Follow-up object vehicle determination method, program, follow-up object vehicle determination device and autonomous mobile body

Abstract

To provide a follow-up object vehicle determination method and the like which can adequately determine an object vehicle which is an object to be followed by a self-vehicle.SOLUTION: The follow-up object vehicle determination method which determines an object vehicle which is an object to be followed by a self-vehicle on the basis of a travel planned route of the self-vehicle includes search step S20 which searches for a candidate vehicle which is a candidate of the object vehicle, prediction step S30 which predicts one or more expectation values corresponding to respective one or more parameters representing a travel mode of the candidate vehicle, calculation step S40 which calculates travel continuation probability which is probability for which the candidate vehicle continues travel of the travel planned route on the basis of one or more measurement values corresponding to respective one or more parameters of the candidate vehicle, and one or more expectation values and determination step S50 which determines the target vehicle on the basis of the travel continuation probability. The one or more expectation values are values of one or more parameters predicted if it is assumed that the candidate vehicle continues to travel on the travel planned route.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to a following target vehicle determination method, a program, a following target vehicle determination device, and an autonomous mobile body.

Background Art

[0002] Currently, research on the automatic driving of mobile bodies such as vehicles is actively underway. In the automatic driving of mobile bodies, in order to avoid collisions with surrounding mobile bodies, techniques for predicting the lanes in which other mobile bodies travel are known. In particular, by specifying other vehicles traveling on the planned travel route of the host vehicle as following targets, it is possible to expect an improvement in the performance of automatic driving.

[0003] For example, Patent Document 1 describes a method for predicting other vehicles that cut into the travel route of the host vehicle. In the invention described in Patent Document 1, an attempt is made to predict the intrusion of other vehicles into the lane in which the host vehicle travels based on predetermined conditions such as the road shape.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the invention described in Patent Document 1, the prediction of the intrusion of other vehicles may be incorrect, and such cases are not assumed. Therefore, the invention described in Patent Document 1 is not suitable for a method for specifying other vehicles that the host vehicle follows.

[0006] Therefore, an object of the present invention is to provide a following target vehicle determination method and the like that can appropriately determine a target vehicle that the host vehicle follows.

Means for Solving the Problems

[0007] In order to achieve the above object, a following target vehicle determination method according to an aspect of the present invention is a following target vehicle determination method for determining a target vehicle to be followed by the host vehicle based on a planned travel route of the host vehicle, including a search step of searching for candidate vehicles that are candidates for the target vehicle, a prediction step of predicting one or more expected values respectively corresponding to one or more parameters representing the driving mode of the candidate vehicle, a calculation step of calculating a driving continuation probability, which is a probability that the candidate vehicle continues to travel on the planned travel route, based on the one or more measured values respectively corresponding to the one or more parameters of the candidate vehicle and the one or more expected values, and a determination step of determining the target vehicle based on the driving continuation probability, wherein the one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route.

[0008] Also, in order to achieve the above object, a program according to an aspect of the present invention is a program for causing a computer to execute the above following target vehicle determination method.

[0009] Also, in order to achieve the above object, a following target vehicle determination apparatus according to an aspect of the present invention is a following target vehicle determination apparatus for determining a target vehicle to be followed by the host vehicle based on a planned travel route of the host vehicle, including a search unit that searches for candidate vehicles that are candidates for the target vehicle, a prediction unit that predicts one or more expected values respectively corresponding to one or more parameters representing the driving mode of the candidate vehicle, a calculation unit that calculates a driving continuation probability, which is a probability that the candidate vehicle continues to travel on the planned travel route, based on the one or more measured values respectively corresponding to the one or more parameters of the candidate vehicle and the one or more expected values, and a determination unit that determines the target vehicle based on the driving continuation probability, wherein the one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route.

[0010] In addition, in order to achieve the above object, an autonomous mobile body according to one aspect of the present invention includes the above following target vehicle determination device and a vehicle information acquisition unit that acquires information on the candidate vehicles.

[0011] These general or specific aspects may be implemented by a system, a method, an integrated circuit, a computer program, or a recording medium such as a non-temporary computer-readable CD-ROM, or may be implemented by any combination of a system, a method, an integrated circuit, a computer program, and a recording medium.

Advantages of the Invention

[0012] According to the present invention, it is possible to provide a following target vehicle determination method or the like that can appropriately determine a target vehicle to be followed by the host vehicle.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

[0014] Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings.

[0015] Note that all of the embodiments described below show comprehensive or specific examples. The numerical values, shapes, materials, constituent elements, arrangement positions and connection forms of the constituent elements, steps, order of steps, etc. shown in the following embodiments are merely examples and are not intended to limit the present invention. In addition, among the constituent elements in the following embodiments, the constituent elements not described in the independent claims indicating the most general concept are described as optional constituent elements.

[0016] In addition, each figure is a schematic diagram and is not necessarily drawn precisely. Also, in each figure, the same constituent members are denoted by the same reference numerals.

[0017] (Embodiment) A following target vehicle determination method, a following target vehicle determination device, and an autonomous mobile body according to the embodiment will be described.

[0018] [1. Configuration of Following Target Vehicle Determination Device and Autonomous Mobile Body] The configuration of the following vehicle determination device and the autonomous mobile body according to the present embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing an example of the configuration of the following vehicle determination device 100 and the autonomous mobile body V10 according to the present embodiment.

[0019] The autonomous mobile body V10 is a mobile body that performs autonomous driving, and includes a vehicle information acquisition unit 10 and a following vehicle determination device 100. In the present embodiment, the autonomous mobile body V10 further includes a travel control unit 70 and a storage unit 80. The autonomous mobile body V10 is not particularly limited as long as it is a mobile body. In the present embodiment, the autonomous mobile body V10 is a vehicle that performs autonomous driving. Hereinafter, the autonomous mobile body V10 is also referred to as the host vehicle (V10).

