JP2009042435A - Safe driving education device and program - Google Patents

Abstract

An object of the present invention is to enhance the effect of safe driving education aimed at preventing accidents.
According to a driving scenario, a simulator control unit 30 displays a virtual vehicle surrounding situation on a display unit 32 during a simulated driving of a teacher. The situation / behavior / error storage unit 18 stores in advance information indicating each of the driver's driving behavior and vehicle surroundings at the time of an accident for each accident pattern. The error estimation unit 16 determines whether or not the learner has performed an incorrect driving action based on the driving action of the learner and the situation around the vehicle. When the learner performs an incorrect driving action, the driving action of the learner and the situation around the vehicle when the situation / action / error comparison unit 22 performs the incorrect driving action are stored in the accident database 20. The accident control pattern that matches the driving behavior and the vehicle surrounding situation is extracted, and the simulator control unit 30 switches the normal driving scenario to the driving scenario of the accident pattern and displays the virtual vehicle surrounding situation on the display unit 32.
[Selection] Figure 1

Description

  The present invention relates to a safe driving education apparatus and a program for performing safe driving education for a teacher by causing a teacher to perform simulated driving.

  A driving simulator is known as a device for supporting driving education of a driver. The driving simulator displays a simulated image that enters the driver's field of view from the state of a virtual vehicle that travels according to the driving operation of the driver during simulated driving by causing the driver to perform simulated driving, or by an acoustic device By generating a sound simulating traveling sound or by swinging the driver's seat with a swinging device, the driver feels close to the feeling of driving an actual vehicle.

  Conventionally, various techniques related to a driving simulator have been proposed in order to perform driving education more effectively.

For example, there is a system that improves the effectiveness of learning by determining the traffic situation that the learner is not good at from the performance (comparison with the reference value) of the learner in the driving simulation, and creating an educational scenario to practice the scene a lot. It is known (for example, refer to Patent Document 1). Also known is a system that improves the effectiveness of the learning by determining the traffic situation that the learner is not good at based on the attention and tension of the learner in the driving simulation, and by creating an educational scenario to practice the scene a lot. (For example, refer to Patent Document 2). In addition, when experiencing a scenario, there is also a device that compares the results of the instructor's experience with the reference data of that scenario, gives a grade, and switches to the next optimal simulated driving scenario when the grade exceeds a certain level. It is known (for example, refer to Patent Document 3).
JP 2003-263098 A JP 2003-263099 A JP 2006-285138 A

  The technologies described in Patent Documents 1 and 2 are devices that determine a difficult scene from the result of the driving operation of the teacher and the degree of burden on the teacher and practice this much. Although this can be expected to improve driving skills, accidents do not occur only in situations where they are not good at it. Accidents often occur from situations where the driver is not very good at the situation, or due to an error in the driving behavior of the driver. Therefore, although the above technique can be applied to training aimed at improving driving skills, it is not suitable for training aimed at preventing accidents.

  The technique described in Patent Document 3 is also a technique for switching to the next simulated driving scenario from the result of the driving operation of the teacher. Is not suitable.

  An object of the present invention is to provide a safe driving education apparatus and program capable of enhancing the effect of safe driving education for the purpose of accident prevention.

  In order to achieve the above object, the safe driving education apparatus according to the present invention provides virtual driving during the simulated driving of the instructor according to the driving scenario for presenting a virtual vehicle surrounding situation to the instructor during the simulated driving. Presentation control means for controlling the vehicle surrounding situation to be presented to the presentation means, measurement means for measuring the driving behavior of the instructor during simulation driving, and the driver's behavior at the time of the accident corresponding to the accident pattern Based on the storage means storing information indicating each of the driving behavior and the vehicle surrounding situation, and the driving behavior measured by the measuring means and the vehicle surrounding situation presented to the presenting means, the driving action mistaken by the teacher An error determination means for determining whether or not an error has been performed, and when the error determination means determines that the instructor has performed an incorrect driving action, the instruction when the erroneous driving action is performed Accident extraction means for extracting an accident pattern in which the driving behavior and vehicle surrounding situation of the person coincides with the driving behavior and vehicle surrounding situation stored in the storage means, and a driving scenario used in the presentation control means, the accident extraction means And switching means for switching to a driving scenario for presenting the vehicle surrounding situation of the accident pattern extracted in step (b).

  According to such a configuration, when there is an error in the driving behavior of the teacher, an accident pattern that is likely to occur from the erroneous driving behavior can be presented to the teacher. The effect of safe driving education can be enhanced.

  The safe driving education device is a driving action that does not correspond to the erroneous driving action based on the driving action measured by the measuring means and the vehicle surrounding situation presented to the presenting means. Priority driving behavior detection means for detecting priority driving behavior that gives priority to the virtual vehicle that the learner performs simulated driving over the object that virtually exists in the virtual vehicle surroundings presented to the learner The accident extraction means is not determined by the error determination means the erroneous driving behavior of the teacher, and when the priority driving behavior of the teacher is detected by the priority driving behavior detection means, One of the accident patterns is extracted from the storage means, and the switching means is the driving used in the presentation control means when the erroneous driving action is not determined and the priority driving action is detected. The vehicle surrounding situation when another vehicle virtually existing in the virtual vehicle surrounding situation performs the driving action stored in the storage means corresponding to the accident pattern extracted by the accident extracting means You may make it switch to the driving | operation scenario for showing.

  According to such a configuration, even if there is no error in the driving behavior of the learner, if the priority driving behavior is detected, the accident pattern due to the driving error of other vehicles virtually existing around the trainer's vehicle Since the so-called accident pattern of the accident can be presented to the teacher, the effect of safe driving education aimed at preventing accidents can be enhanced.

  Further, the program of the present invention causes the computer to present the virtual vehicle surrounding situation to the presenting means according to the driving scenario for presenting the virtual vehicle surrounding situation to the instructor who is performing the simulated driving. Presentation control means for controlling, measuring means for measuring the driving behavior of the instructor during simulation driving, storage means for storing information indicating each of the driving behavior of the driver at the time of the accident and the vehicle surrounding situation corresponding to the accident pattern The error determination means for determining whether or not the teacher has performed an incorrect driving action based on the driving behavior measured by the measuring means and the vehicle surroundings presented on the presentation means; When it is determined that the instructor has performed an incorrect driving behavior, the driving behavior and the vehicle surroundings of the instructor when the erroneous driving behavior is performed are stored in the memory. Accident extraction means for extracting an accident pattern that matches the driving behavior and the vehicle surrounding situation stored in the stage, and the driving scenario used in the presentation control means present the vehicle surrounding situation of the accident pattern extracted by the accident extraction means It is a program for functioning as a switching means for switching to an operation scenario.

