Object Tracking Device

Abstract

To provide an object tracking device capable of suppressing reduction in estimation accuracy of a target's position.SOLUTION: An object tracking device in one aspect of the present disclosure includes a sensor section, a contour calculation section 13, a predicted value calculation section 15, an association section 17, and an estimation section 19. The sensor section 11 acquires a plurality of observed values. The contour calculation section calculates a predicted contour on the basis of a predetermined object shape model and an estimated value calculated in the past. The predicted value calculation section calculates a plurality of predicted values independent of the plurality of acquired observed values and located on the predicted contour and / or within a predetermined range inside the predicted contour. The association section associates each of the plurality of predicted values with at least one of the plurality of observed values and generates a plurality of associated sets. The estimation section calculates a current estimated value on the basis of the plurality of associated sets.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present disclosure relates to a technology for a moving body to track surrounding objects. [Background technology]

[0002] In the extended object tracking (hereinafter referred to as EOT) described in Non-Patent Document 1, under the assumption that the reflected signal received by the radar device occurs at the contour of the target, predicted values ​​on the predicted contour of the target are associated with observed values, and the motion state of the target is estimated in time series. Specifically, in the above-mentioned EOT, the intersection of the predicted contour with a line passing through the center of the predicted contour and the observed value is calculated as a predicted value, and the calculated predicted value is associated with the observed value on the same line as the predicted value. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Soichiro Tokizawa, Keisuke Yoneda, Naoki Suganuma, "Extended Object Tracking with High Stability and Real-Time Performance for Autonomous Driving," Transactions of the Society of Automotive Engineers of Japan, Vol. 52, No. 5, September 2021. Summary of the Invention [Problem to be solved by the invention]

[0004] In the EOT, predicted values ​​are calculated based on observed values. Therefore, once a predicted value is associated with an erroneous observed value, the estimation of the target's motion state continues to be affected by the erroneous association. This ultimately reduces the accuracy of the target position estimation.

[0005] One aspect of the present disclosure provides an object tracking device capable of suppressing a decrease in the accuracy of estimating the position of a target. [Means for solving the problem]

[0006] An object tracking device according to one aspect of the present disclosure includes a sensor unit, a contour calculation unit (13), a predicted value calculation unit (15), an associating unit (17), and an estimating unit (19). The sensor unit is a sensor unit (11) mounted on a moving body (50) and configured to transmit and receive sensor waves around the moving body and acquire multiple observation values ​​corresponding to different reflection positions. The contour calculation unit is configured to calculate a predicted contour, which is a prediction of the current contour of the target, based on a predetermined shape model of the object and previously calculated estimated values ​​of the target's state, including the object's position and orientation. The predicted value calculation unit is configured to calculate multiple predicted values ​​located on the predicted contour and / or within a predetermined range inside the predicted contour, independently of the multiple observation values ​​acquired by the sensor unit. The associating unit is configured to generate multiple association sets by associating each of the multiple predicted values ​​with at least one of the multiple observation values. The estimating unit is configured to calculate a current estimated value based on the multiple association sets generated by the associating unit.

[0007] In an object tracking device according to one aspect of the present disclosure, multiple predicted values ​​are calculated independently from multiple observed values, thereby preventing erroneous associations between predicted values ​​and observed values ​​from continuing and preventing a decrease in the accuracy of estimating the position of a target object. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a hardware configuration of an object tracking device according to a first embodiment. [Figure 2] 1 is a block diagram showing the functional configuration of an object tracking device according to a first embodiment. [Figure 3] 2 is a diagram showing an example of a mounting position and detection range of a sensor unit in a vehicle according to the first embodiment; FIG. [Figure 4] 5 is a diagram showing another example of the mounting position and detection range of the sensor unit in the vehicle according to the first embodiment. FIG. [Figure 5] 4 is a flowchart showing a tracking process executed by the object tracking device according to the first embodiment. [Figure 6] 10 is a flowchart showing a predicted value calculation process executed by the object tracking device according to the first embodiment. [Figure 7] 4 is a flowchart showing an estimation process executed by the object tracking device according to the first embodiment. [Figure 8] FIG. 3 is a diagram showing a predicted contour, observed values, and predicted values ​​according to the first embodiment. [Figure 9] FIG. 3 is a diagram showing associations between one predicted value and multiple observed values ​​according to the first embodiment. [Figure 10] 10 is a diagram showing a state in which a predicted value is calculated in a direct reflection area on the contour of an ellipse according to the first embodiment. FIG. [Figure 11] 10A and 10B are diagrams illustrating a state in which predicted values ​​are calculated in a direct reflection area on a rectangular contour according to the first embodiment. [Figure 12] FIG. 10 is a diagram showing a state in which a predicted value is calculated in a direct reflection region on the contour of an ellipse when a sensor wave is irradiated from behind an object in the first embodiment. [Figure 13] FIG. 10 is a diagram showing a state in which a predicted value is calculated in a direct reflection area on the contour of an ellipse when the sensor is illuminated from the side of an object in the first embodiment. [Figure 14] FIG. 10 is a diagram showing a state in which the closer to the center of the predetermined range on the contour of an ellipse according to the first embodiment, the narrower the position interval of predicted values. [Figure 15] 10A and 10B are diagrams illustrating a state in which predicted values ​​are calculated on and inside the contour of an ellipse when a sensor wave is irradiated from behind an object in the first embodiment. [Figure 16] 10 is a flowchart showing a predicted value calculation process executed by an object tracking device according to a second embodiment. [Figure 17] 10 is a flowchart showing an estimation process executed by an object tracking device according to a second embodiment. [Figure 18] 11 is a flowchart showing a predicted value calculation process executed by an object tracking device according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] (First embodiment) <1. Configuration> The configuration of an object tracking device 10 according to this embodiment will be described with reference to Figures 1 and 2. The object tracking device 10 includes a sensor unit 11 and a processing unit 20, and is mounted on a vehicle 50, which is an automobile. The sensor unit 11 is a radar, a lidar, a sonar, or the like. A radar transmits radio waves such as millimeter waves as sensor waves and receives reflected waves generated when the radar waves are reflected by an object. A lidar transmits light as sensor waves and receives reflected waves generated when the light is reflected by an object. A sonar transmits sound waves as sensor waves and receives reflected waves generated when the sound waves are reflected by an object.

