Signal parameter estimation method, signal parameter estimation device, and signal parameter estimation program

Abstract

This allows for easy estimation of parameters related to the control of traffic signals. [Solution] The method includes the steps of creating intersection information, including the time, based on probe information including the position and time of a vehicle, and performing an estimation process to estimate parameters related to the control of traffic signals at the intersection based on the created intersection information.

Description

Technical Field

[0001] The present disclosure relates to a signal parameter estimation method, a signal parameter estimation device, and a signal parameter estimation program.

Background Art

[0002] Conventionally, technologies for providing information regarding a traffic signal device received from a roadside unit to a vehicle have been developed. For example, Patent Document 1 (Japanese Patent Application Laid-Open No. 2009-252156) discloses the following technology. That is, a driving support device includes a vehicle-road communication means for receiving signal cycle information, which is information regarding a change in a signal at a traffic signal, transmitted from a roadside transmitter installed on a road, a vehicle-vehicle communication receiving means for receiving the signal cycle information of the traffic signal read by another vehicle from the roadside transmitter from the other vehicle, a vehicle group control means for generating vehicle group control information for controlling the other vehicle based on the degree of overlap and the degree of continuity between the signal cycle information received by the vehicle-road communication means and the signal cycle information expected to be received by the vehicle-vehicle communication receiving means, and a vehicle-vehicle communication transmitting means for transmitting the vehicle group control information generated by the vehicle group control means to the other vehicle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Parameters related to the control of a traffic signal device are generally managed by a predetermined institution, and it is difficult for users such as vehicle manufacturers to obtain the parameters. There is a need for a technology that can easily estimate the parameters.

[0005] This disclosure was made to solve the above-mentioned problems, and its purpose is to provide a signal parameter estimation method, a signal parameter estimation device, and a signal parameter estimation program that can easily estimate parameters related to the control of signal lights. [Means for solving the problem]

[0006] The signal parameter estimation method of this disclosure includes the steps of creating intersection information, including time, based on probe information, including the position and time of a vehicle, and performing an estimation process to estimate parameters relating to the control of signal lights at the intersection based on the created intersection information.

[0007] One aspect of this disclosure can be implemented not only as a signal parameter estimation method including such characteristic processing steps, but also as a semiconductor integrated circuit that implements some or all of a signal parameter estimation device that performs such characteristic processing. Furthermore, one aspect of this disclosure can be implemented as a system including a signal parameter estimation device. [Effects of the Invention]

[0008] According to this disclosure, parameters related to the control of traffic signals can be easily estimated. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 shows an example of the configuration of a signal parameter estimation system according to the first embodiment of this disclosure. [Figure 2] Figure 2 shows an example of the configuration of an in-vehicle device according to the first embodiment of this disclosure. [Figure 3] Figure 3 shows an example of the server configuration according to the first embodiment of this disclosure. [Figure 4] Figure 4 shows an example of transit time information created by the server according to the first embodiment of this disclosure. [Figure 5]Figure 5 is a diagram illustrating the server configuration process according to the first embodiment of this disclosure. [Figure 6] Figure 6 is a diagram illustrating the estimation process performed by a server according to the first embodiment of this disclosure. [Figure 7] Figure 7 shows an example of a score evaluation value calculated in the estimation process performed by the server according to the first embodiment of this disclosure. [Figure 8] Figure 8 shows another example of a score evaluation value calculated in the estimation process performed by the server according to the first embodiment of this disclosure. [Figure 9] Figure 9 shows the simulation results illustrating the accuracy of signal parameter estimation by the server according to the first embodiment of this disclosure. [Figure 10] Figure 10 shows the simulation results illustrating the accuracy of signal parameter estimation by the server according to the first embodiment of this disclosure. [Figure 11] Figure 11 is a flowchart illustrating an example of the operation procedure when an in-vehicle device according to the first embodiment of this disclosure performs a process to transmit probe information. [Figure 12] Figure 12 is a flowchart illustrating an example of the operation procedure when the server according to the first embodiment of this disclosure performs estimation processing. [Figure 13] Figure 13 shows an example of a score included in the indication switching information created by the server according to the second embodiment of this disclosure. [Figure 14] Figure 14 shows an example of indication switching information created by a server according to the second embodiment of this disclosure. [Figure 15] Figure 15 is a diagram illustrating the server configuration process according to the second embodiment of this disclosure. [Figure 16] Figure 16 is a diagram illustrating the estimation process by a server according to a second embodiment of the present disclosure. [Figure 17]FIG. 17 is a diagram showing an example of an evaluation value of a score calculated in the estimation process by the server according to the second embodiment of the present disclosure. [Figure 18] FIG. 18 is a diagram showing simulation results for explaining the estimation accuracy of signal parameters by the server according to the second embodiment of the present disclosure. [Figure 19] FIG. 19 is a diagram showing simulation results for explaining the estimation accuracy of signal parameters by the server according to the second embodiment of the present disclosure. [Figure 20] FIG. 20 is a flowchart defining an example of an operation procedure when the server according to the second embodiment of the present disclosure performs an estimation process. [Figure 21] FIG. 21 is a diagram showing an example of the configuration of an in-vehicle device according to the third embodiment of the present disclosure. [Figure 22] FIG. 22 is a diagram showing an example of traffic signal color information created by the server according to the third embodiment of the present disclosure. [Figure 23] FIG. 23 is a diagram showing an example of passing time information created by the server according to the third embodiment of the present disclosure. [Figure 24] FIG. 24 is a diagram showing an example of display change information created by the server according to the third embodiment of the present disclosure. [Figure 25] FIG. 25 is a diagram for explaining the setting process by the server according to the third embodiment of the present disclosure. [Figure 26] FIG. 26 is a diagram for explaining the estimation process by the server according to the third embodiment of the present disclosure. [Figure 27] FIG. 27 is a diagram showing an example of an evaluation value of a score calculated in the estimation process by the server according to the third embodiment of the present disclosure. [Figure 28] FIG. 28 is a diagram showing simulation results for explaining the estimation accuracy of signal parameters by the server according to the third embodiment of the present disclosure. [Figure 29]Figure 29 shows the simulation results illustrating the accuracy of signal parameter estimation by the server according to the third embodiment of this disclosure. [Figure 30] Figure 30 is a flowchart illustrating an example of the operation procedure when an in-vehicle device according to the third embodiment of this disclosure performs a process to transmit probe information. [Figure 31] Figure 31 is a flowchart illustrating an example of the operation procedure when a server according to the third embodiment of this disclosure performs estimation processing. [Modes for carrying out the invention]

[0010] First, the embodiments of this disclosure will be listed and explained. (1) A signal parameter estimation method according to an embodiment of the present disclosure includes the steps of creating intersection information, including time, based on probe information, including the position and time of a vehicle, and performing an estimation process to estimate parameters relating to the control of signal lights at the intersection based on the created intersection information.

[0011] This method allows for the estimation of parameters related to the control of traffic signals at intersections using intersection information based on probe data obtainable directly or indirectly from vehicles. Therefore, parameters related to the control of traffic signals can be easily estimated.

[0012] (2) In (1) above, the intersection information may include the time, the position of the vehicle, and information representing the behavior of the vehicle at the intersection.

[0013] This method allows for the accurate estimation of parameters related to the control of traffic signals, for example, based on the type of road the vehicle was traveling on, the vehicle's behavior at the intersection, and the time when that behavior occurred.

[0014] (3) In (1) or (2) above, the intersection information may further include information representing the time and the color of the signal light.

[0015] This method allows for the estimation of parameters related to the control of traffic signals, for example, by identifying the time when a traffic signal light is blue from vehicle probe data.

[0016] (4) In any of (1) to (3) above, the signal parameter estimation method further includes the step of setting a range of candidate parameters which are candidates for the parameter based on the acquired probe information, and in the step of performing the estimation process, the parameter is determined from among the candidate parameters in the set range.

[0017] This method allows us to limit the range of candidate parameters to be estimated, thereby making it easier to estimate those parameters.

[0018] (5) In any of (1) to (4) above, the intersection information may include passage time information including the time the vehicle passed through the intersection and indication change information including the time the indication of the signal light changed from red to blue, and in the step of performing the estimation process, the parameters may be estimated based on the weighted passage time information and indication change information, and the weight value corresponding to the passage time information may be greater than the weight value corresponding to the indication change information.

[0019] The accuracy of the indication change information based on probe data may be lower than the accuracy of the passage time information based on probe data. By using the method described above, the degree of influence of indication change information on the parameter estimation result can be made smaller than the degree of influence of passage time information on the said estimation result, thus enabling a more accurate estimation of the parameter.

[0020] (6) The signal parameter estimation device according to an embodiment of the present disclosure comprises a creation unit that creates intersection information including time based on probe information including the position and time of a vehicle, and an estimation processing unit that performs estimation processing to estimate parameters relating to the control of signal lights at the intersection based on the intersection information created by the creation unit.

[0021] This configuration allows for the estimation of parameters related to the control of traffic signals at intersections using intersection information based on probe data obtainable directly or indirectly from vehicles. Therefore, parameters related to the control of traffic signals can be easily estimated.

[0022] (7) The signal parameter estimation program according to the embodiment of the present disclosure is a signal parameter estimation program used in a signal parameter estimation device, which causes a computer to function as a creation unit that creates intersection information including time based on probe information including the position and time of a vehicle, and an estimation processing unit that performs estimation processing to estimate parameters relating to the control of signal lights at the intersection based on the intersection information created by the creation unit.

[0023] This configuration allows for the estimation of parameters related to the control of traffic signals at intersections using intersection information based on probe data obtainable directly or indirectly from vehicles. Therefore, parameters related to the control of traffic signals can be easily estimated.

[0024] Embodiments of this disclosure will be described below with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and their descriptions will not be repeated. Furthermore, at least some of the embodiments described below may be combined in any way.