  According to such a program, when there is an error in the driving behavior of the teacher, the accident pattern that is likely to occur from the erroneous driving behavior is presented to the teacher as in the case of the safe driving education device. Therefore, the effect of safe driving education aimed at preventing accidents can be enhanced.

  In addition, the program is a driving action that does not correspond to the erroneous driving action based on the driving action measured by the measuring means and the vehicle surroundings presented on the presenting means, As priority driving behavior detection means for detecting priority driving behavior that prioritizes virtual vehicles simulated by the teacher over objects virtually present in the virtual vehicle surroundings presented to the teacher And the accident extraction means is configured such that when the error determination means does not determine the erroneous driving behavior of the teacher and the priority driving behavior detection means detects the teacher's priority driving behavior. Extracts any accident pattern from the storage means, and the switching means determines that the erroneous driving action is not determined and the priority driving action is detected. Another vehicle in which the driving scenario used in the presentation control unit is virtually present in the virtual vehicle surrounding situation is stored in the storage unit corresponding to the accident pattern extracted by the accident extraction unit May be switched to a driving scenario for presenting a vehicle surrounding situation.

  According to such a program, even if there is no error in the driving behavior of the learner, if the priority driving behavior is detected, an accident pattern due to a driving error of other vehicles virtually existing around the trainer's vehicle Since the so-called accident pattern of the accident can be presented to the teacher, the effect of safe driving education aimed at preventing accidents can be enhanced.

  As described above, according to the safe driving education apparatus and program of the present invention, there is an excellent effect that the effect of the safe driving education for the purpose of accident prevention can be enhanced.

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

  [First Embodiment]

  FIG. 1 is a block diagram showing a schematic configuration of a safe driving education apparatus 10 according to the present embodiment.

  The safe driving education apparatus 10 according to the present embodiment includes a normal driving scenario generation unit 12, a simulated driving execution unit 14, an error estimation unit 16, a situation / action / error storage unit 18, an accident database 20, and a situation / action / error comparison unit. 22 and an accident scenario generation unit 24.

  When a test condition is set from an input device (not shown), the normal operation scenario generation unit 12 generates normal scenario data for simulation operation (hereinafter referred to as a normal operation scenario). Test conditions include, for example, the shape of the road, the presence or absence of a signal, the presence or absence of a pause position, the presence or absence of a visual obstruction, the presence or absence of surrounding objects, and the type / number of objects around the vehicle during simulated driving. Contains information that defines the environment. The normal driving scenario is data specifying the virtual vehicle surroundings to be presented to the instructor during simulated driving. The normal driving scenario simulates the vehicle surroundings that match the test conditions set above. Presented to driving teachers. In addition, accident pattern scenario data (hereinafter referred to as accident scenario), which will be described later, is also data that prescribes the virtual vehicle surroundings to be presented to the instructor during simulation driving. It differs from the normal driving scenario in that the data is intended to be simulated.

  In the present embodiment, the normal operation scenario and the accident scenario are collectively referred to as an operation scenario. However, when both are described separately, description will be made using individual names.

  The simulated driving execution unit 14 functions as a driving simulator. The simulated driving execution unit 14 includes a simulator control unit 30, a display unit 32, an operation unit 34, and a line-of-sight detection unit 36. The simulator control unit 30 realizes a simulated driving environment by displaying the vehicle surroundings during simulated driving on the display unit 32 based on the driving scenario generated by the normal driving scenario generating unit 12 or the accident scenario generating unit 24. .

  Further, the driving behavior of the trainer during the simulated driving is detected by the operation state of the operation unit (including the steering wheel, accelerator, brake, and direction indicator) 34 operated by the trainer and the line-of-sight detection unit (for example, eye camera) 36. Detect from the results. The detection result is output to the simulator control unit 30, and the simulator control unit 30, based on the detection result and the image of the vehicle surroundings displayed on the display unit 32, is a virtual vehicle (hereinafter referred to as a simulated vehicle) The condition of the vehicle and its surroundings are extracted. The extraction result is output to the error estimation unit 16.

  Note that the gaze detection target of the teacher detected by the gaze detection unit 36 is a signal, a front-running vehicle, an oncoming vehicle, a pedestrian, and a bicycle that may collide with the host vehicle (hereinafter simply referred to as a target object). ).

  The error estimation unit 16 estimates a driving error from the information on the driving behavior and the surrounding situation of the teacher output from the simulated driving execution unit 14. When it is determined that the driving behavior of the teacher is a driving error, the situation / behavior / error storage unit 18 stores information indicating the driving behavior determined as the driving error and the surrounding situation at that time. Further, the error estimation unit 16 sequentially stores information indicating the driving behavior and the surrounding situation in the situation / behavior / error storage unit 18 even when it is not determined to be a driving error.

  In the situation / behavior / error storage unit 18, information indicating the driving behavior of the teacher (including the driving behavior resulting in the driving error) and the surrounding situation are stored in time series by the error estimation unit 16.

  The accident database 20 stores information indicating the surrounding environment at the time of the accident, the driving behavior and driving errors of the first party (the driver who caused the accident), and the surrounding situation at the time of the accident for each pattern of the traffic accident. It is a database. This database may be generated from accident data or the like of a traffic accident that has actually occurred, or may be generated from a result of examination on a desk.

  Information indicating the surrounding environment includes information such as weather, road shape, presence / absence of signal, presence / absence of pause position, presence / absence of visual obstruction, its factor, presence / absence of surrounding objects, type / number of objects, etc. Is included. FIG. 2 shows the structure of the accident database 20.

  Description will be made in order from the left in FIG. “Accident number” is an identification number for identifying an accident pattern. “Type” indicates the party type of the accident pattern, “four-to-four” indicates an accident between vehicles, and “four-step” indicates an accident between a vehicle and a pedestrian.

  “Accident type” indicates the accident content of the accident pattern. For example, “rear collision” means rear collision between vehicles, “right straight” means a collision between a right turn vehicle and an object that goes straight, and “meeting” means a collision at the intersection.

