Behavior prediction method and behavior prediction device

Abstract

To provide a behavior prediction method and a behavior prediction device capable of reducing a delay in predicting the behavior of a pedestrian even when the pedestrian speed changes suddenly due to a current display of a pedestrian signal installed at a cross walk.SOLUTION: A behavior prediction method and a behavior prediction device acquire a current display state of a traffic light associated with a cross walk that intersects a route on which a vehicle is scheduled to travel. In this case, the traffic light shall display at least one of the following: a permission sign to permit a pedestrian to cross the cross walk, a prohibition sign to prohibit crossing the cross walk, and an intermediate sign displayed after the permission sign and before the prohibition sign. Then, in calculating a likelihood of a pedestrian crossing the cross walk based on the acquired current display state and a timing at which the pedestrian arrives at the cross walk, a second likelihood calculated when the current display state is the intermediate display is set larger than a first likelihood calculated when the current display state is the permission display.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a behavior prediction method and a behavior prediction device. [Background technology]

[0002] A traffic information providing device has been proposed that detects the presence of pedestrians crossing a vehicle when turning right or left at an intersection, based on the direction and speed of the vehicle after turning right or left and the direction and speed of the pedestrians. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-232412 Summary of the Invention [Problem to be solved by the invention]

[0004] According to Patent Document 1, when a pedestrian attempts to cross a crosswalk, the possibility that the pedestrian's speed may suddenly change due to the pedestrian signal phase installed at the crosswalk is not taken into consideration, which can result in a delay in predicting the pedestrian's behavior.

[0005] The present invention has been made in view of the above-mentioned problems, and its object is to provide a behavior prediction method and a behavior prediction device that can reduce the delay in predicting pedestrian behavior even when pedestrian speeds suddenly change due to the activation of pedestrian signals installed at crosswalks. [Means for solving the problem]

[0006] To solve the above-mentioned problems, a behavior prediction method and a behavior prediction device according to one aspect of the present invention acquire the current signal status of a traffic light associated with a crosswalk that intersects with a route along which a vehicle is scheduled to travel. Here, the traffic light displays at least one of a permission signal that allows pedestrians to cross the crosswalk, a prohibition signal that prohibits pedestrians from crossing the crosswalk, and an intermediate signal that appears after the permission signal and before the prohibition signal. Then, in calculating the likelihood that a pedestrian will cross the crosswalk based on the acquired current signal status and the timing at which the pedestrian will arrive at the crosswalk, the method sets a second likelihood, calculated when the current signal status is the intermediate signal, to be higher than a first likelihood, calculated when the current signal status is the permission signal. [Effects of the Invention]

[0007] According to the present invention, even when the speed of a pedestrian suddenly changes due to the aspect of a pedestrian signal installed at a crosswalk, the delay in predicting the behavior of the pedestrian can be reduced. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing the configuration of a behavior prediction device according to one embodiment of the present invention. [Figure 2] FIG. 2 is a flowchart showing the processing of the behavior prediction device according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0009] Next, an embodiment of the present invention will be described in detail with reference to the drawings. In the description, the same components are designated by the same reference numerals and duplicated explanations will be omitted.

[0010] [Configuration of behavior prediction device] Fig. 1 is a block diagram showing the configuration of a behavior prediction device according to this embodiment. As shown in Fig. 1, the behavior prediction device according to this embodiment includes an environmental information acquisition unit 70 (an image capture unit 71, an on-board sensor 73, and a map information acquisition unit 75) that acquires environmental information about the surroundings of the vehicle, and a controller 100. The controller 100 is connected to the image capture unit 71, the on-board sensor 73, the map information acquisition unit 75, and the vehicle control device 400 via wired or wireless communication paths.

[0011] The imaging unit 71, the on-board sensor 73, and the vehicle control device 400 are mounted on the vehicle itself, but the map information acquisition unit 75 and the controller 100 may be mounted on the vehicle itself or may be installed outside the vehicle.

[0012] The imaging unit 71 captures an image of the surroundings of the vehicle. For example, the imaging unit 71 is a digital camera equipped with a solid-state imaging element such as a CCD or CMOS, and captures an image of the surroundings of the vehicle to obtain a digital image of the surrounding area. The imaging unit 71 captures an image of a predetermined range around the vehicle by setting the focal length, the lens angle of view, the vertical and horizontal angles of the camera, etc.

