JP2005285108A - Unexpected event detection method and unexpected event detection apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an unexpected event detection method with which an unexpected event is detected even without a vehicle detection means on the road and a position where the unexpected event occurs, can be specified. <P>SOLUTION: An unexpected event detection method includes a moving object information collection step for collecting position information representing a position where a moving object is present, and moving object information for finding a speed of the moving object at this position; a traffic information generation step for generating traffic information containing a moving time of the moving object at a predetermined position and the speed of the moving object from the moving object information collected in the moving object information collection step and for storing the traffic information in a traffic information storage means; and an unexpected event determination procedure for determining an unexpected event at a predetermined position from the traffic information stored in the traffic information storage means using a preset threshold. Even if a vehicle detector is not installed on the road, an unexpected event can be detected and a position where the unexpected event occurs, can be specified. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a detection method and a detection system for accidents and regulations that occur suddenly.

As a conventional method for detecting an unexpected event, a method described in Patent Document 1 is known. The conventional method for detecting an unexpected event described in Patent Document 1 is based on traffic state data such as a current speed detected by a vehicle detector installed on a road, and a traffic state after a predetermined time from the present time. Predict the amount, set an upper and lower threshold for abnormal traffic flow determination after a predetermined time based on this predicted value, and determine whether the actual traffic state quantity is outside the range between the set upper and lower threshold However, when it is determined that the traffic flow is out of the range, it is determined that the traffic flow is abnormal.
Japanese Patent No. 2893544

  However, in the conventional method for detecting a sudden event, vehicle detection means such as a vehicle detector needs to be installed on the road, and a sudden event cannot be detected in a section where the sensor is not installed.

  In addition, in the conventional method for detecting a sudden event, in a section where a vehicle sensor is installed, a past statistical value is used as a criterion, so that a traffic jam that occurs regularly and a traffic jam that occurs suddenly are detected. In the section where traffic flow is not regular because the threshold range is constant, that is, in the section where the correlation between the past traffic flow and the current traffic flow is low, The variation in daily traffic flow may be judged as a sudden event.

  In addition, in the conventional method for detecting a sudden event, it is possible to specify between which detectors the sudden event has occurred, but it is not possible to specify at which location between the sensors the sudden event has occurred.

  The present invention has been made to solve these conventional problems, and can detect sudden events even in a section where no sensor is installed, and the variation in daily traffic flow that changes is not determined as a sudden event. It is an object of the present invention to provide a sudden event detection method and a sudden event detection system capable of specifying a place where a sudden event has occurred.

  The sudden event detection method of the present invention includes a mobile object information collection procedure for collecting mobile object information related to position information and time information at a position where a mobile object exists, and a mobile object collected by the mobile object information collection procedure. Using the traffic information generation procedure for generating traffic information including the travel time or speed of the mobile body at a predetermined position from the information and storing the traffic information in the traffic information storage means, and using a predetermined threshold, the traffic information storage means A sudden event determination procedure for determining an abnormal traffic flow suddenly occurring at a predetermined position from the stored traffic information.

  With this configuration, it is possible to determine the sudden event by collecting the moving body information related to the position information and the time information at the position where the moving body exists from the moving body.

  The sudden event detection method of the present invention includes a configuration in which the sudden event determination procedure includes a threshold setting procedure for setting a certain range as a threshold from a statistical value of past traffic information stored in the traffic information storage unit. Yes.

  With this configuration, the threshold value for determining the sudden event can be set within a certain range based on the past statistical value of the moving time or the moving speed of the moving object at a predetermined position.

  Furthermore, the sudden event detection method of the present invention has a configuration in which the threshold setting procedure sets a threshold based on a deviation of past traffic information stored in the traffic information storage means.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range can be determined according to the variation in the deviation of the statistical value of the moving time of the moving body or the speed of the moving body at a predetermined point. it can.

  Furthermore, the sudden event detection method of the present invention has a configuration in which the threshold setting procedure sets a threshold for each point in consideration of a deviation in traffic information depending on a place.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range is set for each point or road according to the variation in the past statistics of the moving time or the moving speed of the moving object at a predetermined point. It can be set optimally depending on the location.

