JP7775843B2 - Mobility support method and mobility support system for mobile objects - Google Patents

Description

This disclosure relates to methods and systems for assisting the movement of mobile objects such as pedestrians.

JP 2006-233503 A discloses a road lighting system attached to existing streetlights or utility poles. This conventional system includes a motion sensor. When a pedestrian is detected by the motion sensor, the conventional system emits multiple spotlights along the pedestrian's direction of travel. The illumination of the multiple spotlights is achieved by multiple illumination devices that make up the conventional system. The multiple spotlights emitted from these illumination devices are reflected by the road surface. Therefore, the conventional system can guide pedestrians along the direction of the array of multiple road surface reflected lights.

In addition to JP 2006-233503 A, examples of documents that demonstrate the state of the art in the technical field related to this disclosure include WO 2019/146008 A and JP 2014-13524 A.

Japanese Patent Application Laid-Open No. 2006-233503 International Publication No. 2019/146008 JP 2014-13524 A

However, in conventional systems, human presence sensors only detect whether a person is present within their detection range, and do not distinguish between multiple pedestrians. As a result, there is a limit to the number of pedestrians that can be guided simultaneously. Furthermore, in conventional systems, the direction of the spotlight emitted by each illumination device is fixed. As a result, the light reflected from the road surface can only encourage pedestrians to move in a fixed direction.

We are considering a method of placing multiple road studs on the road surface and lighting these road studs. With this method, in situations where multiple pedestrians and cyclists are passing by, it is possible to simultaneously guide these moving objects, and it has the potential to encourage each moving object to move in a direction other than their own forward direction. However, on the other hand, in such situations, it is expected that the direction of movement of these moving objects will not be constant. Therefore, there is room for development of a method of simultaneously guiding multiple moving objects while ensuring their traffic safety.

One objective of the present disclosure is to provide technology that can simultaneously guide multiple moving objects while ensuring traffic safety for these objects.

A first aspect of the present disclosure is a method for assisting movement of a moving object, which has the following features.
The method includes the steps of recognizing a moving object included in an image of an infrastructure camera, predicting a future trajectory of the moving object based on the recognition information of the moving object, and outputting an assistance display from a road surface lighting device that assists the movement of the moving object based on the future trajectory.
The method further includes a step of determining whether the future trajectories of the moving bodies predicted in the step of predicting the future trajectories will intersect if the recognition information of the moving bodies includes recognition information of a first and a second moving body, and a step of generating an intersection avoidance trajectory for at least one of the first and second moving bodies if it is determined that the future trajectories of the first and second moving bodies will intersect.
When the intersection avoidance trajectory is generated, in the step of outputting the assistance indication, an assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device.

A second aspect of the present disclosure is a system for assisting movement of a moving object, which has the following features.
The system includes a road lighting device, an infrastructure camera, and a processor. The road lighting device illuminates a road surface. The infrastructure camera captures an image of the road surface. The processor is configured to perform various processes.
The processor is configured to perform the following processes: recognize a moving object included in an image of the infrastructure camera; predict a future trajectory of the moving object based on the recognition information of the moving object; and cause the road surface lighting device to output an assistance display that assists the movement of the moving object based on the future trajectory.
The processor is further configured to perform, when the recognition information of the moving bodies includes recognition information of a first and a second moving body, a process of determining whether the future trajectories of these moving bodies predicted in the process of predicting the future trajectories will intersect, and, when it is determined that the future trajectories of the first and second moving bodies will intersect, a process of generating an intersection avoidance trajectory for at least one of the first and second moving bodies.
When the intersection avoidance trajectory is generated, in the process of causing the road surface lighting device to output the assistance indication, the assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device.

According to the present disclosure, a road surface lighting device outputs an assistance display to assist the movement of a moving object based on the future trajectory of the moving object contained in an image captured by an infrastructure camera. Furthermore, when the object of movement assistance includes a first and a second moving object, it is determined whether the future trajectories of these moving objects will intersect. Furthermore, when it is determined that these future trajectories will intersect, an intersection avoidance trajectory is generated for at least one of the first and second moving objects. Then, when an intersection avoidance trajectory is generated, an assistance display corresponding to this intersection avoidance trajectory is output from the road surface lighting device. Therefore, it is possible to simultaneously guide multiple moving objects while ensuring their traffic safety.

