JP4412261B2 - Traffic simulation apparatus, method and program - Google Patents

Description

  The present invention relates to a traffic simulation apparatus, method, and program, and more particularly, to a traffic simulation apparatus, method, and program for simulating movement of a moving object in consideration of driver recognition and judgment errors.

  When planning a road traffic system, it is necessary to evaluate in advance the effects such as where and what should be maintained to prevent traffic congestion. In view of this, a road traffic simulation device that reproduces the behavior of each vehicle on a computer and simulates traffic flow and traffic congestion has been proposed (see Patent Document 1).

The road traffic simulation apparatus of Patent Document 1 represents a plurality of moving objects and a road traffic environment on a computer and simulates a traffic situation generated by the moving objects. Specifically, each mobile object is a mobile object model that is a combination of a driver model that models a driving operation by a virtual driver and a vehicle motion model that models the physical behavior of each mobile object. It is expressed. The mobile body model passes through the road traffic environment expressed on the computer independently of each other. Further, each vehicle model expresses a detailed vehicle behavior by giving the output value of the driver model to the vehicle motion model in response to various road traffic environments.
JP 2004-199287 A

  As shown in FIG. 6 of Patent Document 1, this driver model calculates a route for a moving body to follow a set route based on data such as a passable area obtained from the traffic environment database 20, and makes it difficult to pass. An apparent difficulty level distribution is generated by converting the difficulty level of an area or another moving object into a field of view from the moving object. Therefore, the driver's behavior is determined by physical restrictions such as buildings and legal restrictions such as the Road Traffic Law, and the recognition state of the driver's surrounding situation and judgment based on the recognition are not considered.

  The driver model is premised on no errors. For this reason, the road traffic simulation apparatus cannot simulate the occurrence of a traffic accident.

  For example, when an accident occurs on a road, the cause may be a driver's recognition error, determination error, or operation error. In the case of an accident at an intersection, the cause is a misrecognition such as `` I did not notice '' or `` I could not see '' the oncoming vehicle, signal, and sign ahead, misjudgment that I could see but missed the speed, handle Operation mistakes such as incorrect operation can be given.

  In particular, the recognition error depends on conditions such as the brightness due to sunlight, the color of the recognition target, the weather, and the background. Also, even if the situation is visible, the driver may not be able to recognize such as “has been overlooked” depending on the driver's arousal level. These are considered to occur probabilistically, although the situation varies depending on the driver.

  In response to such driver mistakes, various safety systems have been put into practical use that assist the driver with equipment and reduce the occurrence of accidents by providing information and controlling the vehicle.

  For example, a safety system that uses infrared light and image processing recognizes and detects surrounding objects from images captured by a camera that is sensitive to infrared light, even under conditions that are difficult for humans to see at night. And provide this information to the driver.

  The radar-based safety system detects the distance to objects around the vehicle with high accuracy, and provides information to the driver if there is a possibility of collision with the object if the course and speed are maintained. And control the vehicle.

  In a safety system using a wireless device, vehicle position and speed information is transmitted from a vehicle equipped with the wireless device, and the information is transmitted to another vehicle via a relay device on the road or directly. The vehicle that has received this information grasps the positions and speeds of the surrounding vehicles with respect to the position of the own vehicle, and determines whether it is in a dangerous state. If it is dangerous, provide information to the driver and control the vehicle.

  In order to promote the spread of such a safety system, it is necessary to verify to what extent the effects of reducing accidents can be reduced by introducing equipment. And these verifications need to be able to evaluate whether the system can cover when humans make mistakes. In addition, even if an accident of the own vehicle can be avoided by such a device, a new accident such as a rear-end collision of a succeeding vehicle may be induced by a sudden avoidance action such as an emergency brake. It is also important to verify.

  The safety system may not be able to recognize surrounding conditions depending on conditions, and may not be able to provide useful information or control the vehicle. The effectiveness of the system must be assumed to operate under all conceivable conditions.

  However, it is difficult in terms of time and cost to confirm the effect with an actual vehicle for various combinations of possible conditions such as traffic conditions and weather conditions in which the system operates. Therefore, it is useful to perform these evaluations using a simulation that can be evaluated under various conditions.

  The present invention has been proposed to solve the above-described problems, and an object of the present invention is to provide a traffic simulation apparatus, method, and program for performing a traffic simulation in consideration of the perceived state and judgment of the surrounding situation.

A traffic simulation apparatus according to the present invention is a traffic simulation apparatus that simulates the movement of a moving object on a road based on the amount of behavior of the moving object, the storage means storing a visual field image, and the storage means Generation means for generating a driver's visual field image corresponding to the position information of the moving body based on the visual field image, and presence recognition degree and recognition of the recognition object based on the driver's visual field image generated by the generating means A recognition means for calculating at least one of speed and recognition distance; and a driver's next action based on the presence recognition degree, recognition speed, and recognition distance of the at least one recognition object calculated by the recognition means. Determination means for determining, based on the driver's next action determined by the determination means, operating means for operating the moving body, And a, and vehicle motion amount calculating means for calculating a motion amount of the moving object according to the operation of the work unit.