  Furthermore, the sudden event detection method of the present invention has a configuration in which the threshold value setting procedure sets a threshold value for each time zone in consideration of deviation of traffic information for each time.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range is optimized according to the time zone according to the variation of the moving time of the moving body at a predetermined point or the past statistical value of the moving body's speed. Can be set to

  Furthermore, in the sudden event detection method of the present invention, the sudden event determination means has a fixed value based on statistical values of past traffic information for each similar pattern of traffic flow, such as weekdays, weekends, holidays, long holidays, and fifty days. A range is set as the threshold value.

  With this configuration, it is possible to optimally set the threshold range according to the day type in the determination of the sudden event.

  Furthermore, in the method for detecting a sudden event according to the present invention, the sudden event determination procedure includes a sudden event occurrence position identifying procedure for identifying a sudden event occurrence position using trajectory information and speed information obtained from vehicle information. .

  With this configuration, the position of the moving object whose vehicle speed information exceeds the threshold range can be specified using the trajectory information and the speed information.

  Furthermore, the sudden event detection method of the present invention has a configuration in which the predetermined position is a position between two predetermined points.

  With this configuration, it is possible to detect a sudden event based on the moving time information of the moving body or the speed information of the moving body between two predetermined points.

  The sudden event detection apparatus of the present invention collects mobile body information collecting means for collecting mobile body information related to position information and time information at a position where the mobile body exists from the mobile body, and the mobile body information collecting procedure. Traffic information generation means for generating traffic information including the travel time or speed of the mobile object at a predetermined position from the mobile object information and storing it in the traffic information storage means, and storing traffic information using a predetermined threshold A sudden event determination means for determining a sudden event at a predetermined position from the traffic information stored in the means.

  With this configuration, it is possible to determine the sudden event by collecting the moving body information related to the position information and the time information at the position where the moving body exists from the moving body.

  The sudden event detection device of the present invention further includes threshold setting means for the sudden event determination means to set a certain range as a threshold from the statistical values of past traffic information stored in the traffic information storage means.

  With this configuration, the threshold value for determining the sudden event can be set within a certain range based on the past statistical value of the moving time or the moving speed of the moving object at a predetermined position.

  Furthermore, the sudden event detection apparatus of the present invention has a configuration in which the threshold setting unit sets the threshold based on a deviation of past traffic information stored in the traffic information storage unit.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range can be determined according to the variation in the deviation of the statistical value of the moving time of the moving body or the speed of the moving body at a predetermined point. it can.

  Furthermore, the sudden event detection apparatus of the present invention has a configuration in which the threshold setting means sets a threshold for each point in consideration of a deviation in traffic information depending on a place.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range is set for each point or road according to the variation in the past statistics of the moving time or the moving speed of the moving object at a predetermined point. It can be set optimally depending on the location.

  Furthermore, the sudden event detection apparatus of the present invention has a configuration in which the threshold setting means sets the threshold for each time zone in consideration of the deviation of traffic information for each hour.

  With this configuration, in setting the threshold range for sudden event determination, the threshold range is optimized according to the time zone according to the variation of the moving time of the moving body at a predetermined point or the past statistical value of the moving body's speed. Can be set to

  Furthermore, in the sudden event detection device of the present invention, the sudden event determination means has a fixed value from statistical values of past traffic information for each similar pattern of traffic flow, such as weekdays, weekends, holidays, long vacations, and fifty days. A range is set as the threshold value.

  With this configuration, it is possible to optimally set the threshold range according to the day type in the determination of the sudden event.

  Furthermore, the sudden event detection device of the present invention has sudden event occurrence position specifying means for the sudden event determination means to specify the occurrence position of the sudden event using the trajectory information and speed information obtained from the vehicle information. .

  With this configuration, the position of the moving object whose vehicle speed information exceeds the threshold range can be specified using the trajectory information and the speed information.

  Furthermore, the sudden event detection means of the present invention has a configuration in which the predetermined position is a position between two predetermined points.

  With this configuration, it is possible to detect a sudden event based on the moving time information of the moving body or the speed information of the moving body between two predetermined points.

  As described above, the sudden event detection method of the present invention collects moving body information related to position information and time information at a position where the moving body exists from the moving body, and time information or speed information of the moving body information. Therefore, it is possible to detect a sudden event based on only so-called vehicle information, and it is possible to detect a sudden event from only vehicle information even in a section where no vehicle sensor is present.

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

(Embodiment 1)
FIG. 1 is a component diagram for explaining a sudden event detection method according to Embodiment 1 of the present invention.