[0020] The following vehicle determination device 100 is a device that determines a target vehicle that the host vehicle V10 follows based on the planned travel route of the host vehicle V10. The following vehicle determination device 100 according to the present embodiment determines the target vehicle using the following vehicle determination method described later. The following vehicle determination device 100 determines, as the target vehicle, the vehicle closest to the host vehicle V10 among the vehicles traveling ahead of the host vehicle V10 on the planned travel route of the host vehicle V10. The planned travel route of the host vehicle is a route determined in advance based on the destination and the like.

[0021] As shown in FIG. 1, the following vehicle determination device 100 includes a search unit 20, a prediction unit 30, a calculation unit 40, and a determination unit 50. Note that the following vehicle determination device 100 may include other processing units and the like. For example, the following vehicle determination device 100 may include a vehicle information acquisition unit 10 and the like.

[0022] The vehicle information acquisition unit 10 acquires information on vehicles located around the host vehicle V10. Here, vehicles that are the targets for the vehicle information acquisition unit 10 to acquire information include, for example, automobiles, motorcycles, and the like. The vehicle information acquired by the vehicle information acquisition unit 10 includes, for example, the direction in which the vehicle is located and the distance to the surface of the vehicle, based on the observation point which is the installation position of the vehicle information acquisition unit 10. The vehicle information acquired by the vehicle information acquisition unit 10 may also include the height of the vehicle. As the vehicle information acquisition unit 10, for example, a three-dimensional information acquisition device such as LiDAR (Light Detection and Ranging) can be used. In the present embodiment, the vehicle information acquisition unit 10 includes LiDAR.

[0023] LiDAR irradiates a vehicle or the like with laser light and measures the distance between the vehicle and the observation point which is the installation position of LiDAR by acquiring the laser light reflected from the vehicle or the like. The vehicle information acquisition unit 10 acquires three-dimensional point cloud data (point cloud) as height information indicating the relationship between the horizontal position and height of the road surface from the measured distance, the position of LiDAR, and the irradiation direction of the laser light. The three-dimensional point cloud data is an example of the vehicle information acquired by the vehicle information acquisition unit 10.

[0024] LiDAR irradiates, for example, 128 laser lights in the vertical direction (that is, the plumb direction) at different irradiation angles. The vertical field of view angle of LiDAR is set with the horizontal direction (that is, the direction perpendicular to the vertical direction) as 0° (0 degree), and is, for example, -25° or more and 15° or less. LiDAR can acquire omnidirectional three-dimensional point cloud data by rotating the irradiation direction of these laser lights 360° in the horizontal direction. The scan rate by the laser light is, for example, 5 Hz or more and 20 Hz or less. In the present embodiment, the scan rate is set to 10 Hz. That is, in the present embodiment, LiDAR acquires 10 frames of three-dimensional point cloud data per second.

[0025] The search unit 20 is a processing unit that searches for candidate vehicles that are candidates for the target vehicle. The search unit 20 searches for candidate vehicles based on the information of the vehicles located around the host vehicle V10 acquired by the vehicle information acquisition unit 10. In the present embodiment, the candidate vehicle is a vehicle located in front of the host vehicle V10. Here, the candidate vehicle will be described with reference to FIG. 2. FIG. 2 is a plan view showing an example of a candidate vehicle according to the present embodiment.

[0026] Vehicles such as the candidate vehicles V90 and V91 shown in FIG. 2, which are located in front of the host vehicle V10 and are traveling on the planned travel route of the host vehicle V10, are included in the candidate vehicles. Note that vehicles located further in front of the candidate vehicle V90 as viewed from the host vehicle V10, such as the candidate vehicle V91, are also included in the candidate vehicles.

[0027] Also, vehicles such as the candidate vehicle V92 shown in FIG. 2, which are located in front of the host vehicle V10 and are stopped on the road, are included in the candidate vehicles. Vehicles stopped on the road include parked vehicles and stopped vehicles. Furthermore, the candidate vehicles include vehicles that stop at a temporary stop position such as a stop line.

[0028] Also, the candidate vehicles located in front of the host vehicle V10 are not limited to vehicles traveling on the planned travel route of the host vehicle V10. For example, the candidate vehicle may be a vehicle located in front of the host vehicle V10 and traveling on a lane outside the planned travel route. Specifically, vehicles such as the candidate vehicle V93 shown in FIG. 2, which are located in front of the host vehicle V10 and traveling on a lane other than the planned travel route, are included in the candidate vehicles. Furthermore, vehicles that move from outside the planned travel route to the planned travel route, such as the candidate vehicle V93 shown in FIG. 2, may also be included in the candidate vehicles. In FIG. 2, the candidate vehicle V93 that is trying to change lanes from a lane other than the planned travel route to the planned travel route is shown, but vehicles that continue to travel on a lane other than the planned travel route are also included in the candidate vehicles. Also, when another route merges into the planned travel route, vehicles located in front of the host vehicle V10 and traveling on the other route are also included in the candidate vehicles.

[0029] The prediction unit 30 is a processing unit that predicts one or more expected values corresponding to one or more parameters representing the driving mode of the candidate vehicle. For example, the one or more parameters may include at least one of the speed of the candidate vehicle, the acceleration of the candidate vehicle, the space (i.e., distance) in the lateral direction (the direction perpendicular to the tangent of the planned travel route) from the edge of the lane in which the candidate vehicle is traveling to the candidate vehicle, and the inter-vehicle distance of the candidate vehicle. Further, the one or more parameters may include all of the speed of the candidate vehicle, the acceleration of the candidate vehicle, the space in the lateral direction from the edge of the lane in which the candidate vehicle is traveling to the candidate vehicle, and the inter-vehicle distance of the candidate vehicle. Here, these parameters will be described with reference to FIG. 3. FIG. 3 is a plan view for explaining the definition of each parameter of the candidate vehicle V90.

