Information processing device and program

Abstract

An information processing device according to one aspect of the present invention comprises an absolute position information acquisition unit, a relative position information acquisition unit, an inter-mobile body corrected position information acquisition unit, and a position information generation unit. The absolute position information acquisition unit acquires absolute position information indicating the absolute positions of a plurality of mobile bodies included in a mobile body group moving as a group. The relative position information acquisition unit acquires relative position information indicating the relative positions of the plurality of mobile bodies with respect to the past absolute positions. The inter-mobile body corrected position information acquisition unit acquires inter-mobile body corrected position information obtained by estimating the positions of the mobile bodies using the relative positions between the plurality of mobile bodies. The position information generation unit generates position information indicating the positions of the mobile bodies on the basis of the absolute position information, the relative position information, and the inter-mobile body corrected position information.

Description

Information processing device and program

[0001] This invention relates to an information processing device and a program.

[0002] Satellite positioning systems, such as GPS (Global Positioning System), are widely used as a means of determining the position (latitude, longitude, and altitude) of a moving object on the ground in real time. In a satellite positioning system, a receiver receives radio waves transmitted from multiple communication satellites, and the receiver determines its two-dimensional or three-dimensional position by using the difference between the transmission time and reception time, the propagation speed of the radio waves, and the position of the satellite (orbital information) to determine the distance from each communication satellite. For example, GPS is a satellite positioning system managed and operated by the United States, while GNSS (Global Navigation Satellite System) is a general term for satellite positioning systems managed and operated by Japan, the United States, Russia, Europe, China, and others. Because GNSS can utilize more satellites, higher accuracy in positioning can be expected compared to GPS.

[0003] In such satellite positioning systems, errors typically occur for several reasons. For example, errors can occur when the time is inaccurate due to the precision of the communication satellite or receiver's clock, when estimation errors are included in orbital information based on satellite position tracking data, when radio waves are delayed or become inaccurate due to refraction, diffraction, or reflection by the ionosphere, troposphere, or surrounding terrain, or when radio waves are absorbed or shielded by tall buildings, making reception impossible. To solve these problems, satellite positioning systems use multiple communication satellites for positioning, but since communication satellites are not geostationary, it is known that their accuracy varies depending on location, time, and environment. Therefore, it is necessary to take measures such as excluding low-accuracy position information using an index value that indicates the degree of error (position accuracy degradation rate).

[0004] Furthermore, there are two types of satellite positioning systems: standalone positioning, which uses a single receiver, and relative positioning, which uses two types of receivers: a base station and a mobile station. Standalone positioning allows for wide-area positioning because it uses a single receiver, but generally results in an error of several meters to over ten meters. Relative positioning requires separate receivers at the base station and mobile station, but is generally a highly accurate system that can keep the error to a few millimeters to about 1 cm. In relative positioning, accuracy is improved by correcting the observation values ​​of the mobile station based on the observation values ​​of the base station (reference station) installed at a known location, thereby reducing the effects of clock errors and refraction. One example of relative positioning is the "Network-type RTK-GNSS system (VRS system)," which uses "electronic reference points" installed by the Geospatial Information Authority of Japan as reference stations and performs corrections at the mobile station in real time via a communication line (Non-Patent Literature 1). In the VRS system, a virtual base station (virtual station) is generated near the mobile station from multiple electronic reference points installed at approximately 1,300 locations nationwide, and the mobile station's position is corrected using this virtual station.

[0005] Furthermore, various methods have been devised for positioning in narrow areas rather than wide areas. For indoor use, a method has been put into practical use that performs triplicate positioning by using multiple Wi-Fi access points, RFID tags, and Bluetooth® beacons installed at known locations, and analyzing the signal strength and arrival time differences of each. There are also smartphone terminals that use ultra-wideband (UWB) wireless communication technology to detect the relative position between terminals with high precision. In addition, in pedestrian autonomous navigation positioning (PDR), a type of dead reckoning, relative movement direction and distance are sequentially measured using an "accelerometer" to measure changes in speed, a "gyroscope (angular velocity sensor)" to measure rotational angular velocity, a "magnetic sensor" to measure the magnitude and direction of the magnetic field, and a "LiDAR sensor (depth sensor)" to measure the distance and shape of surrounding objects, and the position is estimated by applying cumulative corrections to the initial position obtained using GPS, etc. (Non-patent document 2).