[0013] The captured images captured by the imaging unit 71 are output to the controller 100 and stored in a storage unit (not shown) for a predetermined period of time. For example, the imaging unit 71 acquires captured images at predetermined time intervals, and the captured images acquired at the predetermined time intervals are stored in the storage unit as past images. The past images may be deleted after a predetermined period of time has passed since the capture of the past images.

[0014] The on-board sensor 73 is an object detection sensor mounted on the vehicle, such as a laser radar, a millimeter wave radar, a camera, or the like, that detects objects present around the vehicle. The on-board sensor 73 may include a plurality of different types of object detection sensors.

[0015] The on-vehicle sensor 73 detects the environment around the host vehicle. For example, the on-vehicle sensor 73 may detect moving objects, including other vehicles, motorcycles, bicycles, and pedestrians, and stationary objects, including parked vehicles, and detect the position, attitude, size, speed, acceleration, deceleration, yaw rate, etc., of the moving and stationary objects relative to the host vehicle. The on-vehicle sensor 73 may output, as a detection result, the behavior of two-dimensional objects in a zenith view (also called a plan view) viewed from the air above the host vehicle. The on-vehicle sensor 73 may also detect signs (road signs and signs marked on the road surface), guide rails, etc., present around the host vehicle.

[0016] Furthermore, in-vehicle sensor 73 detects the state of the host vehicle in addition to the environment around the host vehicle. For example, in-vehicle sensor 73 may detect the moving speed of the host vehicle (moving speed in the forward / backward direction, left / right direction, turning speed), the steering angle of the wheels of the host vehicle, and the rate of change of the steering angle.

[0017] In addition, the on-board sensor 73 may include a sensor that measures the absolute position of the vehicle, i.e., the position, attitude, and speed of the vehicle relative to a predetermined reference point, using a position detection sensor that measures the absolute position of the vehicle, such as a GPS (Global Positioning System) or odometry.

[0018] The map information acquisition unit 75 acquires map information indicating the structure of the road on which the host vehicle is traveling. The map information acquired by the map information acquisition unit 75 includes road structure information such as absolute lane positions, lane connection relationships, relative position relationships, and crosswalks intersecting with lanes. The map information acquired by the map information acquisition unit 75 may also include facility information such as parking lots and gas stations. Other map information may include traffic light position information, traffic light type, the position of stop lines corresponding to the traffic lights, and the current status of the traffic lights. The map information acquisition unit 75 may own a map database storing map information, or may acquire map information from an external map data server using cloud computing. The map information acquisition unit 75 may also acquire map information using vehicle-to-vehicle communication or road-to-vehicle communication.

[0019] A note is made on the current status of a traffic light included in the map information acquired by the map information acquisition unit 75. For example, if a traffic light is associated with a pedestrian crossing, the current status of the traffic light may indicate at least one of "permitted display," "prohibited display," and "intermediate display."

[0020] Here, a "permission indication" refers to an indication state that allows pedestrians to cross a crosswalk; for example, a permission indication corresponds to a green (or green) light on a traffic light. A "prohibition indication" refers to an indication state that prohibits pedestrians from crossing a crosswalk; for example, a prohibition indication corresponds to a red light on a traffic light. An "intermediate indication" refers to an indication state that appears after a permission indication but before a prohibition indication; for example, an intermediate indication corresponds to a flashing green (or green) light on a traffic light or a yellow light.

[0021] The map information acquisition unit 75 may acquire the current status of the traffic lights via the imaging unit 71 or the in-vehicle sensor 73, or may acquire the information using vehicle-to-vehicle communication or road-to-vehicle communication.

[0022] The environmental information about the surroundings of the vehicle includes at least information about the positions of objects around the vehicle. The information about the positions of objects may be calculated by analyzing images captured by the imaging unit 71 and information obtained by the on-board sensor 73. The environmental information about the surroundings of the vehicle may also include map information obtained by the map information acquisition unit 75. In addition to being acquired by the environmental information acquisition unit 70, the environmental information about the surroundings of the vehicle may also be acquired from outside the vehicle via vehicle-to-vehicle communication or road-to-vehicle communication.