[0010] As shown in FIG. 3, the sensor unit 11 may be mounted in the front center of the vehicle 50 (for example, in the center of the front bumper) and have a detection area A1 in the front center of the vehicle 50. Alternatively, as shown in FIG. 4, the sensor unit 11 may be mounted on the left front, right front, left rear, and right rear of the vehicle 50 (for example, the left and right ends of the front bumper and the left and right ends of the rear bumper) in addition to the front center of the vehicle 50. That is, the sensor unit 11 may have detection areas A2 in the left front, right front, left rear, and right rear of the vehicle 50 in addition to the detection area A1. The sensor unit 11 may be mounted in at least one location among the front center, left front, right front, left rear, and right rear of the vehicle 50.

[0011] The processing device 20 includes a microcomputer including a CPU, a ROM, a RAM, etc. The processing device 20 realizes the functions of the contour calculation unit 13, the predicted value calculation unit 15, the association unit 17, and the estimation unit 19 by the CPU executing a program stored in a non-transient physical recording medium. The processing device 20 is provided with the above-mentioned various functions, thereby performing a tracking process for estimating the motion state of an object in a time series manner. Details of the above-mentioned various functions will be described later. In this embodiment, the ROM corresponds to the non-transient physical recording medium. Note that some or all of the various functions realized by the processing device 20 may be realized using hardware that combines logic circuits, analog circuits, etc.

[0012] <2. Processing> <2-1. Tracking process> The tracking process executed by the object tracking device 10 will be described with reference to the flowchart of Fig. 5. The object tracking device 10 repeatedly executes this tracking process at a predetermined processing cycle.

[0013] In S10, as shown in FIG. 9, the sensor unit 11 transmits and receives sensor waves around the vehicle 50 and acquires a plurality of observation values ​​P3. The sensor unit 11 is a high-resolution sensor that can acquire a plurality of reflected waves reflected at different reflection positions (i.e., reflection points) on a single object by transmitting a sensor wave once, and acquire an observation value P3 based on each reflected wave. Therefore, each of the plurality of observation values ​​P3 corresponds to a different reflection position on a single object. Each of the plurality of observation values ​​P3 includes a reflection position (specifically, a distance from the sensor unit 11 to the reflection point and an orientation of the reflection point relative to the sensor unit 11) as a physical quantity. Each of the plurality of observation values ​​P3 may include a relative velocity in addition to the reflection position as a physical quantity.

[0014] Subsequently, in S20, the contour calculation unit 13 executes a prediction process to calculate a predicted contour L1 of the target. The predicted contour L1 is a predicted value of the contour of the shape of the modeled target. As described above, the sensor unit 11 can obtain multiple observation values ​​P3 from a single object. The shape of the object can be estimated from the distribution of the multiple observation values ​​P3. Therefore, the object tracking device 10 can generate a target having a shape in addition to a motion state based on the multiple observation values ​​P3. The shape here has an area, unlike a mass point which does not have an area. The object tracking device 10 performs extended object tracking (hereinafter referred to as EOT) using a shape model of the object. Extended object tracking is a method of modeling a target by assuming that the target has a shape, and estimating the motion state of the target over time.