FIG. 1 is a diagram illustrating an outline of an embodiment. 10A and 10B are diagrams illustrating a first example of light emission of an LED tack. 10A and 10B are diagrams illustrating a second example of light emission of the LED stud. FIG. 10 is a diagram illustrating an example of a countermeasure when two future trajectories intersect. FIG. 10 is a diagram illustrating another example of a countermeasure when two future trajectories intersect. 10A and 10B are diagrams illustrating examples of measures to be taken when a generated trajectory intersects with a future trajectory of another moving object. 1 is a diagram illustrating a configuration example of a travel assistance system according to an embodiment. 10 is a flowchart illustrating a flow of processing that is performed in a server and that is particularly related to the embodiment. 10 is a flowchart illustrating a flow of processing that is performed in a server and that is particularly related to the embodiment.

Embodiments of the present disclosure will be described below with reference to the drawings. Note that in each drawing, identical or corresponding parts will be designated by the same reference numerals, and their description will be simplified or omitted.

1. Overview Fig. 1 is a diagram illustrating an overview of an embodiment. Fig. 1 depicts a road 1. Examples of the road 1 include a road for pedestrians, and a road for pedestrians as well as moving objects that move at a slower speed than automobiles, such as wheelchairs, self-propelled robots, and bicycles. The shape and width of the road 1 are not particularly limited.

A plurality of LED studs 2 are installed on the road surface of road 1. In the example shown in Figure 1, these LED studs 2 are installed at equal intervals. However, the arrangement of these LED studs 2 is not limited to this example. Furthermore, there is no particular limit to the number of LED studs 2 installed per unit area. Each LED stud 2 illuminates the area around its location, and corresponds to an example of a "road surface lighting device" of the present disclosure. Other examples of road surface lighting devices include a device that illuminates the road surface of road 1 with light to brighten the illumination range, and a device that projects an image onto the road surface of road 1 to brighten the projection range.

For ease of explanation, when referring to a specific one of the multiple LED tacks 2 shown in Figure 1, it will be described as "LED tack 2 (x, y)" using the coordinates (x, y). For example, the lower left LED tack 2 (x, y) will be described as LED tack 2 (1, 1), and the upper left LED tack 2 (x, y) will be described as LED tack 2 (6, 1). Furthermore, the lower left LED tack 2 (x, y) will be described as LED tack 2 (1, 13), and the upper left LED tack 2 (x, y) will be described as LED tack 2 (6, 13).

A number of infrastructure cameras 3 are installed around the road 1. Each infrastructure camera 3 is a camera installed on a structure around the road 1 (for example, a road structure, or a facility structure such as the ceiling, pillar, or wall of a parking lot or factory). Each infrastructure camera 3 also captures an image of a predetermined range set for each infrastructure camera 3. The predetermined image capture range includes, for example, the road surface of the road 1. Part or all of the predetermined image capture range of one infrastructure camera 3 may overlap with that of another infrastructure camera 3.

The server 4 manages the mobility assistance system. The server 4 communicates individually with multiple LED studs 2. In communication with each LED stud 2, the server 4 transmits to each LED stud 2 assistance display information LUM(x,y) that assists the movement of the mobile object. The assistance display information LUM(x,y) includes, for example, instruction information for the LED stud 2(x,y). Examples of instruction information include light emission color information and brightness information. The LED stud 2(x,y) emits light based on the instruction information for the LED stud 2(x,y).

The server 4 also communicates individually with multiple infrastructure cameras 3. In communication with each infrastructure camera 3, the server 4 receives camera information CAM from each infrastructure camera 3. The camera information CAM includes, for example, ID information of the infrastructure camera 3 that sent the camera information CAM and image information acquired by the infrastructure camera 3. The image information may be a video or a still image. If the infrastructure camera 3 performs object recognition processing (described below), information on the recognition results obtained by this object recognition processing may be included in the image information.

Figure 2 is a diagram illustrating a first example of light emission from LED stud 2 (x, y). Figure 2 depicts moving objects 5 and 6 (pedestrians) moving on road 1. Moving objects 5 and 6 are recognized based on images captured by infrastructure camera 3. Trajectory TR5 is the future trajectory of moving object 5 predicted based on the recognition information of moving object 5. Trajectory TR6 is the future trajectory of moving object 6 predicted based on the recognition information of moving object 6. The lengths of trajectories TR5 and TR6 correspond to the distances that moving objects 5 and 6 are predicted to travel a few seconds (2 to 4 seconds) from the current time. This length is calculated, for example, based on the movement speed and direction of moving objects 5 and 6.

