JP7781631B2 - Road congestion reduction support device, road congestion reduction support method, and road congestion reduction support program - Google Patents

Description

Embodiments of the present invention relate to a road congestion reduction support device, a road congestion reduction support method, and a road congestion reduction support program.

Road traffic control systems acquire and analyze various information related to road traffic conditions, and by understanding congestion information, accident information, estimated times of arrival at destinations, etc., they ensure the safety and smooth flow of road traffic and provide drivers with support information that is useful in these areas. Furthermore, research into autonomous driving technology has progressed in recent years, with driving experiments of autonomous vehicles being carried out in various locations. Research is also underway to simulate the impact on traffic conditions when autonomous vehicles and non-autonomous vehicles (hereinafter referred to as manually driven vehicles) are driven together.

JP 2019-114094 A Japanese Patent Application Laid-Open No. 2020-95334

"Highway Traffic Control Technology Handbook," edited by the Highway Traffic Control Technology Handbook Editorial Committee, Denki Shoin, 2005, pp. 7, 8, 48, 121 "Study on road traffic situation evaluation method when autonomous vehicles coexist on expressways," Yoshikazu Oba, Hideki Ueno, 18th ITS Symposium 2020, 1-A-06 "Simulation of Traffic Jams Occurring in a Mixed Environment of Autonomous and Human-Driven Vehicles," by Toda Masaru and Takamatsu Atsuko, Proceedings of the Symposium on Traffic Flow and Self-Accelerated Particle Systems, Vol. 23, Published by the Traffic Flow Mathematics Research Group, 2017. http://traffic.phys.cs.is.nagoya-u.ac.jp/~mstf/article_list/mstf2017-20.html http://traffic.phys.cs.is.nagoya-u.ac.jp/~mstf/pdf/mstf2017-20.pdf

However, when the adoption rate of autonomous vehicles is low and autonomous vehicles with high control capabilities are mixed among many manually driven vehicles, the highly controlled driving of the autonomous vehicles may affect the driving of the manually driven vehicles. For example, the behavior of the autonomous vehicles may affect the driving pace of the manually driven vehicles, increasing the frequency of acceleration and deceleration of the manually driven vehicles, which may become one of the causes of traffic congestion. Therefore, if it were possible to realize road congestion reduction support that minimizes the impact of autonomous vehicles on the driving of manually driven vehicles even in traffic situations where the number of autonomous vehicles on the road is low compared to the number of manually driven vehicles on the road, it would be meaningful as it would enable smoother driving of vehicles.

A road congestion reduction support device according to an embodiment includes an information acquisition unit, a congestion probability determination unit, a congestion reduction support information generation unit, and an output unit. The information acquisition unit acquires traffic condition information for a controlled area and vehicle information including driving function information for vehicles traveling in the controlled area when the information is available. The congestion probability determination unit, when a mixture of autonomous and non-autonomous vehicles is traveling in the controlled area based on the vehicle information, determines whether a predetermined level of congestion or more will occur in the controlled area due to the operation of the autonomous vehicle based on the traffic condition information and the vehicle information. If it is determined that congestion will occur, the congestion reduction support information generation unit generates driving support information that changes the autonomous driving control content of the autonomous vehicle in accordance with the traffic conditions around the autonomous vehicle in order to reduce congestion. The output unit outputs the driving support information.

FIG. 1 is an exemplary schematic block diagram showing a road congestion reduction assistance device according to an embodiment and a road traffic control system capable of transmitting and receiving information to and from the road congestion reduction assistance device. FIG. 2 is an exemplary and schematic explanatory diagram showing the effects that may occur when a mixture of autonomously driven vehicles and manually driven vehicles that are the targets of support by the road congestion reduction support device according to the embodiment passes through a sag on a highway. FIG. 3 is an exemplary schematic block diagram illustrating details of the road congestion reduction assistance device according to the embodiment. FIG. 4 is an exemplary flowchart illustrating a process flow of the road congestion reduction assistance device according to the embodiment. FIG. 5 is an exemplary schematic diagram showing an analysis map that can be used by the road condition analysis unit of the road congestion reduction assistance device according to the embodiment.

Embodiments of the present invention will be described below with reference to the drawings. The configuration of the embodiments described below, as well as the actions and results (effects) achieved by said configurations, are merely examples and are not limited to the contents described below.

Figure 1 is an exemplary schematic block diagram showing a road congestion reduction assistance device 10 according to an embodiment and a road traffic control system 100 capable of sending and receiving information with the road congestion reduction assistance device 10.

The road traffic control system 100 mainly comprises an input-side configuration, a processing-side configuration, and an output-side configuration. The input-side configuration includes road-side sensors 102 and a road-to-vehicle communication system 104. The processing-side configuration includes a traffic control unit 106 and a database 108. The output-side configuration includes an information provider 110 and an other system information provider 112.

The roadside sensors 102 and the road-to-vehicle communication system 104 acquire various data that can be used for road control in areas subject to road control. In this embodiment, the area subject to control may be, for example, one of multiple sections of a highway, and is the area under the jurisdiction of the road traffic control system 100.