[0006] Ministry of Land, Infrastructure, Transport and Tourism, Geospatial Information Authority of Japan, "Various Surveying Methods Using GNSS", [online], [Retrieved February 4, 2025], Internet <URL: https: / / www.gsi.go.jp / denshi / denshi45009.html> Ministry of Land, Infrastructure, Transport and Tourism, "Guidelines for Constructing Indoor Maps / Indoor Positioning Environments", [online], [Retrieved February 4, 2025], Internet <URL: https: / / www.mlit.go.jp / tochi_fudousan_kensetsugyo / tochi_fudousan_kensetsugyo_tk17_000001_00008.html>

[0007] In car navigation systems for automobiles, which are a type of moving object, a single-positioning method is used for positioning. Car navigation systems use map matching methods, which correct the position of the car to align with the centerline of the road using map information (road information), to solve the practical problem of errors in the meter range. On the other hand, if a correction method such as map matching were adopted for a spectator system for, for example, a road race for automobiles or bicycles, all cars would be lined up on the centerline of the road. In a spectator system for a road race for automobiles or bicycles, there are situations where it is necessary to more accurately reflect and understand the position of moving objects, such as the lines taken by each car in corners and the competition with other cars, so a correction method such as map matching is not suitable for a spectator system for a road race for automobiles or bicycles. In such situations, a more accurate positioning method should be used, but a problem is that there is a trade-off between the time interval for positioning (positioning frequency) and positioning accuracy. Many commercially available devices that use relative positioning methods acquire position at 1-second intervals (1 Hz), but generally, accurate measurement requires a measurement time of 10 seconds or more at a fixed point. On the other hand, a moving object traveling at 36 km / h moves 10 meters per second, so even if positioning can be performed at one-second intervals, it becomes difficult to accurately acquire information about the behavior of high-speed moving objects such as bicycles and cars around corners. Some devices allow the positioning frequency to be changed to 5 Hz or 20 Hz, but in that case, the statistical data used for correction decreases, leading to a problem of reduced positioning accuracy.

[0008] To ensure positioning frequency and accuracy, a position estimation method combining wide-area and narrow-area positioning methods has been proposed. In such a method, wide-area positioning acquires highly accurate absolute positions at intervals of a few seconds, while narrow-area positioning acquires relative positions at a higher frequency. By cumulatively correcting the absolute position with the relative position, both positioning frequency and accuracy are ensured. For example, in typical positioning methods for smartphones, the low positioning frequency and accuracy of standalone positioning using GPS or GNSS are compensated for by relative positioning using PDR with various sensors in the smartphone.

[0009] In position estimation methods that combine absolute and relative positions, a problem arises where accuracy decreases due to the cumulative error of the relative position. For example, in PDR, the distance traveled D is calculated by integrating the acceleration obtained from the acceleration sensor twice at a certain time t. ｔ The azimuth angle φ_ is obtained from the angular velocity sensor and magnetic sensor. t By obtaining this data, the relative position (xt+1, yt+1) at time t+1 from the initial position (x0, y0) at time t=0 is estimated using the following equation (1). However, for the sake of simplicity, a position on a two-dimensional plane is used here.

[0010]

[0011] The relative position (xt+1, yt+1) at time t+1 is the same as the position (x) estimated at the previous time t. t ,y t ) and the position (x) estimated at the previous time t-1. t-1 ,y t-1 ) and the position (x) estimated at the previous time t-2. t-2 ,y t-2) and the position is estimated cumulatively from the position estimated at the previous time. Therefore, if past estimated positions are incorrect, a problem arises in which the error in relative position increases cumulatively with the estimation time. For this reason, a method to suppress cumulative errors is necessary, such as resetting the initial position (x0, y0) when the absolute position is obtained by the wide-area positioning method. However, with this method, if the cumulative error is large, resetting the initial position will cause an unnatural movement in the position, such as a jump.

[0012] Thus, the method of measuring relative position using the sensor array of a positioning device presents problems in situations where it is desirable to more accurately reflect and understand the position of a moving object due to cumulative errors. In particular, in road cycling races, sudden acceleration and deceleration occur frequently, so errors tend to accumulate in the distance traveled D, which is obtained by integrating the acceleration twice. One of the highlights of watching road cycling races is watching the competitors vying for position, so it is important for road cycling race viewing systems to acquire the relative positions of moving objects with high accuracy, and a method for acquiring highly accurate relative positions becomes a challenge.

[0013] This invention was made in view of the above circumstances, and its purpose is to improve the positioning accuracy of a position information acquisition system that corrects absolute position information obtained by wide-area positioning using relative position information obtained by narrow-area positioning.

[0014] To achieve the above objective, one embodiment of the information processing apparatus according to the present invention comprises an absolute position information acquisition unit, a relative position information acquisition unit, an inter-moving body correction position information acquisition unit, and a position information generation unit. The absolute position information acquisition unit acquires absolute position information indicating the absolute positions of multiple moving bodies included in a group of moving bodies moving together. The relative position information acquisition unit acquires relative position information indicating the relative positions of the multiple moving bodies to the past absolute positions. The inter-moving body correction position information acquisition unit acquires inter-moving body correction position information, which estimates the position of the moving bodies using the relative positions between the multiple moving bodies. The position information generation unit generates position information indicating the position of the moving bodies based on the absolute position information, the relative position information, and the inter-moving body correction position information.

[0015] According to one aspect of this invention, for example, it becomes possible to improve the positioning accuracy of a location information acquisition system.

[0016] In other words, according to one aspect of this invention, the positioning accuracy of a location information acquisition system can be improved.