In the example shown in FIG. 2, before trajectories TR5 and TR6 are predicted, LED studs 2(x,y) emit light of the same color and at the same level of brightness. When trajectory TR5 is predicted, LED studs 2(x,y) located around trajectory TR5 (i.e., (x,y) = (2,2) to (2,4) and (3,2) to (3,4)) emit light of a specific color and at a high level of brightness. This specific color may be the same as the color of trajectory TR5 before prediction, or it may be a different color. However, the brightness level of this specific color is set to a higher level than the brightness level of trajectory TR5 before prediction. For example, consider a case where brightness levels are expressed in five levels and the brightness level of trajectory TR5 before prediction is "Level 1 to 2." In this case, the brightness level of the specific color is set to "Level 3 to 4."

When trajectory TR6 is predicted, the LED studs 2 (x, y) located around this trajectory TR6 (i.e., (x, y) = (5, 10) to (5, 12) and (6, 10) to (6, 12)) emit light in a specific color at a high level of brightness. This specific color may be the same as the color of trajectory TR6 before prediction, or it may be a different color. However, this specific color is set to a different color from that of the LED studs 2 located around trajectory TR5. Furthermore, the brightness level of this specific color is set to a higher level than the brightness level of trajectory TR6 before prediction.

Figure 3 is a diagram illustrating a second example of light emission from LED stud 2 (x, y). The example shown in Figure 3 is taken at a time slightly later than the example shown in Figure 2, and depicts moving body 5. Figure 3 also depicts moving body 7 (a bicycle). As with the example shown in Figure 2, moving body 5 is recognized based on images captured by infrastructure camera 3. As with moving body 5, moving body 7 is also recognized based on images captured by infrastructure camera 3. Trajectory TR7 is the future trajectory of moving body 7 predicted using the recognition information of moving body 7. Trajectory TR7 is longer than TR5 because moving body 7 is moving at a higher speed than moving body 5.

In the example shown in FIG. 3, the LED studs 2(x,y) located around the periphery of the track TR5 (i.e., (x,y) = (2,7), (2,8), (3,6) to (3,9), (4,7), and (4,8)) emit light in a specific color with a high level of brightness. This specific color is the same as the specific color of the LED studs 2 located around the periphery of the track TR5 described in FIG. 2. Furthermore, the brightness level of this specific color is the same as the specific color of the LED studs 2 located around the periphery of the track TR5 described in FIG. 2.

In the example shown in FIG. 3, upon prediction of trajectory TR7, LED studs 2(x,y) located around this trajectory TR7 (i.e., (x,y) = (3,1) to (3,5) and (4,1) to (4,5)) emit light in a specific color with a high level of brightness. This specific color may be the same as the color of trajectory TR7 before prediction, or it may be a different color. However, this specific color is set to a different color from that of the LED studs 2 located around trajectory TR5.

As shown in Figures 2 and 3, in this embodiment, when a moving object is recognized in an image acquired by the infrastructure camera 3, the future trajectory of the moving object is predicted. Then, LED studs 2 (x, y) located around this future trajectory are caused to emit light of a specific color and at a high level of brightness. When using a device that irradiates light onto the surface of road 1, it is conceivable to irradiate the road surface with a strip of light extending along the future trajectory. Furthermore, when using a device that projects an image onto the surface of road 1, it is conceivable to project an image of any shape corresponding to the future trajectory onto the road surface.

Consider a situation where multiple moving bodies are traveling along road 1. In this case, the illumination of LED studs 2(x,y) described in Figures 2 and 3 is thought to be useful in preventing collisions between these moving bodies (including excessive proximity; the same applies below). This is because the illumination of LED studs 2(x,y) located around the future trajectory of each moving body makes it possible to recognize the approach of other moving bodies. However, in the example shown in Figure 3, moving body 7 approaches moving body 5 from behind. Therefore, from the perspective of moving body 5, it is difficult to notice the illumination of LED studs 2(x,y) located around TR7. Therefore, for example, if moving body 5 suddenly changes direction, there is a possibility that moving body 5 will collide with moving body 7.

In this embodiment, when multiple moving bodies are recognized, it is determined whether a collision will occur between these moving bodies. This collision determination is made based on multiple future trajectories predicted for each of the multiple moving bodies. For example, when there are two moving bodies, it is determined whether the two future trajectories will intersect. If these future trajectories intersect, it can be determined that a collision will occur. When there are three or more moving bodies, it is determined whether two of the three or more future trajectories will intersect. If two future trajectories intersect, it can be determined that a collision will occur between the two moving bodies whose intersecting future trajectories are predicted.