The roadside sensor 102 is, for example, a vehicle detector or roadside camera installed on the roadside. The roadside sensor 102 can acquire vehicle detector information such as traffic volume at the installation location, the speed of traveling vehicles, and time occupancy rate, and sequentially transmits the acquired information to the traffic control unit 106. Note that the vehicle speed of traveling vehicles may also be acquired by image processing using video obtained from a roadside camera, etc.

The road-to-vehicle communication system 104 can use, for example, ETC 2.0. The road-to-vehicle communication system 104 communicates with passing vehicles, acquires various information from passing vehicles, and sequentially transmits this information to the traffic control unit 106. The road-to-vehicle communication system 104 can also provide passing vehicles with various information output from the other system information provider 112. Compared to conventional ETC, ETC 2.0 is capable of sending and receiving larger amounts of information. Therefore, information useful for traffic control and driving can be sent and received smoothly. For example, it is possible to send and receive not only information about entering and exiting highway interchanges, but also route information, etc.

The road-to-vehicle communication system 104 can also acquire vehicle information including driving performance information used by the road congestion reduction assistance device 10. Possible driving performance information includes, for example, information on sudden braking. In this embodiment, it is assumed that the road-to-vehicle communication system 104 is capable of acquiring, as driving performance information, information indicating whether or not a vehicle is capable of autonomous driving. When the vehicle being communicated with is an autonomous vehicle, the road-to-vehicle communication system 104 can acquire autonomous driving status information. Possible autonomous driving status information includes autonomous driving parameter information, the passing position and passing time information of the autonomous vehicle obtained from the GPS (Global Positioning System), etc. It is also possible to assume that similar information is transmitted and received using, for example, a car navigation system installed in the autonomous vehicle.

The traffic control unit 106 acquires vehicle detector information (e.g., traffic volume, speed, time occupancy, etc.) from the roadside sensors 102 and vehicle information (autonomous driving status information, autonomous driving parameters, location information, passing time information, etc.) from the road-to-vehicle communication system 104. The traffic control unit 106 then generates information to be provided to information boards installed on the roadside or other locations in the controlled area via the information provider 110, and information to be provided to road-to-vehicle communication vehicles via the road-to-vehicle communication system 104. The traffic control unit 106 can also use the other system information provider 112 to generate information to be provided to road-to-vehicle communication systems and VICS (registered trademark) (Vehicle Information and Communication System) and other systems targeted at autonomous vehicles.

More specifically, the traffic control unit 106 creates driving time information from speed and road length, and generates information indicating congestion conditions based on speed information, for example. It also generates event information such as road construction information and accident information. Furthermore, the traffic control unit 106 is assumed to have the function of outputting driving support information generated based on various information acquired by the road congestion reduction support device 10 from the traffic control unit 106, such as vehicle detector information (traffic volume, speed, etc.) and vehicle information (autonomous driving status information, autonomous driving parameter information, location information, etc.), via the other system information provider 112. The driving support information generated by the road congestion reduction support device 10 will be described later.

The traffic control unit 106 functions on a central processing unit P1 (e.g., a computer resource such as a general-purpose PC (Personal Computer)) installed in the control center. A large monitor device M is installed in the control center, and multiple control personnel W seated at consoles N can efficiently perform various tasks such as control, operation, and management. The central processing unit P1 is composed of, for example, a central processing unit (CPU), read-only memory (ROM), random access memory (RAM), a storage unit such as a hard disk drive (HDD) or solid state drive (SSD), a communication interface, and an input/output interface. The central processing unit P1, which includes the traffic control unit 106, stores acquired information in a database 108 (DB). The traffic control unit 106 can then reuse the information stored in the database 108 as past data. The database 108 may be installed within the central processing unit P1 or, as shown in Figure 1, may be installed externally.

As described above, the information provider 110 provides various information generated by the traffic control unit 106 to information boards installed on the roadside or other locations in the controlled area, and displays the information to drivers of vehicles traveling in the controlled area, or provides the information to road-to-vehicle communication vehicles via the road-to-vehicle communication system 104.

Similarly, the other system information providing unit 112 provides information generated by the traffic control unit 106 and the road congestion reduction assistance device 10 to autonomous vehicles traveling in the controlled area, directly controlling the autonomous vehicles or providing driving assistance related to autonomous driving to the user using (driving) the autonomous vehicles. In this case, the other system information providing unit 112 may provide information via the road-to-vehicle communication system 104, or may provide information using other communication methods.

Next, we will explain the road congestion reduction support device 10. The road congestion reduction support device 10 is particularly effective in reducing congestion on expressways and other roads that do not have traffic lights. Therefore, in the following explanation, the controlled area will be described as, for example, an expressway.