[0030] In the present embodiment, for the speed and acceleration of the candidate vehicle V90 shown in FIG. 3, the speed v Tan and acceleration a Tan which are the components in the tangent direction of the planned travel route, and the speed v Norm and acceleration a Norm which are the components in the direction perpendicular to the tangent of the planned travel route, among them, the speed v Tan and acceleration a Tan are used as the parameters of the candidate vehicle V90.

[0031] Further, the space in the lateral direction from the edge of the lane in which the candidate vehicle V90 is traveling to the candidate vehicle V90 is defined by the larger value of the spaces w Right and w Left shown in FIG. 3. The space w Right is the distance in the lateral direction from the right edge of the lane in which the candidate vehicle V90 is traveling to the right edge of the candidate vehicle V90. The space w Left is the distance in the lateral direction from the left edge of the lane in which the candidate vehicle V90 is traveling to the left edge of the candidate vehicle V90. Note that the definition of the space is not limited to the above definition. For example, the space may be defined by the smaller value of the spaces w Right and w Left .

[0032] The inter-vehicle distance d of the candidate vehicle V90 is the distance (interval) from the candidate vehicle V90 to the candidate vehicle V91 (hereinafter also referred to as the preceding vehicle) that travels immediately in front of the candidate vehicle V90.

[0033] One or more expected values are values of one or more parameters predicted when it is assumed that the candidate vehicle V90 continues to travel on the planned travel route of the host vehicle V10. In other words, one or more expected values are values of one or more parameters assumed when the candidate vehicle V90 is assumed to be the target vehicle. The method for predicting one or more expected values in the prediction unit 30 will be described later.

[0034] The calculation unit 40 is a processing unit that calculates a travel continuation probability, which is the probability that the candidate vehicle continues to travel on the planned travel route, based on one or more measured values corresponding to one or more parameters of the candidate vehicle and one or more expected values. The calculation unit 40 measures one or more parameters of the candidate vehicle and obtains one or more measured values, for example, based on the position of the candidate vehicle at each time point acquired by the vehicle information acquisition unit 10. Note that one or more measured values may be measured by a processing unit other than the calculation unit 40. For example, one or more measured values may be measured in the vehicle information acquisition unit 10, the search unit 20, or the like.

[0035] In the present embodiment, the calculation unit 40 calculates the travel continuation probability assuming that the travel continuation probability follows a predetermined probability distribution with respect to the difference between each of the one or more measured values and the expected value corresponding to the measured value among the one or more expected values. In the present embodiment, it is assumed that the travel continuation probability follows a normal distribution.

[0036] The calculation unit 40 calculates one or more individual probabilities based on the one or more measured values and the one or more expected values. Each of the one or more individual probabilities indicates the likelihood that the candidate vehicle continues to travel on the planned travel route, calculated based on one of the one or more measured values and the expected value corresponding to the one measured value among the one or more expected values, and the travel continuation probability calculated by the calculation unit 40 is a weighted average of the one or more individual probabilities. Details of the method for calculating the travel continuation probability in the calculation unit 40 will be described later.

[0037] The determination unit 50 is a processing unit that determines the target vehicle based on the running continuation probability calculated by the calculation unit 40. In the present embodiment, the determination unit 50 determines, as the target vehicle, the candidate vehicle closest to the host vehicle V10 among the candidate vehicles whose running continuation probability is equal to or higher than a predetermined threshold probability.

[0038] The travel control unit 70 is a processing unit that controls the travel of the host vehicle V10. In the present embodiment, the travel control unit 70 controls the travel so as to avoid collisions with surrounding vehicles based on information such as the target vehicle determined by the following vehicle determination device 100. The travel control unit 70 controls, for example, the traveling direction of the host vehicle V10 based on the position of the host vehicle V10, the map information stored in the storage unit 80, and the like.

[0039] The storage unit 80 stores information for determining the target vehicle of the host vehicle V10. For example, the storage unit 80 stores map information including information on the road on which the host vehicle V10 travels. Further, the storage unit 80 may store information for autonomous driving.

[0040] [2. Prediction method of expected value] The prediction method for each of one or more expected values according to the present embodiment will be described.

[0041] [2-1. Expected value of speed] In the present embodiment, the prediction unit 30 predicts the expected value of the speed of the candidate vehicle based on the speed of the candidate vehicle that travels immediately before the candidate vehicle (also referred to as the preceding vehicle) (hereinafter also referred to as the preceding vehicle speed), the speed limit of the route during travel, and the design speed calculated based on the curvature of the route at the position of the preceding vehicle.

[0042] Design speed v Curv is represented by the following formula (1) using the curvature κ of the route and the maximum lateral acceleration a LatMax at the position of the preceding vehicle.

[0043] [Equation]

[0044] Here, as the maximum lateral acceleration a LatMax , 0.2G, which is generally felt as strong acceleration, is adopted.

[0045] The speed limit v Limit is the speed limit on the traffic rules of the route at the position of the preceding vehicle.

[0046] In the present embodiment, the prediction unit 30 predicts the minimum speed among the preceding vehicle speed v Prd , the speed limit v Limit , and the design speed v Curv as the expected value of the speed. That is, the expected value of the speed v Exp is represented by the following formula (2).