[0017] Figure 1 is a diagram showing an example of the functional configuration of an information processing device according to an embodiment. Figure 2 is a flowchart showing an example of a processing procedure executed by the information processing device. Figure 3 is a schematic diagram showing an example of a bicycle road race. Figure 4 is a schematic diagram showing the generation of temporary position information for each bicycle. Figure 5 is a schematic diagram showing the inter-body correction position information generated by bicycle 1 shown in Figure 4. Figure 6 is a schematic diagram showing the inter-body correction position information generated by bicycles 1-3 shown in Figure 4. Figure 7 is a schematic diagram showing the inter-body relative center of gravity of bicycle 1 shown in Figure 4. Figure 8 is a schematic diagram showing the inter-body relative center of gravity of bicycle 2 shown in Figure 4. Figure 9 is a schematic diagram showing the inter-body relative center of gravity of bicycle 3 shown in Figure 4. Figure 10 is a diagram showing the relative positions of the temporary position information and inter-body correction position information with respect to the inter-body relative center of gravity shown in Figures 7-9. Figure 11 is a schematic diagram showing the temporary center of gravity of bicycles 1-3 shown in Figure 4. Figure 12 is a schematic diagram showing the position information of bicycles 1-3 shown in Figure 4. Figure 13 shows an example of the hardware configuration of an information processing device related to an application example.

[0018] Hereinafter, an information processing apparatus, an information processing method, and an information processing program according to one embodiment of the present disclosure will be described in detail with reference to the drawings. In the following description, components having substantially the same function and configuration will be denoted by the same reference numerals, and redundant explanations will be given only when necessary.

[0019] (First Embodiment) (Configuration Example) Figure 1 is a diagram showing an example of the functional configuration of an information processing system 1 equipped with an information processing device 10 according to the first embodiment. The information processing system 1 is a positioning system that acquires and displays the location information of a moving object with high accuracy and high frequency.

[0020] A moving object is an object that is the target of a positioning system, such as a mobile device, vehicle, person, or animal. For example, information processing system 1 is a bicycle road race viewing system that displays the position information of each bicycle in real time during a bicycle road race in which multiple bicycles compete, thereby clearly representing the behavior of the bicycles.

[0021] The information processing system 1 comprises an information processing device 10, an absolute position detection unit 20, a relative position detection unit 30, a mobile object correction position detection unit 40, and a position information display unit 50. The information processing device 10 is wirelessly connected to each of the absolute position detection unit 20, the relative position detection unit 30, the mobile object correction position detection unit 40, and the position information display unit 50, and is capable of sending and receiving information to and from each of the absolute position detection unit 20, the relative position detection unit 30, the mobile object correction position detection unit 40, and the position information display unit 50.

[0022] The information processing device 10 includes an absolute position information acquisition unit 11, a relative position information acquisition unit 12, a moving object correction position information acquisition unit 13, and a position information generation unit 14.

[0023] The absolute position information acquisition unit 11 acquires absolute position information of a moving object based on the detection results obtained from the absolute position detection unit 20. The absolute position information is position information obtained by wide-area positioning and is position information in a coordinate space that can uniquely specify the position of the moving object. Typically, the absolute position information includes the latitude, longitude, and altitude of the bicycle. The absolute position detection unit 20 is a means for detecting the absolute position information of the moving object to be positioned, and is attached to each moving object to be positioned, for example. As the absolute position detection unit 20, for example, a receiver such as a GNSS device or GPS device used in a wide-area positioning system can be used. In this case, the absolute position information includes the position of the moving object as determined by a satellite positioning system.

[0024] The relative position information acquisition unit 12 acquires relative position information of a moving object based on the detection results obtained from the relative position detection unit 30. Relative position information is position information obtained by narrow-area positioning and indicates the position relative to a past position. Relative position information is, for example, information indicating the amount of movement or change from the previous position. The relative position detection unit 30 is a means for detecting the relative position information of the moving object to be positioned, and is attached to each moving object to be positioned, for example. As the relative position detection unit 30, for example, an acceleration sensor, a magnetic sensor, or an angular velocity sensor used in a narrow-area positioning system can be used. Note that other narrow-area positioning systems using other types of devices may also be used.

[0025] The inter-moving object correction position information acquisition unit 13 acquires inter-moving object correction position information of a moving object based on the detection results obtained from the inter-moving object correction position detection unit 40. Inter-moving object correction position information is position information in which the position of a moving object is estimated using the relative positions between moving objects included in a group of moving objects. Typically, inter-moving object correction position information is position information in a coordinate space that can uniquely specify the position of a moving object.

[0026] The inter-mobile correction position detection unit 40 is a means for detecting inter-mobile correction position information of the mobile object to be positioned, and is attached, for example, to mobile objects surrounding the mobile object to be positioned. Typically, each of the multiple mobile objects included in a group of mobile objects traveling as a group is designated as a positioning target, and each of the multiple mobile objects included in the group is provided with an inter-mobile correction position detection unit 40 for detecting the positions of other mobile objects included in the group.