The road congestion reduction support device 10 is realized, for example, on a personal computer P2. For example, when a controlled area contains a mixture of conventional vehicles, i.e., non-autonomous vehicles (manually driven vehicles), and autonomous vehicles, and the proportion of autonomous vehicles on the road is low (when the ratio of autonomous vehicles to manually driven vehicles is low), the road congestion reduction support device 10 provides driving support information to autonomous vehicles. For example, autonomous vehicles have higher control performance than manual vehicles operated by a driver, and are controlled to reduce control delays and excessive or insufficient control. In other words, autonomous vehicles detect their driving position, road conditions, etc., while achieving safe and efficient driving control, and drive in a way that minimizes discomfort or annoyance to passengers.

On the other hand, there may be cases where the driver of a manually driven vehicle traveling around an autonomous vehicle is unable to keep up with the autonomous vehicle's appropriate control. Furthermore, even if the autonomous vehicle's behavior is safe and appropriate, the driver of the manually driven vehicle may become unsettled by its behavior and may unnecessarily decelerate (for example, apply sudden braking) or change lanes. In other words, the high-performance control of an autonomous vehicle may affect the driving of surrounding manually driven vehicles (especially those traveling behind), resulting in traffic congestion.

The road congestion reduction assistance device 10 predicts such situations and, if necessary, creates information that prompts autonomous vehicles to drive autonomously with control performance equivalent to that of surrounding manually driven vehicles, or to temporarily switch from autonomous driving to manual driving in order to reduce the possibility of congestion.

Figure 2 is an exemplary and schematic explanatory diagram showing the effects that can occur when an autonomously driven vehicle AC, which is the target of support from the road congestion reduction assistance device 10, and a manually driven vehicle MC traveling around it pass through a sag RC on a highway. A sag RC is a depression where a downhill slope transitions into an uphill slope.

Figure 2 shows, for example, a convoy of five manually driven vehicles MC1-MC5, including one automatically driven vehicle AC, and the movement of each vehicle over time (timings T1-T8). Note that in the example shown in Figure 2, the automatically driven vehicle AC, which is currently operating automatically, is the third vehicle in the convoy. In other words, the ratio of automatically driven vehicles AC to manually driven vehicles MC is 1/6 (approximately 17%), indicating a low ratio of automatically driven vehicles AC. Also, in the example shown in Figure 2, manually driven vehicles MC1-MC5 and automatically driven vehicle AC enter a downhill section R2 from a flat section R1 of highway R, travel along a sag RC, pass an uphill section R3, pass the end point RT of the uphill section, and then travel again on flat section R4.

At timing T1, a convoy including autonomous vehicle AC (third from the front) is traveling on highway R, which has a sag RC (initial stage). At this point, autonomous vehicle AC is traveling on flat section R1.

Next, at timing T2, autonomous vehicle AC is traveling downhill section R2. Because autonomous vehicle AC is driving autonomously, it is slightly affected by downhill section R2 (accelerating its speed), but is quickly controlled to its target speed (for example, a speed set according to the flow of vehicles in the convoy on flat section R1, etc.). Note that manually driven vehicles MC1 and MC2 have a delay in responding to speed changes depending on the driver's skill, so the downhill section R2 causes an increase in speed. If the driver notices the increase in speed, they will apply the brakes after a while. In such a case, the illumination of the brake lights may induce following vehicles to apply the brakes.

Times T3 and T4 are the times when the autonomously driven vehicle AC passes through a sag RC where the road changes from a downhill section R2 to an uphill section R3. Even when driving through a sag RC, the autonomously driven vehicle AC is in autonomous driving mode and is therefore able to drive with high control performance. Therefore, even if the vehicle's speed accelerates or decelerates due to the influence of the sag RC, it can be controlled to the target speed with little reaction delay.

At timings T4 and T5, the autonomously driven vehicle AC is able to quickly achieve a target speed through automatic cruise control. In contrast, the vehicle ahead of the autonomously driven vehicle AC (manually driven vehicle MC2) is manually driven, and so when it enters uphill section R3 from sag RC, the driver may decelerate without realizing that it has entered uphill section R3. As a result, the speed reduction of the manually driven vehicle MC2 and other vehicles due to the influence of uphill section R3 may continue.

In this case, even if the automatically driven vehicle AC itself quickly recovers to the target speed, as shown at timing T5, it is quickly controlled (e.g., deceleration controlled) to achieve the same speed as the manually driven vehicle MC2 in front in order to maintain a sufficient distance from the manually driven vehicle MC2.

Furthermore, at timing T6, the convoy continues traveling and the autonomous vehicle AC reaches the top of the uphill section R3 and is about to reach the uphill end point RT.

In an autonomously driven vehicle AC or a manually driven vehicle MC, once the vehicle reaches the top of uphill section R3, the influence of uphill section R3 disappears, and the vehicle's speed may change slightly in the direction of accelerating. In this case, the driver of a manually driven vehicle MC2 or the like may have a delayed reaction and not realize that the vehicle's speed is increasing due to the disappearance of the influence of uphill section R3 (the accelerator pedal force remains at uphill section R3, and the vehicle speed increases because the influence of the uphill section disappears). In this case, the driver will be late in operating (maneuvering) the vehicle to return it to its previous speed (target speed) after the speed of the vehicle has increased.

In contrast, the automatic driving control of the automatically driven vehicle AC has almost no reaction delay like the driver of a manually driven vehicle MC2, and so the vehicle is controlled to the target speed with almost no acceleration (timing T7).