[0047]

Equation

[0048] In addition, when there is no preceding vehicle within a predetermined threshold distance d Thr in front of the candidate vehicle, the preceding vehicle speed v Exp is not considered in the prediction of the expected value of the speed v Prd . That is, the expected value of the speed v Exp is predicted as the minimum speed among the speed limit v Limit and the design speed v Curv . Also, when there is no preceding vehicle within a predetermined threshold distance d Thr in front of the candidate vehicle, the speed limit v Limit and the design speed v Curv are the values at a position d Thr ahead of the candidate vehicle by the threshold distance.

[0049] As described above, according to the prediction unit 30 according to the present embodiment, the expected value v Exp of the speed considering the interaction between vehicles, traffic rules, and road shape can be predicted.

[0050] [2-2. Expected value of acceleration] The prediction unit 30 predicts the expected value of acceleration aExp is predicted as the acceleration necessary for the candidate vehicle to reach the expected speed v while traveling a distance d from the preceding vehicle. Therefore, the expected acceleration a Exp is expressed by the following formula (3) using the speed v of the candidate vehicle, the expected speed v Exp of the candidate vehicle, and the distance d between vehicles. Exp Note that when there is no preceding vehicle within a predetermined threshold distance d in front of the candidate vehicle, the threshold distance d is used instead of the distance d between vehicles in formula (3).

[0051]

Equation

[0052] In addition, when there is no preceding vehicle within a predetermined threshold distance d in front of the candidate vehicle, the threshold distance d is used instead of the distance d between vehicles in formula (3). Thr is used. Thr

[0053] [2-3. Expected value of space] The prediction unit 30 predicts the expected value w Exp of the space as the space when the candidate vehicle travels in the center of the lane in the lateral direction. That is, the expected value w Exp of the space is the space when the space w Right on the right side of the candidate vehicle and the space w Left on the left side are the same. The expected value w Exp of the space is constant regardless of the presence or absence of a preceding vehicle. Therefore, the expected value w Exp of the space is expressed by the following formula (4) using the lane width w Road of the route on which the candidate vehicle is traveling and the width w Obj of the candidate vehicle.

[0054]

Equation

[0055] [2-4. Expected value of the distance between vehicles] The prediction unit 30 predicts the expected value d Exp of the distance between vehicles based on the distance between vehicles of ordinary vehicles. In this embodiment, the expected value d​Exp is predicted to be 15 m. Note that the expected value d of the inter-vehicle distance Exp is not limited to this. For example, the expected value d of the inter-vehicle distance Exp may be predicted based on the speed limit of the route on which the candidate vehicle travels, etc.

[0056] [3. Method for calculating probability] A method for calculating the driving continuation probability by the calculation unit 40 and a method for calculating the individual probability used for calculating the driving continuation probability will be described.

[0057] [3-1. Method for calculating individual probability] A method for calculating the individual probability by the calculation unit 40 will be described. As described above, the calculation unit 40 calculates one or more individual probabilities based on one or more measurement values and one or more expected values. The calculation unit 40 calculates the individual probability assuming that the individual probability follows a predetermined probability distribution with respect to the difference between each of the one or more measurement values and the expected value corresponding to the measurement value among the one or more expected values. In the present embodiment, a normal distribution is used as the predetermined probability distribution.

[0058] Among one or more parameters, one parameter x, the expected value x of the parameter x Exp and the variance σ of the parameter x x 2 are used, and each individual probability P x is represented by the following formula (5).

[0059]

Equation

[0060] Here, α is a correction value for making each individual probability a value within the range of 0.4 or more and 0.6 or less. In the present embodiment, the calculation unit 40 uses the above formula (5) to calculate the individual probability P x as the individual probability P of speed v , the individual probability P of acceleration a , the individual probability P of space w , and the individual probability P of the inter-vehicle distance d .

[0061] [3-2. Method for calculating running continuation probability] The calculation unit 40 calculates the running continuation probability based on one or more individual probabilities. In the present embodiment, the running continuation probability is a weighted average of one or more individual probabilities. The weight for each individual probability is not particularly limited. In the present embodiment, the highest weight is given to the individual probability P a of acceleration, which is an optimal parameter for predicting the running mode of the candidate vehicle, and the second highest weight is given to the individual probability P w of space, which is a parameter indicating the lateral motion state of the candidate vehicle. At time point t n , the running continuation probability P n is represented by the following formula (6) using the weight W v for speed, the weight W a for acceleration, the weight W w for space, and the weight W d for inter-vehicle distance.

[0062] [Number]

[0063] Note that, in the calculation of the running continuation probability P n , it is not necessarily required to use four individual probabilities. In the calculation of the running continuation probability P n , at least one individual probability may be used. For example, when the individual probability of the inter-vehicle distance is not used, in formula (6), the weight W d for the inter-vehicle distance may be set to zero.

[0064] [3-3. Time-series processing of running continuation probability] The calculation unit 40 may perform time-series processing on the driving continuation probability. Thereby, the reliability of the driving continuation probability can be enhanced. For example, when the measured values of the respective parameters become inaccurate values due to the influence of noise or the like, it is possible to suppress a decrease in the reliability of the driving continuation probability. As the time-series processing, for example, a Binary Bayes Filter (BBF) can be used. Since the BBF processes by converting an arbitrary probability into log odds, prevention of digit loss and acceleration of calculation due to the introduction of the BBF are expected. At time point t n the driving continuation probability P calculated n has a log odds L represented by the following equation (7).

[0065] [Equation]

[0066] Also, if the log odds up to the time point t n immediately preceding time point t n-1 is represented by L n-1 then the log odds L n up to time point t n is represented by the following equation (8) using the BBF.

