Method and program for identifying actual position of observation target

Abstract

To provide a technology capable of improving reproduction accuracy of a traffic situation.SOLUTION: A position specifying method for observing a relatively moving observation target by an observation device includes a step of acquiring a recognition position observed by each observation device, a step of acquiring reliability for each observation device, and a step of specifying a real position of the observation target by weighting each acquired recognition position with each reliability and integrating the recognition positions.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method and a program for identifying the real position of an observed object. [Background technology]

[0002] In recent years, with the spread of autonomous driving technology, techniques have been developed to simulate traffic conditions, etc. For example, in the technique described in Patent Document 1, a traffic flow model is given a range of parameters such as the number of vehicles appearing per unit time, speed, and inter-vehicle distance, and the vehicles behave randomly within the parameter ranges to simulate the generation of random traffic flows. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2024-008144 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Patent Document 1 makes it possible to verify the functions and performance of automated driving in traffic flows that are simulated by simulation. However, the technology described in Patent Document 1 does not consider simulations that reproduce real traffic conditions, and it is thought that there is room for improvement in terms of the accuracy with which real traffic conditions can be reproduced.

[0005] An object of the present invention is to provide a technique that can improve the accuracy of reproducing traffic conditions. [Means for solving the problem]

[0006] One aspect of the present disclosure is a position determination method for observing a relatively moving object using observation equipment, comprising the steps of acquiring perceived positions observed by individual observation equipment, acquiring reliability for each of the observation equipment, and determining the actual position of the object by weighting and integrating each acquired perceived position by its reliability.

[0007] Another aspect of the present disclosure is a location identification program for observing a relatively moving observation target using observation equipment, which causes a computer to execute the steps of acquiring perceived positions observed by each observation device, acquiring reliability for each observation device, and identifying the actual position of the observation target by weighting and integrating each acquired perceived position by its respective reliability. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a technique that can improve the accuracy of reproducing traffic conditions. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating an example of a system configuration of a position specifying system 1 according to an embodiment of the present disclosure. [Figure 2] FIG. 10 is a diagram showing an overview of a position specifying process according to the present embodiment. [Figure 3] FIG. 2 is a functional block diagram of a location specifying server 100 according to the present embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of the hardware configuration of a location specifying server 100 according to the present embodiment. [Figure 5] 10 is a flowchart illustrating an example of a position specifying process according to the present embodiment. [Figure 6A] 10 is a diagram illustrating the association of the observation target 300 according to the present embodiment. FIG. [Figure 6B] 10 is a flowchart illustrating an example of an allocation process. [Figure 7] FIG. 10 is a diagram illustrating integration of recognition results by weighted averaging. [Figure 8] FIG. 10 is a schematic diagram for explaining a method for calculating reliability when introducing an observation device 200. [Figure 9] FIG. 10 is a schematic diagram for explaining a method for calculating reliability when three or more observation devices 200 are present. [Figure 10] 10 is a flowchart illustrating a process of updating a reliability. [Figure 11A] FIG. 10 is a schematic diagram for explaining reliability update processing in the present embodiment. [Figure 11B] FIG. 10 is a schematic diagram for explaining reliability update processing in the present embodiment. [Figure 12] FIG. 10 is a diagram illustrating a simulation result of a position identification process according to an embodiment of the present disclosure. [Figure 13] FIG. 10 is a diagram for explaining simulation conditions. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described with reference to the accompanying drawings, in which the same reference numerals are used to denote the same or similar components.

[0011] <System configuration> 1 is a diagram showing an example of the system configuration of a positioning system 1 according to an embodiment of the present disclosure. The positioning system 1 according to this embodiment includes a positioning server 100 and a plurality of observation devices 200.

[0012] The location determination server 100 is configured to determine the actual location of the observed object 300 based on data relating to the location of the observed object 300 recognized by a plurality of observation devices 200, as will be described later.

[0013] That is, the location determination server according to this embodiment is configured to execute the location determination method according to this embodiment. The location determination method according to this embodiment is a location determination method for observing a relatively moving observation target using observation devices, and is configured to execute the steps of acquiring perceived positions observed by each observation device, acquiring reliability for each observation device, and identifying the real position of the observation target by weighting each acquired perceived position by the respective reliability and integrating them.

[0014] 1 are configured to recognize the positions of observation targets. Note that, hereinafter, the observation devices 200_1, 200_2, ..., 200_n may be collectively referred to as observation devices 200.

[0015] In this embodiment, the observation device 200 is a detection means having the function of detecting the position of an observation target by optical or physical means and outputting information indicating the position of the observation target (also referred to as "recognition" in the present invention), and includes sensors such as visible light or infrared cameras and LIDAR measuring devices. The observation device may also be a control device that combines hardware and software as the detection means, such as a terminal device such as a smartphone. The observation device may be installed as a standalone device or may be incorporated into another device. For example, the observation device may be installed in a communication base station that functions as a repeater that relays communications between a mobile terminal or the like and a core network system. For example, a sensor may be installed in the communication base station and configured to recognize the position of an observation target located near the communication base station (for example, within a radius of several tens of meters).

[0016] Furthermore, in this embodiment, the observed object 300 (also referred to as "observation target") observed by the observation equipment 200 may be, for example, a pedestrian (pedestrians 300_P1, 300_P2, ..., 300_Pm) or a vehicle (vehicles 300_V1, 300_V2, ..., 300_Vn). In this embodiment, for example, a process may be performed on the observation equipment 200, which has a sensor mounted on a utility pole or a communication base station, to identify the position of the observed target 300, such as a pedestrian or a vehicle, that moves relatively.

[0017] The observation device 200 may be configured, for example, by attaching a sensor to a utility pole. The observation device 200 may also be configured, for example, by attaching a sensor to a vehicle. Alternatively, a mobile device carried by a vehicle driver or a pedestrian may recognize other vehicles or pedestrians as observation targets and recognize their positions. Thus, in this embodiment, the observation device 200, or the driver or pedestrian of a vehicle carrying the observation device 200, may also be the observation target 300.

[0018] 1, the positioning system 1 according to this embodiment may further include a database (DB) 600. The database 600 may store, for example, information about the reliability of the observation equipment 200 (described later) and information about the position of the observation target 300 recognized by the observation equipment 200. The database 600 may also store, for example, other data used in the positioning process executed by the positioning server 100.

[0019] Furthermore, the positioning system 1 according to this embodiment may further include an autonomous driving simulation server 510 and an autonomous driving management server 520. In this embodiment, for example, a driving simulation of an autonomous vehicle may be executed in the autonomous driving simulation server 510 using data of an observation target identified by the positioning server 100. In this embodiment, for example, traffic information to be provided to an autonomous vehicle actually traveling may be generated by the autonomous driving management server 520 using data of an observation target identified by the positioning server 100.