[0027] As the inter-moving object correction position detection unit 40, for example, a mobile terminal usable as a Wi-Fi access point (e.g., a smartphone), an RFID tag, or a beacon that emits a Bluetooth signal can be used. Alternatively, as the inter-moving object correction position detection unit 40, a device used in known methods such as ultra-wideband wireless communication technology (UWB) may be used. Furthermore, as the inter-moving object correction position detection unit 40, an RGB camera such as a camera mounted on a smartphone or an ultra-compact 360° camera (action camera) may be used. In this case, the inter-moving object correction position information acquisition unit 13 inputs the surrounding video acquired in real time from the RGB camera into a machine learning model to perform object estimation and depth estimation, and estimates the inter-moving object correction position information of moving objects included in the surrounding video. Furthermore, the accuracy of the estimation of inter-moving object correction position information may be improved by combining the above method with depth sensor information acquired by LiDAR.

[0028] The location information generation unit 14 generates location information for each mobile object using the absolute location information, relative location information, and inter-mobile object correction location information of multiple mobile objects included in the mobile object group. Specifically, when the location information generation unit 14 generates estimated location information by estimating the position of a mobile object using absolute location information obtained by wide-area positioning and relative location information obtained by narrow-area positioning, it corrects the estimated location information using inter-mobile object correction location information with other mobile objects moving in the group. Furthermore, the location information generation unit 14 may generate location information using only information of other mobile objects located within a certain distance from the mobile object. The location information generation unit 14 outputs the generated location information to the location information display unit 50.

[0029] The location information display unit 50 continuously acquires location information generated by the location information generation unit 14 from the information processing device 10 and displays the latest location information of the moving object and a movement history showing changes in location information.

[0030] (Operation Example) Next, an operation example of the information processing system 1 including the information processing apparatus 10 configured as described above will be described. FIG. 2 is a flowchart illustrating an example of a procedure of processing executed by the information processing system 1. Here, as an example, a case where the information processing system 1 is a viewing system for a bicycle road race in which a plurality of bicycles compete will be described. The information processing system 1 displays the position information of each bicycle in the bicycle road race moment by moment. In bicycle road races, it is common to reduce the influence of wind and suppress physical strength consumption, fuel consumption, tire wear and the like by traveling in a group. FIG. 3 is a schematic diagram illustrating an example of a scene of an automobile road race. Each bicycle is equipped with a GNSS device, an acceleration sensor, a magnetic sensor, an angular velocity sensor, and an RGB camera. The GNSS device is an example of the absolute position detection unit 20. The acceleration sensor, the magnetic sensor, and the angular velocity sensor are examples of the relative position detection unit 30. The RGB camera is an example of the inter-moving-body corrected position detection unit 40. Further, the inter-moving-body corrected position information is generated using a machine learning model that executes object estimation and depth estimation on surrounding video acquired from the RGB camera. It is also assumed that n bicycles participating in the bicycle road race are set as positioning targets. The n bicycles are an example of a moving body group that moves collectively.

[0031] Note that the processing procedure in the processing described below is merely an example, and each processing can be appropriately modified as far as possible. In addition, regarding the processing procedure described below, steps can be appropriately omitted, replaced, and added according to the embodiment.

[0032] First, the absolute position information acquisition unit 11 of the information processing apparatus 10 acquires absolute position information of the bicycle (Step S101). At this time, the absolute position information acquisition unit 11 acquires the absolute position p of each bicycle from a receiver such as a GNSS device attached to each bicycle a (i,t). Here, "i" indicates a specific bicycle, and "t" means the time at which positioning is performed. The absolute position information acquisition unit 11 acquires the absolute position p of each of the n bicycles participating in the bicycle road race a (i,t), and acquires the absolute position information P which is the set thereofa (t) = {p a (0,t),p a (1,t), ...,p a The (n,t)} is generated. Furthermore, the absolute position p is determined using a Kalman filter or similar method. a The error in (i,t) may be corrected.

[0033] Next, the relative position information acquisition unit 12 of the information processing device 10 acquires the relative position information of the bicycles (step S102). At this time, the relative position information acquisition unit 12 calculates the distance traveled D for each bicycle by integrating the acceleration acquired from the acceleration sensor twice at time t. ｔ The azimuth angle φ_ is obtained and, using the detection result of the angular velocity sensor or magnetic sensor, is calculated. t And elevation angle θ_ t The relative position information acquisition unit 12 then uses the PDR to determine the travel distance D. ｔ , azimuth φ_ t and elevation / depression angle θ_ t Using the relative position p of each bicycle r Obtain (i,t). In PDR, the relative position p at time t+1. r When obtaining (i,t+1) = (xt+1,yt+1,zt+1), the following equation (2) is used. The relative position information acquisition unit 12 obtains the relative position p of each of the n bicycles participating in the road cycling race. r The relative position information P is obtained by acquiring (i,t) and the set of these (i,t) values. r (t) = {p r (0,t),p r (1,t), ...,p r Generates (n,t).

