used without any load during driving depends on driving ability, the above processing presents simple information that can be intuitively grasped for beginner and intermediate drivers. This is a process for enabling the driver to present more comprehensive information.

  Next, the controller 1 determines the state of surrounding vehicles monitored in step S301, in particular, the state of the rear vehicle, the learned value called in step S303 (the time (Δt) required to complete the merge), and step S310. Based on the obtained allowable deceleration of the rear vehicle, in step S311, the traveling speed of the rear vehicle at the time when the own vehicle ends the merging action is calculated, and in step S312, the own vehicle ends the merging action. The position of the vehicle behind is calculated at the time.

  Further, the controller 1 monitors the state of the surrounding vehicle monitored in step S301, in particular, the state of the surrounding vehicle that travels ahead of the host vehicle on the main lane (hereinafter referred to as the forward vehicle), and step S303. In step S313, the traveling speed of the preceding vehicle at the time when the host vehicle finishes the merging action is calculated based on the learned value (time taken until the merging end (Δt)) called in step S314. The position of the vehicle ahead is calculated at the time when the host vehicle finishes the merging action.

  Next, in step S315, the controller 1 determines the own vehicle position at the merging end time calculated in step S304, the traveling speed of the own vehicle at the merging end time calculated in step S305, and the merging end time calculated in step S311. Based on the traveling speed of the rear vehicle and the position of the rear vehicle at the end of joining calculated in step S312, the TTC between the host vehicle and the rear vehicle at the end of joining is calculated. Note that the TTC (collision margin time) between the host vehicle and the rear vehicle at the end of the merge is obtained by dividing the relative distance between the host vehicle and the rear vehicle at the end of the merge by the relative speed.

  Next, in Step S316, the controller 1 determines whether or not the driving skill of the own vehicle driver called in Step S306 is at a low level, that is, whether or not the own vehicle driver is a beginner driver. If the driver of the host vehicle is a beginner driver, the controller 1 makes an infinite TTC between the host vehicle and the preceding vehicle at the end of the merge in step S317 in order to present simpler information without considering the preceding vehicle. Set to large. On the other hand, if the driver of the own vehicle is an intermediate driver or an advanced driver, in order to be able to present more comprehensive information in consideration of the preceding vehicle, the controller 1 performs the merge calculated in step S304 in step S318. The own vehicle position at the end time, the traveling speed of the own vehicle at the joining end time calculated in step S305, the traveling speed of the preceding vehicle at the joining end time calculated in step S313, and the front at the joining end time calculated in step S314 Based on the position of the vehicle, the TTC between the host vehicle and the preceding vehicle at the end of merging is calculated. Note that the TTC (collision allowance time) between the host vehicle and the preceding vehicle at the end of joining is obtained by dividing the relative distance between the host vehicle and the preceding vehicle at the end of joining by the relative speed.

  Next, in step S319, the controller 1 uses a map for calculating the degree of risk corresponding to the current congestion level of the main lane based on the information indicating the congestion status of the main lane called in step S308, specifically, The map corresponding to the congestion degree of the current main lane is read out from the map showing the correspondence between the surrounding vehicle TTC and the degree of danger, and classified according to the degree of congestion. This is because the standard of risk by TTC with surrounding vehicles also changes depending on the congestion level of the main lane.

  Next, in step S320, the controller 1 uses the map called in step S319 to check the TTC with the rear vehicle at the end of merging calculated in step S315 on the map, and the degree of risk by the rear vehicle at the end of merging. If the driver of the vehicle is an intermediate driver or an advanced driver, the TTC with the preceding vehicle at the end of merging calculated in step S318 is checked against a map to determine the risk level due to the preceding vehicle at the end of merging. to decide. If the driver of the host vehicle is a beginner driver, since the TTC with the preceding vehicle is set to infinity in step S317, the danger level due to the preceding vehicle at the end of merging is not considered.

  Next, in Step S321, the controller 1 integrates the degrees of risk determined separately for the front vehicle and the rear vehicle in Step S320, and calculates a total risk level. At this time, it is desirable to calculate the overall risk level by individually weighting the risk level due to the preceding vehicle and the risk level due to the rear vehicle. Various methods can be considered as a method for calculating the overall risk. In this embodiment, for example, as shown in FIG. 7, the risk according to the TTC with the rear vehicle and the TTC with the front vehicle are calculated. The overall risk level is calculated by comparing the corresponding risk levels with a map that integrates them.