When determining whether a collision has occurred, the length of the future trajectory may be changed. That is, as already mentioned, the length of the future trajectory for illuminating the LED stud 2 corresponds to the distance traveled over several seconds (2 to 4 seconds). In this case, the length of the future trajectory for determining whether a collision has occurred may be set to a distance traveled over a period of time longer than these several seconds (for example, 2 to 6 seconds).

An example of a countermeasure when two future trajectories intersect will be described with reference to Figure 4. Figure 4 corresponds to an example in which the length of the future trajectories for collision determination has been extended in the example described in Figure 3. That is, trajectories TR5# and TR7# depicted in Figure 4 are future trajectories for collision determination between moving body 5 and moving body 7. As can be seen from Figure 4, trajectories TR5# and TR7# intersect. Therefore, in the example shown in Figure 4, the direction of trajectory TR7# is changed to generate trajectory TR7*, which does not intersect with trajectory TR5#. This trajectory TR7* corresponds to the "intersection avoidance trajectory" of this disclosure.

When trajectory TR7* is generated, LED studs 2(x,y) located around trajectory TR7* (i.e., (x,y) = (3,2) to (3,4), (4,1) to (4,6), and (5,3) to (5,5)) emit light in a specific color and at a high level of brightness. In other words, the LED studs 2(x,y) that emit light when trajectory TR7* is generated are different from the LED studs 2(x,y) that emit light when trajectory TR7 described in Figure 3 is generated. This makes it possible to prompt moving body 7 to correct its direction of travel to avoid a collision with moving body 5.

In this way, in the example shown in Figure 4, a trajectory TR7* is generated that does not intersect with trajectory TR5#. As already explained, moving unit 7 is approaching moving unit 5 from behind. Therefore, it is possible that moving unit 7 is aware of moving unit 5. Therefore, it is thought that a proposal to correct its direction of travel based on trajectory TR7* would be easily accepted by moving unit 7.

The proposal to correct the direction of travel can also be made to mobile unit 5 rather than mobile unit 7. Alternatively, the proposal to correct the direction of travel can be made to both mobile units 5 and 7. Figure 5 is a diagram illustrating another example of a measure to be taken when two future trajectories intersect. In the example shown in Figure 5, a trajectory TR7* is generated that does not intersect with trajectory TR5#. This is the same as the example shown in Figure 4. In the example shown in Figure 5, the direction of trajectory TR5# is further changed to generate a trajectory TR5* that does not intersect with trajectory TR7#. This trajectory TR5* also corresponds to the "intersection avoidance trajectory" of this disclosure.

When trajectory TR5* is generated, LED studs 2(x,y) located around trajectory TR5* (i.e., (x,y) = (1,7) to (1,9), (2,7) to (2,9), (3,7) and (3,8)) emit light in a specific color and at a high level of brightness. In other words, the LED studs 2(x,y) that emit light when trajectory TR5* is generated are different from the LED studs 2(x,y) that emit light when trajectory TR5 described in Figure 3 is generated. This makes it possible to prompt moving body 5 to correct its direction of travel to avoid a collision with moving body 7.

In this way, in the example shown in Figure 5, a trajectory TR7* that does not intersect with trajectory TR5# and a trajectory TR5* that does not intersect with trajectory TR7# are generated. Therefore, a correction to the direction of travel can be proposed for both moving bodies 5 and 7.

In this embodiment, when a trajectory TR* is generated that does not intersect with future trajectories TR# (i.e., trajectories TR5# and TR7#) for collision detection (i.e., trajectories TR5* and TR7*), it is determined whether this generated trajectory TR* intersects with the future trajectory of another moving body. The above-mentioned determination method is applied to determine whether this intersection occurs. Then, if it is determined that the trajectory TR* intersects with the future trajectory of another moving body, a trajectory TR** that does not intersect with this future trajectory is generated from the generated trajectory TR*.

An example of measures to be taken when a generated trajectory TR* intersects with the future trajectory of another moving body will be described with reference to Figure 6. Figure 6 corresponds to an example in which the length of the future trajectory TR# for collision detection has been extended in the example described in Figure 3. Figure 6 also depicts moving body 6, which was omitted in the description of Figure 3. Trajectories TR6# and TR7*# depicted in Figure 6 are future trajectories for collision detection between moving body 6 and moving body 7. Note that trajectory TR7*# is also the future trajectory of moving body 7 generated based on trajectory TR7*.