[0020] The positioning system 1 according to this embodiment may be configured as a system capable of simulating real-world events in a virtual space that virtually simulates the real world, such as a digital twin system, and the autonomous driving simulation executed by the autonomous driving simulation server 510 may be executed in the digital twin system. In this case, for example, in the digital twin system, reproductions of current and past traffic conditions in an area of ​​interest may be updated in real time, and a virtual world synchronized with the real world may be reproduced on a computer that constitutes the digital twin system. That is, in this embodiment, a test drive of an autonomous vehicle may be executed on the digital twin based on the actual position of the observed target 300 identified based on the positioning process.

[0021] FIG. 2 is a diagram illustrating an example of an outline of the location identification process executed by the location identification system 1 according to this embodiment.

[0022] First, the location identification server 100 acquires information about the location of the observed target 300 recognized by the observation device 200 from the observation device 200 (S202). At this time, the location identification server 100 may perform preprocessing on the acquired data, such as reducing noise and removing offsets.

[0023] 2, within the target area, for example, there are observation devices 200_T1 and 200_T2 set on utility poles and observation device 200_V1 installed on a vehicle. Furthermore, observation device 200_T1 recognizes pedestrian 300_Pt11, pedestrian 300_Pt12, and vehicle 300_Vt11, observation device 200_T2 recognizes pedestrian 300_Pt22, pedestrian 300_Pt23, and vehicle 300_Vt21, and observation device 200_V1 recognizes pedestrian 300_Pv14 and vehicle 300_Vv11. Note that the vehicle 300_Vv11 recognized by the observation equipment 200_V1, which is a vehicle, is the observation equipment 200_V1 itself, and the observation equipment 200_V1 transmits its own position as the position of the vehicle 300_Vv11 to the location identification server 100. Note that in Figure 2, the direction of movement of the pedestrian 300_P is indicated by an arrow.

[0024] Next, the location server 100 matches the observed objects (S204). In this embodiment, for example, among the pedestrians 300_P and vehicles 300_V recognized by the observation devices 200_T1, 200_T2, and 200_V1 in the target range, the observed objects 300 that can be recognized as identical are matched. In this embodiment, the objects are matched based on the similarity of the trajectories of two objects moving relative to each other. For example, as will be described later, the observed objects 300 may be matched by calculating a score relating to the relevance between the observed objects 300 based on the trajectories of the recognized observed objects 300 using dynamic time warping (DTW).

[0025] 2, as shown by the dashed rectangle in S204, for example, pedestrian 300_Pt12 and pedestrian 300_Pt22 are associated with each other. Also, vehicles 300_Vt11, 300_Vt21, and 300_Vv11 are associated with each other. Also, pedestrian 300_Pt11, pedestrian 300_Pt23, and pedestrian 300_Pv14 are not associated with any other pedestrian 300_P.

[0026] Next, the location identification server 100 integrates the associated observation targets (S206). As shown in Fig. 2, for example, the multiple pedestrians 300_P and vehicles 300_V associated with each other as described above may be integrated into one. That is, as shown in Fig. 2, the observation data obtained by integration shows, for example, pedestrians 300_P1, 300_P2, 300_P3, and 300_P4, and vehicle 300_V1. Here, "integration" refers to a process of having the system recognize multiple observation targets as a single observation target as a group.

[0027] In this embodiment, as will be described later, the results recognized by each observation device 200 are integrated by weighting them according to the reliability of the observation device 200. For example, if the reliability of the observation device 200_T1 is higher than that of the observation device 200_T2, when the associated pedestrians 300_Pt12 and 300_Pt22 are integrated, the recognized position of the pedestrian 300_Pt12 recognized by the observation device 200_T1 is weighted more heavily than the recognized position of the pedestrian 300_Pt22 recognized by the observation device 200_T2. Therefore, in this embodiment, the recognition results of the observation devices 200 with higher reliability are integrated so as to have a greater influence, thereby improving the reproduction accuracy.

[0028] As autonomous driving technology continues to develop, for example, digital twin technology is being used to virtually reproduce actual traffic conditions. For example, by more accurately grasping the positions of observed objects recognized by multiple observation devices such as autonomous vehicles and road infrastructure (electric poles and base stations), this can be utilized to grasp and reproduce situations in emergencies. Therefore, this can also be used in a virtual test drive method for autonomous vehicles, in which a test drive of an autonomous vehicle is performed on a digital twin based on the actual position of the observed object 300 identified by the position identification process according to this embodiment.

[0029] To improve the accuracy of traffic situation reproduction, it is possible to increase the number of observation devices and obtain a final integrated result based on a larger number of recognition results. However, because the observation results from each observation device may contain individual noise, increasing the number of observation devices does not necessarily improve the reproduction accuracy. For example, if mobile devices carried by drivers of ordinary vehicles are used as observation devices, the performance of each observation device may vary greatly, making it difficult to improve the reproduction accuracy. Furthermore, if there is an obstacle within the area of ​​interest that may hinder accurate observation and the obstacle moves dynamically, the influence of the obstacle on the observation results recognized by each observation device may result in the recognition results containing more noise.

[0030] In this embodiment, the reliability of each observation device is calculated based on the position of the observation device, the position of the recognized observation target, the time, etc., and the calculated reliability is updated. Then, the observation results are integrated by calculating a weighted average based on the reliability. This weighted average makes it possible to obtain an integrated result that reduces the influence of observation devices with lower accuracy. Therefore, the accuracy of reproducing traffic conditions can be improved.

[0031] <Function block configuration> The location specification server 100 according to this embodiment will be described with reference to Fig. 3. Fig. 3 is an example of a functional block diagram of the location specification server 100 according to this embodiment. The location specification server 100 includes an observation data acquisition unit 110, a data preprocessing unit 120, a grouping unit 130, a reliability information acquisition unit 140, a recognition result integration unit 150, an integration result output unit 160, a reliability calculation unit 170, and a reliability update unit 180.

[0032] The observation data acquiring unit 110 is configured to acquire information relating to the recognized positions of observation targets such as pedestrians and vehicles from the observation devices 200 (for example, the observation devices 200_1 and 200_2).

[0033] The data preprocessing unit 120 is configured to perform preprocessing such as noise reduction and offset removal on the observation data acquired by the observation data acquisition unit 110 .

[0034] The grouping unit 130 is configured to associate the observation targets 300 with each other based on the observation data that has been preprocessed by the data preprocessing unit 120.