The recognition means imitates the way of recognition of the driver, and based on the driver's visual field image, at least one of the presence recognition degree, recognition speed, and recognition distance of the recognition object reflecting the degree of recognition of the driver is obtained. calculate. This visual field image may be prepared in advance in a storage device, or may be transmitted via a network.

The determination unit imitates the driver's determination method , and determines the next action of the driver based on the calculated presence recognition degree, recognition speed, and recognition distance of at least one recognition object . Therefore, if the driver makes a recognition error, a determination error may occur, and even if the driver does not make a recognition error, a determination error may occur. The operation means operates the moving body based on the determined next action of the driver.

Therefore, the traffic simulation device calculates at least one of the presence recognition degree, the recognition speed, and the recognition distance of the recognition object based on the visual field image, and the calculated presence recognition degree and recognition speed of the calculated at least one recognition object. By determining the next action based on the recognition distance and operating the moving body, the traffic simulation can be performed in consideration of the recognition state and determination of the surrounding situation.

  The present invention can also be applied to a traffic simulation method and program.

The traffic simulation device, method, and program according to the present invention calculate at least one of the presence recognition degree, recognition speed, and recognition distance of a recognition object based on a visual field image, and calculate at least one of the calculated recognition object. By determining the next action based on the presence recognition degree, the recognition speed, and the recognition distance, and operating the moving body, a traffic simulation can be performed in consideration of the recognition state and determination of the surrounding situation.

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

  FIG. 1 is a block diagram showing a configuration of a traffic simulation apparatus according to an embodiment of the present invention. The traffic simulation device displays the result on a display device (not shown), for example, while simulating the movement of each mobile body (for example, a vehicle), and measures the number of traffic accidents.

  Here, the traffic simulation apparatus arranges the moving body model units 10A, 10B, and 10C that model the moving body on the road, and simulates the traffic situation of the moving body model units 10A, 10B, and 10C. In addition, when distinguishing three mobile body model parts, the code | symbol of AC is attached | subjected, but when it is not necessary to distinguish three mobile body model parts, the code | symbol of AC is abbreviate | omitted.

  The traffic simulation apparatus includes a moving body model unit 10 that models a moving body, a traffic state management unit 30 that manages the traffic state of the moving body model unit 10, and other needs such as road network data, traveling vehicle data, and field-of-view images. Installed on the road, a traffic database 40 storing various data, a space placement unit 50 for spatially placing the mobile body model unit 10, a collision determination unit 60 for judging whether or not each mobile body model unit 10 collides. A roadside communication unit 70 that communicates with the mobile body model unit 10 and a road-vehicle communication system unit 80 that controls the roadside communication unit 70 are provided.

  The mobile body model units 10A, 10B, and 10C each include a recognition unit 11 that recognizes an object on the road, a determination unit 12 that determines an action to be moved next, and an operation unit 13 that determines an operation amount of the mobile body. And a behavior calculation unit 14 that calculates the behavior of the moving body based on the determined operation amount.

  The recognizing unit 11 simulates the way the driver recognizes the road situation. Based on the positional relationship between a moving object (for example, a vehicle) or a road object arranged on the road space by the space arranging unit 50, the recognizing unit 11 The vehicle and the road object are recognized at the viewpoint position.

  The determination unit 12 imitates a driver's determination method, and determines an action to be moved next based on the distance to the intersection, the presence / absence / position of a surrounding vehicle, and the speed. The determination unit 12 may, for example, “stop”, “maintain speed”, “slow down” as the action in the traveling direction, “change lane”, “stop to the right”, “turn right” as the action in the lateral direction. Etc.

  The operation unit 13 imitates how the driver operates, and determines the operation amount of the moving body based on the determined action. For example, for the operation amount in the traveling direction, the operation unit 13 determines the acceleration / deceleration for shifting to the target state based on the current vehicle speed, and determines the accelerator and brake operation amounts according to the characteristics of each moving body. . Further, the operation unit 13 determines the next steering amount for the lateral operation amount based on the current vehicle speed and the current steering amount.

  The behavior calculation unit 14 calculates the behavior of the moving body according to the operation amount determined by the operation unit 13. Since each moving body is different in weight, displacement, etc., the behavior differs even with the same operation amount.

  Here, the mobile body model units 10A and 10B further include a safety system that provides traffic information and / or controls the mobile body to assist the driver, but the mobile body model unit 10C does not include a safety system. .

  For example, the moving body model unit 10A further includes an in-vehicle device unit 20A that is the safety system. The in-vehicle device unit 20A includes a sensing unit 21 that measures and recognizes an object with a camera, a radar, and the like, a communication unit 22 that wirelessly communicates with the roadside communication unit 70, and a processing unit that measures, for example, image processing and radio wave intensity. 23, an information providing unit 24 that provides peripheral usage to the driver, and a moving body control unit 25 that controls the operation of the moving body so as to avoid collision with other moving bodies.

  The sensing unit 21 is composed of, for example, a camera or a radar device. For example, in the case of a camera, the sensing unit 21 recognizes a distance to an object by a stereo method or an optical flow method, or recognizes a vehicle, a pedestrian, a bicycle, a white line, a peripheral object, or the like by image processing. In the case of a radar device, the sensing unit 21 recognizes a distance to an object by a laser or a millimeter wave.