  In the following embodiment, a vehicle equipped with an in-vehicle device that can transmit so-called vehicle information such as the vehicle position, vehicle speed, and time at a certain time collected by the vehicle is called a probe car. The information such as the vehicle position and the vehicle speed at a certain time is stored in the vehicle-mounted device for a certain time or a certain distance and transmitted to the center station is called probe information.

  In FIG. 1, an in-vehicle device 1 centers a GPS information receiving unit 11 that receives GPS information from a GPS satellite 3, a vehicle speed sensor 12 that detects a vehicle speed from a vehicle speed pulse, and vehicle information obtained from the GPS information and the vehicle speed sensor 12. It comprises a communication unit 13 for transmitting to the station 2, a map database 14 for displaying traffic jam information and the like, and a control unit 15 for performing signal processing of these units.

  The center station 2 includes an in-vehicle device 1, a vehicle information collecting unit 21 that transmits and receives data such as probe information (vehicle information transmitted from the in-vehicle device 1), and a travel time between two predetermined points based on the received probe information. (Hereinafter referred to as travel time) and a traffic information generation unit 22 that generates a degree of congestion, a traffic information database 25 that stores traffic information generated by the traffic information generation unit 22, and a traffic information generation unit 22 The system includes a sudden event determination unit 23 that determines the occurrence of sudden events based on traffic information, and a map database 24 for identifying the locations of these sudden events on a map.

  The operation of the sudden event determination method of the present invention configured as described above will be described.

  In the in-vehicle device 1, vehicle speed information calculated from vehicle speed pulse information detected by the vehicle speed sensor 12 and information on the own vehicle position acquired from the GPS information receiving unit 11 are accumulated for a certain period, and the communication unit 13 at an arbitrary timing. Is transmitted to the center station 2. For communication between the in-vehicle device 1 and the center station 2, wireless communication such as a mobile phone, an optical beacon, a digital MCA, and a DSRC can be used.

  Note that the vehicle speed information is not calculated from the vehicle speed pulse information from the vehicle speed sensor 12 but can be vehicle speed information calculated from the GPS information received by the GPS information receiving unit 11.

  On the center station 2 side, probe information such as vehicle position information and speed information sent from the vehicle-mounted device 1 is received by the vehicle information collection unit 21, and the traffic information generation unit 22 travels in a specific two-point section. Generate time information and traffic information. Here, the generated traffic information including travel time information and traffic jam information is subjected to statistical processing such as taking the average value of several pieces of probe information in order to reduce the variation in travel of individual vehicles themselves. Smoothed. The generated traffic information is associated with the map database 24 as traffic information of a specific two-point section, and stored in the traffic information database 25 as traffic information for distribution at a specific time. Based on this traffic information, as described below, the sudden event determination unit 23 determines whether or not a sudden event has occurred, and when it is determined that a sudden event has occurred, It is stored in the traffic information database 25.

  Thereafter, when there is a request for inquiries about traffic information from the in-vehicle device 1, necessary information is retrieved and acquired from the traffic information database 25, and the traffic information of the position where the in-vehicle device 1 is inquired through the vehicle information collecting unit 21. Send.

  The operation of the sudden event determination unit 23 of the center station 2 will be described in detail with reference to FIG. FIG. 2 is a configuration diagram showing details of the sudden event determination unit 23 of the center station 2 in the sudden event detection method according to the embodiment of the present invention.

  The sudden event determination unit 23 uses a predetermined threshold value to determine whether or not a sudden event has occurred. When it is determined that a sudden event has occurred, the sudden event determination unit 23 uses the probe information to detect the sudden event. A sudden event occurrence position identifying unit 33 that identifies the position where the accident occurred, and a sudden event information output unit 34 that outputs the sudden event occurrence position and outputs it to the traffic information database 25 as sudden event information.

  About the sudden event determination part 23 comprised as mentioned above, the operation | movement is demonstrated using the flowchart shown in FIG.

  In the sudden event determination unit, when the latest travel time between two points generated by the traffic information generation unit or the average vehicle speed (hereinafter referred to as traffic information T) is input (S10), a predetermined predetermined value is set. It is determined whether the generated traffic information is a normal traffic situation or an unexpected event has occurred (S11). The predetermined threshold value is a fixed threshold value regardless of the location and the time point, and it is possible to detect an unexpected event different from a normal one by an easy configuration. If it is determined as a sudden event, the location of the sudden event is specified (S12), and if not determined, the process is terminated (S13).