[0015] As shown in Fig. 9, the contour calculation unit 13 calculates a predicted contour L1 based on a predetermined shape model of the object and an estimated value P2 calculated in a past processing cycle (specifically, the previous processing cycle). In this embodiment, the object tracked by the object tracking device 10 is an automobile (specifically, a four-wheeled vehicle). Therefore, the contour calculation unit 13 uses, as the shape model of the object, a circular model, an elliptical model shown in Fig. 10, a rectangular model shown in Fig. 11, or the like. Alternatively, the contour calculation unit 13 may estimate the size and shape of the target from the distribution of multiple observed values ​​P3, and select and use an appropriate model from multiple models prepared in advance.

[0016] The estimated value P2 is a value obtained by estimating the state of the target. The state of the target includes the position and orientation of the target. The estimated value P2 has at least one physical quantity at the reference point of the target. The object tracking device 10 estimates the motion state of the reference point in a time series manner based on multiple observed values ​​P3, multiple predicted values ​​P4, a shape model, and a predetermined filter, and calculates the estimated value P2. The predetermined filter is, for example, a Kalman filter. The estimated value P2 has, for example, the x-direction position, y-direction position, velocity, traveling direction, and angular velocity of the reference point. The x-direction corresponds to the length direction of the vehicle 50, and the y-direction corresponds to the width direction of the vehicle 50.

[0017] In this embodiment, the reference point is the center of the rear wheel axle of the automobile. The contour calculation unit 13 calculates a predicted contour L1 of the shape of the target in the current processing cycle based on the past estimated value P2 and the shape model. The interior of the predicted contour L1 corresponds to the predicted target existence area in the current processing cycle. That is, the contour calculation unit 13 predicts the target existence area in the current processing cycle based on the past estimated value P2 and the shape model. Therefore, the predicted contour L1 is likely to be calculated in the vicinity of the multiple observed values ​​P3 acquired in S10.

[0018] Subsequently, in S30, the predicted value calculation unit 15 executes a predicted value calculation process to calculate a plurality of predicted values ​​P4 located on and / or inside the predicted contour L1 calculated in S20, independently of the plurality of observed values ​​P3. That is, the predicted value calculation unit 15 calculates a plurality of predicted values ​​P4 on and / or inside the predicted contour L1 using a calculation method that is not dependent on the plurality of observed values ​​P3 (i.e., independently of the plurality of observed values ​​P3).

[0019] In the reference example shown in Fig. 8, multiple predicted values ​​P4 are calculated depending on multiple observed values ​​P3. Specifically, in the reference example, it is assumed that the sensor wave is reflected by the contour of the target, and straight lines are drawn that pass through the center point P1 and each of the multiple observed values ​​P3. Then, the intersection of each straight line with the predicted contour L1 is calculated as the predicted value P4, and the predicted value P4 and the observed value P3 on the same straight line are associated with each other. The center point P1 is the center of the predicted contour L1.

[0020] As shown in FIG. 8, when the predicted contour L1 deviates from the contour LL of the actual target, the predicted value P4 is associated with an observed value P3 that does not correspond to the predicted value P4. In FIG. 8, the predicted value P4 on the left side of the vehicle is associated with the observed value P3 on the right side of the vehicle. When calculating multiple predicted values ​​P4 based on multiple observed values ​​P3, once an erroneous association occurs, the erroneous association may continue. This may result in a decrease in the calculation accuracy of the estimated value P2. In other words, the object tracking accuracy may decrease.

[0021] Therefore, in this embodiment, the predicted value calculation unit 15 calculates a plurality of predicted values ​​P4 located within a predetermined range on and / or inside the predicted contour L1, independently from the plurality of observed values ​​P3. Details of the predicted value calculation process will be described later.

[0022] Next, in S40, the associating unit 17 associates each of the multiple predicted values ​​P4 calculated in S30 with one or more of the multiple observed values ​​P3 acquired in S10. In this embodiment, the associating unit 17 associates each of the multiple predicted values ​​P4 with two or more of the multiple observed values ​​P3 to generate multiple association sets. Specifically, the associating unit 17 associates one predicted value P4 with two or more observed values ​​P3 located within a set range from the predicted value P4 to generate one association set. Therefore, each association set includes one predicted value P4 and two or more observed values ​​P3 associated with the predicted value P4. In this way, allowing two or more associations for one predicted value P4 increases the likelihood that a correct association will be included among the two or more associations. Therefore, allowing multiple associations increases the influence of correct associations on the estimated value P2 during object tracking, causing the estimated value P2 to converge to a correct value. When the associating unit 17 associates one predicted value P4 with one predicted value P3, if the association is incorrect, the incorrect association may continue.

[0023] Subsequently, in S50, the estimation unit 19 executes an estimation process, and calculates a current estimated value P2 by applying a filter such as a Kalman filter to the plurality of association sets calculated in S40. Details of the estimation process will be described later.

[0024] <2-2. Prediction value calculation process> Next, the details of the predicted value calculation process executed by the predicted value calculation unit 15 will be described with reference to the flowchart of FIG.