  The communication unit 22 transmits and receives information such as the position and speed of surrounding vehicles by road-to-vehicle communication with the roadside communication unit 70 and vehicle-to-vehicle communication between moving bodies equipped with similar devices.

  Based on information from the sensing unit 21 and the communication unit 22, the processing unit 23 determines whether or not to notify the driver of, for example, information on an approaching vehicle, and transmits the information to the information providing unit 24 when the information is transmitted. .

  The information providing unit 24 provides the transmitted information to the driver by a display device or the like. When the information is transmitted from the processing unit 23, the information providing unit 24 inputs the information provided to the driver to the determination unit 12. Thereby, the determination part 12 determines the action which should move to the next with respect to the information provided with a certain probability.

  The moving body control unit 25 performs moving body control, for example, brake control according to the situation processed by the processing unit 23. At this time, the behavior calculation unit 14 calculates the behavior of the moving body according to the operation amount (for example, the brake operation amount) determined by the moving body control unit 25 in addition to the operation amount determined by the operation unit 13.

  On the other hand, the in-vehicle device unit 20B of the moving body model unit 10B is obtained by removing the sensing unit 21 from the configuration of the in-vehicle device unit 20A. Further, the moving body model unit 10C does not include a safety system, that is, an in-vehicle device unit.

  The traffic situation management unit 30 manages the position, speed, and acceleration of each vehicle for each calculation step. It also manages the generation and disappearance of vehicles traveling on the road. The occurrence / disappearance of the vehicle may be given in a scheduled manner, or may be given as a percentage of occurrence in a specific section.

  The space arrangement unit 50 arranges each vehicle on the road space according to the position information and speed information of each vehicle managed by the traffic condition management unit. The collision determination unit 60 determines the collision between the vehicle and the road object and the vehicle and the vehicle by comparing the positional relationship between the vehicle and the road object placed on the road space by the space placement unit. The vehicle and passenger damage due to the collision is calculated from the speed, weight, and collision safety of the collision vehicle.

  The traffic simulation apparatus configured as described above operates as follows. Below, it demonstrates using a vehicle as a moving body.

  FIG. 2 is a diagram illustrating a situation where the vehicle C turns right at the intersection at time T. Each vehicle is assumed to move in the direction of the broken line, and the own vehicle C makes a right turn at this intersection. In the following, the procedure for the driver to determine the amount of operation in the traveling direction based on the recognition of the oncoming vehicle and the determination at that time will be described. In addition, although this embodiment mainly describes recognition of an oncoming vehicle, the recognition, judgment, and operation amount determination of a signal, a pedestrian, a surrounding building, etc. are also performed besides that. Similarly, horizontal determination and operation amount determination are also performed.

  FIG. 3 is a flowchart showing a right turn processing routine of the host vehicle C. The right turn processing routine is executed at every predetermined calculation step. The moving body model unit 10C corresponding to the host vehicle C executes processing as follows.

  First, the recognizing unit 11 calculates the presence recognition degree, the recognition distance, and the recognition speed used for the determination for the host vehicle C to turn right. Here, a case where the driver A is driving the vehicle C will be described as an example.

  Specifically, in step S1, the recognition unit 11 generates a visual field image at the viewpoint of the driver of the host vehicle C based on the information of the host vehicle C placed by the space placement unit 50, and proceeds to step S2.

  FIG. 4 is a diagram illustrating an example of a visual field image at time T. The visual field image is an image that is expressed by projection conversion or the like and corresponds to the visual field of the driver. Only the oncoming vehicle A appears in the field-of-view image of FIG. When each vehicle including the host vehicle C moves with time, the recognition unit 11 generates a visual field image corresponding to a change in the visual field of the driver.

  FIG. 5 is a diagram illustrating an example of a visual field image at time (T + t). In the view image of FIG. 5, not only the oncoming vehicle A but also the oncoming vehicle B appears.

  In step S2, the recognition unit 11 performs contour line extraction processing on the visual field image. In step S3, the recognition unit 11 calculates the area of the recognition target surrounded by the contour line as the area in the visual field, and the process proceeds to step S4.

  6 and 7 are diagrams illustrating the results of performing the contour extraction process on the visual field images of FIGS. 4 and 5. FIG. 8 is a diagram illustrating the in-view area, which is the area of the recognition target. The area in the visual field indicates the area of the recognition target object in the visual field of the driver.

  The ease of recognizing the recognition object also varies depending on the brightness and contrast with the background. In a situation where the contrast is low and difficult to recognize, the contour line is difficult to be extracted, and the visual field area is reduced as well as the recognizability. For this reason, the visual field area can express the difficulty of recognition according to the environment. In the present embodiment, the area to be recognized is calculated by contour extraction. However, the area may be directly obtained by an image processing method such as binarization.

  In step S4, the recognition unit 11 calculates the presence recognition degree of the recognition object. As shown in FIGS. 5 and 7, when there are a plurality of recognition objects, the recognition unit 11 calculates the presence recognition degree of each recognition object. The presence recognition level indicates the recognition level of an object in the driver's field of view, and the smaller the value, the more the presence is not recognized. Note that the presence recognition level is converted into a percentage as necessary. Specifically, the recognition unit 11 calculates the following expression (1) and proceeds to step S5.