  “Party action” indicates a rough action of the first party and the second party of the accident when the accident occurred. For example, in accident number 5, the party action of the first party (the vehicle that caused the accident) is “turn right”, and the party action of the second party (the object in the accident) is “crossing”.

  In addition, as shown in FIG. 2, "weather" at the time of the accident, "road shape" where the accident occurred, "presence / absence of signal", "presence / absence of temporary stop position", "presence / absence of visual obstacles and their factors", Information on the surrounding environment of “type of surrounding objects” is recorded in the accident database 20.

  Also, on the right side of FIG. 2, information on the surrounding situation, driving behavior, and error up to the accident of the first party is shown in time series. FIG. 3 shows the time series data portion. This time-series data is stored in order of “Group No.”, “Behavior Type”, “Location”, “Behavior / Error”, and “Situation” from the top of FIG. 3 for each accident pattern. Here, as will be described later, time-series information is grouped according to a predetermined rule, and “Group No.” is recorded. The “behavior type” is one piece of information indicating driving behavior. In particular, in this “behavior type”, the state of the vehicle that changes according to the driving behavior of the first party (for example, straight ahead, stop, right turn, etc.) Is recorded. “Location” is one piece of information indicating the surrounding situation, and the location of the vehicle that was being driven by the first party at that time is stored. Further, “behavior / error” stores information on driving behavior of the first party. Of the recorded driving behavior information, the driving behavior corresponding to the driving error is recorded so as to clearly indicate that fact. In the “situation”, information on the surrounding situation other than “location” is recorded.

  Hereinafter, information such as the surrounding environment and time series data stored in the accident database 20 for each accident pattern is collectively referred to as accident data.

  The situation / behavior / error comparison unit 22 stores in the accident database 20 the time series data of the driving behavior at the time of the driving error and the surrounding situation at the time of the driving error stored in the situation / behavior / error storage unit 18. The time series data of the first party for each accident pattern being compared is compared, and when the contents and order of the time series data match, the information on the matched accident pattern is output to the accident scenario generator 24. .

  The accident scenario generator 24 generates scenario data of a driving simulator (hereinafter referred to as an accident scenario) for reproducing the accident contents of the accident pattern based on the accident pattern information output from the situation / action / error comparator 22. ) Is generated.

  Each of the above-described components is realized by a computer configured to include a CPU, a RAM, and a ROM. That is, the CPU implements the above-described components by executing a program stored in a ROM or a predetermined storage device, and the processing described below is performed. The situation / action / error storage unit 18 may be constituted by a RAM of a computer, or may be constituted by another recording medium. The accident database 20 may be configured by a recording device such as a hard disk device.

  Here, the flow of processing of the safe driving education apparatus 10 during the simulated driving of the teacher in the present embodiment will be described with reference to the flowchart of FIG.

  In step 100, the time-series data of the teacher stored in the situation / action / error storage unit 18 is cleared, and 0 is set to the variable “accident number”. The accident number is a variable in which an identification number of an accident pattern simulated by the accident scenario is set when the running operation scenario is an accident scenario. When the running operation scenario is a normal operation scenario, 0 remains set.

  In step 102, the normal operation scenario generation unit 12 generates a normal operation scenario.

  In step 104, the simulator control unit 30 switches the driving scenario used in the driving simulator to the normal driving scenario generated by the normal driving scenario generation unit 12, and an image showing the vehicle surroundings according to the normal driving scenario is displayed on the display unit 32. indicate. In this state, the instructor performs simulated driving. The simulator control unit 30 displays the vehicle surroundings on the display unit 32 in accordance with the operation of the operation unit 34 by the teacher according to the progress of the normal driving scenario.

  In step 106, the simulator control unit 30 detects the operation state of the operation unit 34 and measures the driving behavior of the teacher. At the same time, information on the vehicle surroundings (the surroundings of the vehicle) displayed on the display unit 32 is extracted. Information on the measured driving behavior and information on the extracted vehicle surrounding situation are output to the error estimation unit 16.

  In step 108, the error estimation unit 16 determines whether or not there is a change in the driving action information and the vehicle surrounding situation information acquired from the simulator control unit 30. If it is determined that there is no change, the process returns to step 106. If it is determined that there has been a change, the process proceeds to step 110.

  In step 110, it is determined according to a predetermined rule whether or not the teacher has performed an incorrect driving action (driving error) based on the information on the driving action of the learner and the information on the situation around the vehicle (error estimation).

Here, as the definition of the position on the road for error estimation vehicle normal stop by braking distance (normal stop distance) L ₁ and stop by abrupt braking distance (emergency stop distance) L ₂ of the following formula Based on this, the normal stop position and the avoidance limit position are determined.

L _n = τV + V ² / 2d [m]
Where τ is the trainer's brake response time (eg 0.7 seconds)
V: Own vehicle speed (m / s)
d: deceleration during braking (for example, 2.0 m / sec ² when calculating L ₁ ,
L ₂ calculated the time of 6.0m / sec ²⁾

As an example, FIG. 5 shows the definition of the position in the vicinity of the intersection when the vehicle 40 turns right at the intersection with a signal. Normal stop position of FIG. 5, the vehicle 40 is deceleration start position for stopping the intersection edge portion by the brake operation is determined from the normal stopping distance L _1. Avoidance limit position is a deceleration start position for the vehicle 40 is stopped at the intersection end in sudden braking is determined from the emergency stopping distance L _2.

FIG. 6 shows the definition of the position in the vicinity of the intersection of the oncoming straight vehicle when the host vehicle 40 makes a right turn at an intersection with a signal. Normal stop position of FIG. 6, the opposite straight vehicle is deceleration start position for stopping the intersection edge with the brake operation, which is also obtained from a normal stopping distance L ₁ calculated using the above equation. Avoidance limit position is a deceleration start position for opposing straight vehicle stops at an intersection end in sudden braking, which is also obtained from the emergency stopping distance L ₂ that is calculated using the above equation.

  The driving error of the teacher is estimated using the tables shown in FIGS. 7A to 7C according to the following rules (1) to (5).

(1) When an image of a certain object is displayed on the display unit 32 and until the avoidance limit position of the own vehicle 40 with respect to the object, the teacher does not turn his gaze toward the object. And

(2) For objects that may require deceleration of the host vehicle 40 at signalized intersections (such as waiting for an opposite right turn or a crossing pedestrian when driving straight), yellow signals, and red signals Then, after directing the line of sight to the object, estimation is performed according to the table in FIG. 7A from the accelerator and brake operations performed immediately before the intersection.