[0035] The reliability information acquisition unit 140 acquires information relating to the reliability of each observation device 200 stored in a database 600 provided outside the position identification server 100 .

[0036] The recognition result integration unit 150 integrates the recognition results by weighting the recognition positions of the observation objects 300 that have been associated with each other by the grouping unit 130 according to the reliability of each of the observation objects 300 acquired by the reliability information acquisition unit 140.

[0037] The integration result output unit 160 outputs the integration result of the perceived positions of each observed object 300 integrated by the perception result integration unit 150 .

[0038] The reliability calculation unit 170 is configured to calculate the reliability of the observation devices 200. For example, if reliability information for some of the multiple observation devices 200 is not stored in the database 600, the reliability calculation unit 170 may calculate new reliability, as described below.

[0039] The reliability update unit 180 may be configured to update the new reliability calculated by the reliability calculation unit 170. The accuracy of the recognition result by the observation device 200 may vary dynamically depending on the presence of an obstacle that hinders observation, etc. Therefore, the reliability of the observation device 200 may be configured to be updated as a predetermined period of time passes. Furthermore, not only the reliability calculated by the reliability calculation unit 170, but also the reliability acquired by the reliability information acquisition unit 140 from the database 600 may be configured to be updated by the reliability update unit 180 as a predetermined period of time passes.

[0040] The recognition result integration unit 150 may perform integration processing of the recognition results from each observation device 200 based on the reliability calculated by the reliability calculation unit 170, or may perform integration processing of the recognition results based on the reliability updated by the reliability update unit 180.

[0041] The location determination server 100 may further include a storage unit (not shown) that stores, for example, programs executed by the location determination server 100. The storage unit may store, for example, information on the reliability of the observation equipment 200 described above.

[0042] <Hardware configuration> FIG. 4 is a diagram showing an example of the hardware configuration of the location specifying server 100 according to this embodiment.

[0043] The location identification server 100 includes a processor 11 such as a CPU (Central Processing Unit) or a GPU (Graphical Processing Unit), a storage device 12 such as a memory, an HDD (Hard Disk Drive) and / or an SSD (Solid State Drive), a communication IF (Interface) 13 for wired or wireless communication, an input device 14 for accepting input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, a touch panel, a mouse, and / or a microphone. The output device 15 is, for example, a display, a touch panel, and / or a speaker.

[0044] In this embodiment, the location identification server 100 may be configured as one or more physical computers or servers, etc., as needed, or may be configured using a virtual server running on a hypervisor.

[0045] The entirety or a part of the program related to the process of identifying the real position of an observed target executed by the position identification server 100 according to this embodiment may be provided by being stored in a computer-readable storage medium such as the storage device 12. Alternatively, the entirety or a part of the program may be provided from outside the position identification server 100 via a communication network to which the position identification server 100 is connected. In the position identification server 100, for example, the processor 11 executes the position identification program according to this embodiment to realize various operations described below with reference to FIG. 5 etc.

[0046] For example, the storage unit of the location server 100 can be realized using the storage device 12 included in the location server 100. The observation data acquisition unit 110, the data preprocessing unit 120, the grouping unit 130, the reliability information acquisition unit 140, the recognition result integration unit 150, the integration result output unit 160, the reliability calculation unit 170, and the reliability update unit 180 can be realized by the processor 11 of the location server 100 executing a program stored in the storage device 12. The program can be stored in a storage medium as described above, and the storage medium storing the program may be a non-transitory computer-readable medium. The non-transitory storage medium is not particularly limited, and may be, for example, a storage medium such as a USB memory or a CD-ROM.

[0047] Note that these physical configurations are merely examples and do not necessarily have to be independent of each other. For example, the location specification server 100 according to this embodiment may include an LSI (Large-Scale Integration) in which the processor 11 and the storage device 12 are integrated. As described above, the location specification server 100 may include a GPU as the processor 11. In this case, the GPU may execute the above programs to realize various operations described later with reference to FIG. 5 etc.

[0048] The location specification server 100 is not limited to the above-described configuration. For example, some of the functions of the location specification server 100 may be executed by another information processing device or server. Furthermore, the location specification server 100 may be configured using a cloud server.

[0049] Furthermore, with regard to each database (for example, database 600 in which reliability information is stored), for example, some or all of the information and data stored in these databases may be stored in a storage unit provided in location identification server 100.

[0050] <Processing Procedure> The location specification process executed by the location specification server 100 according to this embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing an example of the location specification process executed by the location specification server 100.

[0051] First, in step S502, the observation data acquisition unit 110 acquires observation data, such as the recognition results of the position of the observation target 300, from the observation equipment 200. For example, if multiple observation equipment 200 exist within the area of ​​interest, observation data from all of the observation equipment 200 present within the area of ​​interest may be acquired. Alternatively, observation data may be acquired from some of the multiple observation equipment 200 present within the area of ​​interest. For example, if a relatively large number of observation equipment 200 exist within the area of ​​interest, observation data from highly reliable observation equipment 200 may be selectively acquired. Alternatively, if there is variation in the density of the observation equipment 200, the observation equipment 200 from which observation data is to be selectively acquired may be determined so as to reduce spatial bias in the observation data.

[0052] Furthermore, observation data may be acquired for all observed targets 300 recognized by each observation device 200, or for some of the observed targets 300 recognized by each observation device 200. For example, observation data may be acquired selectively for observed targets 300 recognized to be located along roads, etc., within the area of ​​interest, and for which information regarding traffic conditions is thought to be of particular interest.

[0053] Next, in step S504, the observation data acquired in step S502 may be preprocessed by the data preprocessing unit 120. For example, preprocessing such as noise reduction and offset removal may be performed on the acquired observation data.

[0054] Next, in step S506, the grouping unit 130 associates the observation targets. That is, in this embodiment, one or more observation devices 200 observing the perceived positions of the same observation target 300 are associated with each other based on the similarity of the trajectories indicated by the perceived positions of the observation targets 300 observed over time by different observation devices 200. In this embodiment, for example, the observation targets 300 may be associated with each other based on the trajectories of the recognized observation targets 300 using dynamic time warping (DTW).

[0055] Referring to Fig. 6A, the correspondence of observed objects 300 will be described. As shown in Fig. 6A, utility poles 200_T1 and 200_T2 on which observation equipment 200 is installed are present in the illustrated area of ​​interest. Furthermore, pedestrians 300_Pt11 and 300_Pt12, which are observed objects 300, are recognized by utility pole 200_T1, and pedestrian 300_Pt21 is recognized by utility pole 200_T2. In this embodiment, dynamic time warping is used to determine which of pedestrians 300_Pt11, 300_Pt12, and 300_Pt21, which are observed objects, can be associated.

