JP6436391B2 - Offset search apparatus, computer program, and offset search method - Google Patents

Description

  The present invention relates to an offset search device for searching for an offset of a signal lamp at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections, a computer program for realizing the offset search device, and the offset search device Relates to the offset search method.

  The purpose of traffic signals is to optimize traffic flow at intersections, thereby preventing traffic congestion and improving traffic smoothness, or improving safety such as preventing traffic accidents. As a means of optimizing traffic flow at intersections, there are increasing expectations for traffic simulators, and various technological developments are being carried out. The traffic simulator includes a micro model suitable for evaluating the improvement of intersections, etc., and a macro model suitable for iterative calculations such as offset optimization, which handles detailed vehicle behavior and evaluates improvements in intersections. is there. A micro-model traffic simulator (micro simulator), as input data, provides traffic information such as traffic volume (eg, OD traffic volume) including vehicle start / end information, vehicle travel speed, acceleration / deceleration characteristics, etc. It is handled. The OD traffic volume is the traffic volume between the starting point (starting point) and the ending point (destination point) of the vehicle. For example, survey statistical data obtained as a result of statistical surveys conducted regularly by the country or local government Etc. are used. On the other hand, a macro model traffic simulator (macro simulator) focuses on optimizing an evaluation function by iterative calculation. For this reason, the input data or the vehicle behavior model is simplified such that the reproduction of the vehicle travel is set to two states of stop and travel, and the target route is a single route.

  Such a traffic simulator includes in advance a vehicle movement model, that is, a calculation formula that simulates the behavior of the vehicle. By applying the above input data to the calculation formula, a road network including a plurality of intersections or A desired vehicle evaluation index is obtained by running a simulated vehicle (vehicle on the simulator) on the route.

  For example, using a travel time of a specific vehicle obtained from vehicle detectors provided at both ends of the section and vehicle sensor data obtained in time series for a specific section, data on the time axis is converted to data on the spatial axis. A method for obtaining the length of traffic jam as a traffic evaluation index by projecting onto the traffic is disclosed (see Patent Document 1).

  In addition, in order to optimize traffic flow at intersections, it is important how to control signal lamps (signals) installed at each intersection, and a traffic simulator is used to find the optimal signal control parameters. Has been. For example, when obtaining an offset as one of the signal control parameters, an evaluation value of a vehicle traveling at a preset speed is calculated, and a situation in which the vehicle stops at an intersection with the calculated evaluation value is avoided as much as possible. In such a case, for example, the offset when the evaluation value is minimum is calculated as the optimum value.

Japanese Patent Laid-Open No. 08-161686

  As described above, the conventional traffic simulator (offset calculation simulator) optimizes the evaluation value for a regulated speed compliant vehicle that runs at a set speed (design speed, for example, regulated speed or a speed slightly higher than the regulated speed). The purpose is to become.

  On the other hand, in a section where roads (links) with a distance of about 200 meters are connected, for example, in an urban area, the adjacent intersections stop as the vehicle speed increases. The width of a pass band (also referred to as a through band or a green wave) that can be continuously passed through a blue signal without being widened is increased. For this reason, even if an optimum offset (for example, an offset pattern for each intersection) can be obtained at a predetermined design speed, a high-speed traveling vehicle in which the obtained offset significantly exceeds the regulation speed (for example, a regulation speed violation) Vehicle) may become an offset that allows the vehicle to travel through the intersections without stopping at the intersection, which may promote a violation of the regulation speed.

  The present invention has been made in view of such circumstances, and an offset search device capable of searching for an optimum offset for suppressing the traveling of a high-speed traveling vehicle without interfering with the traveling of the vehicle complying with the regulated speed. It is an object of the present invention to provide a computer program for realizing the offset search device and an offset search method using the offset search device.

An offset search apparatus according to an embodiment of the present invention is an offset search apparatus that searches for an offset of a signal lamp at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections. A first evaluation value calculation unit that calculates a first evaluation value related to the low-speed simulated vehicle when traveling, and a second evaluation value that calculates a second evaluation value related to the high-speed simulated vehicle when traveling The offset is searched so that the first evaluation value calculated by the evaluation value calculation unit and the first evaluation value calculation unit becomes an optimal value, and the second evaluation value calculated by the second evaluation value calculation unit becomes an optimal value. An offset search unit, a first calculation unit that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven, and the low-speed simulated vehicle and the high-speed simulated vehicle are driven When letting A travel calculation unit that calculates a travel time difference or a travel speed difference, wherein the first evaluation value calculation unit is configured to reduce the low speed based on at least one of the first delay time and the first number of stops calculated by the first calculation unit. The first evaluation value related to the simulated vehicle is calculated, and the second evaluation value calculation unit includes a first evaluation value calculated by the first evaluation value calculation unit and a travel time difference calculated by the travel calculation unit or based on the travel speed difference Ru Citea to calculate a second evaluation value according to the high-speed simulated vehicle.

A computer program according to an embodiment of the present invention is a computer program for causing a computer to run a plurality of simulated vehicles on a simulated road network including a plurality of intersections to search for an offset of a signal light device at the intersection. A first evaluation value calculation unit that calculates a first evaluation value related to the low-speed simulated vehicle when the low-speed simulated vehicle is driven, and a second evaluation related to the high-speed simulated vehicle when the high-speed simulated vehicle is driven A second evaluation value calculation unit that calculates a value; an offset search unit that searches for an offset such that the first evaluation value becomes an optimum value and the second evaluation value becomes an optimum value; and the low-speed simulated vehicle is caused to travel A first calculation unit that calculates at least one of the first delay time and the first number of stops at each intersection, and a trip when the low-speed simulated vehicle and the high-speed simulated vehicle are run To function as a time difference or travel calculating section that calculates a travel speed difference, calculated on the basis of at least one of the first delay time and the first number of stops to calculate the first evaluation value according to the low speed simulated vehicle was calculated it calculates a second evaluation value according to the high-speed simulation vehicle based on the travel time difference or travel speed difference and the first evaluation value and the calculation.

An offset search method according to an embodiment of the present invention is an offset search method by an offset search device that searches for an offset of a signal light device at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections, The first evaluation value calculation unit calculates a first evaluation value related to the low speed simulated vehicle when the low speed simulated vehicle is driven, and a second evaluation related to the high speed simulated vehicle when the high speed simulated vehicle is driven. A step in which the second evaluation value calculation unit calculates a value; a step in which the offset search unit searches for an offset so that the calculated first evaluation value becomes an optimal value and the calculated second evaluation value becomes an optimal value; Calculating at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven; and running the low-speed simulated vehicle and the high-speed simulated vehicle Look including the step of calculating the travel time difference or travel speed difference when is, calculates a first evaluation value according to the low speed simulated vehicle based first delay time calculated and at least one of the first number of stops Then, a second evaluation value relating to the high-speed simulated vehicle is calculated based on the calculated first evaluation value and the calculated travel time difference or travel speed difference.

  ADVANTAGE OF THE INVENTION According to this invention, the optimal offset for suppressing driving | running | working of a high-speed driving | running | working vehicle can be searched, without preventing driving | running | working of a regulation speed compliant vehicle.

It is a schematic diagram which shows an example of a vehicle behavior. It is a schematic diagram which shows an example of the road network in the case of searching for the offset by the traffic simulator of this Embodiment. It is a block diagram which shows an example of a structure of the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the offset in the case of simultaneous offset. It is a schematic diagram which shows an example of the offset in the case of the downstream all intersection moving system (1st system) by the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the offset in the case of the cooperation intersection group movement system (2nd system) by the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the method of setting the offset of a search object intersection when the offset set in the case of the cooperation intersection group movement system (2nd system) by the traffic simulator of this Embodiment cannot be absorbed. It is explanatory drawing which shows an example of the search value of the offset of a cooperation intersection group movement system (2nd system). It is a schematic diagram which shows an example of the method of specifying a cancellation | release intersection when there exist multiple intersections which can cancel the offset set in the case of the cooperation intersection group movement system (2nd system) by the traffic simulator of this Embodiment. is there. It is a schematic diagram which shows an example of the offset in the case of 1 intersection movement system. It is a schematic diagram which shows an example of offset inversion. It is explanatory drawing which shows an example of the parameter at the time of setting the tolerance | permissible_range of the offset by the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the restriction | limiting in the case of adjusting an offset to a green signal start. It is a schematic diagram which shows an example of the restriction | limiting in the case of match | combining an offset with the end of a green signal. It is a schematic diagram which shows an example of the restriction | limiting in the case of matching offset to a starting wave. It is explanatory drawing which shows an example of the tolerance | permissible_range of the offset by the traffic simulator of this Embodiment. It is explanatory drawing which shows an example of calculation of the delay time by the traffic simulator of this Embodiment. It is explanatory drawing which shows the concept of the danger level of a vehicle. It is explanatory drawing which shows an example of weighting of the danger level of a vehicle. It is explanatory drawing which shows the 1st Example of the evaluation value which the traffic simulator of this Embodiment calculates. It is a schematic diagram which shows an example of the search of the offset by the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the search of the offset by the traffic simulator of this Embodiment. It is explanatory drawing which shows the 2nd Example of the evaluation value which the traffic simulator of this Embodiment calculates. It is explanatory drawing which shows the travel time difference at the time of setting conditions with a travel speed difference. It is explanatory drawing which shows the travel speed difference at the time of setting conditions by travel time difference. It is a schematic diagram which shows an example of a road network. It is a schematic diagram which shows an example of the selection classification of the search object intersection by the traffic simulator of this Embodiment. It is explanatory drawing which shows the combination of the intersection movement rule at the time of an offset search. It is a schematic diagram which shows an example of the result of the search of the offset by the traffic simulator of this Embodiment. It is a schematic diagram which shows an example of the result of the search of the offset by the traffic simulator of this Embodiment. It is a flowchart which shows an example of the process sequence of the offset search by the traffic simulator of this Embodiment. It is a flowchart which shows an example of the process sequence of the offset search by the traffic simulator of this Embodiment. It is a flowchart which shows an example of the process sequence of the offset search by the traffic simulator of this Embodiment. It is explanatory drawing which shows the comparison with the 1st system regarding the optimization of 1st evaluation value E1, and a 2nd system. It is explanatory drawing which shows the simulation result at the time of making the selection order of a search object intersection into 1 type. It is explanatory drawing which shows the simulation result at the time of making the selection order of a search object intersection into 1 type. It is explanatory drawing which shows the simulation result when the selection order of a search object intersection is made into four types. It is explanatory drawing which shows the simulation result when the selection order of a search object intersection is made into four types. It is explanatory drawing which shows the simulation result in the case of performing an offset search by simultaneously optimizing the 1st evaluation value and the 2nd evaluation value. It is explanatory drawing which shows the concept of the optimization search of an evaluation value. It is explanatory drawing which shows the simulation result at the time of making optimization calculation 400 times. It is explanatory drawing which shows the simulation result at the time of making optimization calculation 400 times. It is explanatory drawing which shows the comparison of the discovery condition of the pattern of a high speed suppression offset.

[Description of Embodiment of Present Invention]
An offset search apparatus according to an embodiment of the present invention is an offset search apparatus that searches for an offset of a signal lamp at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections. A first evaluation value calculation unit that calculates a first evaluation value related to the low-speed simulated vehicle when traveling, and a second evaluation value that calculates a second evaluation value related to the high-speed simulated vehicle when traveling The offset is searched so that the first evaluation value calculated by the evaluation value calculation unit and the first evaluation value calculation unit becomes an optimal value, and the second evaluation value calculated by the second evaluation value calculation unit becomes an optimal value. An offset search unit.

  A computer program according to an embodiment of the present invention is a computer program for causing a computer to run a plurality of simulated vehicles on a simulated road network including a plurality of intersections to search for an offset of a signal light device at the intersection. A first evaluation value calculation unit that calculates a first evaluation value related to the low-speed simulated vehicle when the low-speed simulated vehicle is driven, and a second evaluation related to the high-speed simulated vehicle when the high-speed simulated vehicle is driven A second evaluation value calculation unit that calculates a value; and an offset search unit that searches for an offset so that the first evaluation value becomes an optimum value and the second evaluation value becomes an optimum value.

  An offset search method according to an embodiment of the present invention is an offset search method by an offset search device that searches for an offset of a signal light device at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections, The first evaluation value calculation unit calculates a first evaluation value related to the low speed simulated vehicle when the low speed simulated vehicle is driven, and a second evaluation related to the high speed simulated vehicle when the high speed simulated vehicle is driven. A step in which the second evaluation value calculation unit calculates a value; a step in which the offset search unit searches for an offset so that the calculated first evaluation value becomes an optimal value and the calculated second evaluation value becomes an optimal value; including.

  The first evaluation value calculation unit calculates a first evaluation value related to the low speed simulation vehicle when the low speed simulation vehicle is run. The second evaluation value calculation unit calculates a second evaluation value related to the high-speed simulated vehicle when the high-speed simulated vehicle is run. The low-speed simulated vehicle travels at a practically safe traveling speed such as a regulation speed or a speed slightly higher than the regulation speed (for example, the regulation speed is about +5 km / h and is also referred to as the vicinity of the regulation speed). It is a vehicle that complies with the regulated speed. The simulated vehicle is also simply referred to as a vehicle. The high-speed simulated vehicle is, for example, a high-speed traveling vehicle (also referred to as a regulation speed violation vehicle) that travels at a dangerous traveling speed such as a violation speed (for example, a speed exceeding the regulation speed + 20 km / h) that greatly exceeds the regulation speed. .

  The offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit becomes an optimal value and the second evaluation value calculated by the second evaluation value calculation unit becomes an optimal value. The optimum value of the first evaluation value is, for example, a value that does not hinder the traveling of the low speed simulation vehicle and does not decrease the travel speed. The optimum value of the second evaluation value is a value that suppresses traveling of the high-speed simulated vehicle and decreases the travel speed, for example. The offset refers to, for example, the time difference between the start of green signals of the signal lamps at adjacent intersections.

  The search for the offset can be performed as follows, for example. The initial value of the offset of each intersection in the simulated road network is set in advance. In addition, a selection order of intersections to be searched when searching for an offset is also determined in advance. In addition, a change value N (for example, a ratio with respect to a cycle length such as 2%, 3%, 15%, and 40%, a time such as 1 second) at the time of changing the offset of the search target intersection is set. When the search target intersection offset is not changed according to the selection order of the search target intersection, when + N or −N, evaluation values (first evaluation value and second evaluation value) are calculated and the calculated evaluation The offset is further changed to a value closer to the optimum value (for example, + N or -N), and the evaluation value is repeatedly calculated until the evaluation value becomes the optimum value to search for the offset.