Presence recognition level = [field area] ÷ [total field area] × [field factor] × [wakefulness]
... (1)

  Here, the total area of the visual field indicates the area of the entire visual field. The visual field coefficient indicates the degree of recognition according to the position in the visual field, the central point is high, and becomes lower as it goes to the periphery. This expresses that the vicinity of the gazing point is easy to recognize, and the other peripheral visual fields are difficult to recognize.

  FIG. 9 is a diagram illustrating an example of the visual field coefficient. As shown in the figure, the visual field coefficient is 1.0 at the center point of the image, 0.8 in the surrounding area, and 0.4 at the end. Note that the visual field coefficient may be changed according to the visual acuity of the driver.

  In addition, the awakening level indicates the level of awakening of the driver, and expresses that the recognition level decreases in a blurred state.

  FIG. 10 shows an example of the frequency distribution of the driver's arousal level. In this embodiment, the arousal level is given in a probabilistic distribution for all drivers.

  In step S5, the recognition unit 11 calculates a distance recognition degree and a speed recognition degree by using the presence recognition degree of the recognition target object, the actual distance, and the actual speed. Specifically, the recognition unit 11 calculates the following expressions (2) and (3), and proceeds to step S6.

Recognition distance = [distance] × (1 + k1 × 1 / [existence recognition degree]) (2)
Recognition speed = [speed] × (1-k2 × 1 / [existence recognition degree]) (3)

  k1 and k2 are parameters. The recognition distance and the recognition speed indicate the speed and distance that can be recognized from the visual field with respect to the actual distance and speed. For scenes with a low degree of presence recognition, that is, when it is difficult to recognize the presence, the driver perceives the distance further and perceives the speed slower.

  In step S6, the recognition unit 11 temporarily stores the presence recognition degree, the recognition distance, and the recognition speed in the storage device in order to make the next determination, and proceeds to step S7. Further, the recognition unit 11 may call the presence recognition degree, the recognition distance, and the recognition speed calculated in the previous calculation step depending on the characteristics of the driver, and hold them for the next determination. Using the presence recognition degree, the recognition distance, and the recognition speed calculated in the previous calculation step expresses a recognition delay caused by an elderly person or the like.

  Next, when the vehicle C turns right, the determination unit 12 determines the next action by determining the surrounding situation based on the presence recognition degree, the recognition distance, and the recognition speed calculated by the recognition unit 11. . The recognition unit 11 determines the following four items when turning right.

1. 1. Oncoming car 2. Whether it is a path to collide with your vehicle. 3. Whether to collide when progressing Possibility that oncoming car will come

  And the judgment part 12 uses the following data in order to determine action.

  FIG. 11 is a diagram illustrating the characteristics of the driver. For each driver, “carefulness”, “skilledness”, and “physical ability” indicating the characteristics of the driver are set between 0.0 and 1.0. “Prudentity” indicates driving caution, and the value of drivers who drive violently is low. “Skill level” indicates the level of skill that comes from driving experience, and the value of a beginner driver is low. “Physical ability” indicates physical ability such as visual acuity and reaction time, and the value of an elderly driver is low.

  For example, in the case of driver A who is careful driving with a driving calendar of 20 years in the 40s, the “carefulness”, “skill level”, and “physical ability level” are 0.9, 1.0, 0.7. And In the case of driver B who is driving somewhat violently with a driving calendar of 3 years at the age of 20 years old, the “carefulness”, “skill level”, and “physical ability level” are 0.3, 0.2, and 1.0, respectively. To do. These values are stochastically set based on the results of a survey of the distribution by a questionnaire to the driver.

  Moreover, the recognition part 11 uses a recognition presence degree evaluation value, an arrival time evaluation value, and an oncoming vehicle prediction evaluation value in the step mentioned later. These evaluation values are obtained by the following equation (4).

[Evaluation value] = (α × [Prudentity level] + β × [Proficiency level] + γ × [Physical ability level])
/ (Α + β + γ) (4)

  α, β, and γ are a cautiousness coefficient, a skill level coefficient, and a physical ability level coefficient, respectively. Here, each coefficient has a different value depending on which of the recognition presence degree evaluation value, the arrival time evaluation value, and the oncoming vehicle prediction evaluation value is obtained.

  FIG. 12 is a diagram illustrating a cautiousness coefficient, a skill level coefficient, and a physical ability level coefficient for each evaluation value. For example, when obtaining the presence recognition degree evaluation value, α, β, and γ are 3, 0, and 2, respectively. When obtaining the arrival time evaluation value, α, β, and γ are 2, 2, and 1, respectively.

  In addition, the determination unit 12 sets the presence recognition threshold value, the arrival time threshold value, and the possibility of an oncoming vehicle using an evaluation value in steps to be described later.