[0023] The vehicle control device 400 controls the host vehicle based on the results obtained by the controller 100. For example, the vehicle control device 400 may be configured to drive the host vehicle by automatic driving, which automatically controls acceleration / deceleration and the turning angle of the steered wheels by controlling the accelerator opening and braking force of the brake device of the host vehicle and the steering angle of the steering wheel according to a predetermined driving route. Alternatively, the vehicle control device 400 may assist the driving operation of the occupant of the host vehicle by controlling at least a portion of the acceleration / deceleration, the turning angle of the steered wheels, and the steering force of the steering wheel by controlling the accelerator opening, braking force of the brake device, or a portion of the steering angle and steering reaction force of the steering wheel of the host vehicle. Furthermore, the vehicle control device 400 may assist the driver in driving along the predetermined driving route by displaying the predetermined driving route on a display device or the like.

[0024] The controller 100 (an example of a control unit or processing unit) is a general-purpose microcomputer equipped with a CPU (Central Processing Unit), a memory, and an input / output unit. A computer program (behavior prediction program) for causing the controller 100 to function as part of the behavior prediction device is installed in the controller 100. By executing the computer program, the controller 100 functions as multiple information processing circuits (120, 130, 140, 150, 160) equipped in the behavior prediction device.

[0025] Here, an example is shown in which the multiple information processing circuits (120, 130, 140, 150, 160) provided in the behavior prediction device are realized by software. However, it is also possible to configure the information processing circuits (120, 130, 140, 150, 160) by preparing dedicated hardware for executing each of the information processes described below. Also, the multiple information processing circuits (120, 130, 140, 150, 160) may be configured by individual hardware. Furthermore, the information processing circuits (120, 130, 140, 150, 160) may also be used as electronic control units (ECUs) used for other control related to the vehicle.

[0026] The controller 100 includes a crosswalk detection unit 120, a current signal status acquisition unit 130, a pedestrian information acquisition unit 140, a time measurement unit 150, and a likelihood calculation unit 160 as a plurality of information processing circuits (120, 130, 140, 150, 160).

[0027] The crosswalk detection unit 120 acquires the absolute position of the vehicle, i.e., the current position of the vehicle relative to a predetermined reference point (the position on the map where the vehicle is traveling) via the imaging unit 71 or the in-vehicle sensor 73. Then, the crosswalk detection unit 120 acquires information about crosswalks that exist around the vehicle, based on the map information acquired by the map information acquisition unit 75.

[0028] In particular, when the planned route of the host vehicle is known, the crosswalk detection unit 120 acquires information about crosswalks that intersect with the route the host vehicle is planned to travel. The crosswalk detection unit 120 may acquire information about crosswalks that are within a predetermined distance from the current position of the host vehicle. Alternatively, the crosswalk detection unit 120 may acquire information about crosswalks that the host vehicle can reach within a predetermined distance from the current position of the host vehicle.

[0029] The current indication status acquisition unit 130 acquires the current indication status of a traffic light associated with the crosswalk whose information has been acquired by the crosswalk detection unit 120, via the environmental information acquisition unit 70. The current indication status acquisition unit 130 may determine whether the current indication status of the traffic light is "intermediate indication."

[0030] Note that a "traffic light associated with a crosswalk" may be a traffic light that is installed around the crosswalk and is visible to pedestrians from a sidewalk connected to the crosswalk. Also, a "traffic light associated with a crosswalk" may be a traffic light that is assigned an identification number linked to the crosswalk in the map information.

[0031] The pedestrian information acquisition unit 140 extracts pedestrians located near the crosswalks about which information has been acquired by the crosswalk detection unit 120. Then, the pedestrian information acquisition unit 140 acquires the current positions and movement information of the extracted pedestrians via the environmental information acquisition unit 70. Here, "near the crosswalk" refers to an area within a certain range from the position of the crosswalk. The "near the crosswalk" may be determined based on the range that can be captured by the imaging unit 71 or the distance that can be sensed by the in-vehicle sensor 73.

[0032] The "movement information" includes the pedestrian's movement speed, movement direction, acceleration, and the like.

[0033] Furthermore, the pedestrian information acquisition unit 140 may determine whether the acceleration of a pedestrian is equal to or greater than a predetermined threshold. Here, the pedestrian information acquisition unit 140 may set the predetermined threshold based on the elapsed time from the timing when the current signal state of the traffic light changes from the permit signal to the intermediate signal, as measured by the timing measurement unit 150 described later. In particular, the pedestrian information acquisition unit 140 may set the predetermined threshold to a larger value as the elapsed time increases.

[0034] Alternatively, the pedestrian information acquisition unit 140 may estimate the attributes of the pedestrian. For example, the pedestrian information acquisition unit 140 may estimate the age group of the pedestrian based on the image captured by the imaging unit 71, and use the age group as the attribute of the pedestrian.