  The offset search unit registers (stores) the offset based on the result searched for by the offset search unit. That is, the offset search unit registers the offset when the evaluation value becomes the optimum value as an offset that can be adopted. The registered offset is, for example, an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle, and an offset pattern at each intersection that decreases the travel speed of the high-speed simulated vehicle. The travel speed is a value obtained by dividing the target section by the travel time required to move the section. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  An offset search device according to an embodiment of the present invention includes a first calculation unit that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven, and the high-speed A second calculation unit that calculates at least one of a second delay time and a second number of stops at each intersection when the simulated vehicle is driven, wherein the first evaluation value calculation unit is the first calculation unit A first evaluation value related to the low speed simulation vehicle is calculated based on at least one of the calculated first delay time and the first number of stops, and the second evaluation value calculation unit is the second calculation unit. A second evaluation value relating to the high-speed simulated vehicle is calculated based on at least one of the calculated second delay time and the second number of stops.

  The first calculation unit calculates at least one of the first delay time and the first number of stops at each intersection when the low-speed simulated vehicle is driven. The first delay time is a delay time of the low speed simulation vehicle stopped at the intersection. The delay time is the total delay time at each intersection or a value per intersection obtained by dividing this total value by the number of intersections. If there are multiple low-speed simulated vehicles that have stopped at each intersection, This is the total delay time of multiple low-speed simulated vehicles at the intersection. Further, the first stop count is the number of low-speed simulated vehicles stopped at the intersection. The number of stops is the total number of stops at each intersection or the value per intersection obtained by dividing this total by the number of intersections. If there are multiple stops at each intersection, Is the sum of the number of stops.

  The first evaluation value calculation unit calculates a first evaluation value related to the low speed simulated vehicle based on at least one of the first delay time and the first number of stops calculated by the first calculation unit. When the first delay time is represented by D1 and the first stop count is represented by R1, the first evaluation value E1 can be calculated by an equation of E1 = a1 × D1 + b1 × R1. Further, the first evaluation value E1 may be calculated by an equation of E1 = a1 × D1 or E1 = b1 × R1. a1 and b1 are predetermined coefficients. Note that when calculating the first evaluation value E1, a traffic coefficient such as a balance coefficient or a risk level described later can be added.

  The second calculation unit calculates at least one of the second delay time and the second number of stops at each intersection when the high-speed simulated vehicle is run. The second delay time is a delay time of the high-speed simulated vehicle stopped at the intersection. The delay time is the total delay time at each intersection or the value per intersection obtained by dividing this total value by the number of intersections. If there are multiple high-speed simulated vehicles stopped at each intersection, This is the sum of delay times of multiple high-speed simulated vehicles at the intersection. The second stop count is the number of high-speed simulated vehicles stopped at the intersection. The number of stops is the total number of stops at each intersection or the value per intersection obtained by dividing this total by the number of intersections. If there are multiple stops at each intersection, Is the sum of the number of stops.

  The second evaluation value calculation unit calculates a second evaluation value related to the high-speed simulated vehicle based on at least one of the second delay time and the second number of stops calculated by the second calculation unit. When the second delay time is represented by D2 and the second stop count is represented by R2, the first evaluation value E2 can be calculated by an equation of E2 = a2 × D2 + b2 × R2. Further, the second evaluation value E2 may be calculated by an equation of E2 = a2 × D2 or E2 = b2 × R2. a2 and b2 are predetermined coefficients. Note that when calculating the second evaluation value E2, a traffic coefficient such as a balance coefficient or a risk level described later can be added.

  The offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit becomes an optimal value and the second evaluation value calculated by the second evaluation value calculation unit becomes an optimal value. The optimum value of the first evaluation value is, for example, a value that does not hinder the traveling of the low speed simulation vehicle and does not decrease the travel speed. The optimum value of the second evaluation value is a value that suppresses traveling of the high-speed simulated vehicle and decreases the travel speed, for example.

  The offset search unit registers (stores) the offset based on the result searched for by the offset search unit. That is, the offset search unit registers the offset when the evaluation value becomes the optimum value as an offset that can be adopted. The registered offset is, for example, an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle, and an offset pattern at each intersection that decreases the travel speed of the high-speed simulated vehicle. The travel speed is a value obtained by dividing the target section by the travel time required to move the section. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  The offset search device according to an embodiment of the present invention includes a first calculation unit that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven, and the low-speed simulation device. A travel calculation unit that calculates a travel time difference or a travel speed difference when the simulated vehicle and the high-speed simulated vehicle are run, and the first evaluation value calculation unit includes a first delay time calculated by the first calculation unit and The first evaluation value related to the low speed simulated vehicle is calculated based on at least one of the first number of stops, and the second evaluation value calculation unit calculates the first evaluation calculated by the first evaluation value calculation unit. The second evaluation value relating to the high-speed simulated vehicle is calculated based on the value and the travel time difference or travel speed difference calculated by the travel calculation unit.

  The first calculation unit calculates at least one of the first delay time and the first number of stops at each intersection when the low-speed simulated vehicle is driven. The first delay time is a delay time of the low speed simulation vehicle stopped at the intersection. The delay time is the total delay time at each intersection or a value per intersection obtained by dividing this total value by the number of intersections. If there are multiple low-speed simulated vehicles that have stopped at each intersection, This is the total delay time of multiple low-speed simulated vehicles at the intersection. Further, the first stop count is the number of low-speed simulated vehicles stopped at the intersection. The number of stops is the total number of stops at each intersection or the value per intersection obtained by dividing this total by the number of intersections. If there are multiple stops at each intersection, Is the sum of the number of stops.

  The first evaluation value calculation unit calculates a first evaluation value related to the low speed simulated vehicle based on at least one of the first delay time and the first number of stops calculated by the first calculation unit. When the first delay time is represented by D1 and the first stop count is represented by R1, the first evaluation value E1 can be calculated by an equation of E1 = a1 × D1 + b1 × R1. Further, the first evaluation value E1 may be calculated by an equation of E1 = a1 × D1 or E1 = b1 × R1. a1 and b1 are predetermined coefficients. Note that when calculating the first evaluation value E1, a traffic coefficient such as a balance coefficient or a risk level described later can be added.

  The travel calculation unit calculates a travel time difference or a travel speed difference when the low-speed simulated vehicle and the high-speed simulated vehicle are run. The travel time difference is the difference between the travel time of the high-speed simulated vehicle and the travel time of the low-speed simulated vehicle. The travel speed difference is the difference between the travel speed of the high-speed simulated vehicle and the travel speed of the low-speed simulated vehicle. The travel speed is a value obtained by dividing the target section by the travel time required to move the section.

  The second evaluation value calculation unit calculates a second evaluation value E2 related to the high-speed simulation vehicle based on the first evaluation value E1 calculated by the first evaluation value calculation unit and the travel time difference or the travel speed difference calculated by the travel calculation unit. To do. When the travel time difference is used, the second evaluation value E2 can be calculated by, for example, an equation of E2 = M × E1. Here, for example, M can be a variable that increases in value as the travel time difference increases. Further, when the travel speed difference is used, M can be a variable that increases in value as the travel speed difference increases. Further, as the travel time difference, the larger one of the travel time difference in the up direction and the travel time difference in the down direction, an average value, a mean square, or the like may be used. The same applies to the travel speed difference.

  The offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit becomes an optimal value and the second evaluation value calculated by the second evaluation value calculation unit becomes an optimal value. The optimum value of the first evaluation value is, for example, a value that does not hinder the traveling of the low speed simulation vehicle and does not decrease the travel speed. In addition, since the travel time difference or travel speed difference between the high-speed simulated vehicle and the low-speed simulated vehicle is used as the second evaluation value, for example, the travel time difference or travel speed between the high-speed simulated vehicle and the low-speed simulated vehicle is reduced. Appears in the second evaluation value, it can be easily determined whether or not the search for the offset is approaching the target value. The second evaluation value is a value that suppresses the traveling of the high-speed simulated vehicle and approaches the travel speed or travel time of the low-speed simulated vehicle. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  In the offset search device according to the embodiment of the present invention, the second evaluation value calculation unit, when the travel time difference or the travel speed difference is greater than or equal to a predetermined threshold, the travel time difference or the travel speed difference is greater than the predetermined threshold. The second evaluation value is calculated so as to be a large value as compared with the case where it is small.

  The second evaluation value calculation unit calculates the second evaluation value so that when the travel time difference or the travel speed difference is greater than or equal to a predetermined threshold, the second evaluation value is larger than when the travel time difference or the travel speed difference is smaller than the predetermined threshold. calculate. When the travel time difference or travel speed difference between the high speed simulation vehicle and the low speed simulation vehicle is greater than or equal to a predetermined threshold value, the second evaluation value is increased, and the travel time difference or travel speed difference between the high speed simulation vehicle and the low speed simulation vehicle is the predetermined threshold value. If it is smaller, the second evaluation value is made smaller. Therefore, the second evaluation value shows the degree of optimization for reducing the travel time difference or the travel speed between the high-speed simulation vehicle and the low-speed simulation vehicle. It is possible to easily determine whether is approaching the target value.

  In the offset search device according to the embodiment of the present invention, the second evaluation value calculation unit is larger as the travel time difference calculated by the travel calculation unit is larger than the first evaluation value calculated by the first evaluation value calculation unit. The second evaluation value is calculated by multiplying the coefficient.

  The second evaluation value E2 can be calculated by, for example, an equation of E2 = M × E1. Here, for example, M can be a coefficient (variable) that increases as the travel time difference increases. Since the travel time difference between the high speed simulation vehicle and the low speed simulation vehicle is used as the second evaluation value, for example, the degree of optimization that reduces the travel time difference between the high speed simulation vehicle and the low speed simulation vehicle appears in the second evaluation value. It can be easily determined whether or not the search for the offset is approaching the target value. The second evaluation value is a value that suppresses the traveling of the high-speed simulated vehicle and approaches the travel time of the low-speed simulated vehicle. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  In the offset search device according to an embodiment of the present invention, the offset search unit searches for an offset such that a first evaluation value calculated by the first evaluation value calculation unit is small; and A second offset search unit for searching for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is increased; Is provided.

  The offset search unit includes a first offset search unit and a second offset search unit. The first offset search unit searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is small. For example, when the first evaluation value E1 is calculated by the equation E1 = a1 × D1 + b1 × R1, in order to decrease the first evaluation value E1, the first delay time D1 is decreased or the first stop count R1 Should be reduced. Since reducing the first delay time D1 or the first stop count R1 does not reduce (or suppress) the travel speed of the low-speed simulated vehicle, the offset is searched so that the first evaluation value becomes small. To search for an offset that does not reduce the travel speed of the low-speed simulated vehicle. Note that searching for an offset so that the first evaluation value becomes smaller means that when using the reciprocal of the first evaluation value E1 or an evaluation value proportional to the reciprocal, the offset is increased so that the evaluation value becomes larger. Equivalent to searching.

  The second offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is large. . The first evaluation value being within a predetermined range can be, for example, within a range in which the first evaluation value is not deteriorated. More specifically, the first evaluation value is a temporary value in the vicinity of the optimum value or the optimum value. It can be within a range that does not become larger than the optimum value.

  When the second evaluation value E2 is calculated by the equation E2 = a2 × D2, in order to increase the second evaluation value E2, the second delay time D2 may be increased. Since increasing the second delay time D2 decreases the travel speed of the high-speed simulated vehicle, searching for an offset so as to increase the second evaluation value decreases the travel speed of the high-speed simulated vehicle. The offset will be searched. Note that searching for an offset so that the second evaluation value becomes large is that when the reciprocal of the second evaluation value E2 or an evaluation value proportional to the reciprocal is used, the offset is set so that the evaluation value becomes small. Equivalent to searching.

  With the above-described configuration, the first offset search unit can first search for an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle. The second offset search unit searches for an offset pattern at each intersection that reduces the travel speed of the high-speed simulated vehicle within a range in which the travel speed of the low-speed simulated vehicle does not decrease (or is suppressed). be able to.

  In the offset search device according to an embodiment of the present invention, the offset search unit searches for an offset such that a first evaluation value calculated by the first evaluation value calculation unit is small; and A second offset search unit for searching for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is small; Is provided.

  The offset search unit includes a first offset search unit and a second offset search unit. The first offset search unit searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is small. For example, when the first evaluation value E1 is calculated by the equation E1 = a1 × D1 + b1 × R1, in order to decrease the first evaluation value E1, the first delay time D1 is decreased or the first stop count R1 Should be reduced. Since reducing the first delay time D1 or the first stop count R1 does not reduce (or suppress) the travel speed of the low-speed simulated vehicle, the offset is searched so that the first evaluation value becomes small. To search for an offset that does not reduce the travel speed of the low-speed simulated vehicle. Note that searching for an offset so that the first evaluation value becomes smaller means that when using the reciprocal of the first evaluation value E1 or an evaluation value proportional to the reciprocal, the offset is increased so that the evaluation value becomes larger. Equivalent to searching.

  The second offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is small. . The first evaluation value being within a predetermined range can be, for example, within a range in which the first evaluation value is not deteriorated. More specifically, the first evaluation value is a temporary value in the vicinity of the optimum value or the optimum value. It can be within a range that does not become larger than the optimum value.

  The second evaluation value E2 can be calculated by, for example, an equation of E2 = M × E1. Here, for example, M can be a coefficient (variable) that increases as the travel time difference increases. In order to decrease the second evaluation value E2, the first evaluation value E1 may be decreased or the coefficient M may be decreased. Focusing on the coefficient M, searching for an offset so that the second evaluation value E2 is small means searching for an offset that reduces the travel time difference between the high-speed simulated vehicle and the low-speed simulated vehicle.

  With the above-described configuration, since the degree of optimization for reducing the travel time difference between the high-speed simulated vehicle and the low-speed simulated vehicle appears in the second evaluation value, it is easy to determine whether the search for the offset is approaching the target value. Can do. The second evaluation value is a value that suppresses the traveling of the high-speed simulated vehicle and approaches the travel time of the low-speed simulated vehicle. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  In the offset search device according to the embodiment of the present invention, the first offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value. In the second offset search unit, the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value, and the second evaluation value calculated by the second evaluation value calculation unit becomes a maximum value. Thus, the offset is searched.

  The first offset search unit searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle. The second offset search unit calculates the second evaluation value because the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value of the first evaluation value obtained as a result of searching by the first offset search unit. The offset is searched so that the second evaluation value calculated by the unit becomes a maximum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle and that decreases the travel speed of the high-speed simulated vehicle. That is, it is possible to obtain an offset pattern at each intersection that satisfies both the travel speed of the low-speed simulated vehicle and the travel speed of the high-speed simulated vehicle.

  In the offset search device according to the embodiment of the present invention, the first offset search unit searches for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value. In the second offset search unit, the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value, and the second evaluation value calculated by the second evaluation value calculation unit is a minimum value. Thus, the offset is searched.

  The first offset search unit searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle. The second offset search unit calculates the second evaluation value because the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value of the first evaluation value obtained as a result of searching by the first offset search unit. The offset is searched so that the second evaluation value calculated by the unit becomes a minimum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the low-speed simulated vehicle and that decreases the travel speed of the high-speed simulated vehicle. That is, it is possible to obtain an offset pattern at each intersection that satisfies both the travel speed of the low-speed simulated vehicle and the travel speed of the high-speed simulated vehicle.