[0056] 6A, observed target 300_Pt11 moves to positions a1_0, a1_1, and a1_2 at times t0, t1, and t2, respectively. Similarly, observed target 300_Pt12 and observed target 300_Pt21 move to positions a1_0, a1_1, and a1_2, and positions b1_0, b1_1, and b1_2, respectively, at times t0, t1, and t2. For example, at time t0, the distance between position a1_0 and position b1_0 is smaller than the distance between position a2_0 and position b1_0.

[0057] Therefore, at time t0 alone, observed object 300_Pt11 corresponding to a1_0 and observed object 300_Pt21 are closer to each other than observed object 300_Pt12 corresponding to a2_0 and observed object 300_Pt21 corresponding to b1_0, and it can be assumed that they are corresponding observed objects. However, as time passes from t0 to t1, observed objects 300_Pt12 and 300_Pt21 move leftward in FIG. 6A, while observed object 300_Pt11 moves rightward in FIG. 6A. As time passes from t1 to t2, observed objects 300_Pt12 and 300_Pt21 move further leftward in FIG. 6A, while observed object 300_Pt11 moves further rightward in FIG. 6A.

[0058] From the above, it is suggested that within the area of ​​interest illustrated in Figure 6A, at time t0, observed object 300_Pt11 and observed object 300_Pt21 may be associated, but based on the trajectory moving to times t0, t1, and t2, it is inferred that observed object 300_Pt12 and observed object 300_Pt21 may be associated.

[0059] In this embodiment, the correspondence between the observation targets 300 is determined based on the similarity of the trajectories by focusing on the difference between the trajectories of the respective observation targets 300. For example, using dynamic time warping (DTW), the correspondence between the observation targets 300 can be determined based on the similarity of the trajectories of the observation targets 300, regardless of their proximity in each time slice, based on the score calculated for the observation targets 300. The score can be calculated, for example, by maximizing the total score of bipartite matching or by heuristic allocation using a threshold.

[0060] In the example shown in Figure 6A, using the dynamic time warping method, for example, the score corresponding to the similarity of the trajectories between observed objects 300_Pt11 and 300_Pt21 is 0.1, and the score corresponding to the similarity of the trajectories between observed objects 300_Pt12 and 300_Pt21 is 0.7.In this case, observed objects 300 with higher scores have higher trajectory similarity and are more likely to be matched.

[0061] In this embodiment, by using dynamic time warping, it is possible to improve the accuracy of matching between observed objects 300 compared to using indices without time-series information, such as positions at a certain time cross section or distances between centers of gravity. Furthermore, with dynamic time warping, even if the data sampling cycles of the observation devices 200 differ from one another, it is possible to calculate scores regardless of variations in the number of samples. Therefore, since it is possible to calculate scores regardless of the sampling cycles of the observation devices 200, it is possible to improve the accuracy of matching between observed objects 300.

[0062] In this embodiment, the following method may be used for the heuristic allocation using the threshold value described above: Fig. 6B is a flowchart showing an example of a heuristic allocation process.

[0063] First, for a certain b (e.g., b1 or b2), the score exceeds a predetermined threshold and the a with the largest score (e.g., a1 or a2) is provisionally assigned. Similarly, provisional assignment of a is performed for other b, and provisional assignment is performed for all b (S602).

[0064] Next, for b, the provisional allocation to be confirmed, if the provisional allocation does not conflict with the provisional allocation for another b, it is confirmed; if there is a conflict, the provisional allocation for the one with the higher score is confirmed (S604).

[0065] Next, a and b included in the confirmed allocation are excluded (S606).

[0066] Next, it is determined whether allocation has been completed for all a and b (S608). If it is determined that allocation has been completed for all a and b (YES in S608), the process ends.

[0067] As described above, if it is determined that allocation for all a and b has not been completed (e.g., NO in S608), it is determined whether the score exceeds a predetermined threshold (S610). If it is determined that the score exceeds the predetermined threshold (YES in S610), for that b (e.g., b1 or b2), the a with the highest score is tentatively allocated (S602), and thereafter, the same process (S604 to S610) is repeated.

[0068] The process is repeated until assignment is completed for all a's and b's or until there are no combinations whose scores exceed the threshold. When assignment is completed for all a's and b's (S608) or there are no combinations whose scores exceed the threshold (S610), the process ends.

[0069] For example, we will explain the case where the scores corresponding to the similarity of trajectories between multiple observation targets are 0.4, 0.7, 0.2, and 0.8, respectively, for the score between certain observation targets b1 and a1, the score between certain observation targets b1 and a2, the score between certain observation targets b2 and a1, and the score between certain observation targets b2 and a2.

[0070] For example, the threshold is set to 0.3. As described above, first, matching (association) candidates for b1 are searched, and a2 with a high score with b1 is extracted from a1 and a2, and this is set as a tentative assignment (b1, a2). Similarly, for b2, matching candidates are searched, and a2 with a high score with b2 is extracted from a1 and a2, and this is set as a tentative assignment (b2, a2) (S602). Next, the tentative assignment (b1, a2) is compared with the tentative assignment (b2, a2), and the tentative assignment (b2, a2) with the highest score is confirmed (S604). Next, excluding the confirmed assignment b2 (S606), it is determined whether there is an observation object b whose assignment has not been completed (S608), and for the observation object b1 whose assignment has not been completed, it is determined whether the score of 0.4 of the tentative assignment (b1, a1) exceeds the threshold (S610). The score of the tentative allocation (b1, a1), 0.4, exceeds the threshold value of 0.3, so the allocation (b1, a1) for b1 is confirmed. This completes the process.

[0071] Also, for example, if the threshold is set to 0.5, as described above, first, a tentative assignment (b1, a2) is set for b1, and a tentative assignment (b2, a2) is set for b2 (S602). Next, the tentative assignment (b1, a2) is compared with the tentative assignment (b2, a2), and the tentative assignment (b2, a2) with the higher score is confirmed (S604). After excluding the confirmed assignment b2 (S606), it is determined whether there is an observation object b whose assignment has not been completed (S608). For the observation object b1 whose assignment has not been completed, it is determined whether the score 0.4 of the tentative assignment (b1, a1) exceeds the threshold (S610). Since the score 0.4 of the tentative assignment (b1, a1) does not exceed the threshold 0.5, b1 is not assigned, and the process ends.

[0072] 5, in step S508, the recognition results for the associated observation targets 300 are integrated. In this embodiment, the results recognized by each observation device 200 are weighted and integrated according to the reliability of the observation device 200. In this embodiment, the recognized positions of the observation targets 300 acquired by the observation devices 200 are weighted and averaged based on the reliability of each observation device 200.