[0025] <First Embodiment> [Signal parameter estimation system] Figure 1 shows an example of the configuration of a signal parameter estimation system according to a first embodiment of the present disclosure. Referring to Figure 1, the signal parameter estimation system 501 comprises a plurality of in-vehicle devices 101, a server 301, and a fleet management device 401. The server 301 and each in-vehicle device 101 transmit and receive information via an external network 151, such as the Internet. The in-vehicle devices 101 are mounted on a vehicle 1. The server 301 is an example of a signal parameter estimation device.

[0026] The server 301 and the operation management device 401 are used, for example, by a business operator or individual (hereinafter collectively referred to as a user) that manages the operation of vehicle 1. The server 301 and the operation management device 401 transmit and receive information, for example, via an external network 151.

[0027] Server 301 collects probe information from each vehicle 1 and uses the collected probe information to estimate parameters related to the control of the traffic signal lights L at the intersection where the main road and the secondary road intersect (hereinafter also referred to as "signal parameters"). For example, the probe information includes identification information for identifying vehicle 1 (hereinafter also referred to as "vehicle ID (Identifier)"), the position of vehicle 1, the azimuth angle of vehicle 1, and the time. The vehicle ID is a unique ID for vehicle 1.

[0028] For example, the signal parameters estimated by server 301 are the cycle length, cycle start time, and split seconds of the signal light L.

[0029] The cycle length is the time required for one cycle of the signal indicator light L to complete a full rotation. Specifically, for example, the cycle length is the time from the start of the green light on signal light L to the start of the next green light. Alternatively, the cycle length may be the time from the start of the red light on signal light L to the start of the next red light.

[0030] The cycle start time is the time from the reference time to the start time of the green light illumination at the main road's traffic light L (hereinafter also referred to as "traffic light L1"). The split seconds is the green light time allocated to the main road. The green light time is the time during which vehicle 1 has the right of way at the intersection.

[0031] (In-vehicle device) Figure 2 shows an example of the configuration of an in-vehicle device according to a first embodiment of the present disclosure. Referring to Figure 2, the in-vehicle device 101 comprises an in-vehicle communication unit 11, a creation unit 12, an external communication unit 13, and a storage unit 14. Some or all of the in-vehicle communication unit 11, the creation unit 12, and the external communication unit 13 are implemented by a processing circuit (Circuitry) including, for example, one or more processors. The storage unit 14 is, for example, a non-volatile memory included in the processing circuit.

[0032] For example, the in-vehicle device 101 is connected to multiple in-vehicle devices 202, namely in-vehicle devices 202A and 202B, via a CAN bus 51 that conforms to the CAN (Controller Area Network) standard.

[0033] Furthermore, the in-vehicle device 101 may be connected to the in-vehicle equipment 202 via a transmission line conforming to other communication standards such as Ethernet (registered trademark), rather than being limited to the CAN bus 51.

[0034] Each in-vehicle device 202 transmits a frame containing various types of information, as described later, to the in-vehicle device 101. The in-vehicle devices 202 include various sensors, GNSS (Global Navigation Satellite System) receivers, cameras, etc.

[0035] In the example shown in Figure 2, the in-vehicle device 202A is a GNSS receiver. Hereafter, the in-vehicle device 202A will also be referred to as the GNSS receiver 202A.

[0036] The GNSS receiver 202A receives GNSS signals from one or more satellites, for example, periodically, and determines the position of vehicle 1 based on the received GNSS signals. The position of vehicle 1 is indicated, for example, by latitude and longitude.

[0037] The GNSS receiver 202A periodically transmits location information to the in-vehicle device 101, including the detection result of the vehicle 1's position and the time of detection.

[0038] In the in-vehicle device 101, the in-vehicle communication unit 11 stores the received location information in the storage unit 14 each time it receives location information from the GNSS receiver 202A.

[0039] The creation unit 12 performs a creation process B1 to create probe information for vehicle 1. More specifically, when the processing timing T1 of creation process B1 arrives, the creation unit 12 retrieves multiple location information stored in the storage unit 14 by the in-vehicle communication unit 11 during the period from the previous processing timing T1 to the current processing timing T1.

[0040] The creation unit 12 then creates azimuth angle information that shows the time-series change in the azimuth angle of vehicle 1 over a period that includes multiple detection times corresponding to each of the multiple location information points extracted.

[0041] More specifically, the creation unit 12 calculates the azimuth angle of the vehicle 1 at the detection time included in the location information for each location information retrieved from the storage unit 14.

[0042] Specifically, for example, the creation unit 12 calculates the azimuth angle of vehicle 1 at detection time ta based on the position information of vehicle 1 at a certain detection time ta and the position information of vehicle 1 at another detection time tb that is consecutive to detection time ta.

[0043] The memory unit 14 stores the vehicle ID. When the creation unit 12 creates azimuth angle information, it creates probe information including multiple location information retrieved from the memory unit 14, the azimuth angle information, and the vehicle ID, and outputs it to the external communication unit 13.

[0044] The external communication unit 13 transmits the probe information received from the creation unit 12 to the server 301. More specifically, the external communication unit 13 communicates with the server 301 via the external network 151 by wirelessly communicating with devices such as a wireless base station (not shown) according to a communication method such as Wi-Fi (registered trademark), LTE (Long Term Evolution) (registered trademark), or 5G. The external communication unit 13 may also be configured to communicate with the server 301 via other in-vehicle devices.

[0045] The external communication unit 13 creates an IP packet (hereinafter also referred to as "packet P") that includes probe information received from the creation unit 12 and includes the IP address of its own in-vehicle device 101 and the IP address of the server 301 as the source address and destination IP address, respectively. The external communication unit 13 then transmits the created packet P to the server 301 via the wireless base station device and the external network 151.

[0046] (server) Figure 3 shows an example of the configuration of a server according to the first embodiment of the present disclosure. Referring to Figure 3, the server 301 comprises a communication unit 31, a creation unit 32, a setting unit 33, an estimation processing unit 34, and a storage unit 35. Some or all of the communication unit 31, the creation unit 32, the setting unit 33, and the estimation processing unit 34 are implemented by a processing circuit including, for example, one or more processors. The storage unit 35 is, for example, a non-volatile memory included in the processing circuit. The communication unit 31 is an example of an acquisition unit.

[0047] The communication unit 31 acquires probe information from each vehicle 1. Specifically, the communication unit 31 receives packets P from each vehicle 1 via the wireless base station equipment and the external network 151. The communication unit 31 then stores the probe information contained in the received packets P in the storage unit 35.

[0048] The storage unit 35 stores map information for a region that includes the location of vehicle 1 indicated by probe information transmitted from the in-vehicle device 101.

[0049] The creation unit 32 performs creation process B2 to create intersection information, including the time, based on the probe information acquired by the communication unit 31. More specifically, for example, the creation unit 32 performs creation process B2 using probe information from multiple vehicles 1 received by the communication unit 31.

[0050] For example, intersection information includes information representing the correspondence between time, the position of vehicle 1, and the behavior of vehicle 1 at the intersection. In this embodiment, for example, the creation unit 32 creates passage time information C1 as intersection information, which includes the time when vehicle 1 passed through the target intersection (hereinafter also referred to as "target intersection CS"). Specifically, the passage time information C1 includes at least one of the time when vehicle 1 passed the stop line of the main road of the target intersection CS, and the time when vehicle 1 passed the stop line of the secondary road of the target intersection CS, as the time when vehicle 1 passed through the target intersection CS.

[0051] When the processing timing T2 of the creation process B2 arrives, the creation unit 32 retrieves the probe information of multiple vehicles 1 that were stored in the storage unit 35 by the communication unit 31 during the period from the previous processing timing T2 to the current processing timing T2.

[0052] Then, the creation unit 32 performs a passage determination process for each extracted probe information to determine whether or not the vehicle 1 corresponding to that probe information has passed through the target intersection CS.

[0053] Specifically, for example, the creation unit 32 performs a passage determination process using the extracted probe information and the map information stored in the storage unit 35.

[0054] If the creation unit 32 determines that vehicle 1 has passed through the target intersection CS, it checks whether vehicle 1 passed through the main road of the target intersection CS or through a secondary road of the target intersection CS.

[0055] The creation unit 32 then creates passage time information C1, which includes the time when vehicle 1 passed through the target intersection CS, a score S1 indicating that vehicle 1 passed through the main road of the target intersection CS at that time, and at least one of a score S2 indicating that vehicle 1 passed through the secondary road of the target intersection CS at that time.

[0056] Figure 4 shows an example of transit time information created by the server according to the first embodiment of this disclosure.

[0057] In the example shown in Figure 4, times t0, t1, t8, t11, t12, t17, and t20 are the times when vehicle 1 passed through the main road of the target intersection CS. In this case, for example, the creation unit 32 sets the score S1 at each of these times to "15".

[0058] Furthermore, in the example shown in Figure 4, the times t6, t14, t23, and t24 are the times when vehicle 1 passed through the secondary road of the target intersection CS. In this case, for example, the creation unit 32 sets the score S2 at each of these times to "15".

[0059] When the creation unit 32 creates the passage time information C1, it outputs the created passage time information C1 to the setting unit 33.

[0060] Furthermore, the creation unit 32 may be configured to set a value V1, obtained by multiplying the score S1 by the number of vehicles 1 that passed at that time, as a score indicating that a vehicle passed the main road, if multiple vehicles 1 pass the main road at a given time. The value V1 is, for example, a value less than or equal to a predetermined upper limit.

[0061] Furthermore, the creation unit 32 may be configured to set a score V2, which is obtained by multiplying the score S2 by the number of vehicles 1 that passed at that time, if multiple vehicles 1 pass the secondary road at a certain time. The value V2 is, for example, a value less than or equal to a predetermined upper limit.

[0062] Furthermore, the creation unit 32 is not limited to creating information that indicates that vehicle 1 has passed through the main road or secondary road using scores S1 and S2 as passage time information C1. It may also be configured to create information that indicates that vehicle 1 has passed through the main road or secondary road using other methods such as function approximation and autocorrelation as passage time information C1.