  An offset search device according to an embodiment of the present invention includes: a selection unit that selects a search target intersection when searching for an offset from intersections of the simulated road network in a predetermined order; and a search selected by the selection unit The first setting unit that repeatedly sets only the offset of the signal lamp at the target intersection to a plurality of predetermined values, and the offset of the signal lamp at the search target intersection selected by the selection unit is set to a predetermined value, the set predetermined value A specifying unit for specifying a canceling intersection that can set an offset value for canceling as an offset, and repeatedly setting the offset of the signal lamp at the search target intersection selected by the selecting unit to a plurality of predetermined values, A second setting unit that repeatedly sets the offset of the signal lamp at the cancellation intersection specified by the specifying unit as the cancellation value, and the first offset search unit and the second offset Set search unit is adapted to be searched offset based on a plurality of predetermined values repeatedly set at least one of the first setting portion and the second setting unit.

  A selection part selects the search object intersection at the time of searching an offset from the intersection of a simulation road network in a predetermined order. When the number of intersections in the simulated road network is I, the intersection number M is 1 to I. Since the offset is the time difference between the intersections at the start of the green signal, the intersection number M is 2 to I when the intersection with the intersection number M = 1 is used as a reference. For example, the selection unit selects the intersections to be searched in the order of the intersection number M of 2, 3,.

  The first setting unit repeatedly sets only the offset of the signal lamp at the intersection to be searched selected by the selection unit to a plurality of predetermined values. The plurality of predetermined values are, for example, change values N (for example, ratios to cycle lengths such as 2%, 3%, 15%, and 40%, times such as 1 second) when changing the offset of the search target intersection. . Since the first setting unit sets only the offset of the signal lamp at the search target intersection to a predetermined value, the offset of another intersection group (for example, the downstream intersection group) selected after the search target intersection is not changed. For example, if the offset of the signal lamp at the selected search target intersection is set to + N, only the offset of the signal lamp at the search target intersection will be changed. As a result, one upstream side of the search target intersection The offset of the entire intersection to be searched is set to + N with respect to the (previous) intersection. The setting method by the first setting unit is also referred to as a downstream all intersection moving method or a first method for convenience.

  When the offset of the signal lamp at the intersection to be searched selected by the selection unit is set to a predetermined value, the specifying unit specifies a cancellation intersection that can set an offset value for canceling the set predetermined value as an offset. . For example, when the offset of the signal lamp at the selected intersection to be searched is set to + N, an intersection that can be set as an offset is an offset value for canceling + N (for example, -N with respect to + N). Identifies as an intersection. The specifying unit specifies a canceling intersection every time a search target intersection is selected.

  The second setting unit repeatedly sets the offset of the signal lamp at the intersection to be searched selected by the selection unit to a plurality of predetermined values, and repeatedly sets the offset of the signal lamp at the cancellation intersection specified by the specifying unit as a cancellation value. For example, when the offset of the signal lamp at the selected search target intersection is set to + N, a cancellation intersection that can set an offset value (for example, −N) that cancels + N as an offset is specified. That is, when the offset of the signal lamp at the selected intersection to be searched is set to + N, the offset of the signal lamp at the cancellation intersection is set to -N, and the intersection at one upstream side (one before) of the intersection to be searched is set. On the other hand, the offset of the signal lamp of the intersection group between the search target intersection and the cancellation intersection is set to + N. The setting method by the second setting unit is also referred to as a cooperative intersection group movement method or a second method for convenience.

  The first offset search unit and the second offset search unit search for an offset based on a plurality of predetermined values that are repeatedly set in at least one of the first setting unit and the second setting unit. That is, both the first offset search unit and the second offset search unit may use the first method, and both the first offset search unit and the second offset search unit may use the second method, The first offset search unit may use the first method and the second offset search unit may use the second method, or the first offset search unit uses the second method and the second offset search unit uses the first method. May be used.

  With the above-described configuration, either the first method or the second method can be used for the search for the offset when the first evaluation value is optimized, and the search for the offset when the second evaluation value is optimized. Since either the first method or the second method can be used, the offset search method can be changed to various methods, and the evaluation value is the optimum value as compared with the case where the predetermined offset value is simply changed. The search range for the offset can be expanded.

  In the offset search apparatus according to the embodiment of the present invention, the first offset search unit searches for an offset based on a plurality of predetermined values repeatedly set by one of the first setting unit and the second setting unit. The second offset search unit searches for an offset based on a plurality of predetermined values repeatedly set by the other of the first setting unit and the second setting unit.

  The first offset search unit searches for an offset based on a plurality of predetermined values repeatedly set by one of the first setting unit and the second setting unit, and the second offset search unit includes the first setting unit and the second setting unit. The offset is searched based on a plurality of predetermined values repeatedly set on the other side. For example, the first offset search unit may use the first method (one) and the second offset search unit may use the second method (the other), or the first offset search unit may use the second method and The two-offset search unit may use the first method. With the above-described configuration, the offset search method for optimizing the first evaluation value and the offset search method for optimizing the second evaluation value can be made different. Compared to using the same search method when optimizing the evaluation value, the possibility of finding the optimal value of the evaluation value can be increased and the possibility of finding a new offset pattern can be increased. it can.

  In the offset search apparatus according to the embodiment of the present invention, the first offset search unit searches for an offset based on a plurality of predetermined values repeatedly set by the first setting unit, and the second offset The search unit searches for an offset based on a plurality of predetermined values repeatedly set by the second setting unit.

  The first offset search unit searches for an offset based on a plurality of predetermined values repeatedly set by the first setting unit, and the second offset search unit uses the offset value searched by the first setting unit as an initial value. 2. Search for an offset based on a plurality of predetermined values repeatedly set by the setting unit. That is, the first offset search unit uses the first method, and the second offset search unit searches for the offset using the second method. By searching for the offset using the second method, that is, the cooperative intersection group moving method, there are the following advantages. That is, there is no restriction for preventing the offset inversion because the offset inversion that occurs in the short-distance link does not occur. In addition, when the restrictions on preventing offset inversion are ignored, it is necessary to move the offset other than searching for the optimum value in order to correct the deviation, but there is no restriction on preventing offset inversion. There is also no offset movement other than the optimum value search. When the second offset search unit searches for the offset using the second method, it is possible to move the offset in cooperation with only some of the intersection groups among all the intersections. Thereby, it is possible to increase the possibility of finding an offset pattern that stops the high-speed simulated vehicle at the intersection without reducing the travel speed of the low-speed simulated vehicle.

  In the offset search device according to the embodiment of the present invention, when the second setting unit cannot specify the canceling intersection at the specifying unit, the offset of the signal lamp at the search target intersection selected by the selection unit, The offset value is set within the range of the absolute value of the offset value at the intersection where the offset value with the largest offset amount can be set.

  When the specifying unit cannot specify the canceling intersection, the second setting unit sets the offset of the signal lamp at the intersection to be searched selected by the selecting unit at the intersection at which the canceling value having the maximum canceling amount can be set. Set within the absolute value range of the offset value. For example, when the second offset search unit searches for an offset using the second method, when the offset of the search target intersection is to be changed by +20 seconds, an offset that offsets (absorbs) all of +20 seconds (ie, −20 seconds) When it is not possible to specify a canceling intersection that can be set, the offset offset value of the intersection with the largest offset amount is, for example, −15 seconds. Instead of +20 seconds, the offset is set to +15 seconds so that it is within the range of 15 which is the absolute value of the cancellation value (−15 seconds). Before the search for the offset, the amount of change in the offset of the search target intersection is set according to the cancellation that can be canceled at the offset intersection, thereby canceling the amount of change in the offset of the search target intersection during the offset search. When it is discovered that it cannot be performed, it is possible to prevent the situation where the offset of the search target intersection is corrected and the offset search process is repeated from the beginning, thereby improving the processing capability.

  In the offset searching apparatus according to the embodiment of the present invention, when the specifying unit can specify a plurality of canceling intersections, an offset intersection that is closest to the search target intersection among the plurality of canceling intersections is canceled. As specified.

  When the specifying unit can specify a plurality of cancellation intersections, the specification unit specifies an intersection closest to the search target intersection as a cancellation intersection among the plurality of cancellation intersections. Thereby, when the offset of the search target intersection is set to a predetermined value, the number of intersections where the same predetermined value offset is set in cooperation with the offset of the search target intersection can be minimized. It is possible to increase the possibility of finding an offset that stops the high-speed simulated vehicle by adjusting the offset of only the intersection where the vehicle is not stopped.

  In the offset search device according to the embodiment of the present invention, the selection unit has a plurality of types of the predetermined order.

  The selection unit has a plurality of types of predetermined orders. By having a plurality of types of selection order of search target intersections, it is possible to widen the search range of offsets for finding a required offset pattern.

  In the offset searching apparatus according to the embodiment of the present invention, the selection unit selects a first selection order from the upstream side to the downstream side of the intersection of the simulated road network, and the downstream side to the upstream side of the intersection of the simulated road network. A second selection order for selecting to, a third selection order for selecting in ascending order of distance between intersections of the simulated road network, a fourth selection order for selecting in order of increasing distance between intersections of the simulated road network, and the simulated road A search target intersection is selected by using a plurality of selection orders as the predetermined order among the fifth selection orders for randomly selecting the intersections of the network.

  The selection unit selects a first selection order for selecting from the upstream side to the downstream side of the intersection of the simulated road network, a second selection order for selecting from the downstream side to the upstream side of the simulated road network intersection, and the distance between the intersections of the simulated road network A plurality of selection orders among a third selection order for selecting in order of short distance, a fourth selection order for selecting in order of distance between intersections of the simulated road network, and a fifth selection order for randomly selecting intersections in the simulated road network Are selected as a predetermined order. Thereby, the search range of the offset is expanded, and for example, it is possible to increase the possibility of newly finding a required offset pattern that could not be found in one selection order.

[Details of the embodiment of the present invention]
Hereinafter, an offset search device according to the present invention, a computer program for realizing the offset search device, and an offset search method using the offset search device will be described with reference to the drawings. Hereinafter, a traffic simulator as an example of an offset search device according to the present invention will be described.

  The traffic simulator is a simulated road network (hereinafter also referred to as a road network) including a plurality of intersections or a plurality of simulated vehicles (hereinafter also referred to as vehicles) running on a route to signal lights at intersections (hereinafter also referred to as traffic lights). Signal control parameters (for example, offset, cycle length, split, etc.) are calculated. The traffic simulator can also output traffic evaluation indexes such as traffic jam length, travel time, traffic volume, and queue length.

  The traffic simulator includes a micro model suitable for evaluating the improvement of intersections, etc., and a macro model suitable for iterative calculations such as offset optimization, which handles detailed vehicle behavior and evaluates improvements in intersections. is there. A micro model traffic simulator (hereinafter referred to as a micro simulator) performs the above settings in more detail in order to reproduce changes in vehicle behavior more precisely.

  The micro simulator uses as input data, for example, a traffic volume including start / end point information of vehicle travel (for example, OD table, OD traffic volume, O means Origin, D means Destination), vehicle travel speed, acceleration. Traffic information such as deceleration characteristics is treated as given. In addition, the input data includes signal control parameters such as cycle length, split, offset, link information indicating a road network, network information such as a configuration of subareas (for example, a group of intersections as one group), Information obtained by vehicle detectors such as traffic volume and saturation traffic flow rate is also included.

  On the other hand, a macro model traffic simulator (hereinafter referred to as a macro simulator) greatly simplifies the behavior of the vehicle and does not require information such as an OD table, but instead expresses it as a traffic volume per link. Can do. The input data includes signal control parameters such as cycle length, split, and offset, link information indicating routes, network information such as the configuration of subareas (for example, a group of intersections as one group), traffic Information obtained by vehicle detectors, such as volume and saturation traffic flow rate, is also included. The offset search device, the computer program for realizing the offset search device, and the offset search method according to the present embodiment relate to macro simulation (evaluation by a macro simulator), but micro simulation can also be adopted.

  FIG. 1 is a schematic diagram showing an example of vehicle behavior. FIG. 1 shows an example of vehicle behavior on a macro simulator. As shown in FIG. 1, in the traffic simulator of this embodiment, a vehicle is expressed as a traffic volume in units of links to simplify the behavior of the vehicle. The vehicle is driven on each link, and stops or passes through the intersection according to signal information of the intersection (also referred to as a node).

  FIG. 2 is a schematic diagram showing an example of a road network when searching for an offset by the traffic simulator 100 of the present embodiment. In the example of FIG. 2, intersections 1, 2, 3, 4, 5, 6 and roads (also referred to as links) connecting the intersections (also referred to as nodes) are shown. Note that the example of FIG. 2 is merely an example, and six intersections are illustrated for simplicity, but the number of intersections is not limited to six. Further, the road network in the case of searching for an offset is not limited to the example of FIG. 2, and may be a road network in which one or a plurality of roads intersect.

  As shown in FIG. 2, a vehicle that travels from intersection 1 to 6 is a vehicle that travels downstream, and a vehicle that travels from intersection 6 to 1 is a vehicle that travels upstream. In this way, bidirectional (for example, upstream or upward direction and downstream or downward direction) vehicles are considered. This is because, in many roads, oncoming vehicles run simultaneously, and therefore when finding the offset of a signal lamp between intersections, it is necessary to have an optimum offset for bidirectional traffic.

  Also, as shown in FIG. 2, the difference in green signal start time between intersections 3 and 4 is referred to as an offset. That is, the offset refers to the time difference between the start times of the green lights of the signal lamps at adjacent intersections.

  Assuming that the traveling vehicle has a uniform traffic flow, for example, if the traffic volume per hour is 720 units, the vehicle is generated at a rate of one unit every 5 seconds at the starting point of the link. The traffic volume can be set to a desired value for each link. For example, one vehicle can be generated every five seconds and run toward the intersection 6 according to the traffic volume. Further, depending on the actual traffic volume, it is also possible to generate one vehicle every 6 seconds from the intersection 3 and drive toward the intersection 6. Similarly, one vehicle can be generated and traveled toward intersection 1 every 4 seconds, starting at intersection 6 according to the amount of traffic implemented.

  The traffic simulator 100 travels vehicles (a low speed simulation vehicle and a high speed simulation vehicle) according to the traffic volume on a road network as illustrated in FIG. 2, and delay time (also referred to as stop time), the number of stops, and a balance coefficient, which will be described later. Then, a traffic index such as a risk level is calculated, and evaluation values (first evaluation value E1, second evaluation value E2) are calculated using the calculated results. Then, the traffic simulator 100 searches for an offset so that the first evaluation value E1 is optimal (for example, local minimum) and the second evaluation value E2 is optimal (for example, local maximum), and a required offset (at each intersection). Offset pattern). Hereinafter, the details of the traffic simulator 100 will be described.