  FIG. 7A is a table showing recognition / judgment error determination criteria for a yellow light, a red light, and a crossing object. According to this criterion, the trainer has turned his or her eyes on the object displayed on the display unit 32. It is determined whether it is a recognition error, a determination error, or no error from the three elements of the line of sight, the accelerator, and the brake, such as whether it has increased, whether the brake is turned off, or turned on.

  Specifically, if the trainee's line of sight is directed to the yellow signal, but there is no change in the accelerator operation just before the intersection, a yellow signal recognition error or a determination error that the yellow signal can be ignored Since either one is considered, it is estimated that “yellow signal recognition error or judgment error”. Further, when the accelerator is turned off and the brake is turned on, it is considered that the vehicle is decelerating by recognizing the yellow signal, and therefore, “no error (recognized)” is set.

(3) Regarding the green light, “no error (recognized)” when the line of sight is turned. FIG. 7B shows a table showing recognition error determination criteria for a green light.

(4) Since the vehicle 40 turns right, the yellow signal and the red signal when waiting in the intersection are not recognized as errors even if the line of sight is not directed.

(5) Oncoming straight vehicles, crossing pedestrians, bicycles, and the like when the host vehicle 40 is waiting in the intersection because it makes a right turn are estimated according to the table in FIG.

  FIG. 7C is a table showing recognition / judgment error estimation criteria for an object when the host vehicle 40 is waiting for a right turn. In this criterion, whether or not the line of sight is directed toward the object, or when the line of sight is directed toward the object, the position of the object and the traveling state of the oncoming vehicle that is one of the objects (acceleration Medium, constant speed, decelerating), it is determined whether it is a recognition error, a determination error, or no error. The “position of the object” in this table is the position of each object with respect to the intersection, and is determined from the moving speed of each object.

  Specifically, “Before starting the vehicle's right turn, the trainer's line of sight was facing the oncoming vehicle 42, and when the line of sight was facing, the oncoming vehicle 42 was running at a constant speed just before the opposite intersection (also in FIG. 6). If the instructor starts to turn right on his / her own vehicle 40 afterwards, the instructor has recognized the oncoming straight vehicle 42 but made an erroneous determination that the oncoming straight vehicle 42 will decelerate. This is considered a “judgment error”.

  In step 112, the error estimation unit 16 stores information indicating the driving behavior of the teacher and the vehicle surrounding situation as time series data in the situation / behavior / error storage unit 18. Note that the driving behavior determined to be a driving error in step 110 is stored so as to be clearly indicated as a driving error. FIG. 8 is a specific example of time-series data stored in the situation / action / error storage unit 18. As shown in FIG. 8, information indicating the driving behavior and the surrounding situation is stored in time series. The situation / action / error storage unit 18 stores information on the same items as the time-series data of the first party stored in the accident database 20 (see also FIG. 3).

  In step 114, the situation / behavior / error comparison unit 22 performs the first party's time series data stored in the situation / behavior / error storage unit 18 and the first party for each accident pattern stored in the accident database 20. Compare with time series data.

  FIG. 9 is a flowchart showing the flow of comparison processing performed by the situation / action / error comparison unit 22.

  In step 200, the number Erk of driving errors in the time series data of the teacher stored in the situation / action / error storage unit 18 is counted. As described above, the situation / behavior / error storage unit 18 stores the driving behavior that is a driving error among the driving behaviors of the teacher so as to be clearly indicated. Count this.

  In step 202, it is determined whether Erk is zero. Here, if it is determined that Erk is not 0, in step 204, grouping of the teacher's time-series data is performed.

  FIG. 10 is an explanatory diagram for explaining a rule for group separation of time series data of a teacher. First, the time series data is examined in order from the top, and the place of the time series data that first causes an operation error is determined as grp1 (grp: group) ((1) in FIG. 10). Thereafter, the time series data is examined in order, and the same group number is assigned as long as either “behavior type” or “location” does not change ((2) in FIG. 10). On the other hand, when at least one of “behavior type” and “location” changes, a group number that is added to the immediately preceding group number is added to the time-series data ((3) in FIG. 10). The above is repeated and a group number is assigned to the end of the time series data. This group separation rule is also used when storing the time series data of the first party in the accident database 20.

  Next, iterative processing is performed in the comparison loop 1 from step 206 to step 224. In this comparison loop 1, first, the initial value (1) is set to the variable (n), and the repetition range up to step 224 is processed. In the second and subsequent iterations, the increment value (1) is added to the variable (n), and if it is equal to or less than the closing price (number of accident patterns stored in the accident database 20), the repetition range is processed. This is repeated until the variable (n) exceeds the closing price.

  Details of the iterative process in the comparison loop 1 will be described below.

  In step 208, the number of groups of the teacher's time series data is compared with the number of groups of the first party's time series data of the nth accident pattern in the accident database 20. If the former group number is less than or equal to the latter group number, the match flag is set to 1 in step 210.

  Next, the iterative process is performed in the comparison loop 2 from step 212 to step 218. In this comparison loop 2, first, the initial value (1) is set to the variable (i), and the repeat range up to step 218 is processed. In the second and subsequent iterations, the increment value (1) is added to the variable (i), and if it is equal to or less than the closing price (the number of groups in the teacher's time series data), the iteration range is processed. This is repeated until the variable (i) exceeds the closing price.

  In this comparison loop 2, first, in step 214, the time series data of the group i of the teacher is compared with the time series data of the group i of the nth accident pattern in the accident database 20. If they match, the variable (i) is incremented by 1 without doing anything, and the same comparison process is repeated for the next group. If it is determined in step 214 that the two do not match, the match flag is set to 0 in step 216, then the variable (i) is incremented by 1, and the same comparison process is repeated for the next group. . In this way, the time series data of the teacher is compared with the time series data of the nth accident pattern in the accident database 20. When the comparison is completed for each group, the process exits the comparison loop 2 and proceeds to step 222.

  On the other hand, if the number of groups of the teacher's time series data is greater than the number of groups of the first party's time series data of the nth accident pattern in the accident database 20 in step 208, the match flag is set to 0 in step 220. Is set and the routine proceeds to step 222.