[0034]

[0035] Next, the inter-mobile correction position information acquisition unit 13 of the information processing device 10 acquires inter-mobile correction position information (step S103). At this time, the inter-mobile correction position information acquisition unit 13 uses a trained machine learning model to obtain the positional relationship with other bicycles from the surrounding images acquired from the RGB cameras of each bicycle, and uses that positional relationship to estimate the position of the other bicycles.

[0036] Next, the inter-mobile correction position information acquisition unit 13 targets only other bicycles located within a certain distance from each bicycle and acquires the inter-mobile correction position p of the target bicycle. ro Generates (i,j,t). "j" represents another bicycle located within a certain distance from bicycle i. If the resolution of the surrounding video used is low, false detection of bicycles may occur, or the inter-moving object correction position p may be incorrect. ro The estimation accuracy of (i,j,t) decreases. Therefore, the inter-moving object correction position p is applied only to other bicycles located within a certain distance from the bicycle. ro By obtaining (i,j,t), the final position estimation accuracy is improved. The threshold for determining the target bicycle may be set statically in advance through preliminary experiments, or it may be set dynamically according to the likelihood of detecting a bicycle from the video.

[0037] Here, we assume that out of n bicycles participating in a road race, n' bicycles (n' ≤ n-1) are located within a certain distance from bicycle i. The inter-moving object correction position information acquisition unit 13 determines the inter-moving object correction position s(i,t) = {p ro (i,0,t),p ro (i,1,t),…,p ro The unit 13 then generates {s(i, n', t)}, which is a set of the corrected positions s(i, t) of each of the n bicycles participating in the road race.

[0038] Next, the position information generation unit 14 of the information processing device 10 generates absolute position information P through the processing from step S104 to step S106. a (t), relative position information P r Using (t) and the inter-moving object correction position information S(t), position information for each bicycle is generated. In the actual process, position information is generated in three-dimensional space, but for the sake of simplicity, the following explanation will describe how to generate position information on a two-dimensional plane.

[0039] The position information generation unit 14 first generates provisional position information indicating the estimated position of each bicycle using absolute position information and relative position information according to the following equation (3) (step S104). Provisional position p s (i,t) is estimated position information obtained by estimating the absolute position of each bicycle. In equation (3), "p(i,t)" indicates the position of bicycle i that was generated in the past. The calculation formula in equation (3) is a conventional position estimation method that combines wide-area positioning and narrow-area positioning. As shown in equation (3), when "t" is less than or equal to "0", that is, in the initial state, the absolute position p at time "0" is a (i,0) is used as the temporary position. On the other hand, if "t" is greater than "0", the position p(i,t-1) generated at the previous time t-1 and the relative position p at time t are used. r The provisional position is obtained by adding (i,t). Alternatively, other position estimation methods using the relative position at time t-1 and time t may be used. Furthermore, to avoid the accumulation of relative position errors, the system may initialize the error by using the actually measured absolute position as the provisional position instead of the provisional position generated using equation (3) at regular intervals. The position information generation unit 14 generates the provisional position p of each of the n bicycles participating in the road cycling race. s We obtain (i,t) and the set of these (i,t) is the provisional position information P. s (t) = {p s (0,t), p s (1,t), ..., p s Generates (n,t).

[0040]

[0041] Figure 4 shows the provisional position information P of three bicycles (bicycle 1, bicycle 2, bicycle 3) moving on the road at time t. s (t) = {p s (1,t),p s (2,t),p s This diagram schematically shows (3,t). Bicycles 1-3 are located within a certain distance of each other, and no other bicycles are located within a certain distance of any of bicycles 1-3. That is, bicycle 1 and bicycle 2, bicycle 2 and bicycle 3, and bicycle 1 and bicycle 3 are each located within a certain distance of each other. (Assumed position p)s (i,t) is the relative position p(i,t-1) at time t-1 and p at time t. r It can be determined using (i,t). In Figure 4, the temporary position p s (i,t) is shown as a shaded circle, the position p(i,t-1) generated at time t-1 is shown as a white circle, and the relative position p r (i,t) is indicated by an arrow.

[0042] Figure 5 shows the corrected position between moving objects s(1,t)={p} of bicycles 2 and 3 as seen from bicycle 1 in the example of Figure 4. ro (1,2,t),p ro This is a schematic diagram of (1,3,t). Inter-moving object correction position p ro (1,2,t) are the position coordinates of bicycle 2 estimated using surrounding video footage taken by bicycle 1. Inter-moving object correction position p ro (1,3,t) are the position coordinates of bicycle 3 obtained using the surrounding video captured by bicycle 1. In Figure 5, the inter-moving correction position p of bicycle 2 is shown. ro Correction position p between (1,2,t) and the moving bicycle 3 ro (1,3,t) is shown by a dashed circle.