  As shown in the flowchart of the operation in FIG. 4, the sudden event occurrence position specifying unit 33 determines a certain threshold value Vth based on the vehicle speed information included in the probe information in the section where the sudden event determination unit 32 determines that the sudden event has occurred. The following sections are determined and extracted as traffic jam sections (S20). For example, 10 km, which is used as a general congestion index, can be used as the threshold Vth for determining the congestion. However, it can be changed for each point and every time according to the characteristics of the traffic flow.

  If the vehicle speed information included in the probe information is greater than or equal to a certain threshold value Vth, it is determined that there is no traffic jam in the section where the sudden event has been determined to occur, and the sudden event occurrence position cannot be specified (S21).

  Next, noise other than traffic jams such as waiting for traffic lights and short-term parking is eliminated by removing traffic jam judgment zones that are less than a certain distance from the traffic jam zones of the two points judged as traffic jams. Only the true traffic jam portion is extracted (S22). In the traffic jam section calculated in this way, the point downstream of the traffic flow is specified as the occurrence position of the sudden event (S23).

  Further, when determining a traffic jam, a statistical value obtained by smoothing the transition of past vehicle speed information may be used to determine a traffic jam due to a daily traffic jam and a sudden event and specify a sudden event occurrence position.

  A method for specifying the sudden event occurrence position as described above will be described with reference to FIG. In FIG. 5, the road section 50 between two specific points determined as a sudden event by the sudden event determination unit 32, the traffic light 51 on the road section 50, the sudden event occurrence point 52 on the road section 50, the vehicle speed of the probe information An information transition 53 and a statistical value 54 obtained by smoothing the transition of past vehicle speed information are shown.

  If a section below a certain threshold value Vth is determined as a traffic jam section from the vehicle speed information included in the probe information 53, the vehicle speed sections 55, 56, and 57 are determined as traffic jam sections. However, among these, like the vehicle speed sections 55 and 56, there are those that are stopped by waiting for traffic lights or stopping and stopping for a short time and are not congested due to sudden events. In order to remove these, by performing correction such as removing a traffic jam portion that is less than a certain distance, it is possible to remove a traffic jam portion that is not a traffic jam due to a sudden event and extract only the traffic jam portion 57 due to a sudden event. I can do it. And the downstream side of this traffic jam part is specified as the sudden event occurrence point 58.

  Further, in the transition of the vehicle speed information of the probe information 53, the sudden rise of the vehicle speed may be detected to determine the occurrence position of the sudden event. Furthermore, by comparing the vehicle speed transition of the probe information 53 with a statistical value 54 obtained by smoothing the past vehicle speed transition, a section in which the vehicle speed transition is remarkably different may be determined as a sudden event occurrence section.

  The sudden event information output unit 34 uses the section number or section name indicating the section calculated by the sudden event determination unit 32 and the distance from the end of the section calculated by the sudden event occurrence position specifying unit 33 as the sudden event information. The information is output to the information database 25 (S14).

  After that, when there is a request for inquiries about traffic information from the in-vehicle device 1, the traffic information such as a traffic jam section is searched and extracted from the traffic information database 25 according to the content of the inquiry and transmitted to the in-vehicle device 1.

  With the above configuration, abnormal traffic flow that is different from normal, that is, information on sudden events, is provided to the driver by determining sudden events using predetermined thresholds based on traffic information generated from probe information. Thus, the driver can avoid the point where the sudden event occurs. Moreover, it is possible to detect a sudden event from only probe information even in a section where no vehicle sensor is present, and it is possible to detect a sudden event in a wider range.

(Embodiment 2)
FIG. 6 is used as a component diagram for explaining a sudden event detection method using statistical values in Embodiment 2 of the present invention. In the present embodiment, a sudden event is detected using a statistical value of past traffic information and a fixed determination range. In FIG. 6, the same components as those in FIG. 31 is a threshold value setting unit that uses the past traffic information stored in the traffic information database 25 to generate statistical information obtained by smoothing the past traffic information for each time zone, and sets a threshold for sudden event determination. It is.

  About the sudden event determination part 23 comprised as mentioned above, the operation | movement is demonstrated using the flowchart shown in FIG.