  In step 222, it is determined whether or not the match flag is 1. If the match flag is 1, it is determined that the n-th accident pattern is an accident pattern that matches the teacher's time-series data, and the process exits comparison loop 1 and proceeds to step 228. n is set and the comparison process is terminated.

  On the other hand, if the match flag is 0 in step 222, the nth accident pattern indicates that the number of groups did not match in step 208 or that the time series data did not match in step 214. Can be determined not to be an accident pattern corresponding to the time-series data of the teacher. Accordingly, in the comparison loop 1, the increment value (1) is added to the variable (n), and the processing of the comparison loop 1 is repeated for the next accident pattern.

  In addition, when the variable (n) exceeds the number of accident patterns and exits the comparison loop 1, it is determined that the match flag is 0, or in step 202, Erk is 0 (no driving error of the teacher). If this is the case, in step 226, the corresponding accident number is set to 0 to indicate that there is no corresponding accident pattern, and this comparison process is terminated.

  Here, the comparison process will be described with a specific example. FIG. 11 is an explanatory diagram illustrating a specific example of a comparison process between time series data of a teacher (that is, time series data in the situation / action / error storage unit 18) and time series data in the accident database 20.

  As shown in FIG. 11, the situation / action / error comparison unit 22 compares time-series data of the same group number in the time-series data of the situation / action / error storage unit 18 and the accident database 20.

  For example, when all of the time series elements included in the teacher's time series data apply to the time series elements of the accident database of the group with the same number, this group is assumed to match (ie, the match flag remains 1). . However, it is assumed that there is no time series element (driving behavior that is not a driving error) that is “recognized” in the time series data of the learner that does not apply to the time series element of the accident database.

  When all the groups of the teacher's time series data have an accident pattern that matches the time series data of the accident database 20, the driving behavior of the previous teacher is the driving behavior of the first party of the accident pattern. Therefore, it can be said that the accident pattern is highly likely to occur during driving by the teacher. Therefore, the accident pattern is set as a detection result (that is, the accident pattern number detected in step 228 is set as the accident number), and the comparison process is terminated.

  When the comparison process is completed, the process returns to the main routine of FIG. 4 and the process of step 116 is performed.

  In step 116, the situation / action / error comparison unit 22 refers to the numerical value of the accident number and determines whether or not there is an accident pattern in the accident database 20 that matches the time-series data of the teacher. If the accident number is not 0, it is determined that there is a matching accident pattern, and the process proceeds to step 118.

  In step 118, the situation / action / error comparison unit 22 determines whether or not the corresponding “accident number” is the same as the “accident number”. Here, when it is determined that the “accident number” and the “accident number” are not the same, it indicates that a new accident pattern that is likely to occur in the driving of the teacher is found. Therefore, the situation / action / error comparison unit 22 sets the corresponding “accident number” to “accident number” in step 120, and in step 122, generates the accident scenario information on the accident pattern indicated by the corresponding accident number. An accident scenario is generated that is output to the unit 24 so that the teacher can experience the accident pattern.

  When the accident scenario generation unit 24 receives information on an accident pattern (here, an accident number) from the situation / action / error comparison unit 22, the accident scenario generation unit 24 generates an accident scenario of the accident pattern. Specifically, a scenario table as shown in FIG. 12 is prepared in advance, the information of the accident scenario stored in the scenario table is extracted from the received accident number, and the information on the accident scenario is simulated operation execution unit 14 is converted into an executable format and generated. In addition, the situation / action / error comparison unit 22 outputs accident data (including information on the surrounding environment and time series data) from the situation / action / error comparison unit 22 to the accident scenario generation unit 24. However, an accident scenario may be generated each time from these accident data.

  When the accident scenario generation unit 24 generates an accident scenario, in step 124, the simulator control unit 30 of the simulated operation execution unit 14 switches the operation scenario to be executed to the generated accident scenario. Thereby, the vehicle surroundings according to the accident scenario are displayed on the display unit 32.

  In step 126, the simulator control unit 30 determines whether or not an accident according to the accident scenario has occurred during the simulated driving of the instructor. If no accident has occurred yet, in step 128, the accident scenario It is determined whether or not the process has ended. In step 128, when the simulator control unit 30 determines that the accident scenario has not ended, the simulator control unit 30 returns to step 106, continuously measures the driving behavior of the trainer, and extracts information on the vehicle surrounding situation, The subsequent processing is repeated.

  On the other hand, if the simulator control unit 30 determines in step 126 that an accident has occurred, the simulation operation is terminated there. If it is determined in step 128 that the accident scenario has ended without any accident occurring during execution of the accident scenario, the process returns to step 100 and a signal is sent from the simulator control unit 30 to the normal operation scenario generation unit 12. Then, the normal operation scenario is generated again. The normal operation scenario generation unit 12 generates a new normal operation scenario following the situation after the accident avoidance. The simulator control unit 30 displays an image on the display unit 32 according to the generated normal operation scenario and continues the simulated operation.

  If it is determined that there is no accident pattern in the accident database 20 that matches the teacher's time-series data (step 116, N), or there is a matching accident pattern, the accident pattern is currently being executed. If it is determined that the accident pattern of the accident scenario is (step 118, Y), the process proceeds to step 126 as it is. If the operation scenario currently being executed is a normal operation scenario, a negative determination is made at step 126 or step 128. Therefore, in that case, the process returns to step 106 as it is, and the driving behavior of the teacher is continuously measured, information on the vehicle surrounding situation is extracted, and the subsequent processing is repeated.

  Here, taking a right turn scene at an intersection with a signal (see FIG. 13) as an example, the error estimation unit 16 estimates a driving error from the driving behavior measured and acquired by the simulated driving execution unit 14 and the surrounding situation. The process for switching to the corresponding accident scenario while storing the time series in the situation / action / error storage unit 18 will be specifically described.

(1) When the simulator control unit 30 detects that the trainee has turned his / her line of sight to the traffic light (blue light) on the screen at the vehicle position (a), the error estimation unit 16 follows the table of FIG. It is estimated that “green light is recognized”, and the result is stored in the situation / action / error storage unit 18 (first time-series data in FIG. 8).

(2) Similarly, at the own vehicle position (b), the fact that the teacher's right turn signal and deceleration start operation are detected is stored in the situation / action / error storage unit 18 (second time-series data in FIG. 8). .