[0025] In S100, the predicted value calculation unit 15 determines whether the distance of the target is equal to or greater than a threshold. The distance here is the distance of the estimated value P2 calculated in the previous processing cycle, or the distance calculated based on the position of the estimated value P2. Alternatively, the distance here may be the distance calculated based on the multiple observed values ​​P3 acquired in S10.

[0026] When a target object is located far from the sensor unit 11, the positions of the object's reflection points vary widely, and the reflection points are widely distributed throughout the area of ​​the object. Therefore, when a target object is located far from the sensor unit 11 and the prediction value calculation unit 15 calculates multiple prediction values ​​P4 located within a predetermined range on and / or inside the predicted contour L1, narrowing the predetermined range may reduce the accuracy of associating one prediction value P4 with two or more observation values ​​P3, and may reduce the accuracy of calculating the estimated value P2.

[0027] On the other hand, when a target object is located near the sensor unit 11, the positions of the reflection points of the object vary little, and the reflection points are concentrated in a specific area of ​​the object. Therefore, when a target object is located near the sensor unit 11 and the prediction value calculation unit 15 calculates multiple prediction values ​​P4 located within a predetermined range on and / or inside the predicted contour L1, widening the predetermined range may reduce the accuracy of associating one prediction value P4 with two or more observation values ​​P3, which may reduce the accuracy of calculating the estimated value P2.

[0028] Therefore, the predicted value calculation unit 15 changes the predetermined range depending on the distance of the target. If the predicted value calculation unit 15 determines in S100 that the distance is equal to or greater than the threshold, it proceeds to processing of S110, and if it determines that the distance is less than the threshold, it proceeds to processing of S120.

[0029] In S110, as shown in FIG. 8, a predetermined range is set around the entire periphery of the predicted contour L1. Then, the predicted value calculation unit 15 calculates a plurality of predicted values ​​P4 at predetermined position intervals within the predetermined range. Each of the plurality of predicted values ​​P4 has a coordinate value on the predicted contour L1. The predetermined position intervals may be equal. Alternatively, the predetermined position intervals may be smaller as the distance approaches the center of the predetermined range, so that more predicted values ​​P4 contribute to updating the estimated value P2 as the distance approaches the center of the predetermined range (see FIG. 14).

[0030] In S120, the predicted value calculation unit 15 calculates a direct reflection region of the predicted contour L1. The direct reflection region corresponds to a region where the sensor unit 11 can directly irradiate the sensor wave. As shown in FIG. 10, when the shape model is an ellipse model or a circle model, the predicted value calculation unit 15 calculates the area between the first point Pa and the second point Pb, which is closer to the sensor unit 11, as the direct reflection area. The first point Pa is a point on the predicted contour L1, and is the point of contact between a first tangent line La passing through the sensor unit 11 and the shape model. The second point Pb is a point on the predicted contour L1 different from the first point Pa, and is the point of contact between a second tangent line Lb passing through the sensor unit 11 and the shape model. FIG. 12 shows the direct reflection area when the sensor unit 11 emits a sensor wave from behind an object (specifically, a vehicle). FIG. 13 shows the direct reflection area when the sensor unit 11 emits a sensor wave from the side of the object.

[0031] 11, when the shape model is a rectangular model, the predicted value calculation unit 15 calculates the first side connecting the fourth vertex PP4 to the second vertex PP2 and the second side connecting the fourth vertex PP4 to the third vertex PP3 as the direct reflection region. The second vertex PP2 and the third vertex PP3 are two vertices adjacent to the first vertex PP1, and the first vertex PP1 is the vertex farthest from the sensor unit 11 among the four vertices of the rectangular model. The fourth vertex PP4 is a vertex between the second vertex PP2 and the third vertex PP3 and is different from the first vertex PP1.

[0032] In S130, the predicted value calculation unit 15 sets the direct reflection area to a predetermined range and calculates multiple predicted values ​​P4 at predetermined positional intervals within the predetermined range. As shown in Figures 10 and 11, the predetermined positional intervals may be equal. Alternatively, as shown in Figure 14, the predetermined positional intervals may be smaller the closer to the center of the predetermined range, in order to increase the contribution of predicted values ​​P4 to the estimated value P2 as they are closer to the center of the predetermined range.

[0033] <2-3. Estimation process> Next, the estimation process executed by the estimation unit 19 will be described in detail with reference to the flowchart of FIG.

[0034] In S200, the estimation unit 19 determines whether the following processes of S210 to S230 have been performed on all predicted values ​​P4 calculated in S20. If the estimation unit 19 determines that the processes of S210 to S230 have not been performed on all predicted values ​​P4, it selects one predicted value P4 from among the predicted values ​​P4 for which it has been determined that the processes of S210 to S230 have not been performed, and proceeds to the process of S210. If the estimation unit 19 determines that the processes of S210 to S230 have been performed on all predicted values ​​P4, it proceeds to the process of S240.