  FIG. 13 is a diagram illustrating the presence recognition threshold value, the arrival time threshold value, and the possibility of an oncoming vehicle corresponding to the evaluation value. For example, when the presence recognition level evaluation value is 0.9, the presence recognition level threshold is 3 (%). When the arrival time evaluation value is 0.3, the arrival time threshold is 2.3 (seconds). When the oncoming vehicle prediction evaluation value is 0.5, the possibility of an oncoming vehicle is 50 (%).

  The determination unit 12 executes the processes of the following steps S7 to S14 using the data as described above.

  In step S <b> 7, the determination unit 12 determines whether or not the signal condition, that is, the progress due to the signal lamp is possible. Specifically, the determination unit 12 determines whether or not the signal is blue. When the determination is affirmative, the determination unit 12 proceeds to step S8. When the determination is negative, the determination unit 12 proceeds to step S11. Transition.

  In step S8, the determination unit 12 determines whether there is an oncoming vehicle. When a plurality of oncoming vehicles are recognized, the oncoming vehicle that is closest to the own vehicle C is determined. Specifically, the determination unit 12 determines whether or not the presence recognition level calculated in step S4 exceeds a presence recognition level threshold value. The presence recognition threshold value is obtained as follows.

  The determination unit 12 first calculates the presence recognition level evaluation value according to the equation (4). When the characteristics of the driver A shown in FIG. 11 and the coefficients α, β, and γ for obtaining the presence recognition degree evaluation value shown in FIG. 12 are used, the presence recognition degree evaluation value of the driver A is expressed by Expression (5).

(Evaluation value of presence recognition of driver A) = (3 × 0.9 + 0 × 1.0 + 2 × 0.7) / 5
= 0.8 (5)

  According to FIG. 13, the presence recognition threshold corresponding to the evaluation value 0.8 is 4 (%). Therefore, the determination unit 12 determines whether or not the presence recognition level (%) calculated in step S4 exceeds the presence recognition level threshold 4 (%). If the determination is affirmative, the driver A assumes that the oncoming vehicle has been recognized and proceeds to step S9. If a negative determination is made, it is assumed that the driver A has not recognized the presence of the oncoming vehicle, and the process proceeds to step S12.

  In step S <b> 9, the determination unit 12 determines whether or not the oncoming vehicle is on a path that collides with the host vehicle C. Specifically, the determination unit 12 predicts the course of the oncoming vehicle based on the state of the direction indicator of the oncoming vehicle, the recognition distance and the recognition speed calculated in step S5. When the course of the oncoming vehicle and the course of the own vehicle C intersect (or overlap), it is determined that the oncoming vehicle is on a course that collides with the own vehicle C. For example, when the host vehicle is turning right and the oncoming vehicle is going straight or turning left, it is determined that there is an oncoming vehicle on the course of the host vehicle C. If the determination is affirmative, the process proceeds to step S10. If the determination is negative, the process proceeds to step S14.

  In step S10, the determination part 12 determines whether the own vehicle C and an oncoming vehicle collide. Specifically, the determination unit 12 calculates the arrival time based on the recognition distance and the recognition speed calculated in step S5, and the speed of the host vehicle, and determines whether or not this arrival time exceeds the arrival time threshold value. To do. The arrival time threshold value is obtained as follows.

  The determination unit 12 first calculates the arrival time evaluation value according to the equation (4). When the data of FIGS. 11 and 12 is used, the arrival time evaluation value of the driver A is expressed by Equation (6).

(Driver A arrival time evaluation value) = (2 × 0.9 + 2 × 1.0 + 1 × 0.7) / 5
= 0.9 (6)

  According to FIG. 13, the arrival time threshold value corresponding to the evaluation value 0.9 is 3.8 (seconds). In the case of driver B, the arrival time evaluation value is 0.4, and the arrival time threshold is 2.6. Therefore, the determination unit 12 determines whether or not the arrival time (seconds) already calculated for the driver A exceeds the arrival time threshold value 3.8 (seconds). If the determination is affirmative, it is considered that the vehicle C does not collide with the oncoming vehicle, and the process proceeds to step S14. If a negative determination is made, the driver A is assumed to collide with the oncoming vehicle, and the process proceeds to step S11.

  In step S11, the determination part 12 determines making the speed of the own vehicle C zero, and transfers to step S16.

  On the other hand, in step S12, the determination unit 12 determines to increase the speed of the host vehicle C to the speed v, and proceeds to step S13.

  In step S <b> 13, the determination unit 12 determines whether there is a possibility that an oncoming vehicle exists outside the field of view of the driver A. Here, even if the driver cannot recognize the oncoming vehicle, there is a possibility that the oncoming vehicle will come from a part that is not seen empirically. Therefore, the determination unit 12 determines the presence / absence of an oncoming vehicle stochastically.

  First, the determination unit 12 calculates an oncoming vehicle predicted evaluation value according to Equation (4). When the characteristics of the driver A shown in FIG. 11 and the coefficients α, β, and γ for obtaining the oncoming vehicle prediction evaluation value shown in FIG. 12 are used, the oncoming vehicle prediction evaluation value of the driver A is expressed by Equation (7).