[0067] [Equation]

[0068] The log odds L obtained from equation (8) n is converted into the time-series processed driving continuation probability P using the following equation (9).

[0069] [Equation]

[0070] [3-4. Prediction of Driving Continuation] The determination unit 50, at time point t calculated by the above procedure nUsing the travel continuation probability P up to, it is predicted whether the candidate vehicle will continue to travel on the planned travel route of the host vehicle. The travel continuation probability P decreases as the difference between the assumed behavior when the candidate vehicle continues to travel on the planned travel route and the actual behavior of the candidate vehicle increases. For example, when the travel continuation probability P is calculated to be 0, it can be predicted that the candidate vehicle will not continue to travel on the planned travel route. Also, when the travel continuation probability P is calculated to be 1, it can be predicted that the candidate vehicle will continue to travel on the planned travel route. In the present embodiment, a travel continuation probability P = 0.5, which is difficult to predict whether the candidate vehicle will continue to travel on the planned travel route, is set as the threshold probability. That is, the determination unit 50 according to the present embodiment predicts that the candidate vehicle will continue to travel on the planned travel route if the travel continuation probability P is equal to or greater than the threshold probability 0.5, and predicts that the candidate vehicle will not continue to travel on the planned travel route if the travel continuation probability P is less than the threshold probability 0.5.

[0071] The determination unit 50 determines the target vehicle based on the travel continuation probability P. In the present embodiment, the determination unit 50 determines, as the target vehicle, the candidate vehicle closest to the host vehicle V10 among the candidate vehicles for which the travel continuation probability P is equal to or greater than the threshold probability. Thereby, when the difference between each measured value of one or more parameters and the expected value is large, the travel continuation probability P becomes low, so that the candidate vehicle can be appropriately excluded from the target vehicle.

[0072] With the following vehicle determination device 100 having the above configuration, the target vehicle can be appropriately determined.

[0073] [4. Following target vehicle determination method] The following target vehicle determination method according to the present embodiment will be described with reference to FIG. 4. FIG. 4 is a flowchart showing the flow of the following target vehicle determination method according to the present embodiment.

[0074] The following target vehicle determination method according to the present embodiment is a method for determining a target vehicle that the host vehicle V10 follows based on the planned travel route of the host vehicle V10. In the present embodiment, the target vehicle is determined using the following vehicle determination device 100.

[0075] In the method for determining a following target vehicle according to the present embodiment, first, as shown in FIG. 4, the vehicle information acquisition unit 10 of the following target vehicle determination device 100 acquires information on vehicles located around the host vehicle V10 (vehicle information acquisition step S10).

[0076] Subsequently, the search unit 20 searches for candidate vehicles that are candidates for the target vehicle (search step S20). In the search step S20, the search unit 20 searches for candidate vehicles based on the information on the vehicles located around the host vehicle V10 acquired by the vehicle information acquisition unit 10. In the present embodiment, the candidate vehicle is a vehicle located in front of the host vehicle V10.

[0077] Subsequently, the prediction unit 30 predicts one or more expected values corresponding to one or more parameters representing the driving mode of the candidate vehicle (prediction step S30). As described above, the one or more expected values are the values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route of the host vehicle V10.

[0078] Subsequently, the calculation unit 40 calculates a driving continuation probability, which is the probability that the candidate vehicle continues to travel on the planned travel route of the host vehicle V10, based on one or more measured values corresponding to one or more parameters of the candidate vehicle and the one or more expected values (calculation step S40).

[0079] Subsequently, the determination unit 50 determines the target vehicle based on the calculated driving continuation probability (determination step S50). In the present embodiment, the determination unit 50 determines, as the target vehicle, the candidate vehicle closest to the host vehicle V10 among the candidate vehicles whose driving continuation probability is equal to or greater than a predetermined threshold probability. The threshold probability is set to 0.5, for example.

[0080] Subsequently, the process returns to the vehicle information acquisition step S10, and the above-described steps are repeated.

[0081] By the method for determining a following target vehicle as described above, the target vehicle can be appropriately determined.

[0082] [5. Effect] The effects of the following vehicle determination method (and the following vehicle determination device 100) according to the present embodiment will be described while showing experimental results.

[0083] Here, in order to show the effectiveness of the driving continuation probability used in the present embodiment, the relationship between the behavior of candidate vehicles and the driving continuation probability when the following vehicle determination method and the following vehicle determination device 100 according to the present embodiment are used is shown as a result of verification by experiments.

[0084] The weights applied to each parameter in the calculation of the driving continuation probability used in this experiment and the variances of each parameter are shown in Table 1 below.

[0085] [Table 1]

[0086] Note that the variance values are determined based on actual driving data and the like. Also, in this experiment, the weight for the inter-vehicle distance is set to 0. That is, the inter-vehicle distance is not used as a parameter.

[0087] In this experiment, the transition of the driving continuation probability for two scenes, Scene 1 and Scene 2, was confirmed. Each scene will be described below.

[0088] Scene 1 and Scene 2 will be described with reference to FIGS. 5 and 6. FIGS. 5 and 6 are plan views for explaining the outlines of Scene 1 and Scene 2 in this experiment, respectively.

[0089] Both Scene 1 and Scene 2 are scenes in which the candidate vehicle V90 turns left and thus departs from the planned driving route of the host vehicle V10. In Scene 1, there is no candidate vehicle preceding the candidate vehicle V90, and in Scene 2, there is a preceding candidate vehicle V91.