[0043] An error may occur between the provisional position estimated using data acquired by the user and the inter-body correction position estimated using data acquired by another bicycle. For example, as shown in Figure 5, the provisional position p of bicycle 2. s (2,t) and the inter-body correction position p of bicycle 2, which was acquired using the surrounding video footage taken by bicycle 1. ro There is a positional error at (1,2,t).

[0044] Figure 6 is an extended version of Figure 5, overlaying the provisional positions and inter-body correction positions of three bicycles (bicycle 1, bicycle 2, and bicycle 3). Figure 6 shows that the inter-body correction position generated by another bicycle is inaccurate due to errors caused by individual differences in RGB cameras, etc. For example, the inter-body correction position p of bicycle 2, acquired using surrounding video footage captured by bicycle 1, is shown. ro (1,2,t) and the inter-moving correction position p of bicycle 2, which was acquired using the surrounding video footage taken by bicycle 3. roAn error has occurred in (3,2,t).

[0045] Next, the position information generation unit 14 generates position correction information for each bicycle using the provisional position information and the inter-moving body correction position information (step S105). At this time, the position information generation unit 14 first calculates the inter-moving body relative center of gravity c for each bicycle i using the following equation (4). r Find (i,t). In equation (4), "n" represents the total number of bicycles i and other bicycles j located within a certain distance from bicycle i. Relative center of gravity c between moving objects r (i,t) is the estimated hypothetical position p using the data acquired by the user. s (i,t) and the inter-moving object correction position p estimated using data acquired from other bicycles. ro This is the center of gravity position of (i,j,t). Figure 7 shows the relative center of gravity c between moving parts of bicycle 1 shown in Figure 4. r Figure 8 schematically shows (1,t), and Figure 8 shows the relative center of gravity c between moving parts of the bicycle 2 shown in Figure 4. r Figure 9 schematically shows (2,t), and Figure 9 shows the relative center of gravity c between moving parts of the bicycle 3 shown in Figure 4. r This is a schematic diagram showing (3,t).

[0046]

[0047] Furthermore, using equation (4), the relative center of gravity c between moving bodies is used. r When generating (i,t), a provisional position p is determined based on the distance from the estimated position estimated using a Kalman filter or the like. s (i,t) and the corrected position p between moving objects ro You may assign weights to (i,j,t).

[0048] Next, the position information generation unit 14 calculates the relative center of gravity c between moving bodies for each bicycle i using the following equation (5). r The provisional position p for (i,t) s The relative position p' of (i,t) s The (i,t) is determined. The position information generation unit 14 calculates the relative center of gravity c between moving bodies for each bicycle i using the following equation (6). r Inter-moving object correction position p for (i,t) ro The relative position p' of (i,j,t) ro Find (i,t).

[0049]

[0050]

[0051] Next, the position information generation unit 14 calculates the corrected position r for each bicycle i according to the following equation (7) c (i,t). In equation (7), "n" denotes the total number of other bicycles j located within a certain distance from the bicycle i and the bicycle i itself. The corrected position r c (i,t) is the average of the relative position p' s (i,t) and the relative position p' ro (i,j,t). Figure 10 is a schematic diagram in which the relative position p' s (i,t) and the relative position p' ro (i,j,t) are superimposed with an arbitrary centroid point c(t) as a reference. In the lower left of Figure 10, as an example, the relative position p' of bicycle 1 s (1,t), the relative position p' ro (1,j,t), and the corrected position r c (1,t) are shown in an enlarged manner. Here, the corrected position r c (1,t) is indicated by a thick outlined white circle.

[0052]

[0053] According to the above equation (7), by calculating the corrected position r using the average value of the relative position p' s (i,t) and the relative position p' ro (i,t), the corrected position r c (i,t) in which the error occurring in the temporary position p s (i,t) is reduced is generated. The corrected position r with reduced error occurring in (i,t) c (i,t) is generated. The position information generation unit 14 acquires the corrected position r of each of the n bicycles participating in the bicycle road race c (i,t), and generates the corrected position information R(t)={r which is the set thereof c (0,t), r c (1,t),..., r c (n,t)}.

[0054] Next, the position information generation unit 14 generates position information for each bicycle using the corrected position information (step S106). At this time, the position information generation unit 14 generates a temporary position p for each bicycle i using the following equation (8). s From (i,t), the hypothetical centroid c s Find (i,t). In equation (8), "n" represents the total number of bicycles j located within a certain distance from bicycle i, and the total number of bicycles i. The hypothetical centroid c s (i,t) is the temporary position p of bicycle i. s (i,t) and the hypothetical position p of another bicycle j located within a certain distance from bicycle i. s This is the center of gravity of (j,t). Figure 11 shows the hypothetical center of gravity c of bicycles 1-3 shown in Figure 4. s This is a schematic diagram showing (i,t). In Figure 11, the hypothetical centroid c s (i,t) is indicated by an X mark.