[0035] In S210, the estimation unit 19 determines whether the predicted value P4 selected in S200 is located in a direct reflection region. When the predetermined range is set to the entire periphery of the predicted contour L1, the predetermined range includes a direct reflection region and an indirect reflection region. The indirect reflection region is a region where the sensor unit 11 cannot directly irradiate the sensor wave. If the estimation unit 19 determines that the selected predicted value P4 is located in the direct reflection region, it proceeds to processing of S220. If the estimation unit 19 determines that the selected predicted value P4 is located in the indirect reflection region, it skips processing of S220 and proceeds to processing of S230.

[0036] In S220, the estimation unit 19 increases the contribution of the selected predicted value P4 to the update of the estimated value P2 above a reference value. That is, the estimation unit 19 increases the contribution by adding a predetermined value to the reference value of the contribution. When the selected predicted value P4 is located in the non-direct reflection region, the contribution becomes the reference value. An association set including a predicted value P4 located in the direct reflection region has higher reliability than an association set including a predicted value P4 located in the non-direct reflection region. Therefore, when the selected predicted value P4 is located in the direct reflection region, the estimation unit 19 increases the contribution of the selected predicted value P4 to the update of the estimated value P2.

[0037] In S230, the estimation unit 19 calculates an update amount for the estimated value P2 based on the associated observed value for the selected predicted value P4. Specifically, the estimation unit 19 calculates the update amount using a nonlinear filter such as an extended Kalman filter.

[0038] In S240, the estimated value P2 is updated by weighting and averaging the update amounts of the estimated value P2 calculated for each predicted value P4 according to the contribution of the estimated value P2. The updated estimated value P2 becomes the estimated value P2 for the current processing cycle.

[0039] <3. Effects> According to the first embodiment described above in detail, the following effects are achieved. (1) The object tracking device 10 calculates multiple predicted values ​​P4 independently from multiple observed values ​​P3. Therefore, it is possible to prevent the continued erroneous association of the predicted value P4 with at least one observed value P3, and to prevent a decrease in the estimation accuracy of the target position.

[0040] (2) One predicted value P4 is associated with multiple observed values ​​P3. By allowing multiple associations for one predicted value P4, the correct association is included among the multiple associations, and the estimated position of the target can converge to the correct observed position. This in turn improves the estimation accuracy of the target position.

[0041] (3) When a plurality of predicted values ​​P4 are calculated at predetermined position intervals within a predetermined range, the calculation process of the plurality of predicted values ​​P4 can be simplified. (4) When a plurality of predicted values ​​P4 are calculated at equal position intervals within a predetermined range, the calculation process of the plurality of predicted values ​​P4 can be simplified.

[0042] (5) Since the closer to the center of the specified range, the more likely it is that a larger number of observed values ​​P3 will be obtained, the closer to the center of the specified range, the more densely the predicted values ​​P4 are calculated, thereby improving the accuracy of estimating the target position.

[0043] (6) Many observation values ​​P3 are obtained from the direct reflection area. Therefore, by setting the predetermined range to the direct reflection area, the accuracy of estimating the target position can be improved. (7) When a circular model or an elliptical model is used as the shape model, the direct reflection area can be easily calculated by calculating the area between the first point Pa and the second point Pb as the direct reflection area.

[0044] (8) When a rectangular model is used as the shape model, the direct reflection area can be easily calculated by calculating the first side and the second side as the direct reflection area. (9) When the predetermined range is set to the entire perimeter, the contribution of the predicted value P4 in the direct reflection area to updating the estimated value P2 is calculated to be greater than the contribution of the predicted value P4 in the non-direct reflection area to updating the estimated value P2. This improves the accuracy of estimating the target position.

[0045] (10) By setting a predetermined range around the entire predicted contour L1 far from the sensor unit 11 and setting a predetermined range to the direct reflection area near the sensor unit 11, the position of the target can be estimated with high accuracy from near to far from the sensor unit 11.

[0046] (Second embodiment) <1. Differences from the First Embodiment> The second embodiment has the same basic configuration as the first embodiment, so differences will be described below. Note that the same reference numerals as those in the first embodiment indicate the same configuration, and reference will be made to the preceding description.

[0047] In the first embodiment described above, the predetermined range is changed depending on the distance to the target. In contrast, the second embodiment differs from the first embodiment in that the predetermined range is set regardless of the distance to the target. Specifically, in the second embodiment, the predicted value calculation unit 15 sets the direct reflection area to a predetermined range regardless of the distance to the target. The estimation unit 19 calculates the estimated value P4 without considering the contribution of the predicted value P4 to the update of the estimated value P2. Therefore, in the second embodiment, the predicted value calculation unit 15 executes the flowchart of FIG. 16 instead of the flowchart of FIG. 6. Also, in the second embodiment, the estimation unit 19 executes the flowchart of FIG. 17 instead of the flowchart of FIG. 7.