  FIG. 3 is a block diagram showing an example of the configuration of the traffic simulator 100 of the present embodiment. The traffic simulator 100 includes a simulator engine unit 10, a traffic volume calculation unit 11, a delay time calculation unit 12, a stop frequency calculation unit 13, a risk level calculation unit 14, a balance coefficient that perform calculations based on a calculation formula representing a vehicle movement model. Calculation unit 15, first evaluation value calculation unit 16, second evaluation value calculation unit 17, first offset search unit 18, second offset search unit 19, search target intersection selection unit 20, first offset setting unit 21, second An offset setting unit 22, a storage unit 23 for storing predetermined information such as input data and processing results, a travel calculation unit 24, and the like are provided. Note that the delay time, the number of stops, the balance coefficient, and the degree of danger are also referred to as traffic indicators.

  In the traffic simulator 100, as input data, for example, data such as vehicle speed, acceleration / deceleration characteristics, traffic volume in units of links, saturated traffic flow rate, signal control parameters, link information, network information such as subarea configuration, and the like. Is given.

  The traffic simulator 100 calculates the offset of the signal lamp between the intersections in the road network using the input data. The traffic simulator 100 can also output traffic evaluation indexes (estimated traffic evaluation indexes) such as the congestion length of each link, the travel time of vehicles, the traffic volume, and the queue length (number of queues). The traffic evaluation index is displayed on a map representing the road network. The traffic evaluation index may include an emission amount of environmental pollutants (such as carbon dioxide) (for example, environmental index).

  The traffic volume calculation unit 11 calculates the traffic volume for each link. The traffic volume can be changed to a desired value for each link. The traffic simulator 100 causes the road network illustrated in FIG. 2 to travel according to the traffic volume calculated by the traffic volume calculation unit 11.

  As illustrated in FIG. 2, in the traffic simulator 100, since bidirectional (for example, upward and downward) vehicles are considered, the offset of the intersections 1 to 6 is changed to a predetermined value, and along the downward direction. The vehicle is driven in the order of intersections 1, 2, 3, 4, 5, 6 (links between the intersections), and traffic indicators such as delay time, number of stops, balance coefficient, and risk are calculated for each intersection. Then, along the up direction, the order of the intersections and links is reversed, and the vehicle is driven in the order of intersections 6, 5, 4, 3, 2, 1 (links between the intersections), delay time, stop Traffic indicators such as the number of times, balance coefficient, and risk are calculated for each intersection.

  If the offset of the intersection 2 is set to +10 seconds with respect to the intersection 1, the offset of the intersection 1 becomes −10 seconds with respect to the intersection 2 when the order of the link and the intersection is reversed.

  Next, a method for changing the offset between intersections (that is, links) by the traffic simulator 100 of the present embodiment (also referred to as an intersection movement rule) will be described. The intersection movement rule is a method that indicates how to change the offset of an intersection that is an offset search target (also referred to as a search target intersection) when searching for the offset while changing the offset so that the evaluation value approaches the optimum value. It is.

  First, the case where the offset is not changed will be described. FIG. 4 is a schematic diagram showing an example of offset in the case of simultaneous offset. In FIG. 4, the horizontal axis indicates the distance, and schematically illustrates the intersections 1 to 6 illustrated in FIG. 2. The vertical axis represents time. In the figure, a rectangular frame indicates a red signal at each intersection, and a length of the frame indicates a display time of the red signal. Further, between the frames, a green signal at each intersection is shown, and the length between the frames shows the display time of the green signal. For simplicity, the yellow signal is omitted. The simultaneous offset is a case where the offset between adjacent intersections is 0%, and the blue light start time at each intersection is the same. As shown in FIG. 4, it can be seen that the start time of the green light at each intersection is the same time.

  FIG. 5 is a schematic diagram showing an example of the offset in the case of the all-downstream intersection moving method (first method) by the traffic simulator 100 of the present embodiment. As shown in FIG. 5, the downstream all intersection moving method is a method of changing only the offset of the search target intersection (intersection 2 in the example of FIG. 5) to a predetermined value (+ N in the example of FIG. 5). Since the offsets of the downstream intersections (intersections searched in order) are not changed, as a result, the offsets of the entire downstream intersections (intersections 3, 4, 5, and 6 in the example of FIG. 5) move in the + N direction. To do. In FIG. 5, the offset change is indicated by a rectangular frame with a pattern. For the reverse (upstream) traffic, the offset of the entire downstream intersection moves in the -N direction. In addition, the selection method of a search object intersection is as follows, for example.

  The search target intersection selection unit 20 has a function as a selection unit, and selects a search target intersection when searching for an offset from intersections of the simulated road network in a predetermined order. When the number of intersections in the simulated road network is I, the intersection number M is 1 to I. Since the offset is the time difference between the intersections at the start of the green signal, the intersection number M is 2 to I when the intersection with the intersection number M = 1 is used as a reference. The search target intersection selection unit 20 selects the search target intersection in the order of the intersection number M of 2, 3,. For example, when the selection order of the search target intersections is the order from the upstream side to the downstream side of the intersections of the road network, the search target intersections are intersections 2, 3, 4, 5, 6 in the example of FIG. They will be selected in order.

  The first offset setting unit 21 has a function as a first setting unit, and repeatedly sets only the offset of the signal lamp at the search target intersection selected by the search target intersection selection unit 20 to a plurality of predetermined values. The plurality of predetermined values are, for example, change values N (for example, ratios to cycle lengths such as 2%, 3%, 15%, and 40%, times such as 1 second) when changing the offset of the search target intersection. .

  As shown in FIG. 5, the first offset setting unit 21 sets only the offset of the signal lamp at the search target intersection (intersection 2 in the example of FIG. 5) to a predetermined value, and is therefore selected after the search target intersection. The offsets of other intersection groups (for example, downstream intersection groups) are not changed. For example, if the offset of the signal lamp at the selected search target intersection is set to + N, only the offset of the signal lamp at the search target intersection is changed. As a result, the selection order shown in FIG. Regardless of the search target intersection, the offset of the entire search target intersection is set to + N with respect to the intersection one upstream (one before) of the search target intersection.

  FIG. 6 is a schematic diagram showing an example of an offset in the case of the cooperative intersection group movement method (second method) by the traffic simulator 100 of the present embodiment. As shown in FIG. 6, the cooperative intersection group movement method absorbs the change when the offset of the search target intersection (intersection 2 in the example of FIG. 6) is changed by a predetermined value (+ N in the example of FIG. 6). The intersection (cancellation intersection) that can be offset (for example, set an offset of −N) is identified (in the example of FIG. 6, the intersection 5 is the cancellation intersection), and the offset is determined at the identified intersection (link). This is a method of absorption.

  The second offset setting unit 22 has a function as a specifying unit. When the offset of the signal lamp at the search target intersection selected by the search target intersection selection unit 20 is set to a predetermined value, the second offset setting unit 22 cancels the set predetermined value. An offset intersection that can be set as an offset value is specified. For example, when the offset of the signal lamp at the selected intersection to be searched is set to + N, an intersection that can be set as an offset is an offset value for canceling + N (for example, -N with respect to + N). Identifies as an intersection. The second offset setting unit 22 specifies a cancellation intersection every time a search target intersection is selected.

  The second offset setting unit 22 has a function as a second setting unit, and repeatedly sets the offset of the signal lamp at the search target intersection selected by the search target intersection selection unit 20 to a plurality of predetermined values, and specifies the offset Repeatedly set the offset of the traffic light at the intersection to the offset value. For example, when the offset of the signal lamp at the selected search target intersection is set to + N, a cancellation intersection that can set an offset value (for example, −N) that cancels + N as an offset is specified.

  That is, as shown in FIG. 6, when the offset of the signal lamp at the selected intersection to be searched (intersection 2 in the example of FIG. 6) is set to + N, the cancellation intersection (intersection 5 in the example of FIG. 6) The offset of the signal lamp is set to −N, and the intersection group between the search target intersection and the canceling intersection (one intersection in the example of FIG. 6) with respect to the intersection one upstream (one before) the search target intersection The offset of the signal lamp of 3, 4) will be set to + N.

  The second offset setting unit 22 determines the offset of the signal lamp at the selected search target intersection when the offset intersection cannot be specified (for example, in the example of FIG. 6, when −N offset cannot be set). The offset value having the largest offset amount is set within the absolute value range of the offset value at the intersection.

  FIG. 7 is a schematic diagram showing an example of a method for setting the offset of the search target intersection when the offset set in the case of the cooperative intersection group movement method (second method) by the traffic simulator 100 of the present embodiment cannot be absorbed. FIG. As shown in FIG. 7, when the offset of the search target intersection (intersection 2 in the example of FIG. 7) is to be changed by + N (for example, +20 seconds), an offset −N (for example, for offsetting (absorbing) + N) -20 seconds), the offset cancellation value of the intersection having the largest cancellation amount (intersection 5 in the example of FIG. 7) is, for example, −N + F2 ( For example, if it is −15 seconds, the offset of the signal lamp at the intersection to be searched is set to + N−F2 (+15 seconds), which is the absolute value of the cancellation value, instead of + N.

  Before the search for the offset, the amount of change in the offset of the search target intersection is set according to the cancellation that can be canceled at the offset intersection, thereby canceling the amount of change in the offset of the search target intersection during the offset search. When it is discovered that it cannot be performed, it is possible to prevent the situation where the offset of the search target intersection is corrected and the offset search process is repeated from the beginning, thereby improving the processing capability.

  Next, an offset search value in the cooperative intersection group movement method (second method) will be described based on a calculation example. FIG. 8 is an explanatory diagram showing an example of an offset search value of the cooperative intersection group movement method (second method). The search for an offset in the cooperative intersection group movement method is performed within an offset fluctuation range for preventing offset inversion. In FIG. The offset range indicates the offset fluctuation range of each intersection 2-6. 2. The current temporarily adopted value has the best evaluation value in the search so far, and indicates the temporarily adopted offset value. The current search desired amount indicates that the offset fluctuation value N in the current offset search is ± 40 seconds. 4. After the actual searchable amount, a process in which the search range is calculated according to the condition is shown.

  Since the temporary offset value (provisionally adopted value) of the intersection 2 to be searched is −10 seconds, −20 seconds is the upper limit of fluctuation on the minus side, and the searchable amounts are −20 seconds and +40 seconds. In order to absorb this value, an intersection that can change between +20 seconds and −40 seconds is searched. However, 5. As shown in the absorbable amount, none of the intersections 3 to 6 can absorb (cancel) the fluctuation range of +20 seconds and −40 seconds. Therefore, the intersection 5 where the offset can be changed most greatly absorbs the fluctuation of the offset of the intersection 2, the intersection 5 is set as the canceling intersection, and −13 seconds and +27 seconds that can be absorbed by the intersection 5 are searched. The offset fluctuation value at the intersection 2 is assumed. That is, before starting the search for the current offset, the predetermined value for changing the offset of the search target intersection 2 is corrected from (−20 seconds and +40 seconds) to (−13 seconds and +27 seconds).

  FIG. 9 shows an example of a method for specifying a canceling intersection when there are a plurality of intersections that can cancel the offset set in the case of the cooperative intersection group movement method (second method) by the traffic simulator 100 of the present embodiment. It is a schematic diagram shown. When the second offset setting unit 22 can specify a plurality of cancellation intersections, the second offset setting unit 22 specifies an intersection (an immediately downstream intersection) that is closest to the search target intersection among the plurality of cancellation intersections as the cancellation intersection.

  As shown in FIG. 9, when the offset of the search target intersection 2 is set to + N, when there are a plurality of intersections 4 and 5 that can set an offset to offset + N, the search target intersection An intersection 4 closest to 2 is set as an offset target intersection (absorption target intersection). Thereby, when the offset of the search target intersection is set to a predetermined value, the number of intersections where the same predetermined value offset is set in cooperation with the offset of the search target intersection can be minimized. It is possible to increase the possibility of finding an offset that stops the high-speed simulated vehicle by adjusting the offset of only the intersection where the vehicle is not stopped.

  The aforementioned linked intersection group movement method (second method) can prevent the occurrence of “offset inversion” that occurs in the case of the “one intersection movement method” described later. Therefore, first, the one-intersection moving method will be described.

  FIG. 10 is a schematic diagram showing an example of an offset in the case of the one intersection movement method. As shown in FIG. 10, in the 1-intersection moving method, the offset of the search target intersection (intersection 2 in the example of FIG. 10) is changed to a predetermined value (for example, + N), and the intersection closest to the downstream of the search target intersection ( In the example of FIG. 10, the moving method is such that the offset of the intersection 3) is changed to -N. In the one-intersection moving method as shown in FIG. 10, there is a high possibility that offset inversion occurs between the intersections 2 and 3 (links).

  FIG. 11 is a schematic diagram illustrating an example of offset inversion. If the offset is relatively large in a section where the distance between the intersections (link length) is relatively short, offset reversal occurs and the processing capacity of the downstream intersection decreases. That is, as shown in FIG. 11, if the offset is set large between intersections 2 and 3 having a short distance, the red signal at intersection 3 ends, and if the starting wave reaches intersection 2, the vehicle travels if the signal is green. However, since the offset is increased, a red signal is generated when the starting wave reaches intersection 2, and the vehicle is stopped at intersection 2 until the red signal ends, and can travel toward intersection 3. Can not. Such offset inversion tends to be more influential as the link length is shorter.

  In order to prevent the occurrence of offset inversion as described above, an allowable range for changing the offset to a predetermined value is set when the traffic simulator 100 searches for the offset. Hereinafter, the allowable range of offset will be described.

  FIG. 12 is an explanatory diagram showing an example of parameters when setting the allowable range of offset by the traffic simulator 100 of the present embodiment. An allowable range of the offset between the intersections A and B is represented by Os. The allowable offset range Os can be expressed by Of1 ≦ Os ≦ Of2. Here, Of1 is the lower limit value of the offset in the minus direction when viewed from the intersection A, and Of2 is the upper limit value of the offset in the plus direction when viewed from the intersection A. Since Of1 is a negative value, it is the lower limit of the allowable range, and Of2 is a positive value, which is the upper limit of the allowable range.

  The lower limit value Of1 is the maximum value of Of1a and Of1b. Of1a is a restriction on a vehicle traveling from the intersection B to the intersection A, and can be obtained by adjusting the offset to the start of the green signal and the end of the green signal as described later. Further, Of1b can be obtained by a restriction that does not prevent a starting wave for a vehicle traveling from the intersection A to the intersection B. Of1a and OF1b are parameters related to the offset restriction in the negative direction when viewed from the intersection A.

  Similarly, the upper limit value Of2 is the minimum value of Of2a and Of2b. Of2a can be obtained by the restriction that does not prevent the starting wave for the vehicle traveling from the intersection B to the intersection A. Of2b is the restriction for the vehicle traveling from the intersection A to the intersection B, and the offset starts with a green signal as described later. And by matching the end of the green light. Of2a and OF2b are parameters related to the offset restriction in the plus direction when viewed from the intersection A.