[0090] Regarding the experimental results in each of these scenes, Figures 7 to 14 will be used for explanation. Figure 7 is a graph showing the transition of the driving continuation probability of candidate vehicle V90 in Scene 1. Figures 8, 9, and 10 are graphs showing the transitions of the measured values and expected values of the speed, acceleration, and space of candidate vehicle V90 in Scene 1, respectively. Figure 11 is a graph showing the transition of the driving continuation probability of candidate vehicle V90 in Scene 2. In Figure 11, the driving continuation probability of candidate vehicle V91 is also shown by a dashed line. Figures 12, 13, and 14 are graphs showing the transitions of the measured values and expected values of the speed, acceleration, and space of candidate vehicle V90 in Scene 2, respectively. The horizontal axis of each graph in Figures 7 to 10 indicates the distance along the planned driving route of the host vehicle to the stop line of the intersection where candidate vehicle V90 makes a left turn. Note that the distance in front of the stop line is shown as a negative value, and the distance behind the stop line is shown as a positive value. The horizontal axis of each graph in Figures 11 to 14 indicates the distance along the planned driving route of the host vehicle to the position where candidate vehicle V90 completely departs from the planned driving route of the host vehicle V10. Note that the distance in front of the said position is shown as a negative value, and the distance behind the said position is shown as a positive value.

[0091] In Scene 1, as shown in Figure 8, the speed of candidate vehicle V90 is lower than the expected value predicted based on the speed limit of the planned driving route of the host vehicle. Therefore, when it is assumed that candidate vehicle V90 continues to drive on the planned driving route of the host vehicle V10, since it is assumed that candidate vehicle V90 accelerates based on the speed limit of the planned driving route, as shown in Figure 9, the expected value of the acceleration increases as it approaches the stop line.

[0092] On the other hand, since candidate vehicle V90 decelerates in preparation for a left turn as it approaches the stop line, as shown in Figures 8 and 9, the difference between the measured value and the expected value becomes larger.

[0093] Incidentally, the difference between the measured value and the expected value of the space shown in Fig. 10 does not change significantly until the candidate vehicle V90 approaches the stop line, but the difference increases rapidly at the position after passing the stop line. This is due to the candidate vehicle V90 turning left and moving away from the planned travel route of the host vehicle V10.

[0094] As the above parameters change, as shown in Fig. 7, the running continuation probability falls below the threshold probability of 0.5 at a position 15 m in front of the stop line. Therefore, based on the running continuation probability, it can be said that it was correctly predicted that the candidate vehicle V90 would not continue to travel on the planned travel route of the host vehicle V10, that is, the candidate vehicle V90 should not be the target vehicle. Also, after the candidate vehicle V90 turns left, the difference between the measured value and the expected value in the space increases, and the running continuation probability is maintained at a low value. Thereby, it is possible to reduce the misidentification of a candidate vehicle traveling outside the planned travel route as the target vehicle.

[0095] Also in Scene 2, as shown in Figs. 12 to 14, similar to Scene 1, the difference between the measured value and the expected value of the speed and acceleration of the candidate vehicle V90 increases, and the change in the difference between the measured value and the expected value of the space is small. For this reason, as shown in Fig. 11, when the candidate vehicle V90 is located at a position about 15 m in front of the position where the candidate vehicle V90 completely departs from the planned travel route of the host vehicle V10, the running continuation probability falls below the threshold probability of 0.5. Therefore, also in Scene 2, it can be said that it was correctly predicted based on the running continuation probability that the candidate vehicle V90 would not continue to travel on the planned travel route of the host vehicle V10.

[0096] Also in Scene 2, in front of the candidate vehicle V90, the running continuation probability of the candidate vehicle V91 that continues to travel on the planned travel route of the host vehicle V10 is 0.8 or more and is gradually increasing. Therefore, in Scene 2, it can be said that it was correctly predicted that the target vehicle should be changed from the candidate vehicle V90 to the candidate vehicle V91.

[0097] In the above, an example where the candidate vehicle turns left has been shown, but for other candidate vehicles, it is also possible to correctly predict based on the running continuation probability.

[0098] For example, for a candidate vehicle stopped on the road shoulder, since the difference between the measured values of speed and space and the expected values becomes large, the running continuation probability becomes low and it is excluded from the candidates for the target vehicle. Although such a stopped candidate vehicle should be excluded from the candidates for the target vehicle, a candidate vehicle stopped at a position where a stop is obligatory in front of the host vehicle (hereinafter also referred to as a stop position) should be regarded as the target vehicle. The presence of a stop position can be obtained from map information or the like, and it is possible to set expected values such as speed based on the presence of the stop position. Therefore, it is possible to suppress the difference between the measured values of the speed and acceleration of the candidate vehicle and the expected values from becoming large. Therefore, for such a candidate vehicle, it is possible to suppress the running continuation probability from becoming low, and thus it can be a candidate for the target vehicle.

[0099] Also, according to the present embodiment, it is possible to correctly determine whether or not a candidate vehicle located outside the planned travel route of the host vehicle becomes the target vehicle. An example of such a candidate vehicle will be described with reference to FIG. 15. FIG. 15 is a plan view showing an example in which the candidate vehicle V94 travels on a route that merges into the planned travel route of the host vehicle V10.

[0100] As shown in FIG. 15, when the candidate vehicle V94 is located in front of the host vehicle V10 and is a vehicle that travels on a route that merges into the planned travel route of the host vehicle V10, the fact that the candidate vehicle V94 merges into the planned travel route can be predicted based on map information or the like. Therefore, expected values for each parameter of the candidate vehicle V94 can be appropriately set. For example, the expected value of the speed is set to a low value in front of the stop line. Therefore, also in such a candidate vehicle V94, it is possible to appropriately determine whether or not to regard it as the target vehicle based on the running continuation probability. As described above, in the present embodiment, the target vehicle may be a vehicle that moves from outside the planned travel route of the host vehicle V10 to the planned travel route. Further, the target vehicle may be a vehicle that travels on a route that merges into the planned travel route.