[0048] <2. Processing> <2-1. Prediction value calculation process> The predicted value calculation process executed by the predicted value calculation unit 15 will be described with reference to the flowchart of FIG. In S310, the predicted value calculation unit 15 calculates the direct reflection area of ​​the predicted contour L1, similarly to S120.

[0049] Next, in S320, the predicted value calculation unit 15 sets a direct reflection region to a predetermined range, as in S130, and calculates multiple predicted values ​​P4 at determined position intervals within the predetermined range. That is, in this embodiment, the predicted value calculation unit 15 calculates multiple predicted values ​​P4 in the direct reflection region regardless of the distance to the target.

[0050] <2-2. Estimation process> Next, the estimation process executed by the estimation unit 19 will be described with reference to the flowchart of FIG.

[0051] In S500, similar to S200, the estimation unit 19 determines whether or not the subsequent process of S510 has been executed for all predicted values ​​P4 calculated in S20. If the estimation unit 19 determines that the process of S510 has not been executed for all predicted values ​​P4, the estimation unit 19 selects one predicted value P4 from among the predicted values ​​P4 for which the process of S510 has been determined not to have been executed, and proceeds to the process of S510. If the estimation unit 19 determines that the process of S510 has been executed for all predicted values ​​P4, the estimation unit 19 proceeds to the process of S520.

[0052] In S510, the estimation unit 19 calculates the update amount of the estimated value P2 based on the associated observed value for the selected predicted value P4, similarly to S230. In S520, the estimation unit 19 averages the update amounts of the estimated values ​​P4 calculated for each predicted value P4 to update the estimated value P2. That is, in this embodiment, the estimation unit 19 updates the estimated value P2 by equalizing the weights of all the predicted values ​​P4.

[0053] According to the second embodiment described above, the same effects as the above-mentioned effects (1) to (8) are achieved.

[0054] (Third embodiment) <1. Differences from the second embodiment> The third embodiment has the same basic configuration as the second embodiment, and therefore the differences will be described below. Note that the same reference numerals as those in the second embodiment indicate the same configuration, and reference will be made to the preceding description.

[0055] In the second embodiment described above, the direct reflection region is set to a predetermined range regardless of the distance to the target. In contrast, the third embodiment differs from the second embodiment in that the entire periphery of the predicted contour L1 is fixed to a predetermined range regardless of the distance to the target. The estimation unit 19 calculates the estimated value P4 without considering the contribution of the predicted value P4 to the update of the estimated value P2. Therefore, in the third embodiment, the prediction value calculation unit 15 executes the flowchart of FIG. 18 instead of the flowchart of FIG. 16. Furthermore, in the third embodiment, as in the second embodiment, the estimation unit 19 executes the flowchart of FIG. 17.

[0056] <2. Prediction value calculation process> The predicted value calculation process executed by the predicted value calculation unit 15 will be described with reference to the flowchart of FIG.

[0057] In S400, the predicted value calculation unit 15 sets a predetermined range around the entire periphery of the predicted contour L1, and calculates a plurality of predicted values ​​P4 at predetermined position intervals within the predetermined range. That is, in this embodiment, the predicted value calculation unit 15 calculates a plurality of predicted values ​​P4 around the entire periphery of the predicted contour L1, regardless of the distance to the target.

[0058] According to the third embodiment described above, the same effects as the above-mentioned effects (1) to (5) and (9) are achieved, and the following effects (11) and (12) are also achieved. (11) By fixing the predetermined range, the calculation process of a plurality of predicted values ​​P4 can be simplified. (12) By setting the predetermined range to the entire periphery of the predicted contour L1, it is possible to prevent the estimated position of the target from deviating significantly from the true position.

[0059] (Other embodiments) Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments and can be implemented in various modified forms.

[0060] (a) When the predetermined range is set to the entire periphery of the predicted contour L1, the contribution of the predicted value P4 in the direct reflection region to updating the estimated value P2 may be made larger than the contribution of the predicted value P4 in the indirect reflection region to updating the estimated value P2. Furthermore, the predetermined range may be set to a range corresponding to one or more tires on the predicted contour L1. Since tires tend to have many reflection points, setting the predetermined range to a range corresponding to the tires allows the target position to be estimated with high accuracy.

[0061] (b) In the above embodiment, the predicted value calculation unit 15 changed the predetermined range depending on the distance to the target. However, the predetermined range may also be changed depending on the state of the target. When the sensor unit 11 emits sensor waves from the side of the vehicle being tracked, the observed value P3 corresponding to the reflection position inside the vehicle is not frequently acquired. On the other hand, when the sensor unit 11 emits sensor waves from the rear of the vehicle being tracked, the observed value P3 corresponding to the reflection position inside the vehicle is relatively frequently acquired. Therefore, the predicted value calculation unit 15 may change the predetermined range depending on the direction of the vehicle being tracked. For example, as shown in FIG. 15 , when the sensor unit 11 emits sensor waves from the rear of the vehicle being tracked, the predicted value calculation unit 15 may set the predetermined range to the area above and inside the predicted contour L1. When the sensor unit 11 emits sensor waves from the side of the vehicle being tracked, the predicted value calculation unit 15 may set only the area above the predicted contour L1 as the predetermined range. Dynamically setting the predetermined range depending on the state of the target improves the accuracy of estimating the target position.