  The threshold setting unit 31 uses the past traffic information stored in the traffic information database 25 to generate statistical information T_ave obtained by smoothing the past information for each time zone in each section or point. Further, in each section or point, a threshold value T_ave ± α is set from the statistical value T_ave generated for each time zone and a predetermined determination range α. The predetermined determination range α is a fixed value regardless of the location and time. The generated threshold value is stored in the traffic information database 25. The threshold setting process in the threshold setting unit 31 is performed independently of other processes and is performed at an arbitrary timing. It is desirable to do this regularly to update the statistics of traffic information.

  In the sudden event determination unit 32, when the latest traffic information T generated by the traffic information calculation unit 22 is input (S30), the corresponding section and the corresponding time zone threshold T_ave ± α are obtained from the traffic information database 25. Extract (31). Next, T is compared with T_ave ± α, and when T_ave + α <T or T <T_ave−α, the traffic flow is determined to be an unexpected sudden event. When T_ave−α ≦ T ≦ T_ave + α Then, it is determined that it is within the range of variations in traffic information that occurs daily instead of sudden events (S32). The predetermined determination range α is a fixed value regardless of the location and time, and it is possible to detect an unexpected event that is different from a normal one by using a statistical value.

  FIG. 8 shows a conceptual diagram when the travel time is used for the determination of the sudden event in the sudden event determination unit 32 as described above. FIG. 8 shows a past actual travel time distribution 41, a statistical value (T_ave) 42 obtained by averaging past travel times, a threshold lower limit value (T_ave−α) 43, and a threshold upper limit value (T_ave + α). ) 44 and the current travel time transition T45 generated from the probe information. In this way, by using the past statistical value and the fixed determination range α, it is possible to determine a sudden event at the sudden event determination time point 46.

  When it is determined as a sudden event, the sudden event occurrence position is identified (S33) and the sudden event output (S35) is performed by the sudden event occurrence position identifying unit 33 and the sudden event output unit 34, respectively. Since it is the same, it abbreviate | omits.

  With the above configuration, it is possible to detect natural traffic jams and traffic jams by detecting sudden events using past traffic information statistics and fixed judgment ranges, and detect only traffic jams caused by sudden events. Is possible.

  In the present embodiment, the processing in the threshold setting unit is performed independently of other processing and is performed at an arbitrary timing. However, each time the latest traffic information is input, the statistical information and the threshold are set. It may be set, and the sudden event may be detected based on the set threshold.

(Embodiment 3)
FIG. 6 is used as a component diagram for explaining a sudden event detection method using statistical values in Embodiment 3 of the present invention. In the present embodiment, a sudden event is detected by using a statistical value of past traffic information and a determination range optimally set for each place.
The operation of the threshold setting unit 31 in the sudden event determination unit 23 configured as described above will be described.

  The threshold setting unit 31 uses the past traffic information stored in the traffic information database 25 to generate statistical information T_ave smoothed for each time zone in each section or point. Next, the determination range α is set for each section and point based on the variation in past traffic information. For example, the deviation σ of traffic information for each section and point is calculated using a general method such as standard deviation indicating the degree of variation. When the deviation is large, the sudden event determination range α is increased, and when the deviation is small, the sudden event determination range α is set small. For example, a value proportional to the variation σ is set, such as α = a × σ (a is a positive number). A threshold value T_ave ± α is set based on the statistical value T_ave for each time zone generated as described above and a predetermined determination range α. The generated threshold value is stored in the traffic information database 25. The threshold setting process in the threshold setting unit 31 is performed independently of other processes and is performed at an arbitrary timing. It is desirable to do this regularly to update the statistics of traffic information.

  Since the process in the sudden event determination unit 32 is the same as that in the second embodiment, a description thereof will be omitted.

  As described above, FIGS. 9 and 10 show conceptual diagrams in the case where the sudden event is determined based on the travel time, regarding the processing up to the setting of the threshold and the determination of the sudden event. FIG. 9 shows a conceptual diagram of sudden event determination in a section in which the travel time transition of a specific section has regularity, that is, a section in which the past travel time variation σ is small, and FIG. 10 shows regularity in travel time transition. The outline figure of the sudden event determination in the section where there is no difference, that is, the section where the past travel time variation σ is large is shown.