[0035] Furthermore, the pedestrian information acquisition unit 140 may estimate the age group of the pedestrian based on the movement information of the pedestrian. For example, the pedestrian information acquisition unit 140 may calculate the average movement speed of the pedestrian within a predetermined period, and estimate the age group of the pedestrian as young if the average movement speed is high, and estimate the age group of the pedestrian as middle-aged (senior) if the average movement speed is low.

[0036] Furthermore, the pedestrian information acquisition unit 140 may determine whether or not the pedestrian is accompanied by a child or whether or not the pedestrian is carrying luggage. For example, the pedestrian information acquisition unit 140 may determine whether or not there is an object moving in association with the pedestrian, based on the image acquired by the imaging unit 71, the information acquired by the in-vehicle sensor 73, etc. The presence or absence of an object moving in association with the pedestrian may be determined based on whether or not the movement information of the pedestrian is similar to the movement information of the object to be determined.

[0037] The estimation of pedestrian attributes by the pedestrian information acquisition unit 140 is not limited to the above example. For example, the pedestrian information acquisition unit 140 may use machine learning to estimate pedestrian attributes based on images acquired by the imaging unit 71 or other information acquired by the in-vehicle sensor 73.

[0038] The time measurement unit 150 measures the elapsed time from the timing when the current signal state of the traffic light changed from the permit indication to the intermediate indication, by referring to the current signal state of the traffic light acquired by the current signal state acquisition unit 130. Note that the time measurement unit 150 may acquire the elapsed time from the timing when the current signal state of the traffic light changed from the permit indication to the intermediate indication using road-to-vehicle communication.

[0039] The likelihood calculation unit 160 acquires the current position and movement information of the pedestrian, and calculates the timing at which the pedestrian will arrive at the crosswalk based on the current position and movement information.The likelihood calculation unit 160 then calculates the likelihood that the pedestrian will cross the crosswalk based on the current status of the traffic lights and the timing at which the pedestrian will arrive at the crosswalk.

[0040] For example, the likelihood calculation unit 160 may calculate a lower likelihood that the pedestrian will cross the crosswalk, the later the timing at which the pedestrian arrives at the crosswalk.

[0041] Furthermore, likelihood calculation unit 160 may calculate the position of the pedestrian after a predetermined time has elapsed based on the pedestrian's movement speed, movement direction, and current position, which are included in the movement information. Likelihood calculation unit 160 may calculate a higher likelihood that the pedestrian will cross the crosswalk, the closer the pedestrian's position is to the crosswalk after the predetermined time has elapsed.

[0042] Alternatively, the likelihood calculation unit 160 may set the likelihood calculated when the current signal state of the traffic light is a permit indication as the first likelihood, and the likelihood calculated when the current signal state of the traffic light is an intermediate indication as the second likelihood, with the second likelihood being greater than the first likelihood. This reflects the situation where, compared to when the current signal state of the traffic light is a permit indication, a pedestrian accelerates near a crosswalk and tries to cross the crosswalk in a hurry when the current signal state of the traffic light is an intermediate indication. Note that the likelihood calculation unit 160 may also calculate the likelihood calculated when the current signal state of the traffic light is a prohibition indication as the third likelihood.

[0043] Furthermore, the likelihood calculation unit 160 may set the second likelihood to be greater than the first likelihood when the acceleration of the pedestrian is equal to or greater than a predetermined threshold. More specifically, when the pedestrian information acquisition unit 140 determines that the acceleration of the pedestrian is equal to or greater than a predetermined threshold, the likelihood calculation unit 160 may set the second likelihood to be greater than the first likelihood. This reflects a situation in which a pedestrian actually accelerates near a crosswalk, resulting in the pedestrian arriving at the crosswalk earlier.

[0044] The predetermined threshold used to determine whether the pedestrian's acceleration is equal to or greater than the predetermined threshold may be set to a larger value as the time elapsed since the traffic light's current status changed from a permitted status to an intermediate status increases. The longer the elapsed time, the shorter the time until the traffic light changes to a prohibited status. Therefore, unless the pedestrian accelerates at a higher rate, the pedestrian cannot complete crossing the crosswalk while the traffic light is in the intermediate status. Therefore, the likelihood calculation unit 160 may set the second likelihood to be larger than the first likelihood only when the pedestrian's acceleration is greater as the time elapsed increases.