As can be seen from Figure 6, trajectories TR6# and TR7*# intersect. Therefore, in the example shown in Figure 6, the orientation of trajectory TR7*# is changed to generate trajectory TR7**, which does not intersect with trajectory TR6#. This trajectory TR7** also corresponds to the "intersection avoidance trajectory" (or more accurately, the "corrected intersection avoidance trajectory") of this disclosure.

When trajectory TR7** is generated, LED studs 2(x,y) located around trajectory TR7** (i.e., (x,y) = (3,2) to (3,6), (4,1) to (4,6), (5,4) and (5,5)) emit light of a specific color and at a high level of brightness. Therefore, according to the example shown in Figure 6, it is possible to prompt moving body 7 to correct its direction of travel to avoid a collision with moving bodies 5 and 6.

In this way, according to this embodiment, it is possible to simultaneously guide multiple moving objects while ensuring their traffic safety. The following describes this embodiment in more detail.

2. Mobility Support System 2-1. System Configuration Example Fig. 7 is a diagram showing a configuration example of a mobility support system according to an embodiment. In the example shown in Fig. 7, the mobility support system includes a group of LED studs 2m, a group of infrastructure cameras 3n, and a server 4. The group of LED studs 2m and the group of infrastructure cameras 3n communicate with the server 4 via a communication network 8. The communication network 8 is not particularly limited, and a wired or wireless network may be used.

The LED stud group 2m includes m LED studs 2 (m≧1). The installation locations of the m LED studs 2 are known. Each LED stud 2 operates according to the support display information LUM(x,y) received from the server 4, and illuminates the area around its location. The support display information LUM(x,y) includes, for example, instruction information for the LED stud 2(x,y). Examples of instruction information include luminous color information and brightness information. The luminous color information indicates the color emitted by the light source of the LED stud 2(x,y), such as purple, blue, green, yellow, orange, or red. The brightness information indicates the brightness level of the light source of the LED stud 2(x,y).

The infrastructure camera group 3n includes n infrastructure cameras 3 (n≧1). The installation locations of the n infrastructure cameras 3 are known. Each infrastructure camera 3 captures an image of a predetermined range set for each infrastructure camera 3. The predetermined image capture range is also known. Each infrastructure camera 3 transmits camera information CAM to the server 4. The camera information CAM includes, for example, the ID information of the infrastructure camera 3 that transmitted the camera information CAM and image information acquired by this infrastructure camera 3.

The server 4 comprises an information processing device 41 and a database 42. The information processing device 41 comprises at least one processor 43 and at least one memory 44. The processor 43 includes a CPU (Central Processing Unit). The memory 44 is a volatile memory such as a DDR memory, and expands various programs used in the various processes performed by the processor 43 and temporarily stores various information. The various information used by the processor 43 includes camera information CAM and map information MAP stored in the database 12.

The database 42 is formed in a specified storage device (e.g., a hard disk or flash memory). The database 42 stores map information MAP. The map information MAP includes data on the specifications of man-made objects such as buildings, roads, and railways (e.g., type, size, central position or latitude, longitude, and height of one or more representative positions), as well as data on the specifications of natural objects such as rivers and lakes. The map information MAP also includes data on the installation positions of the LED stud group 2m and the infrastructure camera group 3n. The map information MAP further includes data on the specifications of each infrastructure camera 3 included in the infrastructure camera group 3n. The data on the specifications of each infrastructure camera 3 includes information on the angle of view of each infrastructure camera (i.e., information on the specified imaging range).

8 and 9 are flowcharts illustrating the flow of processing particularly related to the embodiment, which is performed by the server 4 (processor 43). The processing routines shown in FIGS. 8 and 9 are repeatedly executed at predetermined intervals.

In the processing routine shown in FIG. 8, first, the camera information CAM is acquired (step S11). As already explained, the camera information CAM includes the ID information of the infrastructure camera 3 that transmitted the camera information CAM and the image information acquired by this infrastructure camera 3. The image information may also include information about the time when this image information was acquired.

Following the processing of step S11, object recognition processing is performed (step S12). In object recognition processing, moving objects included in the image captured by the infrastructure camera 3 are recognized. There are no particular limitations on the method for recognizing these moving objects, and any known method can be applied. An example of a known method is an object recognition method using a machine learning model. When a moving object is recognized by object recognition processing, identification information is assigned to the recognized moving object. Examples of this identification information include number information assigned to each moving object, type information of the recognized moving object (e.g., pedestrian, wheelchair, self-propelled robot, bicycle, etc.), and feature information of the recognized moving object.