[0062] (c) In the above embodiment, the object tracking device 10 is mounted on an automobile, but it may be mounted on a moving body other than an automobile. For example, the object tracking device 10 may be mounted on a moving body such as a ship, an aircraft, a motorcycle, or a drone.

[0063] (d) The object tracking device 10 and the method described herein may be implemented by a special-purpose computer configured by configuring a processor and memory programmed to perform one or more functions embodied in a computer program. Alternatively, the object tracking device 10 and the method described herein may be implemented by a special-purpose computer configured by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the object tracking device 10 and the method described herein may be implemented by one or more special-purpose computers configured by combining a processor and memory programmed to perform one or more functions with a processor configured with one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by a computer on a computer-readable non-transitory tangible recording medium. The method for implementing the functions of each component included in the object tracking device 10 does not necessarily need to include software; all of the functions may be implemented using one or more hardware components.

[0064] (e) Multiple functions possessed by one component in the above embodiments may be realized by multiple components, or one function possessed by one component may be realized by multiple components. Also, multiple functions possessed by multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Also, part of the configuration of the above embodiments may be omitted. Also, at least part of the configuration of the above embodiments may be added to or substituted for the configuration of another of the above embodiments.

[0065] (f) In addition to the object tracking device 10 described above, the present disclosure can also be realized in various forms, such as a system including the object tracking device 10 as a component, a program for causing a computer to function as the object tracking device, a non-transient physical recording medium such as a semiconductor memory on which this program is recorded, and an object tracking method.

[0066] [Technical idea disclosed in this specification] [Item 1] a sensor unit (11) mounted on a moving body (50), the sensor unit being configured to transmit and receive sensor waves around the moving body and acquire a plurality of observation values ​​corresponding to different reflection positions; a contour calculation unit (13) configured to calculate a predicted contour, which is a prediction of the current contour of the target, based on a predetermined shape model of the object and estimated values ​​of a target state including a position and an orientation of the object, the estimated values ​​being calculated in the past; a predicted value calculation unit (15) configured to calculate a plurality of predicted values ​​located on the predicted contour and / or within a predetermined range inside the predicted contour, independently from the plurality of observed values ​​acquired by the sensor unit; an associating unit (17) configured to associate each of the plurality of predicted values ​​with at least one of the plurality of observed values ​​to generate a plurality of association sets; an estimation unit (19) configured to calculate the current estimate based on the plurality of association sets generated by the association unit; Object tracking device. [Item 2] the associating unit is configured to associate each of the plurality of predicted values ​​with two or more of the plurality of observed values; each of the plurality of association sets includes one of the plurality of predicted values ​​and two or more of the plurality of observed values; Item 1. The object tracking device according to item 1. [Item 3] The predetermined range is determined in advance. Item 3. The object tracking device according to item 1 or 2. [Item 4] The predicted value calculation unit is configured to set the predetermined range depending on a state of the target. Item 3. The object tracking device according to item 1 or 2. [Item 5] the predicted value calculation unit is configured to calculate the plurality of predicted values ​​at predetermined position intervals within the predetermined range. Item 10. An object tracking device according to any one of items 1 to 4. [Item 6] the predicted value calculation unit is configured to calculate the plurality of predicted values ​​at equal position intervals within the predetermined range. 6. The object tracking device according to any one of items 1 to 5. [Item 7] the predicted value calculation unit is configured to calculate the plurality of predicted values ​​such that the position intervals become smaller as the predicted values ​​are closer to the center of the predetermined range. Item 7. An object tracking device according to any one of items 1 to 6. [Item 8] the predetermined range is the entire periphery of the predicted contour; 8. The object tracking device according to any one of items 1 to 7. [Item 9] the predetermined range is a range in which the sensor unit can directly irradiate the sensor wave; The object tracking device according to any one of items 1, 2, and 4 to 7. [Item 10] the shape model is a circular model or an elliptical model, the predetermined range is an area between a first point and a second point on the predicted contour that is closer to the sensor unit, the first point is a point of contact between a first tangent to the predicted contour that passes through the sensor unit and the predicted contour; the second point is a point of contact between a second tangent to the predicted contour that passes through the sensor unit and the predicted contour, and the second point is different from the first point; Item 10. The object tracking device according to item 9. [Item 11] the shape model is a rectangular model having a first vertex farthest from the sensor unit, a second vertex and a third vertex adjacent to the first vertex, and a fourth vertex between the second vertex and the third vertex; the predetermined range includes a first side connecting the fourth vertex to the second vertex and a second side connecting the fourth vertex to the third vertex; Item 10. The object tracking device according to item 9. [Item 12] the predetermined range includes a first range into which the sensor unit can directly irradiate the sensor wave and a second range into which the sensor unit cannot directly irradiate the sensor wave, the plurality of predicted values ​​includes a first predicted value within the first range and a second predicted value within the second range; the estimation unit is configured to make a contribution of the first predicted value to updating of the estimated value higher than a contribution of the second predicted value to the estimated value. Item 9. The object tracking device according to item 8. [Item 13] The predicted value calculation unit is configured to change the predetermined range depending on the distance to the target. 13. The object tracking device according to any one of items 1 to 12. [Item 14] The predicted value calculation unit If the distance to the target is equal to or greater than a predetermined distance, the entire periphery of the predicted contour is set to the predetermined range; When the distance to the target is less than a predetermined distance, the sensor unit sets a range in which the sensor wave can be directly irradiated to the predetermined range. Item 14. The object tracking device according to item 13. [Explanation of symbols]