  Finally, in step S322, the controller 1 converts the risk information calculated in step S321 into an information form that can be easily recognized by the driver of the host vehicle, and transmits the information form to the information presentation device 9. The minute process ends. Thereafter, the process returns to step S301, and the processes from step S301 to step S322 are repeated in the next processing cycle.

[Specific examples of information presentation]
Next, a specific example of information presentation in the merging support apparatus of the present embodiment will be described by taking information presentation by sound and information presentation by image as examples.

(I) Information presentation by sound In the case of information presentation by sound, a speaker is used as the information presentation device 9, and the form of the dial tone output from this speaker is changed according to the risk of the merged result. Specifically, for example, as shown in FIG. 8, a dial tone output from a speaker is used as an intermittent tone, and the interval between the dial tone is controlled to become shorter as the degree of danger increases. That is, as shown in FIG. 8B, the controller 1 of the merging support apparatus according to the present embodiment is intermittent with a short interval from the speaker as the information presentation apparatus 9 when the degree of risk calculated by the above-described processing is high. If the calculated degree of risk is low, intermittent beep sounds with intervals are output from the speaker as the information presentation device 9. In addition, the controller 1 continuously changes the interval of the dial tone output from the speaker as shown in FIG. 8A in accordance with the change in the degree of risk calculated every processing cycle (for example, 100 msec). Let As a result, the driver of the own vehicle outputs from the speaker whether the change in the merging environment accompanying the traveling of the own vehicle has changed to the safe side or the dangerous side, and the direction of the change. Can be grasped from the dial tone.

  In this embodiment, the risk level information is transmitted according to the interval of the intermittent sound. However, the risk level of the merge result is obtained by changing other forms of the dial tone output from the speaker, such as the frequency and volume of the sound. May be presented, or the risk of the merging result may be presented by combining the above methods.

(Ii) Information presentation by image In the case of information presentation by image, a display device is used as the information presentation device 9, and the display form of the image displayed on this display device is changed according to the risk of the merged result. In this case, in order to reduce the line-of-sight load of the driver of the host vehicle, the driver of the host vehicle can recognize the image in the peripheral visual field without looking directly at the image, especially when the risk of the merge result is high. It is desirable to display the image in a form that attracts the driver's attention. Specifically, for example, when the risk of the merge result is high, the display color of the image is red, and the image is displayed when the risk of the merge result is medium, according to the color of the traffic light that the driver is familiar with. The display color of the image displayed on the display device is changed in accordance with the risk of the merge result, such that the display color of the image is yellow and when the risk of the merge result is low, the display color of the image is blue. In addition, the image is displayed blinking on the display device, and the blinking interval (flashing frequency) becomes shorter (the blinking frequency becomes higher) as the degree of danger increases, as in the case of information presentation by intermittent dial tone. You may make it control so. Furthermore, if the image is moved within the screen of the display device, and the direction of movement is high, the flow of the surrounding object in the driver's field of view as the vehicle moves is high. ), The direction is perpendicular to the optical flow when the risk of the merged result is medium, and the direction is the same as the optical flow when the risk of the merged result is low. May be.

  In addition, the brightness of the image displayed on the display device is controlled to increase as the degree of risk increases, or the size of the image displayed on the display device is controlled to increase as the degree of risk increases. The risk of the merge result may be presented by a technique, or the risk of the merge result may be presented by combining the above techniques.

[Effect of the embodiment]
As described above in detail with specific examples, according to the merging support apparatus of the present embodiment, the merging behavior characteristic of the own vehicle is learned, and the own vehicle driver immediately drives based on the learned value. Since the risk level is calculated by predicting the result situation when the action is executed, and the information indicating the risk level of the merged result is presented to the driver of the own vehicle, the driver of the own vehicle It becomes possible to determine a merging strategy while accurately grasping a merging result according to driving characteristics, vehicle characteristics of the host vehicle, and the like, and execute merging at an optimal timing.

  In the merging support apparatus according to the present embodiment, the merging environment in which the merging operation is performed is classified into a plurality of patterns, and the merging behavior characteristics of the own vehicle are learned for each pattern. To learn the characteristics of the merging behavior of the vehicle, including the elements to be controlled, in a fine and accurate manner, and to further improve the prediction accuracy of the merging results and to present the exact risk level to the driver of the vehicle according to the merging environment. Can do.