Following the processing of step S12, it is determined whether or not a moving object has been recognized (step S13). If identification information has been generated in the processing of step S12, the determination result of step S13 is positive. If the determination result of step S13 is positive, tracking processing of the moving object is performed (step S14). This tracking processing of the moving object is not particularly limited, and known methods can be applied. An example of a known method is a method (re-identification) of associating a moving object recognized in multiple frames with the moving object based on feature information of the moving object.

By performing the tracking process, the moving speed and moving direction of the moving object recognized in the process of step S12 are calculated. Tracking information is generated for the moving object whose moving speed and moving direction have been calculated. This tracking information is stored in memory 42, for example, combined with the identification information of the moving object.

In the processing routine shown in Figure 9, tracking information is first obtained (step S21). Tracking information is generated for each moving object. Therefore, in the processing of step S21, tracking information for all moving objects present on road 1 is obtained.

Following the processing of step S21, trajectory TR and trajectory TR# are calculated (step S22). The calculation of trajectory TR and trajectory TR# is performed for each moving object based on the tracking information (movement speed and movement direction) acquired in the processing of step S21. The length of trajectory TR corresponds to the distance the moving object is predicted to travel several seconds from the current time (2 to 4 seconds from now). Trajectory TR# is longer than trajectory TR. The length of trajectory TR# corresponds to the distance the moving object is predicted to travel several seconds from the current time (e.g., 2 to 6 seconds from now). Note that the calculation of trajectory TR# does not have to be performed in the processing of step S22. In this case, the processing of step S23 is performed based on the trajectory TR calculated in the processing of step S22.

Following the processing of step S22, it is determined whether the two trajectories TR# intersect (step S23). For example, if only one trajectory TR# is generated in the processing of step S22, the determination result of step S23 will be negative. Also, for example, even if two or more trajectories TR# are generated in the processing of step S22, if none of these trajectories TR# intersect, the determination result of step S23 will be negative. If the determination result of step S23 is negative, the processing of step S24 is performed. On the other hand, if the determination result is positive, the processing of step S25 is performed.

In the processing of step S24, support display information LUM(x, y) is generated based on at least one trajectory TR generated in the processing of step S22. The support display information LUM(x, y) includes, for example, instruction information for the LED stud 2(x, y). Examples of instruction information include luminous color information and brightness information. If two or more trajectories TR are generated in the processing of step S22, luminous color information is generated so that the luminous color of the LED stud 2(x, y) located around one trajectory TR is different from the luminous color of the LED stud 2(x, y) located around another trajectory TR.

It is desirable that the color assigned as the luminous color of the LED studs 2(x, y) located around a certain track TR remain assigned without change until the moving object corresponding to this track TR is no longer recognized on the road 1. Therefore, when generating the luminous color information, it is desirable that colors such as purple, blue, green, yellow, orange, and red are assigned to each track TR in order, according to the order in which moving objects are recognized on the road 1.

The brightness level included in the brightness information is higher than the brightness level of LED studs 2(x,y) that are not located around the track TR. Here, the brightness level of LED studs 2(x,y) that are not located around the track TR is set, for example, according to the ambient illuminance of the road surface of road 1. For example, consider a situation where brightness levels are expressed in five levels and the ambient illuminance of the road surface of road 1 is low (for example, at night). In this case, the brightness level of LED studs 2(x,y) that are not located around the track TR is set to "Level 2-3." On the other hand, when street lights installed on road 1 are lit even at night, the brightness level is set to "Level 1-2."

In step S25, trajectories TR* and TR*# are generated. Trajectory TR* is generated for at least one of the two trajectories TR# determined to intersect in step S23. Like trajectory TR*, trajectory TR*# is generated for at least one of the two trajectories TR# determined to intersect in step S23. Note that if trajectory TR# was not calculated in step S22, trajectory TR*# does not need to be calculated in step S25. In this case, step S27 is performed based on the trajectory TR* calculated in step S25 and another trajectory TR# calculated in step S22.