  In these figures, a distribution 41 of past travel time data, a statistical value (T) 42 obtained by averaging past travel time data, a lower limit value (T_ave-α) 43 of a determination range, and an upper limit of a determination range A value (T_ave + α) 44 and a current travel time transition 45 generated from the probe information are shown.

  In the section where the travel time variation in FIG. 9 is small, that is, the section where the travel time deviation σ is small, it is possible to determine the sudden event at the sudden event determination time point 46 by setting the determination range α to be small.

  However, even if the statistical value (T) 42 is the same as in FIG. 9, in the section where the travel time variation is large as shown in FIG. As a result, it is impossible to determine a true sudden event.

  Therefore, as in the present embodiment, in a section where daily travel time variation is large, that is, in a section where travel time deviation σ is large, a sudden determination is made at a sudden event determination time point 46 by setting a large determination range α. It can be determined as a sudden event.

  When it is determined as a sudden event, the sudden event occurrence position is identified and the sudden event is output by the sudden event occurrence position identifying unit 33 and the sudden event output unit 34, respectively, but the description is omitted because it is the same as in the first embodiment.

  With the above configuration, it is possible to detect the occurrence of sudden events using statistical values of past traffic information and judgment ranges that are optimally set for each location such as a section or a point. It is possible to set an optimum threshold value in consideration of a small variation in travel time on highways and highways, but a large variation on general roads).

  In the present embodiment, the threshold setting unit 31 sets the determination range α for each section and point based on past traffic information variations, but further changes the determination range α for each time zone. By doing so, it is possible to improve the detection accuracy of the sudden event.

  Here, FIG. 11 shows a conceptual diagram when an unexpected event is determined based on travel time by setting a threshold value for each time zone. The same components as those in FIG. 9 are denoted by the same reference numerals, and description thereof is omitted. As shown in FIG. 11, the optimum determination range α that dynamically changes every time zone, for example, every 5 minutes is set, and the upper and lower determination ranges 44 and 45 are set. It becomes possible to determine as an unexpected event. As a result, the determination range α is small when the traffic condition variation is statistically small, and the determination range α is large when the traffic condition variation is statistically large. It is set, and sudden events can be determined with higher accuracy.

  With the above configuration, it is possible to detect sudden events using statistical values of past traffic information and judgment ranges that are optimally set for each hour, so that variations due to the time zone of traffic flow (no variations in travel time in the daytime) However, it is possible to set an optimum threshold value in consideration of a large variation in travel time at night.

  In the present embodiment, the process in the threshold setting unit is performed independently of other processes and is performed at an arbitrary timing. However, each time the latest traffic information is input, the statistical information and the threshold are set. It may be set, and the sudden event may be detected based on the set threshold.

(Embodiment 4)
FIG. 6 is used as a component diagram for explaining the sudden event detection method using the statistical value of the patterned traffic information in the fourth embodiment of the present invention. In the present embodiment, a sudden event is detected using a statistical value generated for each day type and a predetermined determination range.

  About the sudden event determination part 23 comprised as mentioned above, the operation | movement is demonstrated using the flowchart shown in FIG.

  The threshold value setting unit 31 uses the past traffic information stored in the traffic information database 25 to generate statistical information T_ave in each section or point for each similar pattern of traffic flow. Similar patterns of traffic flow may include weekdays, weekends, holidays, long holidays, and fifty days. Further, in each section or point, a threshold value T_ave ± α is set from the statistical value T_ave generated for each time zone and a predetermined determination range α. The predetermined determination range α may be fixed regardless of all places and times, or may be set for each section or time zone. The generated threshold value is stored in the traffic information database 25. The threshold setting process in the threshold setting unit 31 is performed independently of other processes and is performed at an arbitrary timing. It is desirable to do this regularly to update the statistics of traffic information.

  In the sudden event determination unit 32, when the latest traffic information T generated by the traffic information calculation unit 22 is input (S40), the pattern of statistical values such as weekdays and weekends is calculated from the date and time when the traffic information is generated. A determination is made (S41). After the determination, the corresponding section and the corresponding time zone threshold T_ave ± α are extracted from the traffic information database 25 (42). Next, T is compared with T_ave ± α, and when T_ave + α <T or T <T_ave−α, the traffic flow is determined to be an unexpected sudden event. When T_ave−α ≦ T ≦ T_ave + α Then, it is determined that it is within the range of variations in traffic information that occurs daily instead of sudden events (S43). In this way, it is possible to detect sudden events more accurately by classifying the statistical values to be compared into patterns according to the day type.