[0063] Referring again to Figure 3, for example, the setting unit 33 performs a setting process to set the range of candidate signal parameters (hereinafter also referred to as "parameter candidates") based on the probe information acquired by the communication unit 31.

[0064] More specifically, the setting unit 33 uses the passage time information C1 received from the creation unit 32 to set the candidate range D1 for cycle length, the candidate range D2 for cycle start time, and the candidate range D3 for split seconds.

[0065] Figure 5 is a diagram illustrating the server configuration process according to the first embodiment of this disclosure.

[0066] Referring to Figure 5, the setting unit 33 estimates the time period A1 when the signal light L1 at the target intersection CS is red, and the time period A2 when the signal light L1 is blue, based on the passage time information C1 received from the creation unit 32. Hereinafter, the time when vehicle 1 passes through the main road and the time when vehicle 1 passes through the secondary road will also be referred to as passage time tm and passage time tn, respectively.

[0067] For example, the setting unit 33 estimates time zone A1 as the time period from the time obtained by adding a predetermined time J1 to the passing time tn, up to a predetermined time J2 before the passing time tm closest to that passing time tn. The setting unit 33 also estimates time zone A2 as the time period from the time obtained by adding a predetermined time J3 to the passing time tm, up to a predetermined time J4 before the passing time tn closest to that passing time tm. Times J1, J2, J3, and J4 are predetermined margin times. Times J1, J2, J3, and J4 may be the same value or may be different values.

[0068] In the example shown in Figure 5, the passing times tm are t32, t33, t37, t38, t39, t42, t43, and t44. The passing times tn are t31, t34, t35, t36, t40, t41, t45, t46, and t47.

[0069] The setting unit 33 estimates the following as time zone A1: time zone A11 from the time obtained by adding time J1 to time t31 to a time 2 times before time t32; time zone A12 from the time obtained by adding time J1 to time t36 to a time 2 times before time t37; and time zone A13 from the time obtained by adding time J1 to time t41 to a time 2 times before time t42.

[0070] Furthermore, the setting unit 33 estimates the following as time zone A2: time zone A21 from the time obtained by adding time J3 to time t33 to a time 2 times before time t34; time zone A22 from the time obtained by adding time J3 to time t39 to a time 4 times before time t40; and time zone A23 from the time obtained by adding time J3 to time t44 to a time 4 times before time t45.

[0071] The setting unit 33 then predicts the minimum value M1 and maximum value M2 of the cycle length on the main road based on the estimated time periods A1 and A2.

[0072] In the example shown in Figure 5, the setting unit 33 predicts the length of the period from a time J2 before time t32 to the time obtained by adding time J1 to time t36, and the length of the period from the time obtained by adding time J3 to time t33 to a time J4 before time t40, as the minimum value M1 and the maximum value M2, respectively.

[0073] Then, the setting unit 33 sets the candidate range D1 for cycle length, the candidate range D2 for cycle start time, and the candidate range D3 for split seconds using the predicted minimum value M1 and maximum value M2.

[0074] More specifically, the setting unit 33 calculates the minimum value M11 and maximum value M12 of the candidate cycle length by substituting the predicted minimum value M1 and maximum value M2 into the following equations (1) and (2). Then, the setting unit 33 sets the range from the minimum value M11 to the maximum value M12 as range D1. M11 = min(M1, M2 - 20) ... (1) M12 = max(M2, M1 + 20) ... (2)

[0075] Furthermore, the setting unit 33 calculates the maximum value M21 of the candidate cycle start time by substituting the predicted maximum value M2 into the following equation (3). Then, the setting unit 33 sets the range from zero to the maximum value M21 as range D2. M21 = min(200, M2) ... (3)

[0076] Furthermore, the setting unit 33 sets the value represented by the following equation (4) and the value represented by the following equation (5) as the minimum value M31 and the maximum value M32 of the candidate split seconds, respectively. That is, the setting unit 33 sets the range from the minimum value M31 to the maximum value M32 as range D3. In equations (4) and (5), Lc is the estimated value of the cycle length estimated by the estimation processing unit 34, which will be described later. M31 = 0.25 × Lc ... (4) M32 = 0.75 × Lc ... (5)

[0077] When the setting unit 33 sets the candidate range D1 for cycle length, the candidate range D2 for cycle start time, and the candidate range D3 for split seconds, it outputs the setting range information indicating the set ranges D1, D2, and D3, along with the passage time information C1 received from the creation unit 32, to the estimation processing unit 34.

[0078] The estimation processing unit 34 performs estimation processing to estimate predetermined signal parameters of the signal lights L at the target intersection CS, based on the passage time information C1 created by the creation unit 32.

[0079] More specifically, for example, the estimation processing unit 34 determines the cycle length from among the candidate cycle lengths in the range D1 set by the setting unit 33. The estimation processing unit 34 also determines the cycle start time from among the candidate cycle start times in the range D2 set by the setting unit 33. The estimation processing unit 34 also determines the split seconds from among the candidate split seconds in the range D3 set by the setting unit 33.

[0080] Specifically, the estimation processing unit 34 uses the setting range information received from the setting unit 33 to select a set W of candidate cycle length, candidate cycle start time, and candidate split seconds. The selection of set W is performed, for example, by brute force.

[0081] Figure 6 is a diagram illustrating the estimation process performed by a server according to the first embodiment of this disclosure.

[0082] Referring to Figure 6, when the estimation processing unit 34 selects a set W, it uses the passage time information C1 included in the setting range information received from the setting unit 33 and the set W to estimate the time tm when the light color of signal light L1 and the light color of signal light L on the secondary road (hereinafter also referred to as "signal light L2") are blue and red, respectively, and the time tn when the light color of signal light L1 and the light color of signal light L2 are red and blue, respectively. In Figure 6, the scores corresponding to the times when the light color of signal light L1 or signal light L2 is blue are hatched diagonally to the right. Also, the scores corresponding to the times when the light color of signal light L1 or signal light L2 is red are hatched in a grid pattern.

[0083] In the example shown in Figure 6, the cycle length, cycle start time, and split seconds in the set W selected by the estimation processing unit 34 are assumed to be "8 seconds," "0 seconds," and "5 seconds," respectively. In this case, the estimation processing unit 34 estimates that time t0, t1, t2, t3, t4, t8, t9, t10, t11, t12, t16, t17, t18, t19, t20, and t24 are time tm, and estimates that time t5, t6, t7, t13, t14, t15, t21, t22, and t23 are time tn.

[0084] The estimation processing unit 34 evaluates the scores S1 and S2 indicated by the passage time information C1 included in the set range information created by the creation unit 32 for each selected set W.

[0085] More specifically, for example, when the estimation processing unit 34 estimates times tm and tn, it calculates values ​​H11 and H12 by substituting the value N11 of score S1 at time tm into the following equation (6) and the value N21 of score S2 at time tn into the following equation (7). Then, the estimation processing unit 34 calculates the absolute values ​​of value H11 and value H12 as the evaluation value of score S1 and the absolute value of score S2, respectively. H11 = N11 - 15 ... (6) H12=N21-15 ···(7)

[0086] Furthermore, for example, the estimation processing unit 34 calculates values ​​H13 and H14 by substituting the value N12 of score S1 at time tn into the following equation (8), and the value N22 of score S2 at time tm into the following equation (9). The estimation processing unit 34 then calculates the absolute values ​​of value H13 and value H14 as the evaluation value of score S1 and the absolute value of score S2, respectively. H13 = N12 - (-15) ... (8) H14 = N22 - (-15) ... (9)

[0087] The respective evaluation values ​​of scores S1 and S2 at time tm, and the respective evaluation values ​​of scores S1 and S2 at time tn, are index values ​​that represent the error between the candidate parameters and the actual signal parameters.

[0088] Figure 7 shows an example of a score evaluation value calculated in the estimation process by the server according to the first embodiment of this disclosure. Figure 7 shows the respective evaluation values ​​of scores S1 and S2 when the server 301 selects a set W with a cycle length of "7 seconds", a cycle start time of "zero seconds", and a split time of "4 seconds".

[0089] Figure 8 shows another example of a score evaluation value calculated in the estimation process by the server according to the first embodiment of this disclosure. Figure 8 shows the respective evaluation values ​​of scores S1 and S2 when the server 301 selects a set W with a cycle length of "7 seconds", a cycle start time of "zero seconds", and a split time of "4 seconds".

[0090] The estimation processing unit 34 calculates the evaluation value of the score at each time point, and then calculates a total value K1 by summing up the multiple evaluation values ​​for the period from time t0 to time t24.

[0091] In the example shown in Figure 7, the total score K1 is "615" points, and in the example shown in Figure 8, the total score K1 is "735" points.

[0092] The estimation processing unit 34 calculates multiple sums K1 corresponding to each of the selected sets W, and outputs the set W corresponding to the smallest sum K1 as the result of the estimation process.

[0093] Specifically, the estimation processing unit 34 outputs parameter information to the communication unit 31 indicating the set W corresponding to the smallest total value K1, that is, the set W estimated to have the smallest error with the actual signal parameters.

[0094] In the examples shown in Figures 7 and 8, the estimation processing unit 34 outputs parameter information to the communication unit 31 indicating a set W corresponding to a total value K1 of "615" points shown in Figure 7, which is smaller than the total value K1 of "715" points shown in Figure 8.

[0095] Referring again to Figures 1 and 3, in the server 301, the communication unit 31 transmits the parameter information received from the estimation processing unit 34 to the operation management device 401 via the external network 151.

[0096] The operation management device 401 displays the set W, which represents the parameter information received from the server 301 via the external network 151, on its own monitor or the like.