  In the merging support device of the present embodiment, the congestion situation of the main lane is grasped, and the merging result is predicted in consideration of the braking operation of surrounding vehicles traveling behind the host vehicle on the main lane according to the congestion situation. Therefore, the accuracy of prediction of the merging result can be further improved, and an accurate risk level can be presented to the host vehicle driver according to the merging environment.

  In the merging support apparatus according to the present embodiment, both the risk level due to surrounding vehicles traveling behind the host vehicle in the main lane and the risk level due to surrounding vehicles traveling in front of the host vehicle on the main lane are considered. Since the overall risk level is calculated, it is possible to present appropriate risk level information to the driver of the own vehicle who tends to be conscious of either the front vehicle or the rear vehicle. .

  In the merging support apparatus of this embodiment, since the calculation standard of the degree of risk is changed according to the congestion situation of the main lane, the sensitivity that the own vehicle driver feels dangerous according to the congestion situation of the main lane. Therefore, it is possible to present highly practical information to the driver of the vehicle even on a road with a high degree of congestion.

  In the merging support apparatus according to the present embodiment, the risk level calculation method is changed in accordance with the driving ability of the host vehicle driver. Therefore, even for beginner drivers who can only perform intuitive and simple information processing. Therefore, it is possible to present optimal information without causing confusion due to information.

  Further, in the merging support apparatus of the present embodiment, the driving ability of the own vehicle driver is determined from both sides of the driving tactical aspect and the driving skill aspect, and accordingly, the driving ability of the own vehicle driver is appropriately determined. be able to.

  Further, in the merging support apparatus of the present embodiment, when learning the own vehicle behavior characteristic in the merging environment pattern in which the number of data samples for calculating the learned value of the merging behavior characteristic of the own vehicle is less than the predetermined value, Since the learning value is used instead of the learning value, it is possible to provide assistance for drivers who do not travel frequently at the junction, and for drivers who frequently pass through the junction. The merge result can be predicted using a highly accurate learning value.

  The merging support apparatus described above exemplifies an embodiment of the present invention, and the technical scope of the present invention is not limited to the contents disclosed in the description of each of the above embodiments. Of course, various alternative techniques that can be easily derived from these disclosures are also included.

It is a schematic block diagram of the confluence | merging assistance apparatus to which this invention is applied. It is a functional block diagram which shows the detailed functional structure in the confluence | merging assistance calculating part of a controller. It is a flowchart which shows an example of the process performed by a controller, Comprising: The process relevant to the calculation of the confluence | merging behavior characteristic of the own vehicle. It is a flowchart which shows an example of the process which is a process performed by a controller, and judges the driving capability of the own vehicle driver. It is a figure explaining the method of judging the comprehensive driving capability of the own vehicle driver. It is a flowchart which shows an example of the process which is a process performed by a controller, and is related to a merging result prediction and risk calculation. It is a figure explaining the method of calculating the comprehensive risk degree which match | combined the danger degree by the back vehicle on a main lane, and the danger degree by a front vehicle. It is a figure explaining the example which presents the danger level of a merging result by an intermittent dial tone, (a) is a figure which shows changing a dial tone interval continuously according to a danger level, (b) is danger It is a figure explaining controlling so that a dial tone interval becomes short as a degree increases.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Controller 9 Information presentation apparatus 11 Own vehicle grasping part 12 Surrounding vehicle grasping part 13 Road condition grasping part 14 Confluence support calculating part 21 Own vehicle state temporary memory part 22 Merge behavior characteristic extracting part 23 Surrounding condition temporary memory part 24 In case of merge environment Dividing / determining unit 25 Case-by-case learning calculation unit 26 Case-by-cased learning value storage unit 27 Driving ability determining unit 28 Learning value calling unit that matches the merging environment 29 Own vehicle state calculation unit at the end of merging action 30 Calculation unit for surrounding vehicle conditions at the end of the merging action 31 Risk calculation unit at the end of the merging action 32 Presentation information conversion unit

Claims (11)