[0101] As described above, according to the following vehicle determination method and the following vehicle determination device 100 according to the present embodiment, the target vehicle can be correctly determined. Further, in the present embodiment, since the target vehicle can be determined based on a simple calculation as described above, the time and load required for the calculation process can be reduced. For example, in the present embodiment, the target vehicle can be determined within a time period of about the LiDAR scan cycle (0.1 second) or less.

[0102] (Modifications, etc.) As described above, the following vehicle determination method and the like according to one aspect of the present invention have been described based on the embodiments. However, the present invention is not limited to the embodiments. As long as the gist of the present invention is not deviated from, various modifications conceived by those skilled in the art may also be included in the scope of the present invention when applied to the embodiments.

[0103] Also, the following forms may also be included within the scope of one or more aspects of the present disclosure.

[0104] (1) A part of the components included in the above-described following vehicle determination device 100 may be a computer system including a microprocessor, a ROM, a RAM, a hard disk unit, a display unit, a keyboard, a mouse, and the like. A computer program is stored in the RAM or the hard disk unit. The microprocessor operates according to the computer program to achieve its functions. Here, the computer program is configured by combining a plurality of instruction codes indicating instructions for the computer to achieve a predetermined function.

[0105] (2) Some of the components included in the above target vehicle determination device 100 may be configured by one system LSI (Large Scale Integration). The system LSI is a super multifunctional LSI manufactured by integrating a plurality of components on one chip. Specifically, it is a computer system including a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. When the microprocessor operates according to the computer program, the system LSI achieves its functions.

[0106] (3) Some of the components included in the above target vehicle determination device 100 may be configured by an IC card or a single module detachable from each device. The IC card or the module is a computer system composed of a microprocessor, ROM, RAM, etc. The IC card or the module may include the above super multifunctional LSI. When the microprocessor operates according to the computer program, the IC card or the module achieves its functions. This IC card or this module may have tamper resistance.

[0107] (4) Also, some of the components included in the above target vehicle determination device 100 may be the computer program or the digital signal recorded on a computer-readable recording medium, such as a flexible disk, a hard disk, a CD-ROM, an MO, a DVD, a DVD-ROM, a DVD-RAM, a BD (Blu-ray (registered trademark) Disc), a semiconductor memory, etc. It may also be the digital signal recorded on these recording media.

[0108] Also, some of the components included in the above target vehicle determination device 100 may transmit the computer program or the digital signal via a telecommunication line, a wireless or wired communication line, a network represented by the Internet, data broadcasting, etc.

[0109] (5) The present disclosure may be the following following target vehicle determination method. Further, it may be a computer program for causing a computer to realize the above-described following target vehicle determination method, or may be a digital signal composed of the computer program. Furthermore, the present disclosure may be realized as a non-transitory computer-readable recording medium such as a CD-ROM on which the computer program is recorded.

[0110] (6) Further, the present disclosure may be a computer system including a microprocessor and a memory, wherein the memory stores the above computer program, and the microprocessor operates according to the computer program.

[0111] (7) Further, it may be implemented by another independent computer system by recording and transferring the program or the digital signal to the recording medium, or by transferring the program or the digital signal via the network or the like.

[0112] (8) Each aspect of the above embodiment may be combined.

[0113] (Supplementary Note) Also, the following technique is disclosed by the above description.

[0114] (Technique 1) A following target vehicle determination method for determining a target vehicle that the host vehicle follows based on a planned travel route of the host vehicle, the method including: a search step of searching for candidate vehicles that are candidates for the target vehicle; a prediction step of predicting one or more expected values respectively corresponding to one or more parameters representing a driving mode of the candidate vehicles; a calculation step of calculating a driving continuation probability, which is a probability that the candidate vehicle continues to travel on the planned travel route, based on one or more measured values respectively corresponding to the one or more parameters of the candidate vehicle and the one or more expected values; and a determination step of determining the target vehicle based on the driving continuation probability, wherein the one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route.

[0115] (Technique 2) The following target vehicle determination method according to Technique 1, wherein the candidate vehicle is a vehicle located in front of the host vehicle.

[0116] (Technique 3) The following target vehicle determination method according to Technique 1 or 2, wherein the target vehicle is a vehicle that moves from outside the planned travel route to the planned travel route.

[0117] (Technique 4) The following target vehicle determination method according to Technique 3, wherein the target vehicle is a vehicle that travels on a route that merges into the planned travel route.

[0118] (Technique 5) The following target vehicle determination method according to any one of Techniques 1 to 4, wherein the candidate vehicles include vehicles stopped on the road.

[0119] (Technique 6) The following target vehicle determination method according to any one of Techniques 1 to 5, wherein the candidate vehicles include vehicles stopped at a temporary stop position.

[0120] (Technique 7) The following target vehicle determination method according to any one of Techniques 1 to 6, wherein the one or more parameters include at least one of a speed of the candidate vehicle, an acceleration of the candidate vehicle, a lateral space from an edge of a lane in which the candidate vehicle is traveling to the candidate vehicle, and a distance between the candidate vehicle and a preceding vehicle.

[0121] (Technology 8) The above-mentioned one or more parameters include all of the speed of the candidate vehicle, the acceleration of the candidate vehicle, the space in the lateral direction from the edge of the lane on which the candidate vehicle is traveling to the candidate vehicle, and the inter-vehicle distance of the candidate vehicle. The method for determining a following target vehicle according to any one of Technologies 1 to 7.