[0073] The integration of recognition results by weighted averaging according to reliability will be described with reference to Fig. 7. Fig. 7 shows an example of integrating the observation target 300_Pt12 by the observation device 200_T1 and the observation target 300_Pt21 by the observation device 200_T2, as described above with reference to Fig. 6A. That is, at times t0, t1, and t2, positions a2_0 and b1_0, positions a2_1 and b1_1, and positions a2_2 and b1_2 are weighted and averaged according to reliability, respectively, to obtain positions iab_0, iab_1, and iab_2.

[0074] In this embodiment, the reliability is calculated for each observation device 200 according to the time, the position of the observation device 200, and the position of the recognized observation target 300, and a weighted average of the calculated reliability is calculated to obtain an integrated recognition result.

[0075] That is, the integrated result of the recognition results is calculated by the weighted average of the following Equation 1.

[0076]

number

[0077]

number

[0078] As described above, it is possible to integrate the recognized positions of the observed object 300 using a weighted average according to the reliability, and identify the actual position of the observed object 300. For example, as shown in the above formula 2, the weighting coefficient of an observation device 200 with a relatively low reliability is relatively small, and the weighting coefficient of an observation device 200 with a relatively high reliability is relatively large. Therefore, by calculating a weighted average according to the reliability as shown in formula 1 and multiplying the position of the observed object 300 by a weighting coefficient that varies depending on the reliability, and integrating the results, it is possible to obtain an integrated result that takes reliability into consideration. Therefore, it is possible to improve the accuracy of identifying the actual position of the observed object 300 identified by the positioning server 100.

[0079] In this way, the process of integrating the recognition results is performed (S508 (FIG. 5)), and the location specifying process executed by the location specifying server 100 according to this embodiment, which has been described with reference to FIG. 5, is completed.

[0080] As described above, a location determination method according to an embodiment of the present disclosure is a location determination method for observing a relatively moving observation target using observation devices, and includes the steps of acquiring perceived positions observed by each of the observation devices, acquiring reliability for each of the observation devices, and identifying the actual location of the observation target by weighting and integrating the acquired perceived positions by their respective reliability. The location determination method according to the present embodiment can provide a technology that can improve the accuracy of reproducing traffic conditions, for example.

[0081] The reliability calculation method in this embodiment will be described below. In this embodiment, for example, in the reliability calculation process for calculating the reliability of the observation device 200 for which reliability has not been set, a process for calculating the distance between the actual position of the observation target 300 identified based on the perceived position of each observation target 300 and the perceived position of the observation target 300 identified by the observation device 200 for which reliability has not been set, and a process for calculating the reliability according to the distance may be executed.

[0082] For example, if observation equipment 200 already exists in the area of ​​interest and the reliability of each existing observation equipment 200 has been evaluated, when a new observation equipment 200 is introduced, the reliability of the new observation equipment 200 may be calculated. Alternatively, for example, if no observation equipment 200 exists in the area of ​​interest and a new observation equipment 200 is introduced, the reliability of the new observation equipment 200 may be calculated.

[0083] For example, in a case where an observation device 200 already exists in the area of ​​interest and a new observation device 200 is to be introduced, the reliability of the newly introduced observation device 200 may be calculated based on whether accuracy is improved by adding and evaluating the observation data indicated by the newly introduced observation device 200 to the observation data indicated by the set of original observation devices 200. For example, for the position of any observation target 300 identified by each observation data indicated by the original observation device 200, if the identification accuracy improves by including the observation data of the newly introduced observation device 200, the reliability of the newly introduced observation device 200 may be calculated to be relatively high, and if the identification accuracy decreases, the reliability of the newly introduced observation device 200 may be calculated to be relatively low.

[0084] In this way, the reliability of the newly introduced observation equipment 200 may be calculated based on the contribution of the observation data of the newly introduced observation equipment 200 to the position of the observation target 300 identified by the set of original observation equipment 200. Therefore, since the contribution of the newly introduced observation equipment 200 to the accuracy of the position identified for the observation target 300 is considered to correspond to the reliability of the newly introduced observation equipment 200, the reliability of the newly introduced observation equipment 200 may be calculated, for example, based on a calculation method for the Shapley value exemplified by the following Equation 3.

number

[0085] In this embodiment, for example, when there is no observation device 200 in the area of ​​interest and a new observation device 200 is introduced, the reliability of the new observation device 200 may be calculated, and a method for calculating the reliability in this case will be described below with reference to Fig. 8. Fig. 8 is a schematic diagram for explaining a method for calculating the reliability when a new observation device 200 is introduced.

[0086] 8, a new utility pole is installed within the area of ​​interest, and observation of the observation target 300_Pn1 is started using observation equipment 200_Tn1. The observation equipment 200_Tn1 observes that the position of the observation target 300_Pn1 moves to a0_n-2, a0_n-1, a0_n, a0_n+1, and a0_n+2 at times t=n-2, t=n-1, t=n, t=n+1, and t=n+2, respectively. In this case, the reliability of the observation equipment 200_Tn1 may be calculated based on the distance between the position of the observation target 300_Pn1 observed by the observation equipment 200_Tn1 at the target time t=n and the estimated position of the observation target 300_Pn1 at time t=n on a movement trajectory calculated based on the positions a0_n-2, a0_n-1, a0_n+1, and a0_n+2 of the observation target 300_Pn1 at times t=n-2, t=n-1, t=n+1, and t=n+2 in the observation data.

[0087] The path traveled by the observed object 300_Pn1 from time t=n-2 to t=n-1, t=n+1, and t=n+2 is approximated to a curve cn1 by using an arbitrary polynomial interpolation for positions a0_n-2, a0_n-1, a0_n+1, and a0_n+2. As the arbitrary polynomial interpolation, a known interpolation method such as Lagrange interpolation may be used, for example.

[0088] Assuming that the exemplary observed object 300_Pn shown in FIG. 8 moves along a movement path approximated by a curve cn1, it is estimated to be located at a position a0_n_t on the curve cn1 at time t=n. Furthermore, based on observation data from the observation device 200_Tn1, the observed object 300_Pn1 is located at a position a0_n at time t=n. Because the curve cn1 calculated by polynomial interpolation is considered to be a path close to the actual movement trajectory of the observed object 300_Pn1, as shown in FIG. 8, the distance between the observation result by the observation device 200_Tn1 and the position of the observed object 300_Pn1 on the movement path at the time t=n of interest can be assumed to correspond to the observation error by the observation device 200_Tn1.