(3) In the case where the line of sight is not directed toward the pedestrian 44 who is walking at the position (e) until the own vehicle 40 reaches the avoidance limit position with respect to the intersection at the own vehicle position (c). The error estimation unit 16 estimates “pedestrian recognition error” according to the table of FIG. 7A and stores it in the situation / behavior / error storage unit 18 (third time-series data in FIG. 8).

(4) The situation / behavior / error comparison unit 22 calculates the time-series data (from time-series data 1 to 3 in FIG. 8) of the teacher obtained so far and the time-series data in the accident database 20 (FIG. 3). ) With the comparison algorithm described above. As a result, it is determined that accident number 2 is applicable, and information on the accident pattern of accident number 2 is output to the accident scenario generation unit 24.

(5) The accident scenario generation unit 24 generates an accident scenario based on the transmitted accident pattern information, and the simulator control unit 30 switches the operation scenario to be executed to the generated accident scenario. Based on this accident scenario, the simulator control unit 30 causes the oncoming straight vehicle 42 to appear at the position (g).

(6) The fact that the stop of the host vehicle 40 is detected at the host vehicle position (d) is stored in the situation / action / error storage unit 18 (fourth time-series data in FIG. 8).

(7) When it is detected that the trainee has turned his / her line of sight toward the oncoming vehicle 42 that has advanced to the position (h) while the host vehicle 40 is stopped at the host vehicle position (d), The error estimation unit 16 temporarily holds it.

(8) Further, the signal turns yellow while the host vehicle 40 is stopped at the host vehicle position (d). When it is detected that the teacher has turned his eyes to the yellow signal, the error estimation unit 16 estimates “yellow signal recognition” and stores it in the situation / behavior / error storage unit 18 (the fifth time-series data in FIG. 8). ).

(9) Further, when the host vehicle 40 is stopped at the host vehicle position (d), the trainee has turned his eyes to the oncoming straight vehicle 42 which has proceeded to the position (i) and has been reduced due to a yellow signal. Is detected, the error estimation unit 16 temporarily holds the detection information in the form of updating the detection information temporarily held in (7).

(10) Further, at the own vehicle position (d), when it is detected that the instructor releases the brake and steps on the accelerator to start a right turn, the error estimating unit 16 proceeds to the position of the oncoming straight vehicle 42 and (f). The error of the teacher for the pedestrian 44 who has performed is estimated. Specifically, for the oncoming straight vehicle 42, first, the information held in (9) above (the direction of the line of sight and the position and action of the oncoming straight vehicle at that time) are shown in FIG. It is presumed that there is no error (recognized) according to the table. This estimation result is sent to the situation / action / error storage unit 18 and stored as “recognizing the deceleration of the oncoming straight vehicle” (the sixth time-series data in FIG. 8). Further, the crossing pedestrian 44 has not turned his gaze before the start of the right turn, so it is estimated as a cognitive error according to the table of FIG. 7C and is stored in the situation / action / error storage unit 18. (Time series data 7th in FIG. 8).

(11) Furthermore, the error estimation unit 16 stores the driving action of the vehicle 40 starting to turn right in the situation / action / error storage unit 18 (8th time-series data in FIG. 8).

(12) The situation / behavior / error comparison unit 22 compares the time series data of the teacher obtained so far with the time series data in the accident database 20. As a result, since accident number 5 corresponds, information on the accident pattern of accident number 5 is output to accident scenario generation unit 24.

(13) The accident scenario generation unit 24 newly generates an accident scenario based on the transmitted accident pattern information and outputs the accident scenario to the simulated operation execution unit 14. The simulator control unit 30 of the simulated driving execution unit 14 switches the driving scenario of the simulated driving to the generated accident scenario. Based on this accident scenario, an image that the pedestrian 44 at the position (f) crosses is displayed on the display unit 32.

(14) If the instructor hits the crossing pedestrian 44 without noticing, the simulated driving is ended there. On the other hand, if the teacher notices the crossing pedestrian 44 and stops and does not cause an accident, a signal is sent from the simulator control unit 30 of the simulated driving execution unit 14 to the normal driving scenario generation unit 12, and the normal driving is performed again. Simulate operation is continued by generating a scenario.

  As described above, the driving simulator driving scenario is switched to show the accident scene that is likely to occur based on the driving behavior and driving error of the teacher and the surrounding situation at that time, and the accident is experienced by the teacher. be able to.

  [Second Embodiment]

  In the safe driving education apparatus 50 according to the present embodiment, no error that causes the instructor to cause an accident during simulated driving occurs, and the vehicle is prioritized over objects existing around the vehicle during simulated driving. When priority awareness is detected, an accident scenario is generated in which surrounding vehicles other than the own vehicle operate according to the time-series data of the first party in the accident database 20. In this way, the teacher is given a simulated experience of an accident (accident caused by an error behavior of an object around the vehicle).

  FIG. 14 is a block diagram illustrating a schematic configuration of the safe driving training apparatus 50 according to the second embodiment.

  The safe driving education device 50 of the present embodiment includes a normal driving scenario generation unit 12, a simulated driving execution unit 14, an error estimation unit 16, a situation / action / error storage unit 18, an accident database 20, a priority awareness detection unit 52, a situation A behavior / error comparison unit 54 and an accident scenario generation unit 56 are provided. Here, the constituent elements having the same reference numerals shown in FIG. 1 and FIG. 14 mean the constituent elements having the same functions, and the description thereof will be omitted.

  The priority consciousness detection unit 52 detects the priority consciousness of the learner from the information on the driving behavior and the surrounding situation of the learner received from the simulator control unit 30.

  Presence or absence of priority is determined by whether or not the driving behavior of the teacher registered in the table of FIG. For example, when there is a vehicle or a pedestrian waiting to pass through the intersection and the accelerator is not loosened, it is determined that there is priority. In addition, when passing through an intersection with a signal, if the accelerator is not loosened even though it is a yellow signal, it is determined that there is priority.

  The priority awareness detection unit 52 outputs a detection signal to the situation / behavior / error comparison unit 54 when the priority awareness of the teacher is detected.