[0122] (Technology 9) In the calculating step, assuming that the running continuation probability follows a predetermined probability distribution with respect to the difference between each of the one or more measured values and the expected value corresponding to the measured value among the one or more expected values, the method for determining a following target vehicle according to any one of Technologies 1 to 8 for calculating the running continuation probability.

[0123] (Technology 10) In the calculating step, based on the one or more measured values and the one or more expected values, one or more individual probabilities are calculated. Each of the one or more individual probabilities indicates the likelihood that the candidate vehicle continues to travel on the planned travel route, based on one of the one or more measured values and the expected value corresponding to the one measured value among the one or more expected values. The running continuation probability is the weighted average of the one or more individual probabilities. The method for determining a following target vehicle according to any one of Technologies 1 to 9.

[0124] (Technology 11) A program for causing a computer to execute the method for determining a following target vehicle according to any one of Technologies 1 to 10.

[0125] (Technology 12) A following target vehicle determination device that determines a target vehicle to be followed by the host vehicle based on the planned travel route of the host vehicle, comprising: a search unit that searches for candidate vehicles that are candidates for the target vehicle; a prediction unit that predicts one or more expected values respectively corresponding to one or more parameters representing the driving mode of the candidate vehicle; a calculation unit that calculates a driving continuation probability, which is the probability that the candidate vehicle continues to travel on the planned travel route, based on the one or more measured values respectively corresponding to the one or more parameters of the candidate vehicle and the one or more expected values; and a determination unit that determines the target vehicle based on the driving continuation probability, wherein the one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route.

[0126] (Technology 13) An autonomous mobile body comprising the following target vehicle determination device according to Technology 12 and a vehicle information acquisition unit that acquires information on the candidate vehicle.

Industrial Applicability

[0127] The following target vehicle determination method and the like according to one aspect of the present invention can be applied to vehicles and the like, for example, for autonomous driving.

Explanation of Signs

[0128] 10 Vehicle information acquisition unit 20 Search unit 30 Prediction unit 40 Calculation unit 50 Determination unit 70 Driving control unit 80 Storage unit 100 Following target vehicle determination device V10 Autonomous mobile body (host vehicle) V90, V91, V92, V93, V94 Candidate vehicles

Claims

1. A following target vehicle determination method for determining a target vehicle to be followed by the host vehicle based on a planned travel route of the host vehicle, comprising: a search step of searching for candidate vehicles that are candidates for the target vehicle; a prediction step of predicting one or more expected values respectively corresponding to one or more parameters representing the driving mode of the candidate vehicle; a calculation step of calculating a travel continuation probability, which is the probability that the candidate vehicle continues to travel on the planned travel route, based on one or more measured values respectively corresponding to the one or more parameters of the candidate vehicle and the one or more expected values; a determination step of determining the target vehicle based on the travel continuation probability, wherein the one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route Following target vehicle determination method.

2. The candidate vehicle is a vehicle located in front of the host vehicle The following target vehicle determination method according to Claim 1.

3. The target vehicle is a vehicle that moves from outside the planned travel route to the planned travel route The following target vehicle determination method according to Claim 1 or 2.

4. The target vehicle is a vehicle that travels on a route that merges into the planned travel route The following target vehicle determination method according to Claim 3.

5. The candidate vehicles include vehicles stopped on the road The following target vehicle determination method according to Claim 1 or 2.

6. The candidate vehicles include vehicles that stop at a temporary stop position The following target vehicle determination method according to Claim 1 or 2.

7. The one or more parameters include at least one of the speed of the candidate vehicle, the acceleration of the candidate vehicle, the lateral space from the edge of the lane in which the candidate vehicle is traveling to the candidate vehicle, and the inter-vehicle distance of the candidate vehicle The following target vehicle determination method according to Claim 1 or 2.

8. The one or more parameters include all of the speed of the candidate vehicle, the acceleration of the candidate vehicle, the lateral space from the edge of the lane in which the candidate vehicle is traveling to the candidate vehicle, and the inter-vehicle distance of the candidate vehicle The following target vehicle determination method according to Claim 1 or 2.

9. In the calculation step, assuming that the travel continuation probability follows a predetermined probability distribution with respect to the difference between each of the one or more measured values and the expected value corresponding to the measured value among the one or more expected values, the travel continuation probability is calculated The method for determining a following target vehicle according to claim 1 or 2.

10. In the calculating step, based on the one or more measured values and the one or more expected values, calculate one or more individual probabilities, Each of the one or more individual probabilities is calculated based on one of the one or more measured values and the expected value corresponding to the one measured value among the one or more expected values, and indicates the likelihood that the candidate vehicle will continue to travel on the planned travel route. The probability of continuing to travel is the weighted average of the one or more individual probabilities The method for determining a following target vehicle according to claim 1 or 2.

11. For causing a computer to execute the method for determining a following target vehicle according to claim 1 or 2 Program.

12. A following target vehicle determination device that determines a target vehicle to be followed by the host vehicle based on the planned travel route of the host vehicle, A search unit that searches for candidate vehicles that are candidates for the target vehicle, A prediction unit that predicts one or more expected values corresponding to one or more parameters representing the driving mode of the candidate vehicle, A calculation unit that calculates a probability of continuing to travel, which is the probability that the candidate vehicle will continue to travel on the planned travel route, based on one or more measured values corresponding to the one or more parameters of the candidate vehicle and the one or more expected values, A determination unit that determines the target vehicle based on the probability of continuing to travel, and The one or more expected values are values of the one or more parameters predicted when it is assumed that the candidate vehicle continues to travel on the planned travel route Following target vehicle determination device.

13. The following target vehicle determination device according to claim 12, and An autonomous mobile body comprising a vehicle information acquisition unit that acquires information on the candidate vehicle Autonomous mobile body.

Images