[0055]

[0056] Next, the position information generation unit 14 generates the position p(i,t) of each bicycle i using the following equation (9). Position p(i,t) is the hypothetical center of gravity c s (i,t) Correction position r c This is corrected by (i,t). Figure 12 schematically shows how the position p(i,t) is determined. In Figure 12, the position p(i,t) is shown as a circle with a thick border. Since the position p(i,t) of each bicycle is corrected using measurement data from multiple bicycles, the error is reduced compared to when the measurement data from that bicycle is used alone. The position information generation unit 14 generates position information P(t)={p(0,t), p(1,t),…, p(n,t)}, which is a set of the positions p(i,t) of each of the n bicycles participating in the road race.

[0057]

[0058] Next, the information processing device 10 transmits location information P(t) to the location information display unit 50. The location information display unit 50 updates the display based on the acquired location p(i,t) of each moving object i (step S107). At this time, the location information display unit 50 displays the latest locations of the n bicycles participating in the bicycle road race in a virtual space as shown in Figure 3.

[0059] The information processing system 1 repeatedly executes the processes from step S101 to step S107 at arbitrary time intervals, and repeatedly acquires the latest position information P(t) for each bicycle. In this case, the position information P(t) generated at time t is used as the temporary position p at the next time (t+1). s This is used when determining (i, t+1). The information processing system 1 then displays the latest position of each bicycle on the position information display unit 50 in real time.

[0060] (Function and Effects) The information processing device 10 of this embodiment comprises an absolute position information acquisition unit 11, a relative position information acquisition unit 12, an inter-moving body correction position information acquisition unit 13, and a position information generation unit 14. The absolute position information acquisition unit 11 acquires absolute position information indicating the absolute positions of multiple moving bodies included in a group of moving bodies that move together. The relative position information acquisition unit 12 acquires relative position information of multiple moving bodies included in the group of moving bodies. The inter-moving body correction position information acquisition unit 13 acquires inter-moving body correction position information, which estimates the position of a moving body using the relative positions between multiple moving bodies included in the group of moving bodies. The position information generation unit 14 generates position information indicating the position of a moving body based on the absolute position information, relative position information, and inter-moving body correction position information.

[0061] For example, absolute position information is information obtained by wide-area positioning, and relative position information is information obtained by narrow-area positioning. The position information generation unit 14 corrects the estimated position information of the moving object, which is generated using the absolute position information and relative position information, using the inter-moving object correction position information.

[0062] With the above configuration, the information processing device 10 of this embodiment corrects the absolute position information of a moving object obtained by wide-area positioning using the relative position information of a moving object obtained by narrow-area positioning. It focuses on the fact that a group of moving objects moves as a collective and corrects the position information containing errors by using the relative positions of the moving objects. As a result, the position information of each moving object is corrected using measurement data from multiple moving objects, so the errors contained in the position information can be reduced compared to when the measurement data from each moving object is used individually, and the accuracy of estimating the position of the moving objects can be improved.

[0063] Furthermore, in addition to inter-object correction position information, information from sensors used by the information processing system 1 may also be used. For example, a deep learning-based position estimation method may be used that learns the position information of each mobile object and the detection results of sensors as features to estimate the position of each mobile object. In this case, for example, a machine learning model trained to output the position information of each mobile object by inputting the position information of each mobile object moving in a group (absolute position, relative position, and inter-object correction position) as well as the detection results of various sensors.

[0064] Furthermore, the greater the distance between moving objects, the lower the accuracy of the sensors used to detect inter-object correction position information, and the lower the resolution of the video used to detect inter-object correction position information, which tends to reduce the prediction accuracy of the inter-object correction position information. On the other hand, the information processing device 10 of this embodiment generates position information using inter-object correction position information, using only information from other moving objects located within a certain distance from the moving object. By correcting the position information using only information from other moving objects that are close by, the estimation accuracy of the position information can be further improved. The threshold value that defines the certain distance may be set in advance according to the results of preliminary experiments conducted beforehand, or it may be adjusted in real time using the likelihood of the object detected in the detection of inter-object correction position information.

[0065] Furthermore, to avoid the accumulation of errors in relative position information, the accumulated errors in relative position information may be initialized by using absolute position information as temporary position information at regular time intervals.

[0066] Furthermore, the information processing system 1 using the information processing device 10 can improve the positioning accuracy in the location information acquisition system. The information processing system 1 can display the behavior of moving objects in an easy-to-understand manner by constantly updating the position of each moving object displayed on the location information display unit 50 using the location information acquired from the information processing device 10. For example, in a positioning system where a group of moving objects are traveling as a group, such as in a spectator system for a bicycle road race, it is possible to improve the estimation accuracy by correcting the estimated position of each bicycle using the relative position between bicycles.

[0067] The location information generated by the information processing device 10 of this embodiment may be displayed on a large monitor for spectators to see at a bicycle road race venue, displayed on a mobile terminal used by teams participating in a bicycle road race to understand the physical condition and pacing of their athletes and to formulate strategies, or used for other purposes where the location information of each moving object in a group of moving objects can be utilized.