[0045] Furthermore, the likelihood calculation unit 160 may set the amount of change by subtracting the first likelihood from the second likelihood based on the time elapsed since the timing when the traffic light's current indication changed from a permit indication to an intermediate indication. In particular, the likelihood calculation unit 160 may set the amount of change to be smaller the longer the elapsed time. This reflects the situation in which the longer the elapsed time, the shorter the time until the traffic light changes to a prohibition indication, making it more difficult for pedestrians to start crossing the crosswalk.

[0046] Furthermore, likelihood calculation unit 160 may set the second likelihood to be greater than the first likelihood if the pedestrian starts accelerating within a predetermined time from the timing when the current signal of the traffic light changes from the allow indication to the intermediate indication. This reflects a situation in which the pedestrian actually accelerates near the crosswalk while the current signal of the traffic light is the intermediate indication, and the timing when the pedestrian arrives at the crosswalk becomes earlier.

[0047] On the other hand, the likelihood calculation unit 160 may set the second likelihood to be smaller than the first likelihood when a pedestrian starts to decelerate while the current signal state is intermediate. This reflects a situation in which a pedestrian actually decelerates near a crosswalk while the current signal state of the traffic light is intermediate, delaying the timing at which the pedestrian reaches the crosswalk. Furthermore, when a pedestrian decelerates while the current signal state is intermediate, it is considered that the pedestrian has given up on crossing the crosswalk. Therefore, the calculated second likelihood is set to be small.

[0048] Furthermore, the likelihood calculation unit 160 may set a difference between the first likelihood and the second likelihood based on the attribute of the pedestrian estimated by the pedestrian information acquisition unit 140. More specifically, when the attribute of the pedestrian is age group and the pedestrian is a young person, the likelihood calculation unit 160 may set a large difference. This reflects the fact that young people tend to be relatively active and accelerate near crosswalks.

[0049] On the other hand, when the attribute of the pedestrian is an age group and the pedestrian is in the middle-aged group, the likelihood calculation unit 160 may set the difference amount to a small value. This reflects the fact that when the pedestrian is in the middle-aged group, the activity level of the pedestrian is relatively low and there is a tendency for the pedestrian to accelerate less near a crosswalk.

[0050] Furthermore, when the pedestrian's attribute indicates that the pedestrian is accompanied by a child or is carrying luggage, the likelihood calculation unit 160 may set the difference amount to a small value. This reflects the fact that when a pedestrian is accompanied by a child or is carrying luggage, the pedestrian is less likely to accelerate near a crosswalk.

[0051] On the other hand, if the pedestrian's attribute does not indicate that the pedestrian is accompanied by a child or is carrying luggage, the likelihood calculation unit 160 may set the difference amount to a large value. This reflects the fact that pedestrians tend to accelerate near crosswalks when they are not accompanied by a child or carrying luggage.

[0052] The likelihood calculated by the likelihood calculation unit 160 may be transmitted to the vehicle control device 400, and control based on the likelihood may be performed by a driving assistance method or driving assistance device. For example, the vehicle control device 400 may accelerate the deceleration of the host vehicle as the calculated likelihood increases, thereby stopping the host vehicle before a pedestrian crossing. Accelerating the deceleration of the host vehicle can reduce sudden deceleration of the host vehicle. Such driving assistance can reduce discomfort felt by the occupants of the host vehicle.

[0053] [Processing procedure of behavior prediction device] Next, the processing procedure of the behavior prediction device according to this embodiment will be described with reference to the flowchart of Fig. 2. The processing of the behavior prediction device shown in Fig. 2 may be repeatedly executed at a predetermined cycle. It is assumed that environmental information acquisition unit 70 acquires environmental information around the vehicle at any time in parallel with the processing procedure of the flowchart of Fig. 2.

[0054] In step S101, the crosswalk detection unit 120 acquires information about crosswalks that intersect with the route along which the host vehicle is scheduled to travel.

[0055] In step S103, the pedestrian information acquisition unit 140 extracts pedestrians located near the crosswalk for which information has been acquired.

[0056] In step S105, the pedestrian information acquisition unit 140 acquires the current position and movement information of the extracted pedestrian.

[0057] In step S107, the current signal status acquisition unit 130 acquires the current signal status of the traffic light associated with the crosswalk whose information has been acquired.

[0058] In step S109, the current indication status acquisition unit 130 determines whether the current indication status of the traffic light is "intermediate indication."