  FIG. 11 shows a conceptual diagram in the case where the sudden event is determined based on the travel time regarding the processing up to the threshold setting and the sudden event determination. The same components as those in FIG. 9 are denoted by the same reference numerals, and description thereof is omitted. It is assumed that 45 is the inputted traffic information (weekdays), 47, 48, and 49 are graphs of statistical information patterns and threshold values for weekdays, weekends, and holidays, respectively. Here, the pattern of the statistical value is determined from the day type of the input traffic information, and the weekday pattern 47 which is the day type of the input information is selected. Statistical values and threshold values of weekday patterns are extracted from the traffic information database, and the sudden event occurrence 46 is detected based on these values. In this way, it is possible to detect sudden events more accurately by classifying and using statistical values and threshold values for each day type such as weekdays, weekends, holidays, long holidays, and fifty days.

  When it is determined as a sudden event, the sudden event occurrence position identification (S44) and the sudden event output (S46) are performed by the sudden event occurrence position identifying unit 33 and the sudden event output unit 34, respectively. Since it is the same, it abbreviate | omits.

  With the above configuration, it is possible to set an optimum threshold in consideration of the difference depending on the day of the traffic flow by detecting the sudden event using the statistical value of the patterned traffic information. .

  In the present embodiment, the processing in the threshold setting unit is performed independently of other processing and is performed at an arbitrary timing. However, each time the latest traffic information is input, the statistical information and the threshold are set. It may be set, and the sudden event may be detected based on the set threshold.

  In the above embodiment, the determination of the sudden event is performed based on the travel time between the two points, but the determination of the sudden event may be similarly made using the average vehicle speed between the two points. In other words, there is no sudden event if the average vehicle speed between two specific points is within the range of the threshold calculated in consideration of the variation in the past vehicle speed of the section, and if it is out of the range, for example, it is delayed. For example, by determining that there is a sudden event, the sudden event occurrence section can be specified.

  Further, when the sudden event is determined based on the vehicle speed, the determination can be made by paying attention to any one point. That is, the vehicle speed at a specific point is extracted from the vehicle position information and vehicle speed information included in the probe information, and a threshold is set based on variations in the speeds of a plurality of vehicles that have passed through the point. When the threshold range is exceeded, it can be determined that a sudden event has occurred at that point.

  In the above description, the vehicle speed information is calculated from the position of the vehicle included in the vehicle information and the time at the position. However, the vehicle speed information calculated from the vehicle speed pulse or the vehicle speed information calculated from the GPS information is used. May be used.

  In the above description, the probe car system is taken as an example of the moving body information collecting means. However, the travel time information of the VICS information generated from the fixed sensor capable of collecting the moving body position information and time information is used. Thus, the sudden event may be detected.

  The sudden event determination method of the present invention collects position information indicating the position where the mobile body exists and mobile body information for obtaining the speed of the mobile body at this position, so that the sudden event can be determined. The sudden event can be detected, and the sudden event can be detected only from the vehicle information even in the section where the vehicle sensor is not present. Therefore, it can be used for a system for controlling traffic between the moving body and the center station.

1 is a schematic configuration diagram showing a sudden event determination method according to Embodiment 1 of the present invention. 1 is a schematic configuration diagram showing a sudden event determination unit according to Embodiment 1 of the present invention. The flowchart which shows operation | movement of the sudden event determination part in Embodiment 1 of this invention. The flowchart which shows operation | movement of the sudden event occurrence position specific | specification part in Embodiment 1 of this invention. Conceptual diagram of specifying the occurrence position of sudden event in Embodiment 1 of the present invention Schematic configuration diagram showing a sudden event determination unit in Embodiment 2 of the present invention The flowchart which shows operation | movement of the sudden event determination part in Embodiment 2 of this invention. Conceptual diagram of sudden event determination in Embodiment 2 of the present invention Conceptual diagram of sudden event determination in Embodiment 3 of the present invention Conceptual diagram of sudden event determination in Embodiment 3 of the present invention Conceptual diagram of sudden event determination in Embodiment 3 of the present invention The flowchart which shows operation | movement of the sudden event determination part in Embodiment 4 of this invention. Conceptual diagram of sudden event determination in Embodiment 4 of the present invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Vehicle equipment 2 Center station 3 GPS satellite 11 GPS information receiving part 12 Vehicle speed sensor 13 Communication part 14 Map database 15 Control part 21 Vehicle information collection part 22 Traffic information generation part 23 Sudden event determination part 24 Map database 25 Traffic information database 31 Threshold value Setting unit 32 Sudden event determination unit 33 Sudden event occurrence position specifying unit 34 Sudden event information output unit 41 Distribution of past traffic information 42 Statistical value of past traffic information 43 Lower limit value of threshold value 44 Upper limit value of threshold value 45 Generated from probe information Current traffic information 46 Sudden event determination time point 50 Road 51 Traffic light 52 Sudden event occurrence point 53 Transition of vehicle speed information of probe information 54 Statistical value of transition of past vehicle speed information 55, 56, 57 Congestion judgment unit 58 Sudden event occurrence point