In the merging support device that supports the driving operation by the driver when the own vehicle traveling in the merging lane joins the main lane,
Own vehicle grasping means for grasping the position and behavior of the own vehicle;
Surrounding vehicle grasping means for grasping the position and behavior of surrounding vehicles traveling on the main lane,
Road condition grasping means for grasping the road condition around the own vehicle;
Based on the information from the own vehicle grasping means, the information from the surrounding vehicle grasping means, and the information from the road condition grasping means, a joining environment identifying means for identifying a current joining environment;
At the time of merging, data indicating the merging behavior characteristics of the own vehicle is calculated as a learning value based on the driving characteristics of the driver that change according to the collision margin time of the surrounding vehicle traveling on the main lane with respect to the own vehicle. By means of this, the merging behavior characteristic learning means for learning the merging behavior characteristic of the own vehicle,
Based on the current merging environment identified by the merging environment identifying means and the learning value by the merging behavior characteristic learning means, the situation of the own vehicle at the time of merging end and the surrounding vehicles when the own vehicle immediately executes merging A confluence result prediction means for predicting the situation of
A risk degree calculating means for calculating a risk degree according to a prediction result by the merging result prediction means;
An apparatus for assisting merging, comprising: information presenting means for presenting information representing the risk calculated by the risk calculating means to a driver of the host vehicle. The merging behavior characteristic learning means classifies the merging environment identified by the merging environment identifying means into a plurality of patterns, and learns the merging behavior characteristic of the own vehicle for each pattern,
The merging result predicting means is a merging end time when the own vehicle immediately executes merging based on a learning value of a merging behavior characteristic in a pattern in which the current merging environment identified by the merging environment identifying means is classified. The merging support apparatus according to claim 1, wherein the situation of the host vehicle and the situation of surrounding vehicles are predicted.   The merging behavior characteristic learning means includes a time required from when the own vehicle starts a transition operation from the merging lane to the main lane to the end, a steering behavior of the own vehicle during the transition operation, and the own vehicle during the transition operation. The merging support apparatus according to claim 1, wherein the accelerating / decelerating behavior is learned as the merging behavior characteristic. The road condition grasping means grasps the congestion situation of the main lane,
The merging result prediction means calculates an allowable deceleration of the surrounding vehicle according to the congestion situation of the main lane grasped by the road condition grasping means, and the situation of the surrounding vehicle at the merging end time according to the allowable deceleration The merging support apparatus according to claim 1, wherein:   The risk level calculating means is configured such that when the merging result predicting means predicts that there are surrounding vehicles in both the front and rear of the own vehicle within a predetermined range from the own vehicle at the end of the merging, Claims are provided for calculating a front risk degree by a vehicle and a rear risk degree by surrounding vehicles behind the host vehicle, and calculating a total risk degree by combining the front risk degree and the rear risk degree. Item 3. The merge support device according to Item 1. The road condition grasping means grasps the congestion situation of the main lane,
The risk level calculation means changes a correspondence relationship between a collision margin time and a risk level used for calculation of the risk level according to the congestion situation of the main lane grasped by the road condition grasping means. The merge support apparatus according to claim 1. A driving ability judging means for judging the driving ability of the own vehicle driver based on a learning value by the merging behavior characteristic learning means;
The merging support apparatus according to claim 2, wherein the risk level calculation unit changes a risk level calculation method according to a determination result of the driving ability determination unit. The driving ability determination means calculates the degree of dispersion of the merging behavior characteristics in the merging environment classified into the same pattern, and determines the technical driving ability of the own vehicle driver according to the degree of dispersion .
The merge support apparatus according to claim 7, wherein the risk level calculation unit changes a risk level calculation method according to the technical driving capability determined by the driving capability determination unit . The driving ability determination means calculates the degree of dispersion of the merging behavior characteristics between the merging environments classified into different patterns, and determines the strategic driving ability of the host vehicle driver according to the degree of dispersion .
The merge support according to claim 7 or 8, wherein the risk level calculation unit changes a risk level calculation method in accordance with the strategic driving capability determined by the driving capability determination unit. apparatus.   The merging behavior characteristic learning means, when the number of data samples of past merging behavior characteristics in a pattern in which the current merging environment identified by the merging environment identification means is less than a predetermined value, Learn the merging behavior characteristics by substituting the learning value of the behavior value at the time of lane change or the learning value of the merging environment pattern with the past number of data samples of the merging behavior property exceeding the predetermined value. The merging support apparatus according to claim 2, wherein:   The merging support apparatus according to claim 1, wherein the information presenting means presents the degree of risk calculated by the degree of risk calculating means to the driver of the host vehicle by sound or an image.

Images