[0059] If it is determined that the current status of the traffic light is not "intermediate display" (NO in step S109), in step S119, the likelihood calculation unit 160 calculates the likelihood (first likelihood or third likelihood) that the pedestrian will cross the crosswalk based on the pedestrian's current position and movement information.

[0060] If it is determined that the current indication status of the traffic light is "intermediate indication" (YES in step S109), in step S111, the time measurement unit 150 measures the elapsed time from the time when the current indication status of the traffic light changed from the permitted indication to the intermediate indication.

[0061] In step S113, the pedestrian information acquisition unit 140 acquires the acceleration of the pedestrian. Here, the pedestrian information acquisition unit 140 may determine whether or not the acceleration of the pedestrian is equal to or greater than a predetermined threshold.

[0062] In step S113, the pedestrian information acquisition unit 140 estimates the attributes of the pedestrian.

[0063] In step S117, the likelihood calculation unit 160 calculates the likelihood (second likelihood) that the pedestrian will cross the crosswalk based on the pedestrian's current position and movement information. Note that the likelihood calculation unit 160 may calculate the second likelihood based on the elapsed time from when the current signal state of the traffic light changed from the permit signal to the intermediate signal, the pedestrian's acceleration, and the pedestrian's attributes.

[0064] After the processes of steps S117 and S119, the flowchart of FIG. 2 ends.

[0065] [Effects of the embodiment] As described above in detail, the behavior prediction method and behavior prediction device according to this embodiment acquire the current signal status of a traffic light associated with a crosswalk that intersects with a route along which a vehicle is scheduled to travel. Here, the traffic light displays at least one of a permission signal that allows pedestrians to cross the crosswalk, a prohibition signal that prohibits pedestrians from crossing the crosswalk, and an intermediate signal that appears after the permission signal and before the prohibition signal. Then, in calculating the likelihood that a pedestrian will cross the crosswalk based on the acquired current signal status and the timing at which the pedestrian will arrive at the crosswalk, the second likelihood, calculated when the current signal status is the intermediate signal, is set to be higher than the first likelihood, calculated when the current signal status is the permission signal.

[0066] This reduces the delay in predicting pedestrian behavior even when a pedestrian's speed suddenly changes due to the pedestrian signal phase installed at the crosswalk. Furthermore, when the traffic light's current state is intermediate, the system can respond to situations in which a pedestrian accelerates near the crosswalk and tries to cross the crosswalk in a hurry, improving the accuracy of the calculated likelihood. Even when a pedestrian suddenly accelerates and crosses the crosswalk, it is possible to avoid situations in which the host vehicle approaches or comes into contact with the pedestrian, and also to avoid situations in which the host vehicle needs to suddenly decelerate in order to avoid approaching or coming into contact with the pedestrian.

[0067] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may acquire the acceleration of the pedestrian, and if the acceleration is equal to or greater than a predetermined threshold, set the second likelihood to be greater than the first likelihood. This makes it possible to deal with situations in which the pedestrian actually accelerates near the crosswalk and arrives at the crosswalk earlier, thereby improving the accuracy of the calculated likelihood.

[0068] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may acquire the elapsed time from the timing when the current signal status changes from a permitted signal to an intermediate signal, and set the predetermined threshold based on the elapsed time. Alternatively, the predetermined threshold may be increased as the elapsed time increases. The longer the elapsed time, the shorter the time until the traffic light changes to a prohibited signal, and unless a pedestrian accelerates at a higher rate, the pedestrian will not be able to complete crossing the crosswalk while the traffic light is in the intermediate signal. Setting the predetermined threshold based on the elapsed time allows the likelihood to be calculated accurately, taking into account the time allowed for the pedestrian to cross the crosswalk.

[0069] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may acquire the elapsed time from the timing when the current signal status changed from a permitted signal to an intermediate signal, and set a change amount by subtracting the first likelihood from the second likelihood based on the elapsed time. Alternatively, the change amount may be smaller the longer the elapsed time. The longer the elapsed time, the shorter the time until the traffic light changes to a prohibited signal, which reflects a situation in which it becomes difficult for pedestrians to start crossing the crosswalk. By setting the change amount based on the elapsed time, the likelihood can be calculated accurately, taking into account the time allowed for pedestrians to cross the crosswalk.