[0089] In this way, the reliability may be calculated based on the distance between the position at time t=n of interest on a trajectory calculated based on observed positions at times other than the time of interest and the position at time t=n in the observation data. For example, the reliability may be calculated by multiplying the distance by a predetermined weighting coefficient. Alternatively, the reliability may be calculated by calculating the distance between the position at a time on a movement curve calculated similarly by polynomial interpolation for multiple times of interest and the observed position at the time of interest, and multiplying each distance by a predetermined weighting coefficient.

[0090] Furthermore, the reliability calculated in this manner may be used to calculate the reliability of a newly added observation device 200. For example, even when one observation device 200 exists within the area of ​​interest and a new observation device 200 is to be added, the reliability of the observation device 200 calculated by the above-described method may be used as the reliability of the already-existing observation device 200, and the reliability of the newly added observation device 200 may be calculated by calculating the Shapley value shown in Equation 3 or the like for the observation data of the newly added observation device 200.

[0091] Furthermore, when two observation devices 200_k1 and 200_k2 are present in the area of ​​interest, the reliability of the set of the two observation devices 200_k1 and 200_k2 may be calculated using the method described below. For example, curves ck1 and ck2 indicating the movement trajectory of the position of the same observation target 300 may be calculated by polynomial interpolation for each of the observation data of the observation device 200_k1 and the observation data of the observation device 200_k2, and the deviations from the calculated curves ck1 and ck2 may be calculated for each of the observation devices 200_k1 and 200_k2, and the average of the calculated deviations may be calculated as the reliability of the set of the two observation devices 200_k1 and 200_k2.

[0092] Furthermore, when three or more observation devices 200 exist within the area of ​​interest, the reliability of a set of three or more observation devices 200 may be calculated using the method described below. As shown in Fig. 9, for example, the reliability of a subset consisting of observation devices 200_l1, 200_l2, and 200_l3 within the area of ​​interest is calculated. Note that other observation devices 200 exist within the area of ​​interest.

[0093] As shown in FIG. 9, for an arbitrary observation target 300_l1, it is determined that the observation equipment 200_l1, the observation equipment 200_l2, and the observation equipment 200_l3 are located at positions a2_l1, a2_l2, and a2_l3, respectively, at an arbitrary time t=n. In this embodiment, for example, the center of gravity g_lp of the position (positions a2_l1, a2_l2, and a2_l3) of any observation object 300_l1 observed by each of the observation equipment 200_l1, observation equipment 200_l2, and observation equipment 200_l3 in the area of ​​interest and the center of gravity g_la of the position of the same observation object 300_l1 observed by each of all observation equipment 200 present in the area of ​​interest are calculated, and the reliability of the subset of observation equipment 200 (observation equipment 200_l1, observation equipment 200_l2, and observation equipment 200_l3) may be calculated based on the amount of deviation of the center of gravity g_lp of the position of the observation object 300_l1 observed by each of the observation equipment 200_l1, 200_l2, and 200_l3 from the center of gravity g_la of the position of the observation object 300_l1 observed by each of all observation equipment 200. The reliability of a subset of four or more observation devices 200 may also be calculated using a similar method.

[0094] The reliability for a subset of two observation devices 200 or the reliability for a subset of three or four or more observation devices 200 may be calculated using the above-mentioned method, and the calculated reliability may be used to calculate the above-mentioned Shapley value, for example, when calculating the reliability of a newly added observation device 200 when there are two observation devices 200 in the area of ​​interest, or when calculating the reliability of a newly added observation device 200 when there are three or more observation devices 200.

[0095] The Shapley value exemplified by Equation 3 is generally used to evaluate the contribution of each of multiple participants to an organization, and the greater the contribution of the participant, the greater the Shapley value. In this embodiment, when evaluating the deviation from the curve calculated by polynomial interpolation using the above method, the reliability of the observation device 200 may be evaluated using the Shapley value calculated by the method exemplified by Equation 3, as described below, so that the greater the deviation, the lower the reliability and vice versa.

[0096] For example, as shown in Equation 4, an evaluation value may be calculated by inverting the minimum and maximum Shapley values ​​of each observation device 200.

number

number

number

number

number

[0097] In this embodiment, the evaluation value calculated by Equation 4 may be scaled and weighted so that the sum of all the observation devices 200 becomes 1, as shown in Equation 9 below.

number

[0098] Generally, because traffic volume can change depending on the time of day, the Shapley weights may be updated at predetermined intervals, etc. For example, if a malfunction suddenly occurs in an electrical system within an area of ​​interest that could affect all or many of the observation devices 200 within the area of ​​interest, this could cause noise in the observation data of all or many of the affected observation devices 200. For example, by updating the Shapley weights at predetermined times, such as every predetermined period of time, it is possible to calculate the Shapley value weights of the target observation devices 200 while taking into account the influence of noise, etc., on the observation devices 200 whose reliability has already been calculated.

[0099] That is, in this embodiment, a process of updating the reliability over time is executed, and the process of updating the reliability may include a process of calculating a weighting coefficient based on the Shapley value at a specified observation time for the target observation device 200, a process of obtaining the reliability at the time of the previous update for the target observation device 200, a process of calculating an update amount for the reliability based on the difference between the weighting coefficient calculated at the specified observation time and the reliability at the time of the previous update, and a process of updating the reliability by adding the update amount to the reliability at the time of the previous update.

[0100] Hereinafter, the reliability update process in this embodiment will be described with reference to Fig. 10, Fig. 11A, and Fig. 11B. Fig. 10 is a flowchart illustrating the reliability update process. Fig. 11A and Fig. 11B are schematic diagrams illustrating the reliability update process in this embodiment.

[0101] 11A and 11B are schematic diagrams of an area of ​​interest at time t=t0 and time t=t1, respectively. As shown in FIGS. 11A and 11B, for example, the area of ​​interest is assumed to be a 5x5 mesh, and the reliability is updated. As shown in FIG. 11A, at t=t0, the observation device 200_v1, which is a vehicle, is located at position (2,1), and the observation target 300_p1 is located at position (1,2). Also, as shown in FIG. 11B, at t=t1, the observation device 200_v1 has moved to position (2,2), and the observation target 300_p1 has moved to position (1,3).

[0102] 11A and 11B, the range of influence due to the reliability update is indicated by a dashed line rectangle and a dashed line rectangle, respectively, for the range of influence due to the observation device 200_v1 and the range of influence due to the observation target 300_p1. As shown in FIGS. 11A and 11B, in this embodiment, the range of influence due to the observation device 200_v1 and the range of influence due to the observation target 300_p1 are both within one nearby mesh, a 3x3 mesh range. In the following, FIGS. 11A and 11B will be described as an example in which four observation devices 200 exist within the area of ​​interest.