In other words, at timing T8, the autonomous driving control of the autonomous vehicle AC prevents the speed from increasing even when the influence of the uphill section R3 disappears at the uphill end point RT.

On the other hand, when the influence of the uphill section R3 disappears, the manually driven vehicle MC3 (the vehicle following the automatically driven vehicle AC) traveling behind the automatically driven vehicle AC may increase in speed due to the driver's delayed reaction, as described above.

As a result, the distance between the leading autonomous vehicle AC and the following manually driven vehicle MC3 becomes shorter, and the driver of manually driven vehicle MC3 may apply the brakes in an attempt to maintain an appropriate distance. This braking operation causes the brake lights to light up, and manually driven vehicle MC4 following manually driven vehicle MC3 may react sensitively and apply the brakes even though they are traveling uphill section R3. This braking operation is then transmitted to the following vehicles, causing the line of vehicles to slow down, which is one of the causes of traffic congestion.

On the other hand, if the convoy does not include an autonomous vehicle AC, all manually driven vehicles MC will be affected when they pass the end point RT of the uphill slope. At this time, there may be some difference in the magnitude of the braking force of each manually driven vehicle MC, but because there will be a delay in the driver's reaction, as mentioned above, it is thought that the driver of the following manually driven vehicle MC is unlikely to perform a braking operation due to the shortened inter-vehicle distance.

As such, compared to a convoy consisting only of manually driven vehicles MC, when a convoy also contains automatically driven vehicles AC and the proportion of automatically driven vehicles AC is low, it is thought that a chain reaction of braking caused by the automatic driving of automatically driven vehicles AC (such as speed control using high-performance control) is more likely to occur. In other words, there is an increased likelihood of a chain reaction of braking occurring, which is unlikely to occur with only manually driven vehicles MC.

When the phenomenon described above occurs in a line of vehicles traveling at a high density and on the brink of a traffic jam, even a slight braking operation can trigger the jam, increasing the likelihood of it causing a traffic jam.

A similar phenomenon may also occur when the autonomous vehicle AC makes a sudden lane change. In other words, because the autonomous driving performance of the autonomous vehicle AC is so good, if, for example, there is a fallen object or an accident vehicle on the highway R, the autonomous vehicle AC will react quickly and change lanes at the optimal position just before the accident, taking safety into consideration.

In this case, the driver of the manually driven vehicle MC may feel that the automatically driven vehicle AC in front has suddenly changed lanes, and an obstacle or other object may suddenly come into view, causing the driver to become confused and apply the brakes. The brake lights coming on as a result of this braking operation may cause the following manually driven vehicle MC to react sensitively and apply the brakes as well. This braking operation will propagate to the following vehicles in turn, just as in the area where the sag RC described above exists, resulting in the line of vehicles slowing down and becoming one of the causes of traffic congestion.

As described above, in traffic conditions where the ratio of automatically driven vehicles AC to manually driven vehicles MC is extremely small, when an automatically driven vehicle AC is likely to be the cause of congestion, the road congestion reduction assistance device 10 of this embodiment provides road congestion reduction assistance information to the automatically driven vehicle AC to change the content of its automated driving in order to avoid congestion caused by the automatically driven vehicle AC as much as possible or to reduce the congestion.

The road congestion reduction support information is, for example, information that encourages the automated driving control of the automated vehicle AC to be adjusted to match the driving level of the driver of the surrounding manually driven vehicle MC. For example, the road congestion reduction support information provided recommends changing the automated driving control to take into account the reaction delay of the driver of the manually driven vehicle MC, or temporarily disabling the automated driving control and switching to manual driving.

By having the autonomous vehicle AC change its autonomous driving control in accordance with the road congestion reduction support information, it becomes possible to suppress unintended driving behavior (such as excessive braking by the driver) of the manually driven vehicle MC, thereby contributing to reducing congestion.

The driving level of the driver of the surrounding manually driven vehicle MC can be stored in advance in the road congestion reduction assistance device 10, for example, by analyzing in advance the driving characteristics of the manually driven vehicle MC in the same control area in the past.

The road congestion reduction support information generated by the road congestion reduction support device 10 is provided, for example, to the traffic control unit 106 and output as necessary via the other system information provider 112. The road congestion reduction support information provided to the traffic control unit 106 may be stored in the database 108 and made reusable as past data.

Figure 3 is an exemplary schematic block diagram showing details of the road congestion reduction assistance device 10 according to an embodiment.

The road congestion reduction support device 10 includes a road information acquisition unit 12, a congestion probability determination unit 14, a congestion reduction support information generation unit 16, an output unit 18, etc. Furthermore, the road congestion reduction support device 10 includes a road condition analysis unit 20, a traffic flow simulation unit 22, a database 24, etc.

Note that the configuration in Figure 3 is an example, and for example, the congestion possibility determination unit 14 may include a road condition analysis unit 20, a traffic flow simulation unit 22, a database 24, etc. Note that Figure 3 explains an example in which the traffic flow simulation unit 22 is used to analyze road conditions.