[0097] Next, the relationship between the accuracy of the estimation process using the passage time information C1 by server 301 and the probe rate will be explained. The probe rate is the ratio of the number of vehicles 1 that transmitted probe information F2 to the number of vehicles 1 that entered the target intersection CS during a predetermined period E F1.

[0098] Figures 9 and 10 show simulation results illustrating the accuracy of signal parameter estimation by a server according to the first embodiment of this disclosure. Figures 9 and 10 show estimated values ​​of cycle length, cycle start time, and split seconds when the target intersection CS is an intersection controlled by two indications, and the probe rate is in 10% increments within the range of 10% to 90%. In this embodiment, for example, the predetermined period E is 5 hours, and the number of vehicles 1 entering the target intersection CS per hour is 680. That is, the number of vehicles F1 is 13,600.

[0099] The correct values ​​for cycle length, cycle start time, and split seconds, corresponding to the simulation results shown in Figure 9, are "60 seconds," "1 second," and "30 seconds," respectively. The correct values ​​for cycle length, cycle start time, and split seconds, corresponding to the simulation results shown in Figure 10, are "60 seconds," "1 second," and "17 seconds," respectively.

[0100] In the examples shown in Figures 9 and 10, the estimated cycle length of signal lamp L1 and the estimated cycle length of signal lamp L2 are "60 seconds," the same as the correct value, at each probe rate. Thus, the server 301 according to the first embodiment can correctly estimate the cycle length even if the probe rate is "10%", that is, even if the number of vehicles F2 that transmitted probe information during a predetermined period E is 1360.

[0101] Furthermore, in the examples shown in Figures 9 and 10, the absolute error between the estimated cycle start time of signal light L1 and the correct value of "1 second" for each probe rate is 2 seconds or less. Also, the absolute error between the estimated cycle start time of signal light L2 and the correct value of "1 second" for each probe rate is 2 seconds or less.

[0102] Furthermore, in the examples shown in Figures 9 and 10, the absolute error between the estimated split time of signal light L1 and the correct split time of "30 seconds" is 3 seconds or less at each probe rate. Also, at each probe rate, the absolute error between the estimated split time of signal light L2 and the correct split time of "17 seconds" is 3 seconds or less.

[0103] Thus, in the server 301 according to the first embodiment, even if the probe rate is "10%", that is, even if the number of units F2 is 1360, the cycle start time and split seconds of each signal light L1 and L2 can be estimated with high accuracy.

[0104] [Operation Flow] Figure 11 is a flowchart illustrating an example of the operation procedure when an in-vehicle device according to the first embodiment of this disclosure performs a process to transmit probe information.

[0105] Referring to Figure 11, first, the in-vehicle device 101 receives position information from the GNSS receiver 202A and stores it in the memory unit 14 (step ST101).

[0106] The in-vehicle device 101 stores the new position information received from the GNSS receiver 202A in the storage unit 14 (step ST101) until the processing timing T1 of the creation process B1 for creating probe information arrives (NO in step ST102).

[0107] Then, when the processing timing T1 arrives (YES in step ST102), the in-vehicle device 101 retrieves multiple location information stored in the storage unit 14 during the period from the previous processing timing T1 to the current processing timing T1 (step ST103).

[0108] Next, the in-vehicle device 101 creates azimuth angle information for each retrieved location information, indicating the azimuth angle of vehicle 1 at the detection time included in the location information (step ST104).

[0109] Next, the in-vehicle device 101 performs a creation process B1 to create probe information that includes multiple location information retrieved from the storage unit 14, the created azimuth angle information, and the vehicle ID stored in the storage unit 14 (step ST105).

[0110] Next, the in-vehicle device 101 transmits the created probe information to the server 301 (step ST106), and receives new position information from the GNSS receiver 202A and stores it in the storage unit 14 (step ST101).

[0111] Figure 12 is a flowchart illustrating an example of the operation procedure when the server according to the first embodiment of this disclosure performs estimation processing.

[0112] Referring to Figure 12, first, the server 301 receives probe information from each vehicle 1 and stores it in the storage unit 35 (step ST201).

[0113] Server 301 stores the new probe information received from each vehicle 1 in the storage unit 35 until the processing timing T2 for the estimation process arrives (NO in step ST202) (step ST201).

[0114] Next, when processing timing T2 arrives (YES in step ST202), server 301 retrieves the probe information of multiple vehicles 1 that were stored in the storage unit 35 during the period from the previous processing timing T2 to the current processing timing T2 (step ST203).

[0115] Next, the server 301 creates passage time information C1 based on the probe information of the multiple vehicles 1 that it has retrieved. For example, as described above, the server 301 creates passage time information C1 that includes the time when vehicle 1 passed through the target intersection CS, a score S1 indicating that vehicle 1 passed through the main road, and a score S2 indicating that vehicle 1 passed through the secondary road (step ST204).

[0116] Next, when the server 301 creates the passage time information C1, it performs a setting process using the created passage time information C1 to set the candidate range D1 for the cycle length of the signal light L, the candidate range D2 for the cycle start time of the signal light L, and the candidate range D3 for the split seconds of the signal light L (step ST205).

[0117] Next, when ranges D1, D2, and D3 are set, the server 301 performs estimation processing to estimate the signal parameters. For example, as described above, for each set W of candidate cycle length, candidate cycle start time, and candidate split seconds, the server 301 calculates an evaluation value of the score at each time point included in the transit time information C1. Then, for each set W, the server 301 calculates a total value K1 by summing the evaluation values ​​and determines the set W corresponding to the smallest total value K1 as the estimation result (step ST206).

[0118] Next, the server 301 transmits the estimated signal parameter results to the operation management device 401 (step ST207), and receives new probe information from each vehicle 1 and stores it in the storage unit 35 (step ST201).

[0119] In the signal parameter estimation system 501 according to the first embodiment of this disclosure, the server 301 is configured to estimate the cycle length, cycle start time, and split seconds of the signal lamp L, but it is not limited to this. The server 301 may be configured to estimate only a portion of the cycle length, cycle start time, and split seconds. Furthermore, the server 301 may be configured to estimate other signal parameters instead of some or all of the cycle length, cycle start time, and split seconds, or in addition to these three signal parameters.

[0120] Furthermore, in the signal parameter estimation system 501 according to the first embodiment of this disclosure, the server 301 is configured to create passage time information C1, which includes the time when vehicle 1 passed through the target intersection CS, as intersection information representing the time, the position of vehicle 1, and the behavior of vehicle 1 at the target intersection CS. However, the system is not limited to this configuration. The server 301 may also be configured to create information as intersection information that includes the time when other behaviors, such as vehicle 1 stopping, occurred at the target intersection CS.

[0121] Furthermore, in the signal parameter estimation system 501 according to the first embodiment of this disclosure, the server 301 is configured to set a range of candidate signal parameters based on probe information acquired from the vehicle 1 and to perform estimation processing using that range, but it is not limited to this. The server 301 may be configured to set a range of candidate signal parameters based on information other than probe information. Alternatively, the server 301 may be configured to perform estimation processing using a range that has been set in advance by the user.

[0122] Furthermore, in the signal parameter estimation system 501 according to the first embodiment of this disclosure, the in-vehicle device 101 is configured to transmit probe information to the server 301, but it is not limited to this. The in-vehicle device 101 may be configured to transmit probe information to a device other than the server 301. In this case, when the processing timing T2 of the estimation process arrives, the server 301 acquires the probe information from the other device.

[0123] Next, other embodiments of the present disclosure will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and their descriptions will not be repeated.

[0124] <Second Embodiment> In the first embodiment of the present disclosure described above, the server 301 performs estimation processing using the passage time information C1. In contrast, in the second embodiment of the present disclosure, the server 301 performs estimation processing using, in addition to the passage time information C1, indication change information C2, which includes the time when the indication of the signal lamp L at the target intersection CS changed from red to blue. Except for the contents described below, it is the same as the signal parameter estimation system 501 according to the first embodiment.

[0125] Referring again to Figure 3, in the server 301, the creation unit 32 creates indication switching information C2 based on probe information from multiple vehicles 1 acquired by the communication unit 31. For example, the indication switching information C2 includes at least one of the time tc when the indication of signal lamp L1 switched from red to blue, and the time td when the indication of signal lamp L2 switched from red to blue.

[0126] More specifically, when the creation unit 32 creates the passage time information C1, it selects multiple probe information (hereinafter also referred to as "multiple probe information Pa") containing the same vehicle ID from among the multiple vehicle 1 probe information retrieved from the storage unit 14.

[0127] Then, the creation unit 32 checks whether vehicle 1, whose vehicle ID is included in the probe information Pa, was stopped on a main road or a secondary road, based on the selected probe information Pa and the map information stored in the storage unit 35.

[0128] The creation unit 32 predicts the departure time ts1 when vehicle 1, whose vehicle ID is included in the probe information Pa, leaves the main road if it was stopped on the main road. Specifically, the creation unit 32 checks whether vehicle 1 passed the target intersection CS at a time later than the time tf when vehicle 1 was stopped on the main road.

[0129] The creation unit 32 identifies time tf as departure time ts1 if vehicle 1 passes through the target intersection CS at a time later than time tf. The creation unit 32 then calculates the distance d1 between the position of vehicle 1 at time tf and the center position of the target intersection CS.

[0130] The creation unit 32 calculates the distance d1, and then calculates the time tc by substituting the specified departure time ts1 and the distance d1 into the following equation (10). In equation (10), Q1 and Q2 are constants. In this embodiment, for example, the constants Q1 and Q2 are 134.06 and 5.189, respectively. tc = ts1 - (Q1 - d1) / Q2 ... (10)

[0131] Furthermore, if vehicle 1, whose vehicle ID is included in the probe information Pa, is stopped on the secondary road, the creation unit 32 predicts the departure time ts2 when vehicle 1 leaves the secondary road. Specifically, the creation unit 32 checks whether vehicle 1 passed the target intersection CS at a time later than the time tg when vehicle 1 was stopped on the secondary road.