  FIG. 13 is a schematic diagram showing an example of a restriction when the offset is adjusted to the start of the green signal. As shown in FIG. 13, the link distance between intersections A and B is L (m), and the system speed of the vehicle is V (m / sec). The green light start time at the intersection B is matched so that a vehicle traveling from the intersection B to the intersection A can pass through the green light at the intersection A. In this case, the offset of the intersection B when viewed from the intersection A is −L / V. When the offset moves in the minus direction from -L / V, the start point of the green light becomes earlier, and even if the vehicle travels from the intersection B, it is stopped at the intersection A, and the processing capability at the intersection A decreases. That is, one parameter of Of1a is −L / V.

  FIG. 14 is a schematic diagram showing an example of a restriction when the offset is adjusted to the end of the green signal. As shown in FIG. 14, the link distance between intersections A and B is L (m), and the system speed of the vehicle is V (m / sec). Further, the green signal time at the intersection A is Ga (seconds), and the green signal time at the intersection B is Gb (seconds). The green light end time at the intersection B is adjusted so that a vehicle traveling from the intersection B to the intersection A can pass through the green light at the intersection A. In this case, when viewed from the intersection A, the offset of the intersection B is (Ga−Gb−L / V). When the offset moves in the plus direction from (Ga-Gb-L / V), the end point of the green light is delayed, and even if the vehicle travels from the intersection B, it is stopped at the intersection A and the processing capacity at the intersection A is reduced. To do. That is, one parameter of Of1a is (Ga-Gb-L / V).

  Therefore, Of1a, which is an offset restriction for a vehicle traveling from intersection B to intersection A, is min (-L / V, -L / V + Ga-Gb). The minimum value is due to the offset restriction in the minus direction.

  FIG. 15 is a schematic diagram showing an example of the restriction when the offset is matched to the starting wave. As shown in FIG. 15, the link distance between intersections A and B is L (m), and the starting wave velocity is W (m / sec). A vehicle which has stopped at the red light at the intersection A and has been waiting for a signal starts to start from the vehicle in the front row with the start of the green light, and the starting wave is transmitted toward the intersection B at the starting wave velocity W. Under the constraint that the vehicle traveling from the intersection B to the intersection A is not obstructed by the starting wave, the offset of the intersection B when viewed from the intersection A is −L / W. If the offset moves in the plus direction from -L / W, the signal at the intersection B becomes a red signal when the starting wave reaches the intersection B, and the vehicle at the intersection B cannot travel. That is, Of1b = −L / W.

  Therefore, the lower limit value Of1 in the negative direction when viewed from the intersection A is the maximum value of Of1a and Of1b. Since Of1 is a negative value, one of the two parameters, Of1a and Of1b, is closer to 0, that is, the maximum value.

  The upper limit value Off2 of the offset in the plus direction when viewed from the intersection A can be obtained in the same manner. That is, Of2 = min (Of2a, Of2b), Of2a can be represented by L / W, and Of2b can be obtained as the maximum value of L / V and (L / V + Ga-Gb). Since Of2 is a positive value, it is the closest to 0 of the two parameters, Of2a and Of2b, that is, the minimum value.

  FIG. 16 is an explanatory diagram showing an example of an allowable range of offset by the traffic simulator 100 of the present embodiment. Assuming that the green signal time at the intersection A is Ga (seconds) and the green signal time at the intersection B is Gb (seconds), the allowable range of offset, that is, the lower limit value Of1 of the offset, and the offset according to the magnitude relationship between Ga and Gb. The upper limit value Of2 is different. For example, as shown in FIG. 16, when Ga <Gb, Of1 = max (−L / V + Ga−Gb−L / W) and Of2 = L / V. For example, when the green signal time Ga at the intersection A is 50 seconds, the green signal time Gb at the intersection B is 65 seconds, the link distance L is 150 m, the starting wave speed W is 5 m / second, and the system speed V is 15 m / second, the lower limit of the offset The value Of1 is −25 seconds, and the upper limit value Of2 of offset is 10 seconds. In addition, the lower limit value and upper limit value of the offset in the case of other Ga and Gb magnitude relationships are also as shown in FIG.

  Next, a method for calculating the evaluation value will be described. The traffic simulator 100 of the present embodiment travels a regulated speed compliant vehicle (also referred to as a low-speed simulated vehicle), calculates traffic indicators such as delay time, number of stops, balance coefficient, and risk, and uses the calculated results. A first evaluation value E1 is calculated. In addition, the traffic simulator 100 causes a high-speed traveling vehicle (also referred to as a high-speed simulated vehicle) to travel, calculates traffic indexes such as delay time (stop time), number of stops, balance coefficient, and risk, and uses the calculated results. A second evaluation value E2 is calculated. As will be described later, the first evaluation value E1 and the second evaluation value E2 can also be calculated based on at least one of the delay time and the number of stops. Details will be described below.

  The delay time calculation unit 12 has a function as a first calculation unit, and calculates a first delay time D1 at each intersection when the regulated speed compliant vehicle is driven. A regulated speed compliant vehicle travels at a practically safe running speed, such as a regulated speed or a speed slightly higher than the regulated speed (for example, regulated speed +5 km / h, which is also referred to as the vicinity of the regulated speed). Vehicle. The first delay time D1 is the delay time of the regulated speed compliant vehicle that has stopped at the intersection. The delay time is the sum of the delay time at each intersection or the value per intersection obtained by dividing this total value by the number of intersections. If there are multiple vehicles that comply with the regulated speed at each intersection, The sum of the delay times of a plurality of vehicles that comply with the regulated speed at the intersection.

  The delay time calculation unit 12 has a function as a second calculation unit, and calculates a second delay time D2 at each intersection when the high-speed traveling vehicle is driven. The high-speed traveling vehicle is, for example, a vehicle that travels at a dangerous traveling speed such as a violation speed (for example, a speed exceeding 20 km / h) that greatly exceeds the regulation speed (also referred to as a regulation speed violation vehicle). The second delay time D2 is a delay time of the high-speed traveling vehicle stopped at the intersection. The delay time is the total delay time at each intersection or a value per intersection obtained by dividing this total value by the number of intersections. If there are multiple high-speed traveling vehicles stopped at each intersection, This is the sum of the delay times of a plurality of high-speed traveling vehicles at the intersection.

  FIG. 17 is an explanatory diagram showing an example of calculation of the delay time by the traffic simulator 100 of the present embodiment. As described above, the delay time is the sum of the delay times of the vehicles stopped at the intersection, and is a periodic delay due to the stop of the vehicle due to a red signal at the intersection. The delay time can be obtained from an average queue in one cycle of the traffic light.

  As shown in FIG. 17, it is assumed that five vehicles are stopped by a red light at an intersection. In this case, the delay time is the sum of the delay times of each of the five vehicles, such as the delay time of the first vehicle (front row), the delay time of the second vehicle, and so on. (Sum). Therefore, the longer the queue, the longer the delay time from the last vehicle toward the front row vehicle, so the longer the queue, the longer the delay time.

  As the delay time, a value obtained by converting the total amount of delay times at the respective intersections by the cycle length, for example, converted into a value per second or per hour is used. For example, in the case of one hour, the delay time can be expressed in units of units / second / hour.

  The number-of-stops calculation unit 13 has a function as a first calculation unit, and calculates a first number of stops R1 at each intersection when the regulated speed compliant vehicle is driven. The first stop count R1 is the number of vehicles that comply with the regulated speed stopped at the intersection. The number of stops is the total number of stops at each intersection or the value per intersection obtained by dividing this total by the number of intersections. If there are multiple stops at each intersection, Is the sum of the number of stops.

  The number-of-stops calculation unit 13 has a function as a second calculation unit, and calculates a second number of stops R2 at each intersection when the high-speed traveling vehicle is driven. The second number of stops R2 is the number of high-speed traveling vehicles stopped at the intersection. The number of stops is the total number of stops at each intersection or the value per intersection obtained by dividing this total by the number of intersections. If there are multiple stops at each intersection, Is the sum of the number of stops.

  The risk level calculation unit 14 is based on the light color of the signal lamp at the intersection and the remaining display time of the light color when the vehicle (the regulated speed compliant vehicle and the high-speed traveling vehicle) reaches each intersection. The risk level Q is calculated. Similarly, the risk level calculation unit 14 can also calculate the risk level of a high-speed traveling vehicle.

  If the vehicle reaches the intersection when the lamp color of the signal lamp is switched (for example, when the signal is switched from a green signal to a red signal), the risk of a traffic accident or the like increases. Therefore, a risk factor indicating the degree of danger is determined in advance according to the lamp color (blue, red, etc.) and the remaining display time of the lamp color, and the risk factor when the vehicle reaches the intersection is summed for each vehicle. The degree can be calculated.

  FIG. 18 is an explanatory diagram showing the concept of the risk level of the vehicle. As shown in FIG. 18, it is assumed that the signal lamp at the intersection is switched from a green signal to a red signal at time t. In order to simplify the explanation, the yellow signal is omitted, but the danger level can be expressed in the same manner when there is a yellow signal.

  As shown by the straight line in FIG. 18, when a plurality of vehicles reach the intersection, the danger associated with the traveling of the vehicle according to the lamp color at the time when each vehicle reaches the intersection and the remaining display time of the lamp color. Are considered different. Specifically, it is considered that the risk is greatest when the intersection is reached at time t when the blue signal is switched to the red signal, and before and after time t, the risk gradually decreases as the distance from time t is increased.

  FIG. 19 is an explanatory diagram showing an example of weighting of the risk level of the vehicle. As shown in FIG. 19, when the lamp color of the signal lamp is switched from a blue signal to a red signal, 1 second before and after the switching time is assumed to be the highest risk factor (also called a risk factor), for example, 200. assign. A weighting factor other than 0 is assigned from the start of the red signal to 5 seconds before the start of the red signal, and the weighting factor is decreased as the remaining time of the blue signal is longer. From the start of the red signal to 3 seconds after the start of the red signal, A weighting factor other than 0 is assigned, and the weighting factor is decreased as the remaining red signal time is shorter. Note that the example of FIG. 19 is an example, and the time zone for assigning a weighting factor other than 0 and the value of the weighting factor can be set as appropriate.

  In other words, the degree of risk is calculated by the sum of the arrival traffic volume (number of vehicles) x weighting coefficient for each arrival time for each arrival time (1 second) before and after the time when the green light changes to the red light. Can do.

  The balance coefficient calculation unit 15 calculates a balance coefficient B of the regulated speed compliant vehicle. Moreover, the balance coefficient calculation part 15 can also calculate the balance coefficient of a high-speed traveling vehicle. The balance coefficient is, for example, {K1 × (delay time per upstream vehicle−delay time per downstream vehicle) × sum of upstream and downstream traffic volume} + {K2 × (number of stops per upstream vehicle) -The number of stops per down vehicle) x sum of up and down traffic volume}.

  FIG. 20 is an explanatory diagram showing a first example of evaluation values calculated by the traffic simulator 100 of the present embodiment. The first evaluation value calculation unit 16 calculates the first evaluation value E1 related to the vehicle that complies with the regulated speed according to an equation: E1 = a1 × first delay time D1 + b1 × first number of stops R1 + c1 × balance coefficient B + d1 × risk Q. . Here, a1, b1, c1, and d1 are predetermined coefficients.

  Further, the first evaluation value calculation unit 16 may calculate the first evaluation value E1 by an equation of E1 = a1 × first delay time D1 + b1 × first stop count R1, and E1 = a1 × first delay time. D1 or E1 = b1 × first stop count R1 may be used.

  The second evaluation value calculation unit 17 calculates a second evaluation value E2 related to the high-speed traveling vehicle by an expression of E2 = a2 × second delay time D2. Here, a2 is a predetermined coefficient. The second evaluation value calculation unit 17 is intended to find an offset pattern that does not change the travel time of the high-speed traveling vehicle from the travel speed of the vehicle that complies with the regulated speed, and is considered to contribute most to the travel time. The second evaluation value E2 can be calculated using only the second delay time D2 that is obtained.

  Further, the second evaluation value calculation unit 17 may calculate the second evaluation value E2 by an expression of E2 = a2 × second delay time D2 + b2 × second stop count R2, or E2 = b2 × second You may calculate by the formula of the frequency | count R2 of stop. b2 is a predetermined coefficient. In addition, when calculating the 2nd evaluation value E2, traffic indicators, such as a balance coefficient or a risk, can also be considered.

  The first offset search unit 18 and the second offset search unit 19 constitute an offset search unit. In the first offset search unit 18 and the second offset search unit 19, the first evaluation value E1 calculated by the first evaluation value calculation unit 16 becomes an optimum value, and the second evaluation value E2 calculated by the second evaluation value calculation unit 17 The offset is searched so that becomes the optimum value. The optimum value of the first evaluation value E1 is, for example, a value that does not hinder travel of the regulated speed compliant vehicle and does not decrease the travel speed. The optimum value of the second evaluation value E2 is a value that suppresses traveling of the high-speed traveling vehicle and decreases the travel speed, for example. The travel speed is a value obtained by dividing the target section by the travel time required to move the section. For example, the travel speed St can be obtained by the following formula: travel speed St (km / hour) = route length (m) / travel time T (seconds) × 3.6.

  The offset search units (first offset search unit 18 and second offset search unit 19) register (store) the offset based on the search result. That is, the offset search unit registers the offset when the evaluation value becomes the optimum value as an offset that can be adopted. The registered offset is, for example, an offset pattern at each intersection that does not decrease the travel speed of the vehicle that complies with the regulated speed, and an offset pattern at each intersection that decreases the travel speed of the high-speed traveling vehicle. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  21 and 22 are schematic diagrams showing an example of an offset search by the traffic simulator 100 of the present embodiment. In the example of FIGS. 21 and 22, assuming that the selection order of the intersections to be searched is selected from the upstream side to the downstream side of the intersection group, the intersections to be searched in the order of intersections 2, 3, 4, 5, 6 (search target intersections) ) Is selected. FIG. 21 shows a state where intersection 2 is selected as the first search target intersection.

  The search for the offset can be performed as follows, for example. The initial value of the offset of each intersection in the simulated road network is set in advance. In addition, a selection order of intersections to be searched when searching for an offset is also determined in advance. In addition, a change value N (for example, a ratio with respect to a cycle length such as 2%, 3%, 15%, and 40%, a time such as 1 second) at the time of changing the offset of the search target intersection is set. In the case where the offset of the search target intersection (intersection 2 in the example of FIG. 21) is not changed in accordance with the selection order of the search target intersection, the evaluation value (the first evaluation value, the second evaluation value) is set to + N or −N. (Evaluation value) is calculated, the offset is further changed to the one where the calculated evaluation value approaches the optimum value (for example, + N or -N), and the evaluation value is repeatedly calculated until the evaluation value becomes the optimum value. Explore.