[0103] As shown in FIG. 10, first, the Shapley weight at the current observation point is calculated (S1002). Here, for example, the Shapley weight at the observation point at t=t0 shown in FIG. 11A is calculated. In this embodiment, the observation point is represented as (t, (ia, ja), (io, jo)). t, i, and j represent time, the coordinate along the vertical axis in FIG. 11A, and the coordinate along the horizontal axis in FIG. 11A, respectively. In FIG. 11A, the observation device 200_v1 is located at coordinates (ia, ja)=(2,1), and the observation target 300_p1 is located at coordinates (io, jo)=(1,2). Therefore, the observation point for which the Shapley weight is calculated here is represented as (t, (ia, ja), (io, jo))=(t0, (2,1), (1,2)). In the example shown in FIG. 11A, the time t=t0=0, and the Shapley weight at this observation point (t0, (2, 1), (1, 2)) is 0.5.

[0104] Next, observation points that meet the influence range update conditions are extracted (S1004). Here, the combination of the position of observation equipment 200_v1 and the influence range of the recognized position of observation target 300_p1 at the next time (t=t1) is extracted as the observation point. For example, 9 x 9 = 81 points of (t,(ia,ja),(io,jo))=(t1,(1,0),(0,1)), (t1,(1,1),(0,1)), ..., (t1,(3,2),(2,3)) may be extracted as the observation points.

[0105] Next, an observation point of interest is determined for each of the extracted observation points (S1006). Here, for example, (t,(ia,ja),(io,jo))=(t1,(2,2),(1,3)) is determined as the observation point of interest.

[0106] Next, the presence or absence of reliability for the past is determined by referring to the database 600 (S1008). For example, the presence or absence of reliability at the immediately previous time is determined by referring to the database 600.

[0107] If it is determined that the reliability of the past data is stored in the database 600 (YES in S1008), the reliability of the past data is acquired (S1010). Then, the difference between the acquired reliability of the past data and the Shapley weight at the observation point calculated in S1002 is calculated (S1014).

[0108] On the other hand, if it is determined that past reliability values ​​are not stored in the database 600 (NO in S1008), for example, 1 / the number of observation devices 200 is calculated (S1012). As described above, in the example shown in FIG. 11A, there are four observation devices 200, and 1 / the number of observation devices 200 is calculated as 0.25. Thereafter, the difference between the calculated value of 1 / the number of observation devices 200 and the Shapley weight at the observation point calculated in S1002 is calculated (S1014).

[0109] 11A shows an example where t=t0, and the reliability before the current time has not been calculated, and no past reliability is stored in the database 600. Therefore, since the answer is NO in S1008, 1 / number of observation devices 200=0.25 is calculated in S1012 (S1012), and the difference between 0.25 and the Shapley weight=0.5 calculated in S1002 is calculated as 0.25 (S1014).

[0110] Next, a coefficient is calculated from the difference between the current observation point and the observation point of interest (S1016). For example, the coefficient is calculated from the difference between the current observation point (t0, (2,1), (1,2)) and the observation point of interest (t0, (2,2), (1,3)). At this time, for example, a coefficient may be determined that decreases as the total distance between the difference in the position of the observation device 200 (here, for example, the difference between (2,1) and (2,2)) and the difference in the perceived position of the observed object 300 (here, for example, the difference between (1,2) and (1,3)) increases. The rule for determining the coefficient may be selected arbitrarily. For example, the coefficient is calculated as 0.9.

[0111] The coefficient may be calculated based on, for example, the following Equation 10.

[0112]

number

[0113] Next, the reliability update amount is calculated (S1018). The reliability update amount may be calculated based on the reliability update amount = coefficient x difference using the difference value calculated in S1014 and the coefficient value calculated in S1016. Here, the reliability update amount = 0.9 x 0.25 = 0.225 is calculated.

[0114] Next, the reliability is updated based on the reliability update amount (S1020). The updated reliability may be calculated, for example, based on the reliability update amount calculated in S1018 and the past reliability acquired in S1010 (or 1 / number of observation devices 200 if past reliability is not stored), based on the following equation: reliability (reliability of the observation point of interest) = past reliability (or 1 / number of observation devices 200) + reliability update amount. Here, the reliability at the observation point of interest (t1,(2,2),(1,3)) = 0.475 is calculated. The calculated reliability at the observation point of interest may be stored in database 600, for example.

[0115] Next, it is determined whether the reliability has been updated for all observation points (S1022). For example, it is determined whether the reliability has been updated for all 9 x 9 = 81 observation points of (t, (ia, ja), (io, jo)) = (t1, (1, 0), (0, 1)), (t1, (1, 1), (0, 1)), ..., (t1, (3, 2), (2, 3)). If there is an observation point whose reliability has not been updated, the process returns to S1006, and the reliability update process described above is executed.

[0116] In this way, the reliability update process at time t=t0 is completed, and next, the reliability update process at time t=t1 is executed.

[0117] As shown in FIG. 11B, at time t=t1, the observation equipment 200_v1 moves to a position (ia, ja)=(2, 2), and the observation target 300_p1 moves to a position (io, jo)=(1, 3).

[0118] The Shapley weight is calculated to be 0.5 at the current observation point (t1, (2, 2), (1, 3)) (S1002).

[0119] Next, as observation points that meet the influence range update conditions, for example, 9x9=81 points of (t,(ia,ja),(io,jo))=(t2,(1,0),(0,1)), (t2,(1,1),(0,1)), ..., (t2,(3,2),(2,3)) are extracted as observation points (S1004).

[0120] Next, an observation point of interest is determined for each of the extracted observation points (S1006), and for example, (t, (ia, ja), (io, jo)) = (t2, (2, 2), (1, 3)) is determined as the observation point of interest.

[0121] Next, the database 600 is referenced to determine whether or not there is reliability for the past data (S1008), and at time t=t1, the reliability for time t=t0 = 0.475 is obtained (S1010), and the difference between the obtained reliability for the past data and the Shapley weight at the observation point calculated in S1002 is calculated as 0.5-0.475=0.025 (S1014).

[0122] Next, a coefficient is calculated from the difference between the current observation point and the observation point of interest (S1016), and in this case, it is calculated to be 0.9, for example.

[0123] Next, the reliability update amount is calculated as reliability update amount=coefficient×difference=0.9×0.025=0.023.

[0124] Next, the reliability is updated based on the reliability update amount (S1020), and the updated reliability is calculated as 0.475 + 0.023 = 0.498 based on the reliability update amount + past reliability. 0.498 is stored as the reliability at the observation point of interest (t2, (2, 2), (1, 3)).