[0132] The creation unit 32 identifies time tg as departure time ts2 if vehicle 1 passes through the target intersection CS at a time later than time tg. The creation unit 32 then calculates the distance d2 between the position of vehicle 1 at time tg and the center position of the target intersection CS.

[0133] The creation unit 32 then calculates the time td by substituting the specified time ts and the calculated distance d2 into the following equation (11). In equation (11), Q3 and Q4 are constants. In this embodiment, for example, the constants Q3 and Q4 are 134.06 and 5.189, respectively. td = ts² - (Q3 - d²) / Q4 ... (11)

[0134] The creation unit 32 calculates at least one of the time tc and time td, and then creates indication change information C2 which includes the calculation result and at least one of the following: score S3 indicating that the color of signal lamp L1 has changed from red to blue, and score S4 indicating that the color of signal lamp L2 has changed from red to blue. In the following description, scores S3 and S4 will be collectively referred to as score Sa.

[0135] Figure 13 shows an example of a score included in the indication switching information created by the server according to the second embodiment of this disclosure.

[0136] Referring to Figure 13, the creation unit 32 sets the score Sa at times tc, td, tc-1, tc+1, td-1, and td+1 to "3". Time tc-1 is a time before time tc and is the closest time to time tc. Time tc+1 is a time after time tc and is the closest time to time tc. Time td-1 is a time before time td and is the closest time to time td. Time td+1 is a time after time td and is the closest time to time td.

[0137] Furthermore, the creation unit 32 sets the score Sa at times tc-2, tc+2, td-2, and td+2 to "2". Time tc-2 is a time before time tc and is the second closest time to time tc. Time tc+2 is a time after time tc and is the second closest time to time tc. Time td-2 is a time before time td and is the second closest time to time td. Time td+2 is a time after time td and is the second closest time to time td.

[0138] Furthermore, the creation unit 32 sets the score Sa to "1" at times tc-3, tc+3, td-3, and td+3. Time tc-3 is a time before time tc and is the third closest time to time tc. Time tc+3 is a time after time tc and is the third closest time to time tc. Time td-3 is a time before time td and is the third closest time to time td. Time td+3 is a time after time td and is the third closest time to time td. Hereafter, each of the times shown in Figure 13 will also be referred to as the transition time.

[0139] Figure 14 shows an example of indication switching information created by a server according to the second embodiment of this disclosure.

[0140] Referring to Figure 14, the creation unit 32 creates information including scores S3 and S4 in the created passage time information C1 as indication switching information C2. Specifically, for example, if score S1 is not set at a time corresponding to time tc, time tc-1, or time tc+1 in the passage time information C1, the creation unit 32 sets score S3 to "3". In the example shown in Figure 14, time t19 corresponds to time tc, time t7 corresponds to time tc-1, and times t2 and t9 correspond to time tc+1.

[0141] Furthermore, the creation unit 32 sets a score of "2" as the score S3 if the score S1 is not set at the time corresponding to time tc-2 or time tc+2 in the passage time information C1. In the example shown in Figure 14, time t6 corresponds to time tc-2, and times t3, t10, and t21 correspond to time tc+2.

[0142] Furthermore, the creation unit 32 sets a score of "1" as the score S3 if the score S1 is not set at the time corresponding to time tc-3 or time tc+3 in the passage time information C1. In the example shown in Figure 14, times t5 and t16 correspond to time tc-3, and times t4 and t22 correspond to time tc+3.

[0143] Furthermore, the creation unit 32 sets a score of "3" as the score S4 if the score S2 is not set at the time corresponding to time td, time td-1, or time td+1 in the passage time information C1. In the example shown in Figure 14, times t12, t11, and t13 correspond to time td, time td-1, and time td+1, respectively.

[0144] Furthermore, if the creation unit 32 has not set a score S2 at a time corresponding to time td-2 or time td+2 in the passage time information C1, it sets a score of "2" as the score S4. In the example shown in Figure 14, time t10 corresponds to time td-2.

[0145] Furthermore, the creation unit 32 sets a score of "1" as the score S4 if the score S2 is not set at the time corresponding to time td-3 or time td+3 in the passage time information C1. In the example shown in Figure 14, time t9 and time t15 correspond to time td-3 and time td+3, respectively.

[0146] Furthermore, if a score S1 is set at a time corresponding to any of the switching times in the passing time information C1, the creation unit 32 may set the sum of the score S1 and the score S3 corresponding to that switching time as the score for that time.

[0147] Furthermore, if a score S2 is set at any of the switching times in the passing time information C1, the creation unit 32 may set the sum of the score S2 and the score S4 corresponding to that switching time as the score for that time.

[0148] Referring again to Figure 3, once the creation unit 32 has created the indication switching information C2, it outputs the created indication switching information C2 to the setting unit 33.

[0149] Furthermore, the creation unit 32 is not limited to a configuration in which it creates information representing the change in the indication of signal lights L1 and L2 from red to blue using scores S3 and S4 as indication change information C2. It may also be configured to create information representing the change in the indication of signal lights L1 and L2 from red to blue using other methods such as function approximation and autocorrelation as indication change information C2.

[0150] Figure 15 is a diagram illustrating the server configuration process according to the second embodiment of this disclosure.

[0151] Referring to Figure 15, the setting unit 33 performs a setting process to set the candidate range D1 for cycle length, the candidate range D2 for cycle start time, and the candidate range D3 for split seconds, based on the indication switching information C2 created by the creation unit 32.

[0152] More specifically, for example, the setting unit 33 estimates time zones A1 and A2 based on the indication switching information C2 received from the creation unit 32. Then, based on the estimated time zones A1 and A2, the setting unit 33 predicts the minimum value M1 and maximum value M2 of the cycle length on the main road. Once the setting unit 33 has predicted the minimum value M1 and maximum value M2 of the cycle length, it sets a candidate range D1 of the cycle length using the equations (1) and (2) described above.

[0153] Furthermore, for example, when the setting unit 33 receives indication switching information C2 from the creation unit 32, it sets the remainder obtained by dividing the number of seconds at time tc in the indication switching information C2 by the estimated cycle length Lc as the candidate range D2 for the cycle start time. In the example shown in Figure 15, time tc is time t50, which is between time t36 and time t37.

[0154] Furthermore, for example, when the setting unit 33 receives indication switching information C2 from the creation unit 32, it sets the candidate range D3 of split seconds using the above-described equations (4) and (5).

[0155] Then, once the setting unit 33 has completed the setting process, it outputs the setting range information indicating the set ranges D1, D2, and D3, along with the indication switching information C2 received from the creation unit 32, to the estimation processing unit 34.

[0156] When the estimation processing unit 34 receives setting range information from the setting unit 33, it uses the setting range information to select a set W of candidate cycle length, candidate cycle start time, and candidate split seconds.

[0157] Figure 16 is a diagram illustrating the estimation process by a server according to a second embodiment of the present disclosure.

[0158] Referring to Figure 16, when the estimation processing unit 34 selects set W, it estimates the times tm and tn using the indication switching information C2 included in the setting range information received from the setting unit 33, and the set W.

[0159] In the example shown in Figure 16, the cycle length, cycle start time, and split seconds in the set W selected by the estimation processing unit 34 are assumed to be "8 seconds," "0 seconds," and "5 seconds," respectively. In this case, the estimation processing unit 34 estimates that times t0, t1, t2, t3, t4, t8, t9, t10, t11, t12, t16, t17, t18, t19, t20, and t24 are time tm, and estimates that times t5, t6, t7, t13, t14, t15, t21, t22, and t23 are time tn. In Figure 16, scores corresponding to times when the light color of signal lamp L1 or signal lamp L2 is blue are marked with a diagonal hatch to the right. Scores corresponding to times when the light color of signal lamp L1 or signal lamp L2 is red are marked with a grid-like hatch.

[0160] For example, the estimation processing unit 34 estimates signal parameters based on weighted passage time information C1 and indication change information C2. The weighting value W1 corresponding to the passage time information C1 is greater than the weighting value W2 corresponding to the indication change information C2. In this embodiment, value W1 corresponds to the scores S1 and S2, and value W2 corresponds to the scores S3 and S4.

[0161] When the estimation processing unit 34 estimates the times tm and tn, it evaluates the scores S1, S2, S3, and S4 indicated by the indication switching C2 included in the setting range information received from the setting unit 33.

[0162] More specifically, for example, the estimation processing unit 34 calculates the respective evaluation values ​​of scores S1 and S2 using equations (6) to (9) described above. Furthermore, for example, the estimation processing unit 34 calculates values ​​H31 and H41 by substituting the value N31 of score S3 at time tm into the following equation (12), and substituting the value N41 of score S4 at time tn into the following equation (13). The estimation processing unit 34 then calculates the absolute values ​​of value H31 and value H41 as the evaluation value of score S3 and the absolute value of score S4, respectively. H31 = N31 - 15 ... (12) H32=N41-15 ···(13)

[0163] Furthermore, for example, the estimation processing unit 34 calculates values ​​H33 and H34 by substituting the value N32 of score S3 at time tn into the following equation (14) and substituting the value N42 of score S4 at time tm into the following equation (15). The estimation processing unit 34 then calculates the absolute values ​​of value H33 and value H34 as the evaluation value of score S3 and the absolute value of score S4. H33 = N32 - (-15) ... (14) H34 = N42 - (-15) ... (15)

[0164] Figure 17 shows an example of a score evaluation value calculated in the estimation process performed by the server according to the second embodiment of this disclosure.

[0165] Referring to Figure 17, the estimation processing unit 34 calculates the score evaluation value at each time point and then calculates a total value K2 by summing the multiple evaluation values ​​for the period from time t0 to time t24. In the example shown in Figure 17, the total value K2 is "611" points.

[0166] The estimation processing unit 34 calculates multiple sums K2 corresponding to each of the selected sets W, and outputs to the communication unit 31 the set W corresponding to the smallest sum K2, i.e., the set W that is estimated to have the smallest error with the actual signal parameters, as the result of the estimation process.