  When the evaluation value is not improved, the next search target intersection (intersection 3 in the example of FIG. 22) is selected and the offset of the search target intersection (intersection 3 in the example of FIG. 22) is not changed. When + N and −N, evaluation values (first evaluation value, second evaluation value) are calculated, and the calculated evaluation value approaches the optimum value (for example, + N or −N). Further, the offset is changed, and the calculation of the evaluation value is repeated until the evaluation value becomes the optimum value, and the offset is searched. Thereafter, the same processing is performed up to the intersection 6.

  Next, details of the first offset search unit 18 and the second offset search unit 19 will be described.

  The first offset search unit 18 searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit 16 is small. For example, as shown in FIG. 20, when the first evaluation value E1 related to the vehicle complying with the regulated speed is calculated by an equation of E1 = a1 × D1 + b1 × R1 + c1 × B + d1 × Q, the first evaluation value E1 is reduced. For example, the first delay time D1 may be reduced or the first stop count R1 may be reduced. Decreasing the first delay time D1 or the first number of stops R1 is not to reduce the travel speed of the vehicle that complies with the regulated speed (or to suppress the decrease), so the offset is set so that the first evaluation value E1 becomes small. To search for an offset that does not reduce the travel speed of the vehicle that complies with the regulated speed. Note that searching for an offset so that the first evaluation value E1 is small means that when an inverse value of the first evaluation value E1 or an evaluation value proportional to the inverse number is used, the offset is searched so that the evaluation value becomes large. Is equivalent to.

  The second offset search unit 19 is configured such that the first evaluation value E1 calculated by the first evaluation value calculation unit 16 is within a predetermined range, and the second evaluation value E2 calculated by the second evaluation value calculation unit 17 is increased. Search for an offset. The first evaluation value E1 being within a predetermined range can be, for example, within a range that does not deteriorate the first evaluation value E1, and more specifically, the first evaluation value E1 is an optimum value or an optimum value. It can be within a range that does not become larger than the temporary provisional optimum value in the vicinity.

  As shown in FIG. 20, when the second evaluation value E2 related to the high-speed traveling vehicle is calculated by the equation E2 = a2 × D2, the second delay time D2 is increased to increase the second evaluation value E2. That's fine. Since increasing the second delay time D2 decreases the travel speed of the high-speed traveling vehicle, searching for the offset so that the second evaluation value E2 increases increases the travel speed of the high-speed traveling vehicle. The offset to be searched is to be searched. Note that searching for an offset so that the second evaluation value E2 is large means that when an inverse value of the second evaluation value E2 or an evaluation value proportional to the inverse number is used, the offset is searched so that the evaluation value becomes small. Is equivalent to.

  With the above-described configuration, first, the first offset search unit 18 can search for an offset pattern at each intersection that does not decrease the travel speed of the regulated speed-compliant vehicle. Then, the second offset search unit 19 sets an offset pattern at each intersection that reduces the travel speed of the high-speed traveling vehicle within a range in which the travel speed of the vehicle complying with the regulated speed does not decrease (or is suppressed). Can be explored.

  More specifically, the first offset search unit 18 searches for an offset so that the first evaluation value E1 calculated by the first evaluation value calculation unit 16 becomes a minimum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the vehicle that complies with the regulated speed. In the second offset search unit 19, the first evaluation value E1 calculated by the first evaluation value calculation unit 16 is not larger than the minimum value of the first evaluation value E1 obtained as a result of searching by the first offset search unit 18, The offset is searched so that the second evaluation value E2 calculated by the second evaluation value calculation unit 17 becomes a maximum value. Thereby, it is possible to search for an offset pattern at each intersection that does not decrease the travel speed of the vehicle that complies with the regulated speed and that decreases the travel speed of the high-speed traveling vehicle. That is, it is possible to obtain an offset pattern at each intersection that satisfies both the travel speed of the vehicle that complies with the regulated speed and the travel speed of the high-speed traveling vehicle.

  The second evaluation value E2 is not limited to that illustrated in FIG. FIG. 23 is an explanatory diagram showing a second example of evaluation values calculated by the traffic simulator 100 of the present embodiment. As shown in FIG. 23, the first evaluation value E1 is the same as the example of FIG. The second evaluation value calculation unit 17 calculates the second evaluation value E2 by an equation of E2 = M × E1. Details will be described below.

  The travel calculation unit 24 calculates a travel time difference TT or a travel speed difference ST when the regulated speed compliant vehicle and the high speed traveling vehicle are driven. The travel time difference TT is a difference (TT = Tt2−Tt1) between the travel time Tt2 of the high speed simulation vehicle and the travel time Tt1 of the low speed simulation vehicle. The travel speed difference ST is a difference (ST = St2-St1) between the travel speed St2 of the high-speed simulated vehicle and the travel speed St1 of the low-speed simulated vehicle. The travel speed is a value obtained by dividing the target section by the travel time required to move the section.

  The second evaluation value calculation unit 17 uses the first evaluation value E1 calculated by the first evaluation value calculation unit 16 and the travel time difference TT or the travel speed difference ST calculated by the travel calculation unit 24 to relate to the second high speed simulation vehicle. An evaluation value E2 is calculated.

  When the travel time difference TT is used, the second evaluation value E2 can be calculated by, for example, an equation of E2 = M × E1. Here, M is calculated according to the following rules. The travel time difference per 1 km in the upward direction is Tu (seconds / km), and the travel time difference per 1 km in the downward direction is Td (seconds / km). The travel time difference TT can be expressed as TT = max (Tu, Td). The travel time difference threshold is T (for example, 2 seconds / km). As condition 1, if max (Tu, Td) <T, M = 1. Further, as condition 2, when max (Tu, Td) ≧ T, M = {max (Tu, Td) −T + 1} × k. Here, k is a coefficient for emphasizing travel time, and k ≧ 1.

  For example, when the difference between Tu and Td is less than the travel time difference threshold T, M = 1 and the second evaluation value E2 = E1. Further, when k = 1, when the difference between Tu and Td is equal to or greater than the travel time difference threshold T, for example, when T + 1 (that is, max (Tu, Td) = T + 1), M = 2 (= 1 + 1) and E2 = 2 × E1. Similarly, when the difference between Tu and Td is, for example, T + 2 (that is, max (Tu, Td) = T + 2), M = 3 (= 2 + 1), and E2 = 3 × E1.

  Thus, the value of the variable M can be increased as the travel time difference TT increases. Even when the travel speed difference is used, the value of the variable M can be increased as the travel speed difference increases.

  The travel time is a value obtained by adding the average delay time D1 per vehicle to the travel time when traveling on the target route at the regulated speed (50 km / h). For example, 2500 m / (50 km / h / 3 .6). A value obtained by dividing this value by 2.5 km is a travel time per 1 km.

  Further, the travel time difference TT is not limited to the larger of the travel time difference Tu in the upward direction and the travel time difference Td in the downward direction. For example, an average value of Tu and Td, a mean square of Tu and Td, or the like may be used. Good. The same applies to the travel speed difference ST.

  In the offset search unit, the first evaluation value E1 calculated by the first evaluation value calculation unit 16 becomes an optimal value (for example, a minimum), and the second evaluation value E2 calculated by the second evaluation value calculation unit 17 is an optimal value (for example, , Find the offset so that

  More specifically, as in the case of FIG. 20, the provisional search value of the first evaluation value E1 is E1op, the provisional search value of the second evaluation value E2 is E2op, and the first evaluation value E1 is optimized. After the search, two-stage optimization is performed in which a search is performed with the second evaluation value E2 as an optimization. The first stage optimization is a new search result when E1 <E1op, and the second stage optimization is a new search result when E1 ≦ E1op and E2 <E2op.

  The optimum value of the first evaluation value is, for example, a value that does not hinder the traveling of the low speed simulation vehicle and does not decrease the travel speed. In addition, since the travel time difference or travel speed difference between the high-speed simulated vehicle and the low-speed simulated vehicle is used as the second evaluation value, for example, the travel time difference or travel speed between the high-speed simulated vehicle and the low-speed simulated vehicle is reduced. Appears in the second evaluation value, it can be easily determined whether or not the search for the offset is approaching the target value. That is, as described above, by using the second evaluation value E2 shown in FIG. 20, particularly the variable M, the goal of making the travel time difference or the travel speed difference less than a predetermined value becomes clearer. The second evaluation value is a value that suppresses the traveling of the high-speed simulated vehicle and approaches the travel speed or travel time of the low-speed simulated vehicle. With the above-described configuration, it is possible to obtain an optimum offset for suppressing the traveling of the high-speed traveling vehicle (high-speed simulated vehicle) without hindering the traveling of the regulated speed-compliant vehicle (low-speed simulated vehicle).

  Next, a comparison result when using the travel time difference TT and the travel speed difference ST will be described. FIG. 24 is an explanatory diagram showing a travel time difference when conditions are set by a travel speed difference, and FIG. 25 is an explanatory diagram showing a travel speed difference when conditions are set by a travel time difference. In the figure, the compliant vehicle is a vehicle that complies with the regulated speed, and the high-speed vehicle is a high-speed traveling vehicle. FIG. 24 shows the travel time (seconds / km) of the regulated speed compliant vehicle when the travel speed difference between the travel speed of the regulated speed compliant vehicle and the travel speed of the high speed traveling vehicle is set to a condition of 0.8 km / h. It shows how the travel time (seconds / km) of a high-speed traveling vehicle and the travel time difference change when the travel speed changes. FIG. 25 shows the travel speed (km / h) of the regulated speed compliant vehicle when the travel time difference between the travel time of the regulated speed compliant vehicle and the travel time of the high-speed traveling vehicle is set to 2.0 seconds / km. ) And the travel speed (km / h) of the high-speed traveling vehicle, and how the travel speed difference changes as the travel time changes.

  As shown in FIG. 24 and FIG. 25, in the region where the speed of the regulated speed compliant vehicle and the high-speed traveling vehicle is high, the travel speed difference when the travel time difference is a condition is more than the travel time difference when the travel speed difference is a condition. Therefore, it is more preferable to use the travel time difference for the condition setting in the region where the speed of the regulated speed compliant vehicle and the high speed traveling vehicle is high.

  Next, the selection order of the search target intersection will be described. FIG. 26 is a schematic diagram illustrating an example of a road network, and FIG. 27 is a schematic diagram illustrating an example of selection types of intersections to be searched by the traffic simulator 100 according to the present embodiment. As shown in FIG. 26, the intersections 1-2, 2-3, 3-4, 4-5, and 5-6 (link distances) are 300 m, 80 m, 100 m, 200 m, and 130 m, respectively. . FIG. 26 is an example, and the link distance is not limited to the example of FIG.

  The search target intersection selection unit 20 has a plurality of types of predetermined orders for selecting a search target intersection. By having a plurality of types of selection order of search target intersections, it is possible to widen the search range of offsets for finding a required offset pattern.

  That is, as a method of expanding the search range, a method of giving an initial value of an offset input to the simulator with a random number is generally used. However, in an actual operation system, it is necessary to set the offset currently being executed as an initial value. This is because if a better offset pattern than the current operational offset pattern cannot be found, and if the improvement effect is small even with the new offset pattern, it is necessary to adopt the current operational offset. In addition, since the current operational offset pattern is based on the result of the continuous search in the simulation, there is a high possibility that the current operational offset pattern exists in an area that does not change significantly from the optimum value. By having a plurality of types of selection order of search target intersections, it is possible to expect the same effect as the case where the initial value of the offset is given as a random number while the offset currently being executed is set as the initial value.

  More specifically, the search target intersection selection unit 20 selects a first selection order from the upstream side to the downstream side of the intersection of the simulated road network (intersections 2, 3, 4, 5 in the examples of FIGS. 26 and 27). , 6), the second selection order for selecting from the downstream side to the upstream side of the intersection of the simulated road network (in the example of FIGS. 26 and 27, the order of the intersections 6, 5, 4, 3, 2), the simulated road The third selection order (in the example of FIGS. 26 and 27, the order of intersections 3, 4, 6, 5, and 2) in the order of the shortest distance between the intersections of the network, and the longest distance between the intersections of the simulated road network Among the fourth selection order to be selected (in the example of FIGS. 26 and 27, the order of intersections 2, 5, 6, 4, and 3) and the fifth selection order to randomly select the intersection of the simulated road network A search target intersection is selected with a selection order as a predetermined order. Thereby, the search range of the offset is expanded, and for example, it is possible to increase the possibility of newly finding a required offset pattern that could not be found in one selection order.

  Next, combinations of intersection movement rules during offset search will be described. As described above, the intersection movement rules include the downstream all intersection movement method (first method) exemplified in FIG. 5 and the linked intersection group movement method (second method) exemplified in FIG. In addition, in the offset search, the offset search performed by the first offset search unit 18 so that the first evaluation value E1 becomes the optimum value and the second offset search unit 19 so that the second evaluation value E2 becomes the optimum value. There is an offset search to be performed.

  FIG. 28 is an explanatory diagram showing combinations of intersection movement rules during offset search. The first offset searching unit 18 and the second offset searching unit 19 search for an offset based on a plurality of predetermined values that are repeatedly set at least one of the first offset setting unit 21 and the second offset setting unit 22. That is, both the first offset search unit 18 and the second offset search unit 19 may use the first method (pattern 1 in FIG. 28), and any of the first offset search unit 18 and the second offset search unit 19 28 may use the second method (pattern 2 in FIG. 28), the first offset search unit 18 may use the first method, and the second offset search unit 19 may use the second method (FIG. 28). Pattern 3), or the first offset search unit 18 may use the second method, and the second offset search unit 19 may use the first method (pattern 4 in FIG. 28).

  With the above-described configuration, either the first method or the second method can be used for the offset search when the first evaluation value E1 is optimized, and the offset search when the second evaluation value E2 is optimized. In addition, since either the first method or the second method can be used, the search method for the offset can be changed to various methods, and the evaluation value is smaller than when the predetermined value of the offset is simply changed. The search range for the offset that is the optimum value can be expanded.

  The first offset searching unit 18 searches for an offset based on a plurality of predetermined values that are repeatedly set by one of the first offset setting unit 21 and the second offset setting unit 22, and the second offset searching unit 19 The offset is searched based on a plurality of predetermined values repeatedly set by the other of the first offset setting unit 21 and the second offset setting unit 22. For example, the first offset search unit 18 may use the first method (one) and the second offset search unit 19 may use the second method (the other), or the first offset search unit 18 may use the second method. The second offset search unit 19 may use the first method.

  With the above-described configuration, the offset search method when optimizing the first evaluation value E1 and the offset search method when optimizing the second evaluation value E2 can be made different, so the first evaluation value E1 When the same evaluation method is used when optimizing the second evaluation value E2, the possibility of finding the optimum evaluation value can be increased, and the possibility of finding a new offset pattern can be increased. Can be high.