Claims (16)

A mobile body information collection procedure for collecting mobile body information related to position information and time information at a position where the mobile body exists from the mobile body, and a mobile body information collected in the mobile body information collection procedure. The traffic information generation procedure for generating traffic information including the travel time of the mobile body or the speed of the mobile body and storing it in the traffic information storage means, and the traffic stored in the traffic information storage means using a predetermined threshold value A sudden event detection method comprising: a sudden event determination procedure for determining an abnormal traffic flow that suddenly occurs at the predetermined position from information. 2. The sudden event determination procedure according to claim 1, wherein the sudden event determination procedure includes a threshold value setting procedure for setting a certain range as the threshold value from a statistical value of past traffic information stored in the traffic information storage means. Event detection method. The sudden event detection method according to claim 2, wherein the threshold setting procedure sets the threshold based on a deviation of past traffic information stored in the traffic information storage means. 4. The sudden event detection method according to claim 3, wherein the threshold setting procedure sets the threshold for each point in consideration of a deviation of traffic information depending on a place. 4. The sudden event detection method according to claim 3, wherein the threshold setting procedure sets the threshold for each time zone in consideration of a deviation of traffic information for each time zone. The threshold setting procedure is characterized in that a predetermined range is set as the threshold value from statistical values of past traffic information for each similar pattern of traffic flow, such as weekdays, weekends, holidays, long holidays, and fifty days. The sudden event detection method according to any one of claims 2 to 5. The sudden event determination procedure according to claim 1, wherein the sudden event determination procedure includes a sudden event occurrence position specifying procedure for specifying the occurrence position of a sudden event using trajectory information and speed information obtained from the mobile object information. Event detection method. The sudden event detection method according to any one of claims 1 to 7, wherein the predetermined position is a position between two predetermined points. From the moving body information collecting means for collecting the moving body information related to the position information and the time information at the position where the moving body exists from the moving body, and the moving body information collected in the moving body information collecting procedure, Traffic information generating means for generating traffic information including the travel time of the mobile body or the speed of the mobile body and storing the traffic information in the traffic information storage means; and traffic stored in the traffic information storage means using a predetermined threshold value An unexpected event detection device comprising: an unexpected event determination means for determining an abnormal traffic flow suddenly occurring at the predetermined position from the information. 10. The sudden event determination unit according to claim 9, further comprising a threshold value setting unit configured to set a certain range as the threshold value from a statistical value of past traffic information stored in the traffic information storage unit. Event detection device. 10. The sudden event detection device according to claim 9, wherein the threshold setting unit sets the threshold based on a deviation of past traffic information stored in the traffic information storage unit. 11. The sudden event detection apparatus according to claim 10, wherein the threshold value setting means sets the threshold value for each point in consideration of a deviation of traffic information depending on a place. 11. The sudden event detection device according to claim 10, wherein the threshold value setting means sets the threshold value for each time zone in consideration of a deviation of traffic information for each time zone. The threshold value setting means sets a certain range as the threshold value from statistical values of past traffic information for each similar pattern of traffic flow, such as weekdays, weekends, holidays, long holidays, and fifty days. The sudden event detection device according to any one of claims 10 to 13. 10. The sudden event determination unit according to claim 9, wherein the sudden event determination unit includes a sudden event occurrence position specifying unit that specifies a generation position of a sudden event using trajectory information and speed information obtained from the mobile object information. Event detection device. 16. The sudden event detection device according to claim 9, wherein the predetermined position is a position between two predetermined points.

Images