[0167] Next, we will explain the relationship between the accuracy of the estimation process using the passage time information C1 and the indication change information C2 from server 301 and the probe rate.

[0168] Figures 18 and 19 show simulation results illustrating the accuracy of signal parameter estimation by a server according to a second embodiment of the present disclosure. Figures 18 and 19 show estimated values ​​of cycle length, cycle start time, and split seconds when the target intersection CS is an intersection controlled by two indications, and the probe rate is in 10% increments within the range of 10% to 90%. The value of the probe rate parameter, i.e., the number of vehicles 1 F1 that entered the target intersection CS during a predetermined period E, is 13,600, similar to the simulation results shown in Figures 9 and 10.

[0169] The correct values ​​for cycle length, cycle start time, and split seconds, corresponding to the simulation results shown in Figure 18, are "60 seconds," "1 second," and "30 seconds," respectively. The correct values ​​for cycle length, cycle start time, and split seconds, corresponding to the simulation results shown in Figure 19, are "60 seconds," "1 second," and "17 seconds," respectively.

[0170] In the examples shown in Figures 18 and 19, the estimated cycle length of signal lamp L1 and the estimated cycle length of signal lamp L2 are "60 seconds," the same as the correct value, at each probe rate. Thus, the server 301 according to the second embodiment can correctly estimate the cycle length even when the probe rate is "10%", i.e., even when the number of units F2 is 1360 units, similar to the server 301 according to the first embodiment, which estimates the cycle length using passing time information C1 without using indication switching information C2.

[0171] Furthermore, in the example shown in Figure 18, the estimated split time of signal lamp L1 is "30 seconds," which is the same as the correct value, at each probe rate. In other words, the server 301 according to the second embodiment can improve the accuracy of the split time estimation compared to the server 301 according to the first embodiment.

[0172] Furthermore, in the example shown in Figure 18, the absolute difference between the estimated cycle start time of signal lamp L1 for each probe rate and the correct value of "1 second" for that cycle start time is 2 seconds.

[0173] Furthermore, in the example shown in Figure 19, similar to the simulation results shown in Figure 10, the absolute error between the estimated cycle start time of signal lamp L2 and the correct value of "1 second" for each probe rate is 2 seconds or less. Also, in the example shown in Figure 19, similar to the simulation results shown in Figure 10, the absolute error between the estimated split seconds of signal lamp L2 and the correct value of "17 seconds" for each probe rate is 3 seconds or less.

[0174] Thus, in the server 301 according to the second embodiment, just like the server 301 according to the first embodiment, even if the probe rate is "10%", that is, even if the number of units F2 is 1360, the cycle start time of signal lamp L1, and the cycle start time and split seconds of signal lamp L2 can be estimated with high accuracy.

[0175] [Operation Flow] Figure 20 is a flowchart illustrating an example of the operation procedure when the server according to the second embodiment of this disclosure performs estimation processing.

[0176] Referring to Figure 20, the process from step ST301 to step ST304 is the same as the process from step ST201 to step ST204 shown in Figure 12.

[0177] Next, after creating the passage time information C1, the server 301 creates the indication change information C2 based on the probe information of multiple vehicles 1 retrieved from the storage unit 35. For example, as described above, the server 301 creates the indication change information C2 by including the scores S3 and S4 in the created passage time information C1 (step ST305).

[0178] Next, the server 301 performs configuration processing based on the created indication switching information C2 (step ST306).

[0179] The processing in steps ST307 and ST308 is the same as the processing in steps ST206 and ST207 shown in Figure 12.

[0180] In the signal parameter estimation system 501 according to the second embodiment of this disclosure, the server 301 is configured to perform estimation processing using signal indication changeover information C2 in addition to the passage time information C1, but it is not limited to this configuration. The server 301 may also be configured to perform estimation processing using the passage time information C1 and other information related to the target intersection CS other than the signal indication changeover information C2.

[0181] Furthermore, in the estimation process by the server 301 according to the second embodiment of this disclosure, the weighting value corresponding to the passage time information C1 is greater than the weighting value corresponding to the indication switching information C2, that is, the values ​​of scores S1 and S2 are greater than the values ​​of scores S3 and S4, but this is not limited to this. The values ​​of scores S1 and S2 and the values ​​of scores S3 and S4 may be the same as each other. Alternatively, the values ​​of scores S3 and S4 may be smaller than the values ​​of scores S1 and S2.

[0182] <Third Embodiment> In the first embodiment of the present disclosure described above, the server 301 performs estimation processing using the passage time information C1. In contrast, in the third embodiment of the present disclosure, the server 301 performs estimation processing using, in addition to the passage time information C1, indication switching information C2 and signal light color information C3 that represents the correspondence between the time and the light color of the signal light L. Except for the contents described below, it is the same as the signal parameter estimation system 501 according to the first embodiment and the signal parameter estimation system 501 according to the second embodiment.

[0183] [In-vehicle device] Figure 21 is a diagram showing an example of the configuration of an in-vehicle device according to a third embodiment of the present disclosure. Referring to Figure 21, the in-vehicle device 102 further includes a light color determination unit 15 compared to the in-vehicle device 101 shown in Figure 2. Some or all of the in-vehicle communication unit 11, creation unit 12, external communication unit 13, and light color determination unit 15 are implemented by a processing circuit including, for example, one or more processors. The storage unit 14 is, for example, a non-volatile memory included in the processing circuit.

[0184] In the example shown in Figure 21, the in-vehicle device 202B is a camera. Hereafter, the in-vehicle device 202B will also be referred to as camera 202B.

[0185] Camera 202B performs an imaging process to capture images of the area in front of vehicle 1. Then, camera 202B transmits the captured image and image information indicating the time of capture to the in-vehicle device 102. Camera 202B performs the imaging process and transmits image information periodically, for example.

[0186] In the in-vehicle device 102, the in-vehicle communication unit 11 stores the received image information in the storage unit 14 each time it receives image information from the camera 202B.

[0187] For example, the in-vehicle device 102 transmits probe information to the server 301, including color determination information that indicates the result of determining the color of the signal lamp L.

[0188] More specifically, the light color determination unit 15 performs a light color determination process to determine the light color of the signal lamp L based on the image information acquired by the in-vehicle communication unit 11.

[0189] Specifically, when the processing timing T1 of the creation process B1 arrives, the light color determination unit 15 retrieves multiple image information stored in the storage unit 14 by the in-vehicle communication unit 11 during the period from the previous processing timing T1 to the current processing timing T1.

[0190] The light color determination unit 15 then checks whether the image information contains a traffic light L for each image information it retrieves. If the image contains a traffic light L, the light color determination unit 15 determines the light color of the traffic light L. The light color determination unit 15 then creates light color determination information that shows the determination result and the time the image was taken.

[0191] The light color determination unit 15 outputs one or more created light color determination information to the creation unit 12. Here, we assume that the light color determination unit 15 outputs multiple light color determination information to the creation unit 12.

[0192] When the creation unit 12 creates probe information, it outputs the probe information, along with multiple color determination information received from the color determination unit 15, to the external communication unit 13.

[0193] The external communication unit 13 transmits a packet P containing probe information received from the creation unit 12 to the server 301 via the wireless base station device and the external network 151.

[0194] [server] Referring again to Figure 3, when the communication unit 31 receives a packet P from each vehicle 1, it stores the probe information contained in the received packet P in the storage unit 14.

[0195] Figure 22 shows an example of signal light color information created by a server according to the third embodiment of this disclosure.

[0196] Referring to Figures 3 and 22, for example, in creation process B2, the creation unit 32 creates signal light color information C3 representing the time and at least one of the light color of signal light L1 and signal light L2, based on the light color determination information included in the probe information acquired by the communication unit 31.

[0197] More specifically, for example, when the processing timing T2 of the creation process B2 arrives, the creation unit 32 retrieves multiple probe pieces of information stored in the storage unit 35 by the communication unit 31 during the period from the previous processing timing T2 to the current processing timing T2.

[0198] The creation unit 32 then checks whether at least one of the extracted probe information contains color determination information (hereinafter also referred to as "color determination information H") indicating the determination result of the color of the signal lamp L at the target intersection CS.

[0199] Specifically, for each piece of extracted probe information, the creation unit 32 checks whether vehicle 1, whose vehicle ID is included in the probe information, has passed through the target intersection CS, based on the probe information and the map information stored in the storage unit 35. If vehicle 1 has passed through the target intersection CS, the creation unit 32 then checks whether the extracted probe information contains light color determination information H, which indicates the light color at the shooting time closest to the time vehicle 1 passed through the target intersection CS.

[0200] If none of the extracted probe information contains the light color determination information H, the creation unit 32 creates the passage time information C1 shown in Figure 4 and the indication switching information C12 shown in Figure 14 based on the extracted probe information. On the other hand, if one of the probe information contains the light color determination information H, the creation unit 32 creates the signal light color information C3.

[0201] More specifically, if the probe information includes light color determination information H, the creation unit 32 checks, based on the probe information and the map information stored in the storage unit 35, whether vehicle 1 with the vehicle ID included in the probe information has passed through the main road or the secondary road of the target intersection CS.

[0202] If vehicle 1, whose vehicle ID is included in the extracted probe information, was passing through the main road, the creation unit 32 creates signal light color information C3, which shows the time of passage tm and a score S5 related to the color of the signal light L1.

[0203] In the example shown in Figure 22, the signal light L1 is blue at the times t2, t3, t4, t9, t10, t18, and t19 when a vehicle 1 passes through the main road. In this case, for example, the creation unit 32 sets the score S5 for each of these times to "30 points".

[0204] Furthermore, in the example shown in Figure 22, the signal light L1 is red at the times t5, t15, and t23 when a certain vehicle 1 passes through the main road. In this case, for example, the creation unit 32 sets the score S5 for each of these times to "-30 points".