(Driver A's oncoming vehicle prediction evaluation value) = (2 × 0.9 + 3 × 1.0 + 0 × 0.7) / 5
= 1.0 (7)

  According to FIG. 13, the possibility of an oncoming vehicle corresponding to the evaluation value 1.0 is 100 (%). Therefore, the determination unit 12 determines whether or not the possibility of an oncoming vehicle exceeds a predetermined threshold (for example, 30%). If an affirmative determination is made, it is considered that an oncoming vehicle is coming, and the process proceeds to step S11.

  As described above, the determination unit 12 performs traffic determination such as the presence / absence of an oncoming vehicle and whether or not there is a collision based on the presence recognition degree and the arrival time. However, the degree of presence recognition and the arrival time vary greatly depending on the degree of driver recognition. Therefore, the determination unit 12 can simulate traffic determination according to the characteristics of the driver by setting an optimal threshold value for each driver and comparing the threshold value with the presence recognition level and arrival time.

  In step S16, the operation unit 23 calculates the brake operation amount or the accelerator operation amount using the driver data and the vehicle data so as to achieve the target speed determined in step S11 or step S12, and proceeds to step S17.

  For example, when the speed determined by the determination unit 12 is zero, the operation unit 13 calculates the brake operation amount so that the stop state of the host vehicle C can be maintained or the vehicle can be decelerated to the stop state. When the speed determined by the determination unit 12 is v, the operation unit 13 calculates the accelerator opening so that the host vehicle C has the speed v. The target speed v may be set to a different value for each driver within an appropriate range.

  In step S <b> 17, the behavior calculation unit 14 calculates the speed and acceleration of the host vehicle C based on the operation amount obtained by the operation unit 13. The speed and acceleration of the host vehicle C calculated in this way are supplied to the traffic condition management unit 30.

  As a result, the space placement unit 50 rearranges the own vehicle C on the road space according to information such as position information, speed, and acceleration managed by the traffic condition management unit 30 and simulates the own vehicle C. be able to.

  Here, before one calculation step, the determination unit 12 determines that “no oncoming vehicle” (negative determination in step S8), sets a target speed (step S12), and as a result, the host vehicle C accelerates. Even in the state, in the next current calculation step, the determination unit 12 may recognize the oncoming vehicle (affirmative determination in step S8) and determine to stop (step S11). In this case, the host vehicle C may suddenly brake from a certain speed to a stopped state. At this time, whether or not the own vehicle C collides with the oncoming vehicle varies depending on the speed and the determination delay time in the previous calculation step. In other words, the host vehicle C and the oncoming vehicle may collide, and may not collide due to braking.

  Here, the right turn processing routine of the moving body model unit 10C of the vehicle C has been described, but the moving body model unit 10B of the opposite vehicle B can similarly determine the speed and acceleration of the own vehicle. For example, even if the vehicle is traveling straight, if it is determined that the vehicle is traveling on the course of the vehicle within the field of view, it can be determined that the vehicle is decelerating and the speed can be reduced so as not to collide.

  The space arrangement unit 50 can re-arrange each vehicle on the road space according to information such as position information, speed, acceleration, and the like managed by the traffic condition management unit 30, and simulate the movement of each vehicle. it can.

  The collision determination unit 60 determines the collision between the vehicle and the road object and between the vehicle and the vehicle by comparing the positional relationship between the vehicle and the road object placed on the road space by the space placement unit 50. When the collision determination unit 60 determines that the car collides, the collision determination unit 60 calculates the magnitude of the collision.

For example, when the speed V _C and the weight M _C just before the collision of the host vehicle _C , the speed V _B and the weight M _B just before the collision of the oncoming vehicle B, and the collision safety coefficient K _C of the host vehicle C, The energy E received by the collision is obtained by the following equation (8), for example.

E = K _C [(1/2) M _C V _C ² + (1/2) M _B V _B ² ] (8)

  The collision determination unit 60 classifies the magnitude of the collision due to the accident based on the energy E of Expression (8).

  As described above, the traffic simulation according to the embodiment of the present invention calculates the behavior amount of the vehicle based on the driver's recognition, determination, and operation, and simulates the movement of each moving body based on the behavior amount. Thereby, the said traffic simulation apparatus can simulate the motion of a moving body in consideration of a driver's recognition mistake and judgment error.

  The traffic simulation device calculates the presence recognition level, the recognition speed, and the recognition distance of the recognition target object that reflects the degree of recognition of the driver based on the driver's visual field image, and uses these calculation results to determine the next of the driver. Since the action is determined, it is possible to reproduce a driver's recognition error and a determination error caused thereby.

  Furthermore, since the traffic simulation apparatus sets a threshold for determining the next action according to each characteristic of the driver, a more realistic traffic simulation is performed by using the statistical characteristics of the driver. be able to.

[Evaluation of safety devices]
A vehicle accident occurs due to a driver's judgment, for example, a recognition error of an oncoming vehicle. In order to prevent such a vehicle accident, there is also a vehicle equipped with a safety system that assists the driver's safe driving. Below, the simulation of the vehicle carrying various safety systems is explained.

(For communication systems)
In the inter-vehicle communication system, even if there is a shield between the transmission point and the reception point, radio waves may be diffracted and reflected by the building, so that wireless communication may be possible. Here, it is assumed that the mobile body model units 10A and 10B perform wireless communication.