  The situation / behavior / error comparison unit 54, like the situation / behavior / error comparison unit 22 of the first embodiment, is a driving operation at the time of the driving error of the teacher stored in the situation / action / error storage unit 18. When the time series data of the surrounding situation at the time of behavior and driving error is compared with the time series data of the first party for each accident pattern stored in the accident database 20, and the contents and order of the time series data match The information on the matched accident pattern is output to the accident scenario generation unit 24. The situation / behavior / error comparison unit 54 of the present embodiment is a case in which no driving error that causes an accident occurs in the driving for a certain time or a certain distance, and the priority awareness detecting unit 52 When priority awareness is detected, the accident scenario generator 56 outputs a signal for generating an accident scenario of the accident.

  The accident scenario generation unit 56 generates an accident scenario for reproducing the accident contents of the accident pattern based on the accident pattern information output from the situation / action / error comparison unit 54. When the accident / scenario scenario generation signal received from the situation / action / error comparison unit 54 is received, the vehicle around the host vehicle is the first party in the accident pattern stored in the accident database 20. Accident scenarios that drive according to the series data are generated. In other words, an accident scenario is generated for reproducing an accident pattern that will be an accident when viewed from a teacher.

  Here, the operation of the safe driving education apparatus 50 during the simulated driving of the instructor in the present embodiment will be described with reference to the flowchart of FIG.

  In step 300, 0 is set to the driving time, the travel distance, and the number of errors.

  The processing from step 302 to step 308 is the same as the processing from step 100 to step 106 in FIG. 4 in the first embodiment, and a description thereof will be omitted.

  In step 310, the driving time and travel distance are updated according to the driving operation of the teacher.

  In step 312, the error estimation unit 16 determines whether or not there is a change in the driving action information and the vehicle surrounding information acquired from the simulator control unit 30. Since the determination here is the same as the processing in step 108 of FIG. 4 in the first embodiment, detailed description thereof is omitted.

  If it is determined in step 312 that there is no change, the process returns to step 308. If it is determined that there has been a change, in step 314, the error estimation unit 16 determines that the teacher has erroneous driving behavior (driving error) based on the information on the driving behavior of the teacher and the information on the vehicle surroundings. Is determined according to a predetermined rule (error estimation). The error estimation method is the same as that in the first embodiment.

  Next, in step 316, the error estimation unit 16 determines whether or not an operation error has occurred as a result of the error estimation. If it is determined that an operation error has occurred, +1 is added to the number of errors in step 318, and the process proceeds to step 320.

  Since the processing from step 320 to step 336 is the same as the processing from step 112 to 128 in FIG. 4 in the first embodiment, the description thereof is omitted. That is, the time series data of the teacher and the time series data for each accident pattern in the accident database 20 are compared, and if they match, an accident scenario of the accident pattern is generated and the display unit 32 follows the accident scenario. If an image is displayed and they do not match, the image display according to the normal operation scenario is continued.

  On the other hand, if the error estimation unit 16 determines in step 316 that a driving error has not occurred, the process proceeds to step 340, where the priority awareness detection unit 52 sets a predetermined threshold value for driving time or traveling time. It is determined whether it has been exceeded. If a negative determination is made in step 340, the process proceeds to step 320.

  If an affirmative determination is made in step 340, the priority awareness detection unit 52 determines the presence or absence of priority awareness in step 342. As described above, the presence or absence of priority awareness is determined based on whether or not the driving behavior indicating the priority awareness registered in the table shown in FIG. 15 is detected. If it is determined in this determination that no driving action indicating priority is detected, a negative determination is made in step 344 and the routine proceeds to step 320. When the driving behavior indicating priority awareness is detected by the priority awareness detection unit 52, an affirmative determination is made in step 344, and a detection signal indicating detection of priority awareness is output to the situation / action / error comparison unit 54. Then, the process proceeds to step 346.

  In step 346, the situation / behavior / error comparison unit 54 determines whether or not the driving scenario currently being executed is a normal driving scenario. If the driving scenario being executed is a normal driving scenario, in step 348, the situation / action / error comparing unit 54 extracts one of the accident patterns registered in the accident database 20, and the extracted accident pattern. An image generated by the surrounding vehicle of the host vehicle on the display unit 32 is displayed on the display unit 32, and a signal for generating an accident scenario for causing the learner to simulate the “getting accident” is displayed in the accident scenario generation unit 56. Output to.

  There are various methods for extracting the accident pattern at this time. For example, information on the surrounding environment of the trainee's vehicle during simulated driving (information such as “weather” and “road condition” shown in FIG. 2) ) And the surrounding environment of the driving scenario that is currently being executed may be extracted from the accident database 20. Thereby, when switching from a normal driving scenario to an accident scenario, it is possible to make a transition in a natural manner without greatly changing the surrounding environment.

  The situation / action / error comparison unit 54 outputs the accident number information indicating the accident pattern extracted from the accident database 20 and a signal for generating the accident scenario of the accident to the accident scenario generation unit 56.

  When the accident scenario generation unit 56 receives the signal for generating the accident scenario of the accident and the information of the accident number, the accident in which the surrounding vehicle of the own vehicle performs wrong driving action as the first party of the accident pattern of the accident number. Generate a scenario. As described in the first embodiment, the generation method prepares an accident scenario table as shown in FIG. 12 in advance, and extracts information on the accident scenario stored in the table from the received accident number. The accident scenario information may be generated by converting it into a format that can be executed by the simulated driving execution unit 14, or the accident data of the corresponding accident pattern (the surrounding environment or (Including information on time-series data) may be output to the accident scenario generation unit 24, and the accident scenario generation unit 24 may generate an accident scenario of an accident each time from these accident data.

  After generating the accident scenario of the accident received in step 348, in step 350, the simulator control unit 30 of the simulated driving execution unit 14 switches the driving scenario to be executed to the accident scenario of the accident that has been generated. As a result, the situation around the vehicle according to the accident scenario of the accident is displayed on the display unit 32.

  Thereafter, the process proceeds to step 334, and simulated driving training according to the accident scenario is performed until the accident scenario is completed or until an accident occurs.

  As described above, when the trainee does not cause any error that may lead to an accident during the simulated driving and further detects the priority consciousness prioritizing the subject vehicle over the objects existing around the vehicle during the simulated driving. Has been configured to generate accident scenarios for accidents, so that the teacher can have a simulated experience of accidents.

  The present invention is not limited to the above-described embodiments, and various design changes can be made within the scope of the invention described in the claims.

  For example, in the first and second embodiments described above, the display unit 32 that displays an image has been described as an example of the presentation unit that presents the vehicle surroundings. However, not only the display unit 32 but also a driving scenario. A safe driving education device including various types of presentation means, such as providing an acoustic device that generates sound effects and sounds as one of the presentation means, or providing a vibration device that generates vibration in the teacher's seat, etc. It may be configured.