[0205] If vehicle 1, whose vehicle ID is included in the extracted probe information, was passing through the secondary road, the creation unit 32 creates signal light color information C3, which shows the time of passage tn and a score S6 related to the color of the signal light L2.

[0206] In the example shown in Figure 22, the signal light L2 is blue at times t5 and t22 when a vehicle 1 passes through the secondary road. In this case, for example, the creation unit 32 sets the score S6 at times t5 and t22 to "30 points".

[0207] The creation unit 32 sets the score S6 for the time when a certain vehicle 1 passes through the secondary road to "-30 points" if the color of the signal light L2 is red at that time.

[0208] Figure 23 shows an example of transit time information created by a server according to the third embodiment of this disclosure.

[0209] Referring to Figures 3 and 23, the creation unit 32 creates signal light color information C3 and then creates passing time information C11. More specifically, for example, the creation unit 32 creates passing time information C11 by including scores S1 and S2 in the created signal light color information C3.

[0210] Specifically, for example, the creation unit 32 sets a score S1 if, in the signal light color information C3, a score has not been set for the time tm when vehicle 1 passed through the main road of the target intersection CS.

[0211] In the example shown in Figure 23, the times t0, t1, t8, t11, t12, t17, and t20 are the same as the passing time information C1 shown in Figure 4, which is the passing time tm. In this case, the score S1 at each of these times is "15".

[0212] Furthermore, for example, the creation unit 32 sets a score S2 if, in the signal light color information C3, a score has not been set for the time tn when vehicle 1 passed through the secondary road of the target intersection CS.

[0213] In the example shown in Figure 23, times t6, t14, t23, and t24 are the same as the passing time information C1 shown in Figure 4, which is the passing time tn. In this case, the score S2 at each of these times is "15".

[0214] Furthermore, the creation unit 32 may be configured to set score S1 at the time corresponding to the passing time tm in the signal light color information C3, even if score S5 is set at that time. In other words, the score at that time may be the sum of score S5 and score S1.

[0215] Furthermore, the creation unit 32 may be configured to set score S2 at the time corresponding to the passing time tn in the signal light color information C3, even if score S6 is already set at that time. In other words, the score at that time may be the sum of score S6 and score S2.

[0216] Figure 24 shows an example of indication switching information created by a server according to the third embodiment of this disclosure.

[0217] Referring to Figures 3 and 24, when the creation unit 32 creates the passage time information C11, it creates the indication change information C12. More specifically, for example, the creation unit 32 creates the indication change information C12 by including the scores S3 and S4 in the created passage time information C11.

[0218] Specifically, for example, the creation unit 32 sets a score of "3" as the score S3 if a score has not been set at the time corresponding to time tc, time tc-1, or time tc+1 in the transit time information C11. In the example shown in Figure 24, time t7 corresponds to time tc-1.

[0219] Furthermore, if a score has not been set at the time corresponding to time tc-2 or time tc+2 in the transit time information C11, the creation unit 32 sets the score S3 to "2". In the example shown in Figure 24, time t6 corresponds to time tc-2.

[0220] Furthermore, if a score has not been set at the time corresponding to time tc-3 or time tc+3 in the passage time information C11, the creation unit 32 sets the score S3 to "1". In the example shown in Figure 24, time t16 and time t22 correspond to time tc-3 and time tc+3, respectively.

[0221] Furthermore, the creation unit 32 sets a score of "3" as the score S4 if no score has been set at the time corresponding to time td, time td-1, or time td+1 in the passage time information C11. In the example shown in Figure 24, times t12, t11, and t13 correspond to time td, time td-1, and time td+1, respectively.

[0222] Furthermore, if a score has not been set at the time corresponding to time td-2 or time td+2 in the transit time information C11, the creation unit 32 sets the score S4 to "2". In the example shown in Figure 24, time t10 corresponds to time td-2.

[0223] Furthermore, if a score has not been set at the time corresponding to time td-3 or time td+3 in the transit time information C1, the creation unit 32 sets the score S4 to "1". In the example shown in Figure 24, time t9 and time t15 correspond to time td-3 and time td+3, respectively.

[0224] When the creation unit 32 creates the indication switching information C12, it outputs the created indication switching information C12 to the setting unit 33.

[0225] Furthermore, the creation unit 32 is not limited to creating information representing the respective light colors of signal lights L1 and L2 using scores S5 and S6 as signal light color information C3. It may also be configured to create information representing the respective light colors of signal lights L1 and L2 using other methods such as function approximation and autocorrelation as signal light color information C3.

[0226] Figure 25 is a diagram illustrating the server configuration process according to the third embodiment of this disclosure.

[0227] Referring to Figure 25, the setting unit 33 performs a setting process to set the candidate range D1 for cycle length, the candidate range D2 for cycle start time, and the candidate range D3 for split seconds, based on the indication switching information C12 created by the creation unit 32.

[0228] More specifically, when the setting unit 33 receives the signal change information C12 from the creation unit 32, it estimates the time zones A1 and A2 based on the signal change information C12. Then, based on the estimated time zones A1 and A2, the setting unit 33 predicts the minimum value M1 and maximum value M2 of the cycle length on the main road.

[0229] The setting unit 33 predicts the minimum and maximum values ​​M1 and M2 of the cycle length on the main road, and then uses equations (1) and (2) described above to calculate the minimum and maximum values ​​M11 and M12 of the candidate cycle length.

[0230] The setting unit 33 then sets the range D1 of the candidate cycle length as a divisor of the value R that falls within the range from the minimum value M11 to the maximum value M12. The value R is expressed by the following equation (16). In equation (16), the value U is greater than or equal to zero and less than or equal to the value Y. The value Y is the value obtained by subtracting 1 from the number of time tc W included in the indication switching information C12. W is, for example, an integer greater than or equal to 2. R = min(tc + 1 - tc, U) ... (16)

[0231] In the example shown in Figure 25, time tc is time t50 between time t36 and time t37, and time t51 between time t41 and time t42, etc.

[0232] Furthermore, for example, when the setting unit 33 receives indication switching information C2 from the creation unit 32, it sets the remainder obtained by dividing the number of seconds at each time tc in the indication switching information C2 by the estimated cycle length Lc as the candidate range D2 for the cycle start time.

[0233] Furthermore, for example, when the setting unit 33 receives indication switching information C2 from the creation unit 32, it sets the candidate range D3 of split seconds using the above-described equations (4) and (5).

[0234] Then, once the setting unit 33 has completed the setting process, it outputs the setting range information indicating the set ranges D1, D2, and D3, along with the indication switching information C12 received from the creation unit 32, to the estimation processing unit 34.

[0235] When the estimation processing unit 34 receives setting range information from the setting unit 33, it uses the setting range information to select a set W of candidate cycle length, candidate cycle start time, and candidate split seconds.

[0236] Figure 26 is a diagram illustrating the estimation process by a server according to a third embodiment of the present disclosure.

[0237] Referring to Figure 26, when the estimation processing unit 34 selects set W, it estimates the times tm and tn using the indication switching information C12 included in the setting range information received from the setting unit 33, and the set W.

[0238] In the example shown in Figure 26, the cycle length, cycle start time, and split seconds in the set W selected by the estimation processing unit 34 are assumed to be "8 seconds," "0 seconds," and "5 seconds," respectively. In this case, the estimation processing unit 34 estimates that time t0, t1, t2, t3, t4, t8, t9, t10, t11, t12, t16, t17, t18, t19, t20, and t24 are time tm, and estimates that time t5, t6, t7, t13, t14, t15, t21, t22, and t23 are time tn.

[0239] When the estimation processing unit 34 estimates the times tm and tn, it evaluates the scores S1, S2, S3, S4, S5, and S6 indicated by the indication switching information C12 included in the setting range information received from the setting unit 33.

[0240] More specifically, for example, the estimation processing unit 34 changes the value of the second term in each of the above-mentioned equations (6), (7), (12), and (13) from "15" to "30". Also, for example, the estimation processing unit 34 changes the value of the second term in each of the above-mentioned equations (8), (9), (14), and (15) from "-15" to "-30". Then, the estimation processing unit 34 uses equations (6) to (9) and equations (12) to (15) to calculate the evaluation value of the corresponding score.

[0241] Furthermore, for example, the estimation processing unit 34 calculates values ​​H51 and H52 by substituting the value N51 of score S6 at time tm into the following equation (17) and substituting the value N61 of score S7 at time tn into the following equation (18). The estimation processing unit 34 then calculates the absolute values ​​of value H51 and value H52 as the evaluation values ​​of score S6 and score S7, respectively. H51 = N51 - 30 ... (17) H52 = N52 - 30 ... (18)

[0242] Furthermore, for example, the estimation processing unit 34 calculates values ​​H53 and H54 by substituting the value N52 of score S6 at time tn into the following equation (19) and the value N62 of score S7 at time tm into the following equation (20). The estimation processing unit 34 then calculates the absolute values ​​of value H53 and value H54 as the evaluation values ​​of score S6 and score S7, respectively. H53 = N52 - (-30) ... (19) H54 = N62 - (-30) ... (20)

[0243] Figure 27 shows an example of a score evaluation value calculated in the estimation process performed by the server according to the third embodiment of this disclosure.

[0244] Referring to Figure 27, the estimation processing unit 34 calculates the evaluation value of the score at each time point, and then calculates a total value K3 by summing the multiple evaluation values ​​for the period from time t0 to time t24. In the example shown in Figure 27, the total value K3 is "1017" points.

[0245] The estimation processing unit 34 calculates multiple sums K3 corresponding to each of the selected sets W, and outputs the set W corresponding to the smallest sum K3, i.e., the set W estimated to have the smallest error with the actual signal parameters, as the result of the estimation process.