  FIG. 14 is a diagram showing diffraction and reflection of radio waves. In the case of diffraction, the radio wave intensity is obtained by the knife-edge method and in the case of reflection by the ray tracing method. Therefore, for example, the in-vehicle device unit 20A of the moving body model unit 10A receives the radio wave transmitted from the in-vehicle device unit 20B of the moving body model unit 10B, which is an oncoming vehicle, by the own vehicle through diffraction and reflection. If the radio wave intensity exceeds the threshold value, communication is assumed to be successful, and information such as the position and speed of the oncoming vehicle can be recognized by communication.

  In such a vehicle-to-vehicle communication system, the mobile body model unit 10A can predict whether a collision will occur based on the position and speed of the other vehicle B transmitted from the mobile body model unit 10B and the position and speed of the own vehicle. . For this reason, even if the moving body model unit 10 cannot recognize the oncoming vehicle B in the field of view, it can recognize the oncoming vehicle B through wireless communication, and therefore can change the judgment.

  For example, when the information providing unit 24 issues a driver alarm and provides information, the recognition unit 11 changes the arousal level of the expression (1) to a large value (for example, by multiplying by a constant of 1 or more) and recognizes the presence. What is necessary is just to calculate a degree.

  When the information providing unit 24 provides information regarding the presence or absence of a surrounding vehicle by voice, characters, images, or the like, the recognizing unit 11 may change the presence recognition degree obtained by Expression (1) to a large value. When the information providing unit 24 provides the position and speed information, the recognition unit 11 may change the recognition distance and the recognition speed to more accurate values.

  When evaluating the reliability of information provided by the driver, drivers who believe in the information may be probabilistically distributed. For drivers who do not trust the information, the presence recognition level does not change. Also good.

(Vehicle control system)
In addition, when the vehicle control system, for example, the moving body control unit 25 predicts that the target vehicle travels on the course of the host vehicle, the host vehicle is stopped before the collision by determining and operating the braking operation amount. . In this case, the operation unit 13 gives priority to the operation amount determined by the moving body control unit 25 over the operation amount determined by the driver. Whether or not wireless communication such as radio wave intensity is established may be calculated in the present apparatus, or a result calculated by an external apparatus may be used.

(At night)
In the case of nighttime, the space arrangement unit 50 sets the road space to brightness corresponding to nighttime, and sets the state in which the vehicle emits a light source corresponding to headlights.

  FIG. 15 is a diagram illustrating an example of a night vision image. When the space arrangement unit 50 sets the brightness of the road space as described above, the recognition unit 11 obtains a driver's visual field image by processing such as a path tracing method and a ray tracing method. As a result, the determination unit 12 can simulate determination of a driver that can be placed at night.

  On the other hand, in the case of a night shooting system (night vision) in which the vehicle projects infrared light and shoots with a camera, the space placement unit 50 can adjust the road space according to conditions such as brightness of infrared light and projection. Set the amount of light.

  FIG. 16 is a diagram illustrating an example of a visual field image captured by a camera by projecting infrared light. As described above, in the night photographing system, the recognition unit 11 processes the visual field image as an output from the camera, and the determination unit 12 evaluates the effect of the night photographing system by making a driver's determination using the processing result. be able to. Similarly, in a stereo safety system using two cameras, field-of-view images with respective camera positions as viewpoints are obtained and evaluated using these field-of-view images.

  Also in a safety system using a camera, the recognition unit 11 may change the value of the presence recognition level by providing information and perform vehicle control, as in the safety system using wireless communication.

In a system that assists night vision with infrared light or the like, the visual field coefficient may be increased.

(Radar system)
The radar system can determine the distance to the object with respect to the bearing.

  FIG. 17 is a diagram illustrating the relationship between the distance to the target and the direction calculated by the radar system. Specifically, the space placement unit 50 calculates the azimuth at the vehicle position and the distance to the object in that direction. In the radar system, the radar system can be evaluated by processing the relationship between the azimuth and the distance as an output from the radar.

  The traffic simulation apparatus simulates the movement of a moving body not equipped with a safety system, calculates the behavior amount of the moving body equipped with the safety system as described above, and moves each movement based on the behavior amount. Simulate body movements. Thereby, the said traffic simulation apparatus can evaluate a safety system by comparing the accident occurrence number of the vehicle carrying the safety system with the vehicle which is not carried.

  It should be noted that the present invention is not limited to the above-described embodiment, and it is needless to say that the present invention can also be applied to a design modified within the scope described in the claims.

  For example, the data used in this embodiment may be stored in advance in a database or may be transmitted via a network. Moreover, what was shown in FIG. 10 thru | or FIG. 13 is only an example, and does not limit this invention.