It is a block diagram which shows the schematic structure of the safe driving training apparatus which concerns on 1st Embodiment. It is a figure which shows the structure of an accident database. It is a figure which shows the time series data part stored in the accident database. It is a flowchart which shows the flow of a process of the safe driving training apparatus in the driving | running | working simulation of the teacher in 1st Embodiment. It is a figure explaining the definition of the position in the vicinity of an intersection when the own vehicle turns right at a signalized intersection. It is a figure explaining the definition of the position in the intersection vicinity of an oncoming straight vehicle when the own vehicle turns right at a signalized intersection. (A) is a table showing recognition / judgment error determination criteria for yellow light, red light and crossing objects, (B) is a table showing recognition error determination criteria for green light, and (C) It is a table | surface which shows the recognition and judgment error criteria with respect to the target object when the vehicle is waiting for a right turn. It is a specific example of the time series data memorize | stored in a condition, action, and error memory | storage part. It is a flowchart which shows the flow of the comparison process which a condition, action, and error comparison part performs. It is explanatory drawing explaining the rule of group separation of a teacher's time series data. It is explanatory drawing explaining the specific example of the comparison process of a teacher's time series data (namely, time series data of a situation, action, error storage part), and the time series data of an accident database. It is an example of the scenario table referred when an accident scenario production | generation part produces | generates an accident scenario. It is explanatory drawing explaining the process which switches from a normal driving scenario to an accident scenario in the right turn scene of the intersection with a signal. It is a block diagram which shows the schematic structure of the safe driving training apparatus which concerns on 2nd Embodiment. It is a table which shows the judgment criteria for detecting priority awareness. It is a flowchart which shows the flow of a process of the safe driving training apparatus in the driving | running | working simulation of the instructor in 2nd Embodiment.

Explanation of symbols

10 Safety Driving Education Device 12 Normal Driving Scenario Generation Unit 14 Simulated Driving Execution Unit 16 Error Estimation Unit 18 Situation / Behavior / Error Storage Unit 20 Accident Database 22 Situation / Behavior / Error Comparison Unit 24 Accident Scenario Generation Unit 30 Simulator Control Unit 32 Display Unit 34 operation unit 36 gaze detection unit 50 safe driving education device 52 priority awareness detection unit 54 situation / behavior / error comparison unit 56 accident scenario generation unit

Claims (4)

A presentation that controls the virtual vehicle surroundings to be presented to the presenting means during the simulated driving of the instructor according to the driving scenario for presenting the virtual vehicle surroundings to the trained driver Control means;
A measuring means for measuring the driving behavior of the teacher during the simulated driving;
Storage means for storing information indicating each of the driving behavior of the driver at the time of the accident and the situation around the vehicle corresponding to the accident pattern;
An error determination unit that determines whether or not the teacher has performed an incorrect driving behavior based on the driving behavior measured by the measuring unit and the vehicle surroundings presented on the presentation unit;
When it is determined by the error determination means that the learner has performed an incorrect driving action, the driving action and the vehicle surroundings of the learner when the incorrect driving action is performed are stored in the storage means. Accident extraction means for extracting an accident pattern that matches the stored driving behavior and the situation around the vehicle;
Switching means for switching the driving scenario used in the presentation control means to a driving scenario for presenting the vehicle surroundings of the accident pattern extracted by the accident extracting means;
Including safe driving education equipment. Based on the driving behavior measured by the measuring means and the vehicle surroundings presented on the presenting means, the driving behavior does not correspond to the erroneous driving behavior and is presented to the learner. Priority driving behavior detection means for detecting priority driving behavior that prioritizes the virtual vehicle simulated by the instructor over an object that virtually exists in a vehicle surrounding situation,
The accident extraction means is the storage means when the error determination means does not determine the driver's erroneous driving behavior and the priority driving behavior detection means detects the teacher's priority driving behavior. Extract any accident pattern from
The switching means corresponds to the accident pattern extracted by the accident extraction means when the erroneous driving action is not determined and the priority driving action is detected, and the driving scenario used by the presentation control means is selected. The safe driving according to claim 1, wherein the driving behavior stored in the storage means is switched to a driving scenario for presenting a vehicle surrounding situation when another vehicle virtually existing in the virtual vehicle surrounding situation is performed. Educational equipment. Computer
A presentation control means for controlling the virtual vehicle surrounding situation to be presented to the presenting means in accordance with a driving scenario for presenting the virtual vehicle surrounding situation to a trainee during simulated driving;
A measuring means for measuring the driving behavior of a teacher during simulated driving,
Storage means for storing information indicating each of the driving behavior of the driver at the time of the accident and the situation around the vehicle corresponding to the accident pattern;
An error determination unit that determines whether or not the teacher has performed an incorrect driving behavior based on the driving behavior measured by the measuring unit and the vehicle surroundings presented on the presentation unit;
When it is determined by the error determination means that the learner has performed an incorrect driving action, the driving action and the vehicle surroundings of the learner when the incorrect driving action is performed are stored in the storage means. Accident extraction means for extracting an accident pattern that matches the stored driving behavior and vehicle surrounding situation, and a driving scenario used by the presentation control means for presenting the vehicle surrounding situation of the accident pattern extracted by the accident extraction means Switching means to switch to the operating scenario of
Program to function as. Based on the driving behavior measured by the measuring means and the vehicle surroundings presented on the presenting means, the computer is a driving behavior that does not correspond to the erroneous driving behavior, and for the learner While also functioning as a priority driving behavior detection means for detecting priority driving behavior that prioritizes the virtual vehicle that the learner performs simulated driving over the object that virtually exists in the virtual vehicle surroundings presented,
The accident extraction means is the storage means when the error determination means does not determine the driver's erroneous driving behavior and the priority driving behavior detection means detects the teacher's priority driving behavior. Extract any accident pattern from
The switching means corresponds to the accident pattern extracted by the accident extraction means when the erroneous driving action is not determined and the priority driving action is detected, and the driving scenario used by the presentation control means is selected. The program according to claim 3, wherein the driving behavior stored in the storage means is switched to a driving scenario for presenting a vehicle surrounding situation when another vehicle virtually existing in the virtual vehicle surrounding situation is performed.

Images