[0125] Next, it is determined whether the reliability has been updated for all observation points (S1022), and once it has been updated for all observation points, the reliability update process at time t=t1 is completed, and then the reliability update process at time t=t2 may be performed.

[0126] For example, if noise or the like affecting many observation devices 200 is present at a certain time, the reliability calculated based on the Shapley weight at that time may deviate significantly from the actual reliability due to the influence of the noise. Furthermore, for example, if multiple observation devices 200 or multiple observation targets 300 are relatively close to each other at a certain time, the reliabilities calculated at that time may be close to each other. Even in this case, even if there is an actual difference in reliability between the observation devices 200, it is considered that the difference in reliability may be less apparent. In this embodiment, the reliability can be updated using the method described above with reference to FIG. 10, etc., and the reliability is calculated based on observation data from the observation devices 200 during a predetermined update period. This makes it possible to obtain a more reliable reliability than when the reliability is calculated based only on observation data at a certain time.

[0127] 12 is a diagram illustrating the results of a simulation of a position identification process according to an embodiment of the present disclosure. In the simulation shown in FIG. 12, the simulation of the comparative example shows a case where observation data from the observation equipment 200 is integrated by simple averaging (dashed line). In addition, the simulation of the example shows a case where observation data from the observation equipment 200 is integrated by weighting averaging with reliability (solid line).

[0128] FIG. 13 is a diagram illustrating the simulation conditions. As shown in FIG. 13, in the simulation, a position determination process was performed based on observation data acquired by recognizing the position of the observation target 300_s using observation devices 200 (three observation devices 200_vs1, 200_vs2, and 200_vs3 are shown in FIG. 13), which are vehicles moving at a constant speed on a circumference Cs. The root mean square error (RMSE) was evaluated for errors in the recognized position of the observation target 300 depending on the number of observation devices 200. As shown in FIG. 12, the simulation was performed for cases where the number of observation devices 200 was two, three, four, five, and six. In the simulation, noise was periodically mixed into the recognition of each observation device 200. Also, as shown in Figure 13, in the simulation, an obstruction 700s was placed within the area of ​​interest, and when there was an obstruction on the line between the observation equipment 200 and the observation target 300 (observation equipment 200_vs2 in Figure 13), noise with 5x gain was applied.

[0129] 12, regardless of the number of observation devices 200, the RMSE was smaller in the example (solid line) than in the comparative example (dashed line). Therefore, it was confirmed that in the position identification process of this embodiment, by integrating the recognition results of the observation data by taking a weighted average of the reliability, it is possible to improve the accuracy of the recognized position of the observed object 300. It was also confirmed that the accuracy of the recognized position can be further improved by increasing the number of observation devices 200.

[0130] FIG. 12 shows the results obtained by simulation, but in experiments conducted by the inventors in the field, it was also confirmed that the accuracy of the recognized position can be improved by the position identification process according to this embodiment.

[0131] <Summary> According to the embodiment described above, the position determination method for observing a relatively moving observation target using observation equipment is configured to execute the steps of acquiring perceived positions observed by individual observation equipment, acquiring reliability for each of the observation equipment, and identifying the actual position of the observation target by weighting and integrating each acquired perceived position by the reliability.

[0132] This makes it possible to provide technology that can improve the accuracy of reproducing traffic conditions. Furthermore, the technology according to this embodiment can contribute to achieving Goal 9 of the Sustainable Development Goals (SDGs), "Build resilient infrastructure, promote inclusive and sustainable industrialization," and Goal 11 of the Sustainable Development Goals (SDGs), "Make cities and human settlements inclusive, safe, resilient, and sustainable."

[0133] The above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The flowcharts, sequences, elements included in the embodiments, and their arrangements, materials, conditions, shapes, sizes, etc., described in the embodiments are not limited to those illustrated and can be modified as appropriate. Furthermore, the calculation methods described in the above-described embodiments are not limited to those illustrated. Furthermore, configurations shown in different embodiments can be partially substituted or combined with each other. [Explanation of symbols]

[0134] 1...location determination system, 100...location determination server, 110...observation data acquisition unit, 120...data preprocessing unit, 130...grouping unit, 140...reliability information acquisition unit, 150...recognition result integration unit, 160...integration result output unit, 170...reliability calculation unit, 180...reliability update unit, 200...observation equipment, 300...observation target, 300...observation equipment, 510...autonomous driving simulation server, 520...autonomous driving management server, 600...database

Claims

1. A position identification method for observing a relatively moving observation target using an observation device, comprising: obtaining perceived positions observed by each observation device; obtaining a reliability for each of the observation devices; determining a real position of the object by weighting each of the acquired perceived positions by the respective reliability levels and integrating the weighted perceived positions; A location determination method comprising:

2. A step of associating one or more of the observation devices observing the perceived positions of the same observation target based on the similarity of trajectories indicated by the perceived positions observed over time by different observation devices; The method of claim 1 , comprising:

3. The step of determining the actual position includes a step of weighting and averaging the obtained perceived positions based on the respective reliability levels. The location determination method of claim 1 .

4. A step of calculating the reliability of the observation device for which the reliability has not been set, calculating a distance between the actual position of the observation target identified based on each of the perceived positions and the perceived position of the observation target identified by the observation device for which the reliability has not been set; The method of claim 3 , further comprising the step of: calculating the reliability, the step including the step of: calculating the reliability according to the distance.

5. The step of calculating the reliability includes: and a step of calculating the reliability for the perceived positions acquired by each of the observation devices based on a Shapley value when the perceived positions acquired by the observation devices for which the reliability is not set are obtained. The location determination method according to claim 3 .

6. updating the confidence over time, calculating a weighting coefficient based on the Shapley value at a predetermined observation time for the observation instrument of interest; acquiring the reliability of the observation equipment at the time of the previous update; calculating an update amount of the reliability based on a difference between the weighting coefficient calculated at the predetermined observation time point and the reliability at the time of the previous update; updating the reliability by adding the update amount to the reliability at the time of the previous update, further comprising updating the confidence level. The location determination method according to claim 5 .

7. A virtual test drive method for an autonomous vehicle, comprising: performing a test drive of the autonomous vehicle on a digital twin based on the actual position of the observed object identified based on the method according to any one of claims 1 to 6.

8. A position identification program for observing a relatively moving observation target using an observation device, obtaining perceived positions observed by each observation device; obtaining a reliability for each of the observation devices; determining a real position of the object by weighting each of the acquired perceived positions by the respective reliability levels and integrating the weighted perceived positions; A location identification program for causing a computer to execute the above.

Images