  Further, the first offset search unit 18 searches for an offset based on a plurality of predetermined values repeatedly set by the first offset setting unit 21, and the second offset search unit 19 is searched by the first offset setting unit 21. The offset is searched based on a plurality of predetermined values repeatedly set by the second offset setting unit 22 using the offset value as an initial value. That is, the first offset search unit 18 uses the first method, and the second offset search unit 19 searches for the offset using the second method (pattern 3 in FIG. 28).

  By searching for the offset using the second method, that is, the cooperative intersection group moving method, there are the following advantages. That is, there is no restriction for preventing the offset inversion because the offset inversion that occurs in the short-distance link as illustrated in FIGS. 10 and 11 does not occur. In addition, when the restrictions on preventing offset inversion are ignored, it is necessary to move the offset other than searching for the optimum value in order to correct the deviation, but there is no restriction on preventing offset inversion. There is also no offset movement other than the optimum value search. When the second offset searching unit 19 searches for the offset using the second method, it is possible to move the offset in cooperation with only some of the intersection groups among all the intersections. Accordingly, it is possible to increase the possibility of finding an offset pattern that stops the high-speed traveling vehicle at the intersection without reducing the travel speed of the vehicle that complies with the regulated speed.

  FIG. 29 and FIG. 30 are schematic diagrams showing an example of the result of the search for the offset by the traffic simulator 100 of the present embodiment. FIGS. 29 and 30 schematically show the results when the offset is searched using the pattern 3 illustrated in FIG. In FIG. 29, the first offset search unit 18 uses the first method (downstream all intersection movement method) to search for an offset at which the first evaluation value E1 related to the regulated speed observing vehicle becomes a minimum value and perform the first evaluation. The state at the end of optimization of the value E1 is shown. For example, with the offset when the first evaluation value E1 is minimized, the regulated speed compliant vehicle can travel without reducing the average travel speed. However, the high-speed traveling vehicle is in a state where it can pass without stopping at a red light.

  FIG. 30 shows the result of the first offset search unit 18 searching for the first evaluation value E1 when the second offset search unit 19 searches for the offset using the second method (cooperative intersection group movement method). An offset at which the second evaluation value E2 does not become larger than the minimum value of the obtained first evaluation value E1 and the second evaluation value E2 becomes a maximum value is searched to show a state at the end of optimization of the second evaluation value E2. That is, the red light time of an intersection irrelevant to the travel of the regulated speed compliant vehicle is adjusted (that is, the offset changes) so that the high-speed traveling vehicle stops at the intersection. As a result, an offset pattern that reduces the average travel speed of the high-speed traveling vehicle can be obtained without reducing the average travel time of the regulated speed-compliant vehicle.

  Next, operation | movement of the traffic simulator 100 of this Embodiment is demonstrated. FIGS. 31, 32 and 33 are flowcharts showing an example of an offset search processing procedure by the traffic simulator 100 of the present embodiment. The traffic simulator 100 sets the initial value of the offset of each intersection (S11), and determines the selection order of search target intersections (S12). The selection order of the search target intersection is illustrated in FIG.

  The traffic simulator 100 assigns an intersection number M according to the determined order (S13). If the number of intersections is I, M is an integer from 2 to I. The traffic simulator 100 sets an intersection movement method (S14). For example, the first method (downstream all intersection movement method) can be set.

  The traffic simulator 100 sets an offset change amount (predetermined value) N1 (S15). The amount of change (predetermined value) N1 is, for example, a ratio to the cycle length of 15%, 40%, 15%, 40%, 3%, 2%, etc. in the order of setting the offset, 1 second (only the last final adjustment is in seconds) However, it is not limited to this.

  The traffic simulator 100 sets the intersection number M to 2 (S16), and if the offset of the search target intersection is not changed, + N1, and −N1, the vehicle travels and the first delay time D1, The second delay time D2 and the first evaluation value E1 are calculated (S17). The traffic simulator 100 determines whether or not an offset pattern that satisfies the current first evaluation value E1 <provisional first optimum value is found at + N1 or −N1 (S18). The provisional first optimum value can be a value closest to the optimum value among the first evaluation values E1 calculated so far.

  When the offset pattern is found (YES in S18), the traffic simulator 100 further changes the offset in that direction to calculate the first delay time D1, the second delay time D2, and the first evaluation value E1 (S19). ), The process of step S18 is repeated. If no offset pattern is found (NO in S18), the traffic simulator 100 determines whether the intersection number M is equal to I (S20). If the intersection number M is not equal to I (NO in S20), 1 is added to the intersection number M (S21), and the process after step S17 is repeated.

  If the intersection number M is equal to I (YES in S20), the traffic simulator 100 determines whether there is another change amount N1 of the offset (S22), and if there is another change amount N1 (YES in S22). ), And repeats the processing from step S15. When there is no other change amount N1 (NO in S22), the traffic simulator 100 calculates the travel time Tt1 of the regulated speed compliant vehicle and the travel time Tt2 of the high speed traveling vehicle (S23).

  The traffic simulator 100 determines whether or not Tt2−Tt1 ≦ time threshold Td1 (S24). The time threshold value Td1 is, for example, 2 seconds / km, but is not limited thereto.

  One of the purposes of the traffic simulator 100 is to reduce the difference between the travel time of a high-speed travel vehicle and the travel time of a regulated speed-compliant vehicle by stopping the high-speed travel vehicle at a red light at an intersection and increasing the delay time. (For example, approaching 0 km / hour).

  If Tt2−Tt1 ≦ time threshold Td1 is not satisfied (NO in S24), the traffic simulator 100 determines whether or not the number of simulations has reached the predetermined number C (S25), and does not reach the predetermined number C (S25). NO), the processing after step S11 is repeated. If the predetermined number C is reached (YES in S25), the traffic simulator 100 ends the process.

  When Tt2−Tt1 ≦ time threshold Td1 (YES in S24), the traffic simulator 100 ends the search by the first offset search unit 18 and sets the intersection movement method (S26). For example, the second method (cooperative intersection group movement method) can be set.

  The traffic simulator 100 sets an offset change amount (predetermined value) N2 (S27). The amount of change (predetermined value) N2 can be set to 3%, 3%, 1 second, 1 second, for example, in the offset setting order, but is not limited to this.

  The traffic simulator 100 sets the intersection number M to 2 (S28), and when the offset of the search target intersection is not changed, when it is set to + N2, and when it is set to -N2, the vehicle travels and the first delay time D1, The second delay time D2, the first evaluation value E1, and the second evaluation value E2 are calculated (S29). The traffic simulator 100 determines whether an offset pattern that satisfies the current first evaluation value E1 ≦ the provisional first optimum value and the current second evaluation value E2 ≧ the provisional second optimum value is found at + N2 or −N2. Is determined (S30). The provisional second optimum value can be a value closest to the optimum value among the second evaluation values E2 calculated so far.

  When the offset pattern is found (YES in S30), the traffic simulator 100 further changes the offset in that direction to change the first delay time D1, the second delay time D2, the first evaluation value E1, and the second evaluation value. E2 is calculated (S31), and the process of step S30 is repeated. If no offset pattern is found (NO in S30), the traffic simulator 100 determines whether or not the intersection number M is equal to I (S32). If the intersection number M is not equal to I (NO in S32), 1 is added to the intersection number M (S33), and the process after step S29 is repeated.

  If the intersection number M is equal to I (YES in S32), the traffic simulator 100 determines whether there is another change amount N2 of the offset (S34), and if there is another change amount N2 (YES in S34). ), And repeats the processing from step S27. When there is no other change amount N2 (NO in S34), the traffic simulator 100 calculates the travel time Tt1 of the regulated speed compliant vehicle and the travel time Tt2 of the high speed traveling vehicle (S35).

  The traffic simulator 100 determines whether or not Tt2−Tt1 ≦ time threshold Td2 (S36). The time threshold Td2 is, for example, 2 seconds / km, but is not limited to this. If Tt2−Tt1 ≦ time threshold Td2 is not satisfied (NO in S36), the traffic simulator 100 determines whether or not the number of simulations has reached the predetermined number C (S38), and does not reach the predetermined number C (S38). NO), the processing after step S11 is repeated. When the predetermined number C is reached (YES in S38), the traffic simulator 100 ends the process.

  When Tt2−Tt1 ≦ time threshold Td2 (YES in S36), the traffic simulator 100 registers the offset of each intersection obtained by the search of the second offset search unit 19 as a candidate pattern (required offset pattern candidate). (S37), and the process of step S38 is performed.

  In steps S23 and S35 described above, the travel speed may be calculated instead of the travel time. The travel speed St1 of the vehicle complying with the regulated speed can be obtained by the following formula: St1 = route length / {route length / (S1 / 3.6) + first delay time D1} × 3.6. The speed St2 can be obtained by the following formula: St2 = route length / {route length / (S2 / 3.6) + second delay time D2} × 3.6. Here, S1 is the speed (for example, 50 km / hour) of the vehicle that complies with the regulated speed, and S2 is the speed of the high-speed traveling vehicle (for example, S1 + 20 km / hour). In this case, the speed threshold Sd1 in step S24 can be, for example, 6 km / hour, and the speed threshold Sd2 in step S36 can be, for example, 3 km / hour, but is not limited thereto. It is not something.

  The traffic simulator 100 of the present embodiment can also be realized using a general-purpose computer that includes a CPU (processor), a RAM, and the like. That is, as shown in FIG. 31, FIG. 32 and FIG. 33, a computer program that defines the procedure of each process is loaded into a RAM provided in the computer, and the computer program is executed by a CPU (processor). Thus, the traffic simulator 100 can be realized.

  Next, a simulation result by the traffic simulator 100 of the present embodiment will be described.

  FIG. 34 is an explanatory diagram showing a comparison between the first method and the second method regarding optimization of the first evaluation value E1. In FIG. 34, the horizontal axis represents the average travel speed (km / hour) in the first method (downstream all intersection movement method), and the vertical axis represents the average travel speed (km) in the second method (cooperative intersection group movement method). / Hour). For the following hypothetical 100 routes, the first method and the second method were compared for optimization of the first evaluation value E1.

The simulation conditions are as follows.
Number of intersections and cycle length: 15 intersections, 120 seconds Traffic volume: Daytime Up and down 700 units / hour (saturated traffic flow rate 1800 units / hour)
Traveling speed: 50km / h Large vehicle mixture rate: 0%
Main road blue hour ratio at each intersection: 40% to 70% within a range of 1% determined by random numbers Link length: 100m to 400m (10m units determined by random numbers)
Offset initial value: 0% (simultaneous offset)
Sample route: Virtual 100 route created under the above conditions Selection order of search target intersections: Shortest link distance

  The large vehicle mixture rate is used when large vehicle correction is performed. When controlling the offset, there are cases where priority is given to a direction in which a large vehicle has a large amount of traffic in consideration of acceleration / deceleration performance of large vehicles and carbon dioxide emissions. If the large vehicle mixing ratio is 50% and the weighting factor is 2.0, the traffic volume is 1.5 times, the delay time and the number of stops are also 1.5 times, and the offset is given priority accordingly. For large vehicle correction, for example, large vehicle correction = delay time (unit / second / hour) × (100−large vehicle mixture rate (%))) / 100 + delay time (unit / second / hour) × large vehicle mixture Rate (%) / 100 × Large vehicle weight coefficient + 30 × Number of stops (times / hour) × (100−Large vehicle mixture rate (%)) / 100 + Number of stops (times / hour) × Large vehicle mixture rate (%) / 100 × Large car weighting factor.

  The simulation result is shown in FIG. In the case of simultaneous offset, the average travel speeds in the case of the first method and in the case of the second method were 26.5 km / hour, 28.7 km / hour, and 28.4 km / hour, respectively. The results of the first method were slightly better, but as shown in FIG. 34, the results of the first method and the second method are distributed almost evenly around the 45 degree line. There is no difference between the two systems. However, in the second method, for example, in a special case where there is only one link with a long distance, such as the number of intersections on the target route is as small as 3 to 5, there is a possibility that the search range of this link may be restricted. For this reason, it is more preferable to use the first method in optimizing the first evaluation value E1.

  FIG. 35 and FIG. 36 are explanatory diagrams showing simulation results when the selection order of search target intersections is one type. FIGS. 37 and 38 show simulation results when the selection order of search target intersections is four types. It is explanatory drawing shown. FIG. 35 and FIG. 36 show a case where the selection order of the search target intersections is selected in ascending order of the link distance. In FIG. 37 and FIG. 38, the search target intersections are selected in four ways: from the upstream side to the downstream side, from the downstream side to the upstream side, from the shortest link distance, and from the longest link distance.

In addition, the following four types of comparison are made for the intersection movement rules. That is,
A. Simultaneous offset [regulated speed vehicle optimization (optimization of first evaluation value E1) and high-speed traveling vehicle delay adjustment (optimization of second evaluation value E2) no change in offset],
B. Non-consideration of high-speed traveling vehicle [Adopting the first method for regulation speed vehicle optimization (optimization of the first evaluation value E1), no change in offset for high-speed traveling vehicle delay adjustment (optimization of the second evaluation value E2)],
C. High-speed restraint offset first method [Adopting first method for regulation speed vehicle optimization (optimization of first evaluation value E1) and high-speed traveling vehicle delay adjustment (optimization of second evaluation value E2)] (pattern in FIG. 28 1),
D. High-speed inhibition offset first method + second method [First method is adopted for regulation speed vehicle optimization (optimization of first evaluation value E1), and high-speed traveling vehicle delay adjustment (optimization of second evaluation value E2) is first. 2 method is adopted] (corresponding to pattern 3 in FIG. 28).

The simulation conditions are as follows.
Number of intersections and cycle length: 15 intersections, 120 seconds Traffic volume: Night Up and down direction 100 units / hour (saturated traffic flow rate 1800 units / hour)
Traveling speed: 50 km / hour for vehicles that comply with the regulation speed, 70 km / hour for high-speed traveling vehicles Large vehicle mixture rate: 0%
Main road green time ratio at each intersection: 40% to 70% (determined by random numbers in 1% increments)
Link length: 100m to 400m (determined by random number in 10m increments)
Offset initial value: 0% (simultaneous offset)
Sample route: Virtual 100 route created under the above conditions

  The evaluation index in this analysis is “travel speed of a high-speed traveling vehicle−travel speed of a vehicle that complies with a regulated speed (km speed)”, and the closer this value is to 0 km / hour, the better the result. As shown in FIGS. 35 and 36, the best results are obtained when the first method is used to minimize the first evaluation value E1 and the second method is used to maximize the second evaluation value E2. I understand that. In addition, as shown in FIGS. 37 and 38, it can be seen that a better result can be obtained by selecting an optimum value from four calculation results.

  FIG. 39 is an explanatory diagram showing a simulation result when an offset search is performed by simultaneously optimizing the first evaluation value E1 and the second evaluation value E2. Since the results differ depending on whether the first evaluation value E1 and the second evaluation value E2 are AND or OR, simulations were performed in both cases.