[0246] Next, we will explain the accuracy of the estimation process performed by server 301 using passing time information C11, indication change information C12, and signal light color information C3, and the relationship between the probe rate and the inclusion rate K of light color determination information. The inclusion rate K represents the ratio of the number of vehicles F3 that transmitted probe information containing light color determination information out of the total number of vehicles F2 that transmitted probe information.

[0247] Figures 28 and 29 show simulation results illustrating the accuracy of signal parameter estimation by a server according to a third embodiment of the present disclosure. Figures 28 and 29 show estimated values ​​of cycle length, cycle start time, and split seconds when the target intersection CS is an intersection controlled by two indications, the probe rate is in 1% increments from 1% to 5%, and the intrusion rate K is 3%, 5%, or 10%. The value of the probe rate parameter, i.e., the number of vehicles 1 F1 that entered the target intersection CS during a predetermined period E, is 13,600, similar to the simulation results shown in Figures 9 and 10.

[0248] The correct values ​​for cycle length, cycle start time, and split seconds for signal light L1, corresponding to the simulation results shown in Figure 28, are "60 seconds," "1 second," and "30 seconds," respectively. The correct values ​​for cycle length, cycle start time, and split seconds for signal light L2, corresponding to the simulation results shown in Figure 29, are "60 seconds," "1 second," and "17 seconds," respectively.

[0249] In the examples shown in Figures 28 and 29, when the probe rate is 4% or higher, the estimated values ​​of each signal parameter of signal lamp L1 and signal lamp L2 for each value of the contamination rate K are the same as the correct values. Also, when the probe rate is 3% and the contamination rate K is 10%, the estimated values ​​of each signal parameter of signal lamp L1 and signal lamp L2 are the same as the correct values. In other words, the server 301 according to the third embodiment of this disclosure can correctly estimate each signal parameter even when the probe rate is small, compared to the server 301 according to the first embodiment and the server 301 according to the second embodiment.

[0250] [Operation Flow] Figure 30 is a flowchart illustrating an example of the operation procedure when an in-vehicle device according to the third embodiment of this disclosure performs a process to transmit probe information.

[0251] Referring to Figure 30, first, the in-vehicle device 102 receives position information from the GNSS receiver 202A and image information from the camera 202B and stores them in the storage unit 14 (step ST401).

[0252] The in-vehicle device 102 stores the new position information received from the GNSS receiver 202A and the new image information received from the camera 202B in the storage unit 14 until the processing timing T1 of the creation process B1 for creating probe information arrives (NO in step ST402) (step ST401).

[0253] The processing in steps ST403 and ST404 is the same as the processing in steps ST103 and ST104 shown in Figure 11.

[0254] Next, the in-vehicle device 102 retrieves multiple image information stored in the storage unit 14 during the period from the previous processing timing T1 to the current processing timing T1 (step ST405).

[0255] Next, the in-vehicle device 102 checks whether there is image information including an image G including the signal lamp device L among the extracted multiple pieces of image information (step ST406).

[0256] And when there is image information including the image G (YES in step ST406), the in-vehicle device 102 creates lamp color determination information based on the image information (step ST407).

[0257] Next, the in-vehicle device 102 performs creation process B1 to create probe information including a plurality of position information retrieved from the storage unit 14, the created azimuth information and lamp color determination information, and the vehicle ID stored in the storage unit 14 (step ST408).

[0258] The process of step ST409 is the same as the process of step ST107 shown in FIG. 11.

[0259] On the other hand, when there is no image information including the image G (NO in step ST406), the in-vehicle device 102 performs creation process B1 to create probe information including a plurality of position information retrieved from the storage unit 14, the created azimuth information, and the vehicle ID stored in the storage unit 14 (step ST408).

[0260] FIG. 31 is a flowchart defining an example of an operation procedure when the server according to the third embodiment of the present disclosure performs an estimation process.

[0261] Referring to FIG. 31, the processes from step ST501 to step ST503 are the same as the processes from step ST201 to step ST203 shown in FIG. 12.

[0262] Next, the server 301 checks whether at least any one of the multiple pieces of probe information retrieved from the storage unit 35 includes lamp color determination information H indicating the determination result of the lamp color of the signal lamp device L at the target intersection CS (step ST5 $

[0263] Then, when the server 301 determines that the light color determination information H is included in the probe information (YES in step ST504), it creates the traffic signal color information C3 based on the light color determination information H (step ST505).

[0264] Next, when the server 301 creates the traffic signal color information C3, it creates the passing time information C11. For example, as described above, the server 301 creates information including the scores S1 and S2 in the created traffic signal color information C3 as the passing time information C11 (step ST506).

[0265] Next, when the server 301 creates the passing time information C11, it creates the display switching information C12. For example, as described above, the server 301 creates information including the scores S3 and S4 in the created passing time information C11 as the display switching information C12 (step ST507).

[0266] Next, the server 301 performs a setting process based on the created display switching information C12 (step ST508).

[0267] The processes of step ST509 and step ST510 are the same as the processes of step ST206 and step ST207 shown in FIG. 12.

[0268] On the other hand, when the server 301 determines that the light color determination information H is not included in all of the plurality of probe information retrieved from the storage unit 35 (NO in step ST504), it does not create the traffic signal color information C3 and creates the passing time information C1 (step ST506).

[0269] In the signal parameter estimation system 501 according to the third embodiment of the present disclosure, the in-vehicle device 102 is configured to perform the light color determination process. However, the present disclosure is not limited to this configuration. The server 301 may be configured to perform the light color determination process. In this case, the in-vehicle device 102 transmits the probe information including the image information received from the camera 202B to the server 301.

[0270] Furthermore, in the signal parameter estimation system 501 according to the third embodiment of this disclosure, the server 301 is configured to perform estimation processing using indication switching information C12 and signal light color information C3 in addition to the passage time information C11, but it is not limited to this. The server 301 may also be configured to perform estimation processing using the passage time information C11 and signal light color information C3 without using the indication switching information C12.

[0271] Furthermore, in the signal parameter estimation system 501 according to the third embodiment of this disclosure, the server 301 is configured to create signal light color information C3 representing the correspondence between time and the light color of the signal light L in the creation process B2, but it is not limited to this. The signal light color information C3 may also represent the correspondence between time and other information related to the signal light L other than the light color.

[0272] Furthermore, some or all of the functions of the server 301 according to each embodiment of this disclosure may be provided by cloud computing. That is, the server 301 according to each embodiment of this disclosure may be a cloud server composed of multiple servers.

[0273] The embodiments described above should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims rather than the above description, and all modifications within the meaning and scope equivalent to the claims are intended to be included.

[0274] Each process (each function) of the above-described embodiment is implemented by a processing circuit including one or more processors. The processing circuit may consist of one or more memories, various analog circuits, various digital circuits, and other integrated circuits in addition to the one or more processors. The one or more memories store programs (instructions) that cause the one or more processors to execute each of the above processes. The one or more processors may execute each of the above processes according to the programs read from the one or more memories, or they may execute each of the above processes according to logic circuits that have been pre-designed to execute each of the above processes. The processors may be various processors suitable for computer control, such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit). Furthermore, the physically separated multiple processors may cooperate with each other to execute each of the above processes. For example, the processors installed in each of several physically separate computers may cooperate with each other via a network such as a LAN (Local Area Network), WAN (Wide Area Network), and the Internet to perform the above processes. The program may be installed in the memory via the network from an external server device, or it may be distributed on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), and semiconductor memory, and then installed in the memory from the recording medium.

[0275] The above description includes the following features. [Note 1] The steps include creating intersection information, including the time, based on probe information, which includes the location and time of the vehicle, The process includes the step of performing an estimation process to estimate parameters related to the control of traffic signals at an intersection based on the intersection information that has been created, A signal parameter estimation method that, in the step of creating the intersection information, creates the intersection information based on probe information from multiple vehicles. [Explanation of Symbols]

[0276] 1 vehicle 11. In-vehicle communications unit 12,32 Creation Department 13. External communications unit 14,35 Storage part 15 Light color judgment section 31 Communications Department 33 Settings Section 34 Estimation Processing Unit 51 Communications Bus 101,102 On-vehicle equipment 202,202A,202B Automotive equipment 151 External Network 301 Server 401 Traffic control device 501 Signal Parameter Estimation System

Claims

1. The steps include creating intersection information, including the time, based on probe information, which includes the location and time of the vehicle, A signal parameter estimation method comprising the step of performing an estimation process to estimate parameters related to the control of signal lights at an intersection based on the intersection information created.

2. The signal parameter estimation method according to claim 1, wherein the intersection information includes the time, the position of the vehicle, and information representing the behavior of the vehicle at the intersection.

3. The signal parameter estimation method according to claim 1 or claim 2, wherein the intersection information further includes information representing the time and the color of the signal light.

4. The signal parameter estimation method further includes: The step of setting a range of parameter candidates, which are candidates for the parameter, based on the acquired probe information, The signal parameter estimation method according to claim 1 or claim 2, wherein in the step of performing the estimation process, the parameter is determined from among the parameter candidates within the set range.

5. The intersection information includes passage time information, which includes the time the vehicle passed through the intersection, and indication change information, which includes the time the indication of the signal light changed from red to blue. In the step of performing the estimation process, the parameters are estimated based on the weighted passage time information and the indication change information. The signal parameter estimation method according to claim 1 or claim 2, wherein the weighting value corresponding to the passage time information is greater than the weighting value corresponding to the indication switching information.

6. A creation unit that creates intersection information including the time based on probe information including the vehicle's position and time, A signal parameter estimation device comprising an estimation processing unit that performs estimation processing to estimate parameters related to the control of signal lights at an intersection based on the intersection information created by the creation unit.

7. A signal parameter estimation program used in a signal parameter estimation device, Computers, A creation unit that creates intersection information including the time based on probe information including the vehicle's position and time, An estimation processing unit performs estimation processing to estimate parameters related to the control of traffic signals at the intersection based on the intersection information created by the creation unit. A signal parameter estimation program to function as such.

Images