The road congestion reduction assistance device 10 acquires traffic condition information for the controlled area of the expressway R, and vehicle information including driving function information for the autonomously driven vehicles AC and manually driven vehicles MC (vehicles) traveling in the controlled area. The road congestion reduction assistance device 10 acquires, for example, information acquired and managed by the road traffic control system 100, such as vehicle detector information and autonomously driven vehicle information, from the traffic control unit 106 of the road traffic control system 100.

Here, vehicle detector information can be acquired, such as traffic volume, speed, and time occupancy rate, which are measurements taken by roadside sensors 102 installed on the roadside. Furthermore, autonomous vehicle information can be acquired from the road-to-vehicle communication system 104, such as information about the autonomous vehicle AC itself, including whether the autonomous vehicle AC is operating autonomously, and the location and time at which the autonomous vehicle AC passed (information created from GPS data, etc.).

The road congestion reduction assistance device 10 may also connect to the road traffic control system 100 via a network and acquire this information as information managed by the traffic control unit 106.

When it is determined based on the vehicle information that the autonomous vehicle AC is included in the control area, the congestion possibility determination unit 14 determines, based on the traffic condition information and vehicle information, whether congestion of a predetermined level or higher (a level of congestion determined in advance by analyzing past traffic conditions) will occur in the control area due to the autonomous vehicle AC's travel. The congestion possibility determination unit 14 first provides the acquired information to the road condition analysis unit 20, which then analyzes the road conditions.

The road condition analysis unit 20 divides the control area into multiple divided areas (e.g., several hundred meters) based on traffic condition information and performs a vehicle speed analysis for each vehicle traveling in each divided area. The road condition analysis unit 20 then determines whether or not congestion will occur based on the acquired vehicle speed analysis results.

The rate of autonomous vehicles AC in the managed area of highway R is estimated from vehicle detector information and autonomous vehicle information. To estimate the rate of autonomous vehicles AC, the traffic volume and speed from the vehicle detector information are used to calculate the density of the relevant divided area (relevant section) from the relationship traffic volume = density x speed, and the number of autonomous vehicles AC in the divided area is obtained from the autonomous vehicle information, and from these, the rate of autonomous vehicles AC in the divided area can be calculated.

In addition, the road condition analysis unit 20 can calculate the current number of autonomously driven vehicles AC and the current number of manually driven vehicles MC based on this information.

The road condition analysis unit 20 can also estimate the position of the manually driven vehicle MC based on the number of autonomous vehicles AC, the positions of autonomous vehicles AC obtained as autonomous vehicle information, and the vehicle detector passing count and average speed. By providing this information to the traffic flow simulation unit 22 and performing a simulation, it is possible to create information about future traffic conditions (including, for example, congestion forecast information, such as the likelihood of future congestion).

The traffic flow simulation unit 22 adjusts and sets parameters, and sets and simulates the initial states, such as the position of the automatically driven vehicle AC (with time information when it passed that position) and the position of the manually driven vehicle MC (with time information when it passed that position), created by the road condition analysis unit 20. From the results of this simulation, the traffic flow simulation unit 22 performs simulation calculations to determine what state will transition from the initial setting state, and uses this to understand the traffic situation.

In particular, by clearly defining the differences between autonomous vehicles AC and manually driven vehicles MC and clearly simulating the impact of their coexistence, it will be possible to create information for understanding future traffic conditions when autonomous vehicles AC and manually driven vehicles MC coexist.

For example, the traffic flow simulation unit 22 can conduct a simulation in which the mixing rate of autonomously driven vehicles AC is low, and if it is found that the autonomous driving behavior of the autonomously driven vehicles AC affects the driving of the surrounding manually driven vehicles MC, and that in this case there is a possibility that the manually driven vehicles MC may perform unexpected braking operations (apply the brakes), it can analyze that this could be a factor in traffic congestion.

In this way, the traffic flow simulation unit 22 generates information about future traffic conditions (traffic simulation results). The road condition analysis unit 20 uses this information to analyze road conditions. Note that the road condition analysis unit 20 divides the control area (expressway R) into regions of several hundred meters each and calculates the mixing rate of autonomous vehicles AC, etc. By checking the traffic simulation results as future traffic condition information for that situation, the road condition analysis unit 20 can confirm how traffic conditions will change in the future.

If the traffic simulation results for a divided section show a decrease in speed, it is considered to be in the direction of congestion, indicating a tendency for congestion to occur (depending on how you look at it, this can also be considered a prediction of the possibility of future congestion (congestion prediction)).

Furthermore, if the ratio of autonomously driven vehicles AC to manually driven vehicles MC is low, the impact that autonomously driven vehicles AC have on surrounding manually driven vehicles MC may increase the risk of accidents or affect manually driven vehicles MC. As a result, if there is a possibility that manually driven vehicles MC may take deceleration actions that they would not normally take, these factors may lead to a tendency for congestion to occur.

If the speed in the divided area does not decrease, it can be analyzed as not being congested. For example, a speed of 40 km/h or less on intercity expressways and 20 km/h or less on urban expressways can be used as an indicator of congestion (predetermined level of congestion).