  The AND processing is a new search when E1 <E1Op and E2> E2Op with respect to the optimum value E1Op of the first evaluation value E1 and the optimum value E2Op of the second evaluation value E2 obtained in the search so far. As a result. In addition, the OR process sets a new search result when E1 <E1Op or E2> E2Op.

  As shown in FIG. 39, while these search rules can reduce the travel speed of a high-speed traveling vehicle, the speed improvement amount of a vehicle that complies with the regulated speed is small. For the purpose of preferentially reducing the travel speed of the high-speed traveling vehicle, it is also possible to perform the offset search by simultaneously optimizing the first evaluation value E1 and the second evaluation value E2. However, if optimization of the first evaluation value E1 (for example, minimization) and optimization of the second evaluation value E2 (for example, maximization) are performed at the same time, both optimization directions interfere with each other, and compliance with the regulation speed is observed. It may reduce the travel speed of the vehicle. Therefore, when priority is given to not reducing the travel speed of the vehicle that complies with the regulated speed, the first evaluation value E1 is optimized as the first stage, and the first evaluation value found in the first stage as the second stage. It is preferable to search for an offset by optimizing the second evaluation value E2 while maintaining a condition that does not deteriorate the optimum value of E1.

  FIG. 40 is an explanatory diagram showing the concept of evaluation value optimization search. FIG. 40 shows a state in which the search value falls into one of an infinite number of local minimum values. In the traffic simulator 100 of the present embodiment, the search range is limited to search for an offset that maximizes the second evaluation value E2 within a range that does not deteriorate the first evaluation value E1. Rather than using the first method for minimizing the first evaluation value and maximizing the second evaluation value, the first method is used to minimize the first evaluation value, and the second method is used to maximize the second evaluation value. If the method is used, the search method is changed and it is possible to move to another distribution, so that the possibility of finding a new offset pattern is increased. Specifically, as shown in FIG. 30 described above, in the second method, only the intersection where the regulated speed compliant vehicle is not stopped is adjusted to find a new offset pattern that increases the delay of the high-speed traveling vehicle. Can do.

  Next, a case where the number of optimization calculations is increased will be described. 41 and 42 are explanatory views showing simulation results when the optimization calculation is performed 400 times. In FIG. High-speed inhibition offset first method + second method [First method is adopted for regulation speed vehicle optimization (optimization of first evaluation value E1), and high-speed traveling vehicle delay adjustment (optimization of second evaluation value E2) is first. Among the results obtained in [Adopting 2 systems], the worst offset pattern has a speed difference in the 7 km / h range. 41 and 42 show the results when the number of optimization calculations is further increased from 4 times to 400 times for this virtual route.

The simulation conditions are as follows.
Number of intersections and cycle length: 15 intersections, 120 seconds Traffic volume: Night Up and down direction 100 units / hour (saturated traffic flow rate 1800 units / hour)
Traveling speed: 50 km / hour for vehicles that comply with the regulation speed, 70 km / hour for high-speed traveling vehicles Large vehicle mixture rate: 0%
Main road blue hour rate and link length at each intersection: 1 case with the worst results on virtual 100 routes Offset initial value: Initial value of 100 patterns is determined by random numbers Sample route: Speed difference is 7km / hour in Fig. 37 Virtual 1 route Selection order of intersections to be searched: Up and down the intersections, 4 types in order of short and long link distance

  In the case of FIGS. 41 and 42, how many good offset patterns are found is important. In actual operation, it is preferable that the number of searches is smaller from the viewpoint of time reduction. FIG. 42 shows the result of performing the optimization calculation 400 times for the same virtual route while changing the initial offset value (100 patterns) and the selection order of the intersections to be searched (four patterns). In FIG. 42, the number of searches is an average value of the number of simulations performed in one optimization calculation, and in the fast deterrence offset design, each of 93.8 times, 120.4 times and 170.0 times are added 263. .7 times and 290.3 times are the total number of searches for obtaining results. The number of improvement value discovery times per time is the number of times an offset pattern in which the evaluation value is newly improved by moving the offset value by ± N is discovered. The number of offset patterns with a speed difference of less than 3 km / hour indicates how many offset patterns with a speed difference of less than 3 km / hour have been found during the optimization calculation performed 400 times. When the second method is selected for the search for the high-speed deterrence offset, it is more likely that an improved value will be found, and the number of patterns with a speed difference of less than 3 km / hour is 16 cases, which is 2.7 times that of other methods. .

  FIG. 43 is an explanatory diagram showing a comparison of discovery statuses of fast deterrence offset patterns. In FIG. 43, the horizontal axis indicates the travel speed difference when the high-speed traveling vehicle is not considered, the left vertical axis indicates the travel speed difference in the case of the first high-speed deterrent offset method, and the right vertical axis indicates The travel speed difference in the case of the high speed inhibition offset first method + second method is shown. In the figure, the point on the line of 45 degrees indicates that no offset pattern that suppresses the speed of the high-speed traveling vehicle was found for the offset pattern searched when the high-speed traveling vehicle was not considered. Show. It can be seen that more right offset patterns are found in the right side of FIG.

  On the other hand, a pattern with a large speed difference in the search when the high-speed traveling vehicle is not considered can be expected to obtain a better result when a pattern search is performed when the high-speed traveling suppression offset is subsequently used. This means that it is more efficient to perform a search in the case of high-speed deterrence offset based on a pattern that results in a result that is close to the target to some extent when a search is performed without consideration of high-speed vehicles. .

  In the above-described embodiment, the formula for calculating the evaluation value is configured to use traffic indicators such as the delay time, the number of stops, the balance coefficient, the degree of risk, but the present invention is not limited to this. An index that considers environmental measures such as the above can be added, and another index may be added.

  In the above-described embodiment, the speed of the high-speed traveling vehicle is not limited to one type, for example, three types of 70 km / h, 80 km / h, and 90 km / h with respect to the regulation speed of 50 km / h, Four types of simulated vehicles can be driven in accordance with the speed of the regulated speed compliant vehicle, and five or more types of simulated vehicles may be driven.

  The disclosed embodiments are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

DESCRIPTION OF SYMBOLS 10 Simulator engine part 11 Traffic volume calculation part 12 Delay time calculation part 13 Stop frequency calculation part 14 Risk degree calculation part 15 Balance coefficient calculation part 16 1st evaluation value calculation part 17 2nd evaluation value calculation part 18 1st offset search part 19 Second offset search unit 20 Search target intersection selection unit 21 First offset setting unit 22 Second offset setting unit 23 Storage unit 24 Travel calculation unit

Claims (17)

An offset search device that searches for an offset of a signal lamp at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections,
A first evaluation value calculation unit that calculates a first evaluation value related to the low speed simulation vehicle when the low speed simulation vehicle is driven;
A second evaluation value calculation unit for calculating a second evaluation value related to the high-speed simulation vehicle when the high-speed simulation vehicle is driven;
An offset search unit for searching for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is an optimal value, and the second evaluation value calculated by the second evaluation value calculation unit is an optimal value ;
A first calculator that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven;
A travel calculation unit that calculates a travel time difference or a travel speed difference when the low-speed simulated vehicle and the high-speed simulated vehicle are run;
With
The first evaluation value calculation unit includes:
A first evaluation value related to the low speed simulation vehicle is calculated based on at least one of the first delay time and the first number of stops calculated by the first calculation unit;
The second evaluation value calculation unit
The first evaluation value and the travel calculated Citea Ru to calculate a second evaluation value according to the high-speed simulation vehicle based on the travel time difference or travel speed difference calculated in part offset calculated by the first evaluation value calculation unit Search device. The second evaluation value calculation unit
When the travel time difference or the travel speed difference is greater than or equal to a predetermined threshold, the second evaluation value is calculated to be a larger value than when the travel time difference or the travel speed difference is smaller than the predetermined threshold. The offset search device according to claim 1 . The second evaluation value calculation unit
Claim 1 or claim are to calculate the second evaluation value by multiplying the large factor as the travel time difference calculated by the trip calculator to the first evaluation value calculated by the first evaluation value calculation unit is larger offset searching apparatus according to 2. The offset search unit
A first offset search unit that searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is small;
A second offset search unit for searching for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is large. The offset search device according to claim 1, comprising: The offset search unit
A first offset search unit that searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is small;
A second offset search unit that searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is small. offset searching device according to any one of up to claim 1 or we claim 3 comprising and. The first offset search unit includes:
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value,
The second offset search unit
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value and the second evaluation value calculated by the second evaluation value calculation unit becomes a maximum value. The offset search device according to claim 4 . The first offset search unit includes:
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value,
The second offset search unit
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value and the second evaluation value calculated by the second evaluation value calculation unit is a minimum value. The offset search device according to claim 5 . A selection unit that selects, in a predetermined order, intersections to be searched when searching for an offset from the intersections of the simulated road network;
A first setting unit that repeatedly sets only the offset of the signal lamp at the intersection to be searched selected by the selection unit to a plurality of predetermined values;
When the offset of the signal lamp at the intersection to be searched selected by the selection unit is set to a predetermined value, a specifying unit that specifies a cancellation intersection that can set an offset value for canceling the set predetermined value as an offset; ,
A second setting unit for repeatedly setting the offset of the signal lamp at the intersection to be searched selected by the selection unit to a plurality of predetermined values, and repeatedly setting the offset of the signal lamp at the cancellation intersection specified by the specifying unit as a cancellation value; With
The first offset search unit and the second offset search unit are:
The offset search according to any one of claims 4 to 7, wherein an offset is searched based on a plurality of predetermined values repeatedly set at least one of the first setting unit and the second setting unit. apparatus. The first offset search unit includes:
The offset is searched based on a plurality of predetermined values repeatedly set in one of the first setting unit and the second setting unit,
The second offset search unit
The offset search device according to claim 8 , wherein an offset is searched based on a plurality of predetermined values repeatedly set by the other of the first setting unit and the second setting unit. The first offset search unit includes:
The offset is searched based on a plurality of predetermined values repeatedly set by the first setting unit,
The second offset search unit
The offset search device according to claim 8 or 9 , wherein an offset is searched based on a plurality of predetermined values repeatedly set by the second setting unit. The second setting unit includes:
If the canceling intersection cannot be specified by the specifying unit, the offset of the signal lamp at the search target intersection selected by the selecting unit is set to the absolute value of the canceling value of the intersection at which the canceling value having the maximum canceling amount can be set. The offset search device according to any one of claims 8 to 10 , wherein the offset search device is set within a range of values. The specific part is:
If you can identify a plurality of offset intersections, one of claims 8 whose distance from the search target intersection of the offset crossing said plurality of are to be identified as offset intersection intersection close to claim 10 The offset search apparatus according to item 1. The selection unit includes:
The offset search device according to any one of claims 8 to 12 , which has a plurality of types of the predetermined order. The selection unit includes:
The first selection order for selecting from the upstream side to the downstream side of the intersection of the simulated road network, the second selection order for selecting from the downstream side to the upstream side of the intersection of the simulated road network, and the distance between the intersections of the simulated road network A plurality of selections among a third selection order for selecting in order of shortness, a fourth selection order for selecting in order of long distance between intersections of the simulated road network, and a fifth selection order for randomly selecting intersections of the simulated road network 14. The offset search device according to claim 13 , wherein a search target intersection is selected with the order as the predetermined order. A computer program for causing a computer to run a plurality of simulated vehicles on a simulated road network including a plurality of intersections to search for an offset of a signal lamp at an intersection,
Computer
A first evaluation value calculation unit that calculates a first evaluation value related to the low speed simulation vehicle when the low speed simulation vehicle is driven;
A second evaluation value calculation unit for calculating a second evaluation value related to the high-speed simulation vehicle when the high-speed simulation vehicle is driven;
An offset search unit for searching for an offset such that the first evaluation value becomes an optimum value and the second evaluation value becomes an optimum value ;
A first calculator that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven;
The slow simulated vehicle and fast simulated vehicle to function as a travel calculation unit for calculating a travel time difference or travel speed difference when is traveling,
Calculating a first evaluation value relating to the low speed simulation vehicle based on at least one of the calculated first delay time and the first number of stops;
Calculating first evaluation values and the calculated travel time difference or the high-speed simulation computer program that calculates the second evaluation value of the vehicle based on the travel speed difference was. An offset search method by an offset search device for searching for an offset of an intersection signal lamp by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections,
A first evaluation value calculation unit calculating a first evaluation value related to the low speed simulation vehicle when the low speed simulation vehicle is driven;
A second evaluation value calculation unit calculating a second evaluation value related to the high-speed simulation vehicle when the high-speed simulation vehicle is driven;
An offset search unit searching for an offset such that the calculated first evaluation value becomes an optimum value and the calculated second evaluation value becomes an optimum value ;
Calculating at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven;
Look including the step of calculating the travel time difference or travel speed difference when caused to travel the slow simulated vehicle and fast simulation vehicle,
Calculating a first evaluation value relating to the low speed simulation vehicle based on at least one of the calculated first delay time and the first number of stops;
An offset search method for calculating a second evaluation value relating to the high-speed simulated vehicle based on the calculated first evaluation value and the calculated travel time difference or travel speed difference . An offset search device that searches for an offset of a signal lamp at an intersection by running a plurality of simulated vehicles on a simulated road network including a plurality of intersections,
A first evaluation value calculation unit that calculates a first evaluation value related to the low speed simulation vehicle when the low speed simulation vehicle is driven;
A second evaluation value calculation unit for calculating a second evaluation value related to the high-speed simulation vehicle when the high-speed simulation vehicle is driven;
An offset search unit for searching for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is an optimal value, and the second evaluation value calculated by the second evaluation value calculation unit is an optimal value;
A first calculator that calculates at least one of a first delay time and a first number of stops at each intersection when the low-speed simulated vehicle is driven;
A second calculator that calculates at least one of a second delay time and a second number of stops at each intersection when the high-speed simulated vehicle is driven,
The first evaluation value calculation unit includes:
A first evaluation value related to the low speed simulation vehicle is calculated based on at least one of the first delay time and the first number of stops calculated by the first calculation unit;
The second evaluation value calculation unit
Ri Citea to calculate a second evaluation value according to the high-speed simulation vehicle based on at least one of the second delay time and a second number of stops calculated by the second calculator,
The offset search unit
A first offset search unit that searches for an offset such that the first evaluation value calculated by the first evaluation value calculation unit is small;
A second offset search unit for searching for an offset so that the first evaluation value calculated by the first evaluation value calculation unit is within a predetermined range and the second evaluation value calculated by the second evaluation value calculation unit is large. When
With
The first offset search unit includes:
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is a minimum value,
The second offset search unit
The offset is searched so that the first evaluation value calculated by the first evaluation value calculation unit is not larger than the minimum value and the second evaluation value calculated by the second evaluation value calculation unit becomes a maximum value. some offset search device.

Images