Following the processing of step S25, it is determined whether the trajectory TR*# intersects with another trajectory TR# (step S26). The processing of step S26 is basically the same as the processing of step S23. The difference between the processing of these steps lies in the trajectory that is the subject of the intersection determination. That is, in the processing of step S26, it is determined whether the trajectory TR*# calculated in step S25 intersects with another trajectory TR# calculated in the processing of step S22. Note that the other trajectory TR# is a trajectory TR# different from the two trajectories TR# that were the subject of the intersection determination in the processing of step S23. If the determination result of step S26 is positive, the processing of step S27 is performed. On the other hand, if the determination result is negative, the processing of step S28 is performed.

In step S27, the trajectory TR* is modified to generate a trajectory TR**. The modified trajectory TR* is the trajectory TR* that was generated together with the trajectory TR*# when the trajectory TR*# that was determined to intersect with another trajectory TR# in step S26 was generated in step S25.

In the processing of step S27, trajectory TR**# may be generated along with trajectory TR**. In this case, the processing of step S26 may be performed for trajectory TR**#. In other words, a determination may be made as to whether trajectory TR**# generated along with trajectory TR** intersects with another trajectory TR#. By repeating the processing of step S27 and step S26, it is possible to generate an appropriate trajectory TR** that does not intersect with another trajectory TR#.

In the processing of step S27, instead of correcting the trajectory TR*, another trajectory TR may be corrected. Both the trajectory TR* and the another trajectory TR may also be corrected.

In step S28, support display information LUM(x, y) is generated based on the trajectory TR* generated in step S25 or the trajectory TR* corrected in step S27. The processing in step S28 is essentially the same as the processing in step S24.

Following the processing of step S28, the support display information LUM(x, y) is sent to each LED stud 2 (step S29). Upon receiving the support display information LUM(x, y), the LED stud 2(x, y) emits light based on the instruction information contained in this information.

1 Road 2 LED stud 2m LED stud group 3 Infrastructure camera 3n Infrastructure camera group 4 Server 5, 6, 7 Mobile object 41 Information processing device 42 Database 43 Processor 44 Memory TR5, TR6, TR7, TR5*, TR7*, TR7**, TR5#, TR6#, TR7#, TR7*# Track CAM Camera information LUM(x, y) Support display information

Claims (6)