The congestion possibility determination unit 14 determines the possibility of congestion based on the results of analysis by the road condition analysis unit 20. Specifically, it checks the congestion trend for each section obtained by the road condition analysis unit 20, and determines that there is a possibility of congestion if congestion is on the rise.

When the congestion possibility determination unit 14 determines that there is a possibility of congestion in the divided area, the congestion reduction support information generation unit 16 generates driving support information for the autonomous vehicle AC to prevent congestion (to reduce congestion) based on that result and the analysis results of the road condition analysis unit 20.

As mentioned above, the generated driving assistance information is information that recommends that the autonomously driven vehicle AC in question cancel autonomous driving or operate autonomously with control performance equivalent to that of the manually driven vehicle MC, for example, when it is determined that the autonomous driving status of the autonomously driven vehicle AC is affecting the surrounding manually driven vehicle MC and that this effect may cause traffic congestion.

When recommending control performance equivalent to that of a manually driven vehicle MC, the congestion reduction support information generation unit 16 may create driving support information based on the current traffic conditions (such as the driving conditions of surrounding manually driven vehicles MC) obtained from the congestion reduction support information generation unit 16.

If the congestion possibility determination unit 14 determines that there is no possibility of congestion, the congestion reduction support information generation unit 16 may generate information indicating that there is no possibility of congestion reduction. Also, if it is determined that there is no possibility of congestion, driving support information may not be created.

The output unit 18 outputs the driving assistance information created by the congestion reduction assistance information generation unit 16 to the traffic control unit 106 in the road traffic control system 100. Regarding output, the road traffic control system 100 and the road congestion reduction assistance device 10 can be connected via a network, and information can be output via network communication. The output unit 18 can also provide information directly from the road traffic control system 100 to individual smartphones or individual car navigation systems, for example, via a web application.

Note that while Figure 3 shows an example in which the road condition analysis unit 20 performs road condition analysis using the traffic flow simulation unit 22, road condition analysis may also be performed based on various data stored in the database 24. Use of the database 24 will be described later.

Figure 4 is an exemplary flowchart showing the processing flow of the road congestion reduction assistance device 10 configured as described above.

First, in step S100, the road information acquisition unit 12 acquires information such as vehicle detector information and vehicle information about the autonomous vehicle AC acquired and managed by the road traffic control system 100 from the traffic control unit 106 in the road traffic control system 100 (information acquisition step, information acquisition process). As described above, the traffic control unit 106 acquires, with regard to the vehicle detector information, traffic volume, speed, and time occupancy rate, which are the measurement results of roadside sensors 102 installed on the roadside.

In addition, with regard to vehicle information, it is assumed that information about the autonomous vehicle AC itself (such as whether the autonomous vehicle is driving autonomously, and the position and time at which the autonomous vehicle AC passed) can be obtained through communication between the autonomous vehicle AC and the road-to-vehicle communication system 104. Furthermore, the road information acquisition unit 12 also acquires information about roads subject to control managed by the traffic control unit 106 (information about road shapes such as the managed section, section length, and number of lanes).

Next, in step S102, the road condition analysis unit 20 analyzes the situation of the controlled area (expressway R) based on the information acquired by the road information acquisition unit 12 (congestion possibility determination step, congestion possibility determination process). The road condition analysis divides the controlled area, expressway R, into regions of several hundred meters each, and analyzes whether there is congestion based on the speed and other information from vehicle detector information for each divided region. This analysis can be performed by the traffic flow simulation unit 22.

In addition, when determining whether or not there is a traffic jam, the road condition analysis unit 20 can obtain similar results by analyzing traffic conditions based on the acquired information using analysis maps created in advance and stored in the database 24, rule-based analysis, statistical models, machine learning models, similar pattern searches using past performance data, etc.

When using an analytical map, it is possible to first calculate traffic volume and speed information from a theoretical perspective in traffic engineering, and then create and use an analytical map CM like the one shown in Figure 5.

Figure 5 shows an example of how an analysis map can be created by calculating density from traffic volume and speed using the relationship traffic volume = density x speed, and then calculating the inter-vehicle distance assuming an average vehicle length (e.g., 7m including the distance between the front and rear of the vehicle). Based on this inter-vehicle distance, the system sets "reduction of congestion required," "warning required," or "no assistance information required."

By creating and using an analysis map in advance based on these theoretical characteristics, traffic conditions can be analyzed efficiently and easily. In this way, analysis using a road traffic condition analysis map created based on past performance data and the results of traffic flow simulations has the advantage of allowing analysis results to be obtained with relatively little processing.

Note that Figure 5 shows an example of how the degree of need for congestion reduction support information is obtained from a map. When considering this based on inter-vehicle distance, for example, if the inter-vehicle time is 2 seconds or more, support information is not required, if it is 1.5 to 2 seconds, a warning is required, and if it is 1.5 seconds or less, congestion reduction is required.