[0067] 10...object tracking device, 11...sensor unit, 13...contour calculation unit, 15...prediction value calculation unit, 17...association unit, 19...estimation unit, 20...processing device, 50...vehicle.

Claims

1. a sensor unit (11) mounted on a moving body (50), configured to transmit and receive sensor waves around the moving body and acquire a plurality of observation values ​​corresponding to different reflection positions; a contour calculation unit (13) configured to calculate a predicted contour, which is a prediction of the current contour of the target, based on a predetermined shape model of the object and estimated values ​​of a target state including a position and an orientation of the object, the estimated values ​​being calculated in the past; a predicted value calculation unit (15) configured to calculate a plurality of predicted values ​​located on the predicted contour and / or within a predetermined range inside the predicted contour, independently from the plurality of observed values ​​acquired by the sensor unit; an associating unit (17) configured to associate each of the plurality of predicted values ​​with at least one of the plurality of observed values ​​to generate a plurality of association sets; an estimator (19) configured to calculate the current estimate based on the plurality of association sets generated by the associator and an estimation filter; Object tracking device.

2. the associating unit is configured to associate each of the plurality of predicted values ​​with two or more of the plurality of observed values; each of the plurality of association sets includes one of the plurality of predicted values ​​and two or more of the plurality of observed values; The object tracking device according to claim 1 .

3. The predetermined range is determined in advance. The object tracking device according to claim 1 or 2.

4. The predicted value calculation unit is configured to set the predetermined range depending on a state of the target. The object tracking device according to claim 1 or 2.

5. the predicted value calculation unit is configured to calculate the plurality of predicted values ​​at predetermined position intervals within the predetermined range. The object tracking device according to claim 1 or 2.

6. the predicted value calculation unit is configured to calculate the plurality of predicted values ​​at equal position intervals within the predetermined range. The object tracking device according to claim 1 or 2.

7. the predicted value calculation unit is configured to calculate the plurality of predicted values ​​such that the position intervals become smaller as the predicted values ​​are closer to the center of the predetermined range. The object tracking device according to claim 1 or 2.

8. the predetermined range is the entire periphery of the predicted contour; The object tracking device according to claim 5 .

9. the predetermined range is a range in which the sensor unit can directly irradiate the sensor wave; The object tracking device according to claim 5 .

10. the shape model is a circular model or an elliptical model, the predetermined range is an area between a first point and a second point on the predicted contour that is closer to the sensor unit, the first point is a point of contact between a first tangent to the predicted contour that passes through the sensor unit and the predicted contour; the second point is a point of contact between a second tangent to the predicted contour that passes through the sensor unit and the predicted contour, and the second point is different from the first point; The object tracking device of claim 9.

11. the shape model is a rectangular model having a first vertex farthest from the sensor unit, a second vertex and a third vertex adjacent to the first vertex, and a fourth vertex between the second vertex and the third vertex, the predetermined range includes a first side connecting the fourth vertex to the second vertex and a second side connecting the fourth vertex to the third vertex; The object tracking device of claim 9.

12. the predetermined range includes a first range in which the sensor unit can directly irradiate the sensor wave and a second range in which the sensor unit cannot directly irradiate the sensor wave, the plurality of predicted values ​​includes a first predicted value within the first range and a second predicted value within the second range; the estimator is configured to make a contribution of the first predicted value to updating of the estimated value higher than a contribution of the second predicted value to the estimated value. The object tracking device according to claim 8 .

13. The predicted value calculation unit is configured to change the predetermined range depending on the distance to the target. The object tracking device according to claim 1 .

14. The predicted value calculation unit If the distance to the target is equal to or greater than a predetermined distance, the entire periphery of the predicted contour is set to the predetermined range; When the distance to the target is less than a predetermined distance, the sensor unit sets a range in which the sensor wave can be directly irradiated to the predetermined range. The object tracking device of claim 13.

Images