A method for supporting movement of a mobile object , performed by a server, comprising:
A step of recognizing a moving object included in an image of an infrastructure camera;
predicting a future trajectory of the moving object based on the recognition information of the moving object;
a step of outputting an assistance display for assisting the movement of the moving object from a road surface lighting device based on the future trajectory;
Including,
a step of determining whether the future trajectories of the moving bodies predicted in the step of predicting the future trajectories intersect when the recognition information of the moving bodies includes recognition information of a first moving body and a second moving body;
generating an intersection avoidance trajectory for at least one of the first and second moving bodies when it is determined that the future trajectories of the first and second moving bodies will intersect;
Further comprising:
When the intersection avoidance trajectory is generated, in the step of outputting the assistance indication, an assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device ;
a step of determining whether or not the future trajectory of the third moving body predicted in the step of predicting the future trajectory intersects with the intersection avoidance trajectory for at least one of the first and second moving bodies when the recognition information of the moving body includes recognition information of a third moving body;
a step of correcting the crossing avoidance trajectory that intersects with the future trajectory of the third moving body when it is determined that the future trajectory of the third moving body and the crossing avoidance trajectory for at least one of the first and second moving bodies intersect;
Further comprising:
a step of outputting the assistance display when the intersection avoidance trajectory that intersects with the future trajectory of the third moving body is corrected, wherein the road surface lighting device outputs an assistance display corresponding to the corrected intersection avoidance trajectory . A method for supporting movement of a mobile object, performed by a server, comprising:
A step of recognizing a moving object included in an image of an infrastructure camera;
predicting a future trajectory of the moving object based on the recognition information of the moving object;
outputting, from a road surface lighting device, an assistance display that assists movement of the moving object based on the future trajectory;
Including,
a step of determining whether the future trajectories of the moving bodies predicted in the step of predicting the future trajectories intersect when the recognition information of the moving bodies includes recognition information of a first moving body and a second moving body;
generating an intersection avoidance trajectory for at least one of the first and second moving bodies when it is determined that the future trajectories of the first and second moving bodies will intersect;
Further comprising:
When the intersection avoidance trajectory is generated, in the step of outputting the assistance indication, an assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device;
a step of determining whether or not the future trajectory of the third moving body predicted in the step of predicting the future trajectory intersects with the intersection avoidance trajectory for at least one of the first and second moving bodies when the recognition information of the moving body includes recognition information of a third moving body;
generating an intersecting avoidance trajectory for the third moving body when it is determined that the future trajectory of the third moving body and the intersecting avoidance trajectory for at least one of the first and second moving bodies will intersect;
Further comprising:
When the intersection avoidance trajectory for the third moving body is corrected, in the step of outputting the assistance display, an assistance display corresponding to the corrected intersection avoidance trajectory for the third moving body is output from the road surface lighting device.
A method for supporting movement of a moving body, comprising: 3. The method of claim 1 or 2 ,
a step of outputting an assistance display, in which if the crossing avoidance trajectory is generated for at least one of the first and second moving bodies in the step of generating the crossing avoidance trajectory, the color of the assistance display that assists the movement of the first and second moving bodies is different from that of the assistance display corresponding to the crossing avoidance trajectory for at least one of the first and second moving bodies. 3. The method of claim 1 or 2,
a step of outputting the support display in which, when the crossing avoidance trajectory is generated for both the first and second moving bodies in the step of generating the crossing avoidance trajectory, the color of the support display corresponding to the crossing avoidance trajectory for the first moving body is different from that of the support display corresponding to the crossing avoidance trajectory for the second moving body. A system for supporting movement of a moving body,
Infrastructure cameras installed around the road,
a road surface lighting device that illuminates the road surface;
a processor configured to perform various processes;
Equipped with
the processor:
A process of recognizing a moving object included in an image of the infrastructure camera;
A process of predicting a future trajectory of the moving object based on the recognition information of the moving object;
a process of causing the road surface lighting device to output an assistance display for assisting the movement of the moving object based on the future trajectory;
configured to:
The processor further comprises:
a process of determining whether the future trajectories of the moving bodies predicted in the process of predicting the future trajectories intersect with each other when the recognition information of the moving bodies includes recognition information of a first moving body and a second moving body;
a process of generating an intersection avoidance trajectory for at least one of the first and second moving bodies when it is determined that the future trajectories of the first and second moving bodies will intersect;
configured to:
When the intersection avoidance trajectory is generated, in the process of causing the road surface lighting device to output the assistance indication, an assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device ,
The processor further comprises:
a process of determining whether or not the future trajectory of the third moving body predicted in the process of predicting the future trajectory intersects with the intersection avoidance trajectory for at least one of the first and second moving bodies when the recognition information of the moving body includes recognition information of a third moving body;
a process of correcting the intersection avoidance trajectory that intersects with the future trajectory of the third moving body when it is determined that the future trajectory of the third moving body and the intersection avoidance trajectory for at least one of the first and second moving bodies intersect;
configured to:
a process of outputting the assistance display, in which when the intersection avoidance trajectory that intersects with the future trajectory of the third moving body is corrected, an assistance display corresponding to the corrected intersection avoidance trajectory is output from the road surface lighting device . A system for supporting movement of a moving body,
Infrastructure cameras installed around the road,
a road surface lighting device that illuminates the road surface;
a processor configured to perform various processes;
Equipped with
the processor:
A process of recognizing a moving object included in an image of the infrastructure camera;
A process of predicting a future trajectory of the moving object based on the recognition information of the moving object;
a process of causing the road surface lighting device to output an assistance display for assisting the movement of the moving object based on the future trajectory;
configured to:
The processor further comprises:
a process of determining whether the future trajectories of the moving bodies predicted in the process of predicting the future trajectories intersect with each other when the recognition information of the moving bodies includes recognition information of a first moving body and a second moving body;
a process of generating an intersection avoidance trajectory for at least one of the first and second moving bodies when it is determined that the future trajectories of the first and second moving bodies will intersect;
configured to:
When the intersection avoidance trajectory is generated, in the process of causing the road surface lighting device to output the assistance indication, an assistance indication corresponding to the intersection avoidance trajectory is output from the road surface lighting device,
The processor further comprises:
a process of determining whether or not the future trajectory of the third moving body predicted in the process of predicting the future trajectory intersects with the intersection avoidance trajectory for at least one of the first and second moving bodies when the recognition information of the moving body includes recognition information of a third moving body;
a process of generating an intersection avoidance trajectory for the third moving body when it is determined that the future trajectory of the third moving body and the intersection avoidance trajectory for at least one of the first and second moving bodies will intersect;
configured to:
When the intersection avoidance trajectory for the third moving body is corrected, in the process of outputting the assistance display, an assistance display corresponding to the corrected intersection avoidance trajectory for the third moving body is output from the road surface lighting device.
A mobility support system for a mobile object.