In the case of a rule-based approach, the analysis map can be implemented by formulating rules using well-known IF-THEN rules, etc. Furthermore, when using a statistical model, a statistical model can be created and used to calculate the congestion level in a given situation using past traffic volume and speed data (actual data). In this case, analysis results based on the statistical analysis of past data can also be obtained.

Furthermore, if there is a large amount of data available, it is possible to use a well-known machine learning model instead of a statistical model, and analyze traffic conditions using a model trained through machine learning. With machine learning models, it is relatively easy to perform additional training when new information is obtained, and it is relatively easy to create a model trained with the latest data.

When determining the possibility of congestion, similar traffic condition information similar to past traffic condition information for the area under control may be acquired, and the similar traffic condition information may be used to perform vehicle speed analysis under current traffic conditions. In this way, analysis using similar pattern search using past performance data has the effect of obtaining analysis results that are equivalent to conditions on similar dates and times in the past.

As mentioned above, if traffic simulation is available in the traffic flow simulation unit 22, a traffic simulation can be performed using the state (traffic volume, etc.) at the time of calculation as the initial state, and the results can be used as reference information for future traffic conditions to analyze the traffic situation.

Returning to FIG. 4, next, in step S104, the traffic congestion possibility determination unit 14 analyzes the traffic conditions on Expressway R using the road condition analysis unit 20, and then determines the possibility of congestion of a predetermined level or greater based on the traffic condition analysis results. For example, based on the obtained road traffic condition analysis results, the traffic congestion tendency and the degree of need for congestion reduction support information for each divided area are confirmed on the analysis map CM in FIG. 5. The traffic congestion possibility determination unit 14 determines that there is a possibility of congestion when the degree of need for congestion reduction support information indicates a need for caution or congestion reduction.

If it is determined in step S104 that there is a possibility of congestion (Yes in S104), in step S106, the congestion reduction support information generation unit 16 generates congestion reduction support information for reducing congestion (congestion reduction support information generation step, congestion reduction support information generation process). When there is a possibility of congestion, the congestion reduction support information is generated based on the analysis results of road traffic conditions to prevent congestion (to reduce congestion). Here, for example, when the analysis results of a traffic simulation of road traffic conditions confirm that the autonomous driving of the autonomous vehicle AC may be affecting the drivers of surrounding manually driven vehicles MC and it is determined that this impact may be a cause of congestion, the congestion reduction support information generation unit 16 generates congestion reduction support information for the autonomous vehicle AC, such as "information urging the cancellation of autonomous driving" or "information recommending autonomous driving with control performance equivalent to that of a manually driven vehicle."

If it is determined in the processing of step S104 that there is no possibility of traffic congestion (No in S104), "information indicating no possibility of traffic congestion" is generated.

Finally, in step S108, the created congestion reduction support information is output to the road traffic control system 100 via the information provider 110 so that it can be used by a central processing unit P1 or other device with traffic control functions (output step, output process). Note that if the congestion reduction support information is not required, a message to that effect is output.

In this way, the road congestion reduction assistance device 10 of this embodiment provides the generated congestion reduction assistance information to the autonomously driven vehicle AC when the autonomous driving behavior of the autonomously driven vehicle AC affects the driving behavior of the surrounding manually driven vehicles MC, potentially resulting in congestion, for example, in cases where the mixture rate of autonomously driven vehicles AC is low. The congestion reduction assistance information can prompt the autonomously driven vehicle AC to cancel autonomous driving, or enable the autonomous driving control situation (behavior of the autonomously driven vehicle AC under automatic control) to be automatically driven with the same performance (behavior) as the surrounding manually driven vehicles MC. This can contribute to reducing the possibility of congestion.

Note that the above-described embodiment illustrates a situation in which autonomously driven vehicles AC and manually driven vehicles MC coexist, where the number of autonomously driven vehicles AC present is very small compared to the number of manually driven vehicles MC present.

In other embodiments, the above-described technology is also possible in situations where at least one manually driven vehicle MC follows an autonomously driven vehicle AC. In other words, by ensuring that the high-performance control of the autonomously driven vehicle AC does not affect the following manually driven vehicle MC, it is possible to prevent the following manually driven vehicle MC from performing unnecessary braking operations or changing lanes, thereby contributing to reducing the burden on the driver of the following vehicle.

In addition, in the above-described embodiment, expressway R was shown as the controlled area. In other embodiments, the invention can be applied to other types of roads, such as general roads. Note that in the case of general roads, traffic lights and the like are present, so there is a possibility that congestion may be alleviated to a lesser extent than on expressway R. However, the highly sophisticated control of the autonomous vehicle AC on surrounding manually driven vehicles MC can have the effect of reducing the impact (such as an increase in the frequency of acceleration and deceleration of the vehicle speed). Reducing the increase in the frequency of acceleration and deceleration of the manually driven vehicle MC can contribute to suppressing the occurrence of congestion.

In the above-described embodiment, an example was described in which the road congestion reduction assistance device 10 is configured on a personal computer P2 separate from the road traffic control system 100. However, the same effects as those of the above-described embodiment can also be obtained when the functions of the road congestion reduction assistance device 10 are implemented as a program within the road traffic control system 100.