It is a block diagram which shows the structure of the traffic simulation apparatus which concerns on embodiment of this invention. It is a figure which shows the condition where the own vehicle C makes a right turn at the intersection at time T. It is a flowchart which shows the right turn process routine of the own vehicle C. It is a figure which shows an example of the visual field image in the time T. FIG. It is a figure which shows an example of the visual field image in time (T + t). It is a figure which shows the result of having performed the outline extraction process with respect to the visual field image of FIG. It is a figure which shows the result of having performed the outline extraction process with respect to the visual field image of FIG. It is a figure which shows the area in a visual field which is an area of recognition object. It is a figure which shows an example of a visual field coefficient. An example of the frequency distribution of the driver's arousal level is shown. It is a figure which shows the characteristic of a driver. It is a figure which shows the cautiousness coefficient for every evaluation value, a skill level coefficient, and a physical ability level coefficient. It is a figure which shows the possibility that the presence recognition threshold value corresponding to an evaluation value, an arrival time threshold value, and an oncoming vehicle will come. It is a figure which shows the diffraction and reflection of an electromagnetic wave. It is a figure which shows an example of the night vision image. It is a figure which shows an example of the visual field image image | photographed with the camera by projecting infrared light. It is a figure which shows the distance to the target object and the relationship of an azimuth | direction calculated by the radar system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10, 10A, 10B, 10C Mobile body model part 11 Recognition part 12 Judgment part 13 Operation part 14 Behavior calculation part 20A, 20B Car-mounted apparatus part 21 Sensing part 22 Communication part 23 Processing part 24 Information provision part 25 Mobile body control part

Claims (10)

A traffic simulation device that simulates the movement of a moving object on a road based on the behavior amount of the moving object,
Storage means for storing a field-of-view image;
Generating means for generating a driver's visual field image corresponding to the positional information of the moving body based on the visual field image stored in the storage means ;
A recognition unit that calculates at least one of the presence recognition level, the recognition speed, and the recognition distance of the recognition target object based on the visual field image of the driver generated by the generation unit ;
A determination unit that determines a driver's next action based on a presence recognition degree, a recognition speed, and a recognition distance of the at least one recognition object calculated by the recognition unit ;
Based on the driver's next action determined by the determining means, an operating means for operating the moving body;
Vehicle behavior amount calculating means for calculating the behavior amount of the moving body according to the operation of the operating means;
Traffic simulation device with The recognition means calculates the presence recognition degree of the recognition object,
The determination means determines whether or not the recognition object exists in the vicinity of the moving body based on whether or not the presence recognition degree calculated by the recognition means exceeds a threshold set according to the characteristics of the driver. The traffic simulation apparatus according to claim 1. The recognition means calculates a recognition speed and a recognition distance of the recognition object;
The determination means calculates an arrival time based on the speed of the moving body, the recognition speed and the recognition distance calculated by the recognition means, and sets a threshold set according to the characteristics of the driver. The traffic simulation device according to claim 1, wherein it is determined whether or not the object to be recognized collides, based on whether or not the number is exceeded. The recognition means calculates a recognition speed and a recognition distance of the recognition object;
The determination means predicts a course of the recognition object based on a recognition speed and a recognition distance calculated by the recognition means and a movement direction of the recognition object, and the course of the moving object is the recognition object. The traffic simulation device according to any one of claims 1 to 3, wherein it is determined whether or not the vehicle crosses the path. The said recognition means calculates the presence recognition degree of the said recognition target object based on the area of the said visual field image, and the area of the recognition target object in the said visual field image at least. The traffic simulation device described. The recognition means further calculates the presence recognition degree of the recognition object using at least one of a visual field coefficient indicating a recognition degree according to a position in the visual field image and a wakefulness degree indicating a driver's wakefulness degree. The traffic simulation device according to claim 5. It further comprises information providing means for providing the driver with peripheral information of the mobile object,
The said recognition means changes at least one value of presence recognition degree, recognition speed, and recognition distance of a recognition target object using the surrounding information provided by the said information provision means. The traffic simulation apparatus according to item 1. It further comprises information providing means for providing the driver with peripheral information of the mobile object,
The traffic simulation apparatus according to claim 6, wherein the recognizing unit changes at least one value of a visual field coefficient and an arousal level using the peripheral information provided by the information providing unit. A traffic simulation method for simulating the movement of a moving object on a road based on the amount of behavior of the moving object,
Based on the visual field image stored in the storage means storing the visual field image, generate a visual field image of the driver corresponding to the position information of the moving body,
Based on the generated visual field image of the driver , calculate at least one of the presence recognition degree, the recognition speed, and the recognition distance of the recognition target object ,
Based on the calculated presence recognition level, recognition speed, and recognition distance of the at least one recognition object , a next action of the driver is determined,
Based on the determined next action of the driver, the mobile body is operated,
A traffic simulation method for calculating a behavior amount of a moving object according to the operation. A traffic simulation program for causing a computer to simulate the movement of a moving object on a road based on the amount of movement of the moving object,
In the computer,
Based on the visual field image stored in the storage means storing the visual field image, the driver generates a visual field image of the driver corresponding to the position information of the moving body,
Based on the generated visual field image of the driver , at least one of the presence recognition degree, the recognition speed, and the recognition distance of the recognition object is calculated ,
Based on the calculated presence recognition degree, recognition speed, and recognition distance of the at least one recognition object , the driver's next action is determined,
Based on the next action of the determined driver, the mobile body is operated,
A traffic simulation program for calculating the amount of behavior of a moving object according to the operation.