[0068] (Application Example) Next, an example of the hardware configuration of the information processing device 10 according to the above embodiment and modified example will be described with reference to the block diagram in Figure 13.

[0069] As shown in Figure 13, the information processing device 10 includes, for example, a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a communication interface 104, and storage 105.

[0070] The CPU 101 is an integrated circuit capable of executing various programs and controls the overall operation of the information processing device 10. The ROM 102 is, for example, a non-volatile semiconductor memory and stores programs and control data for controlling the information processing device 10. The RAM 103 is, for example, a volatile semiconductor memory and is used as a workspace for the CPU 101. The communication interface 104 is an input / output component for communicating information with the outside world, such as USB, HDMI (registered trademark), or a network interface. The storage 105 is a non-volatile storage device. The storage 105 stores the system software of the information processing device 10. The CPU 101 may also be called a "processor". The information processing device 10 may also be equipped with a display. The display may be, for example, a liquid crystal display or a touch panel display, and displays information.

[0071] In the above embodiment, the CPU 101 of the information processing device 10 may be other circuits (or processors). For example, the information processing device 10 may have a GPU (Graphics Processing Unit), an NPU (Neural Network Processing Unit), an MPU (Micro Processing Unit), etc. instead of a CPU. Each of the processes described in each embodiment may be implemented by dedicated hardware. The processing of the information processing device 10 may be a mixture of processes executed by software and processes executed by hardware, or it may be one or the other.

[0072] (Other Embodiments) The functions of the elements disclosed herein may be implemented using circuits or processing circuits that include general-purpose processors, special-purpose processors, integrated circuits, ASICs (Application Specific Integrated Circuits), FPGAs (Field Programmable Gate Arrays), other conventional general-purpose circuits, and / or combinations thereof that are programmed using one or more programs stored in one or more memories, or otherwise configured to perform the disclosed functions. A processor is considered a processing circuit or circuit because it includes transistors and other circuits. A processor may be a programmed processor that executes programs stored in memory.

[0073] In this disclosure, a circuit, unit, or means is hardware that performs the enumerated functions, or hardware programmed to perform the enumerated functions. Hardware may be any hardware disclosed herein that is programmed or configured to perform the enumerated functions.

[0074] Furthermore, the computer instructions contained within the computer program provide logic and routines that enable hardware (e.g., processing circuits or circuits) to perform the methods disclosed herein. This computer program can be implemented in commonly known forms, such as computer-readable storage media, computer program products, memory devices, recording media such as CD-ROMs and DVDs, and / or memory for FPGAs and ASICs.

[0075] This invention is not limited to the embodiments described above, and in practice, the components can be modified and implemented without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the embodiments. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components from different embodiments may be appropriately combined.

[0076] 1... Information processing system 10... Information processing device 11... Absolute position information acquisition unit 12... Relative position information acquisition unit 13... Inter-moving object correction position information acquisition unit 14... Position information generation unit 20... Absolute position detection unit 30... Relative position detection unit 40... Inter-moving object correction position detection unit 50... Position information display unit 101... CPU 102... ROM 103... RAM 104... Communication interface 105... Storage p... Position p a ...absolute position p r ...relative position p ro ...Correction position between moving objects c r ...relative centroid between moving objects p' s, p' ro ...relative position r c ...correction position c s ... False center of gravity

Claims

1. An information processing device comprising: an absolute position information acquisition unit that acquires absolute position information indicating the absolute positions of multiple moving objects included in a group of moving objects; a relative position information acquisition unit that acquires relative position information indicating the relative positions of the multiple moving objects relative to the past absolute positions; an inter-moving object correction position information acquisition unit that acquires inter-moving object correction position information that estimates the position of the moving objects using the relative positions between the multiple moving objects; and a position information generation unit that generates position information indicating the position of the moving objects based on the absolute position information, the relative position information, and the inter-moving object correction position information.

2. The information processing apparatus according to claim 1, wherein the absolute position information is information obtained by wide-area positioning, the relative position information is information obtained by narrow-area positioning, and the position information generation unit generates the position information by correcting the estimated position information obtained by correcting the absolute position information using the relative position information, and then correcting the inter-moving-body correction position information.

3. The information processing apparatus according to claim 2, wherein the position information generation unit generates the position information of each of the moving bodies using the estimated position information of other moving bodies whose distance from the moving body is less than or equal to a threshold and the inter-moving body correction position information.

4. A program that causes a computer to execute the following functions: a function to acquire absolute position information indicating the absolute positions of multiple moving objects included in a group of moving objects; a function to acquire relative position information indicating the past relative positions of the multiple moving objects with respect to the absolute positions; a function to acquire inter-moving object correction position information which estimates the position of the moving objects using the relative positions between the multiple moving objects; and a function to generate position information indicating the position of the moving objects based on the absolute position information, the relative position information, and the inter-moving object correction position information.

Images