The road congestion reduction assistance program executed by the CPU of the road congestion reduction assistance device 10 of this embodiment may be configured to be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM, flexible disk (FD), CD-R, or DVD (Digital Versatile Disk).

Furthermore, the road congestion reduction support program may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the road congestion reduction support program executed in this embodiment may be configured to be provided or distributed via a network such as the Internet.

Although several embodiments of the present invention have been described above, the above embodiments and variations are merely examples and are not intended to limit the scope of the invention. The above embodiments can be implemented in various forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The above embodiments and variations thereof are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as set forth in the claims.

10 Road congestion reduction support device 12 Road information acquisition unit 14 Congestion probability determination unit 16 Congestion reduction support information generation unit 18 Output unit 20 Road condition analysis unit 22 Traffic flow simulation unit 24 Database 100 Road traffic control system 102 Road side sensor 104 Road-to-vehicle communication system 106 Traffic control unit 108 Database 110 Information provision unit 112 Other system information provision unit AC Automatically driven vehicles MC, MC1, MC2, MC3, MC4 Manually driven vehicle P1 Central processing unit RC Sag

Claims (10)

an information acquisition unit that acquires traffic condition information of a control target area and vehicle information including driving function information of a vehicle traveling in the control target area when the information is available;
a congestion possibility determination unit that, when it is determined based on the vehicle information that autonomous vehicles and non-autonomous vehicles are traveling in a mixed state in the control target area, determines based on the traffic condition information and the vehicle information whether congestion of a predetermined level or more will occur in the control target area due to the traveling of the autonomous vehicles; and
a congestion reduction support information generation unit that generates driving support information for changing the content of autonomous driving control of the autonomous vehicle in accordance with the traffic conditions around the autonomous vehicle in order to reduce the congestion when it is determined that the congestion will occur; and
an output unit that outputs the driving support information;
A road congestion reduction support device comprising: the congestion possibility determination unit acquires a mixture rate of the automatically driven vehicles in the control target area based on the traffic condition information and the vehicle information,
2. The road congestion reduction support device according to claim 1, wherein, when the mixing rate is equal to or less than a predetermined mixing rate, the congestion reduction support information generation unit generates, as the driving support information, information for changing the autonomous driving state of the autonomous vehicle to driving that corresponds to the driving state of the non-autonomously driving vehicles that are driving around the autonomous vehicle. The road congestion reduction support device described in claim 1 or 2, wherein the congestion possibility determination unit includes a road condition analysis unit that divides the control target area into multiple divided areas based on the traffic condition information and performs a vehicle speed analysis for each vehicle traveling in each divided area, and determines whether or not congestion will occur based on the acquired vehicle speed analysis results. The road congestion reduction support device according to claim 3, wherein the road condition analysis unit performs the vehicle speed analysis by a traffic flow simulation using the current traffic condition information. The road congestion reduction support device of claim 3, wherein the road condition analysis unit performs the vehicle speed analysis based on current traffic condition information using a statistical model generated based on past traffic condition information for the control target area. The road congestion reduction support device of claim 3, wherein the road condition analysis unit performs the vehicle speed analysis in current traffic conditions using an analysis map generated based on past traffic condition information for the control target area. The road congestion reduction support device described in claim 3, wherein the road condition analysis unit acquires similar traffic condition information similar to past traffic condition information for the control target area, and uses the similar traffic condition information to perform the vehicle speed analysis for current traffic conditions. The road congestion reduction support device of claim 3, wherein the road condition analysis unit performs the vehicle speed analysis in current traffic conditions using a machine learning model that has learned past traffic condition information for the control target area. an information acquisition step of acquiring traffic condition information of a control target area and vehicle information including driving function information of a vehicle traveling in the control target area;
a congestion possibility determination step for determining, based on the traffic condition information and the vehicle information, whether or not congestion of a predetermined level or more will occur in the control area due to the driving of the autonomous vehicles, if autonomous vehicles and non- autonomous vehicles are traveling in the control area based on the vehicle information;
a congestion reduction support information generation step of generating driving support information that changes the content of autonomous driving control of the autonomously driven vehicle in accordance with the traffic conditions around the autonomously driven vehicle in order to reduce the congestion when it is determined that the congestion will occur;
an output step of outputting the driving support information;
A road congestion reduction support method comprising: an information acquisition process for acquiring traffic condition information of a control target area and vehicle information including driving function information of vehicles traveling in the control target area;
a congestion possibility determination process that, when it is determined based on the vehicle information that autonomous vehicles and non-autonomous vehicles are traveling in a mixed state in the control target area, determines whether congestion of a predetermined level or more will occur in the control target area due to the traveling of the autonomous vehicles, based on the traffic condition information and the vehicle information; and
a congestion reduction support information generation process that generates driving support information for changing the content of autonomous driving control of the autonomously driven vehicle in accordance with the traffic conditions around the autonomously driven vehicle in order to reduce the congestion when it is determined that the congestion will occur; and
an output process for outputting the driving support information;
A road congestion reduction support program that runs the above on a computer.