[0070] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may set the second likelihood to be greater than the first likelihood when a pedestrian starts accelerating while the current display state is intermediate. This allows the likelihood to be calculated in a way that reflects a situation in which a pedestrian actually accelerates near a crosswalk and arrives at the crosswalk earlier. As a result, the accuracy of the calculated likelihood is improved.

[0071] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may set the second likelihood to be greater than the first likelihood if the pedestrian starts accelerating within a predetermined time from the time when the current indication changes from the permit indication to the intermediate indication. This allows the likelihood to be calculated in a way that reflects the situation in which the pedestrian actually accelerates near the crosswalk and arrives at the crosswalk earlier. As a result, the accuracy of the calculated likelihood is improved.

[0072] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may set the second likelihood to be smaller than the first likelihood when the pedestrian starts to decelerate while the current indication is intermediate. This allows the likelihood to be calculated to reflect a situation in which the pedestrian decelerates near the crosswalk and the timing of the pedestrian's arrival at the crosswalk is delayed. As a result, the accuracy of the calculated likelihood is improved. Note that when the pedestrian decelerates while the current indication is intermediate, it is considered that the pedestrian has given up on crossing the crosswalk.

[0073] Furthermore, the behavior prediction method and behavior prediction device according to this embodiment may acquire attributes of the pedestrian and set a difference between the first likelihood and the second likelihood based on the attributes. This allows for the calculation of a likelihood that reflects the attributes of the pedestrian. As a result, the accuracy of the calculated likelihood is improved.

[0074] Furthermore, the driving assistance method and driving assistance device according to this embodiment may accelerate the deceleration of the vehicle as the likelihood calculated by the above-described behavior prediction method or behavior prediction device increases. This reduces sudden deceleration of the host vehicle and reduces discomfort felt by the vehicle occupants. In particular, even if a pedestrian suddenly accelerates to cross a crosswalk, it is possible to avoid a situation in which the host vehicle approaches or comes into contact with the pedestrian, and it is also possible to avoid a situation in which the host vehicle needs to suddenly decelerate in order to avoid approaching or coming into contact with the pedestrian.

[0075] Each function described in the above embodiments may be implemented by one or more processing circuits, including programmed processors, electrical circuits, and even devices such as application specific integrated circuits (ASICs), circuit components arranged to perform the described functions.

[0076] Although the present invention has been described above based on the embodiments, it will be apparent to those skilled in the art that the present invention is not limited to these descriptions and that various modifications and improvements are possible. The descriptions and drawings that form part of this disclosure should not be understood as limiting the present invention. Various alternative embodiments, examples, and operating techniques will become apparent to those skilled in the art from this disclosure.

[0077] The present invention naturally includes various embodiments not described herein. Therefore, the technical scope of the present invention is defined only by the invention-specifying matters according to the scope of the claims that are appropriate from the above description. [Explanation of symbols]

[0078] 70 Environmental Information Acquisition Department 71 Imaging unit 73 In-vehicle sensors 75 Map information acquisition unit 100 Controllers 120 Crosswalk detection unit 130 Current status acquisition unit 140 Pedestrian Information Acquisition Unit 150 Time measurement section 160 Likelihood calculation unit 400 Vehicle control device

Claims

1. Obtain information on pedestrian crossings that intersect with the route the vehicle is scheduled to travel, a traffic light associated with the crosswalk, the traffic light providing a warning to pedestrians; A permission sign permitting crossing of the crosswalk; a prohibition sign prohibiting crossing the crosswalk; an intermediate display displayed after the permission display and before the prohibition display; and acquiring a current status of the traffic signal indicating at least one of the following: Acquire the current location and movement information of the pedestrian; calculating a timing at which the pedestrian will arrive at the crosswalk based on the current position and the movement information; In calculating the likelihood that the pedestrian will cross the crosswalk based on the current status and the timing, a likelihood calculated when the current indication state is the permission indication is defined as a first likelihood; a likelihood calculated when the current display state is the intermediate display is defined as a second likelihood, setting the second likelihood to be greater than the first likelihood; A behavior prediction method characterized by:

2. 2. A behavior prediction method according to claim 1, Acquire the acceleration of the pedestrian; When the acceleration is equal to or greater than a predetermined threshold, the second likelihood is set to be greater than the first likelihood. A behavior prediction method characterized by:

3. 3. A behavior prediction method according to claim 2, acquire the elapsed time from the timing when the current display state changed from the permission display to the intermediate display; setting the predetermined threshold value based on the elapsed time; A behavior prediction method characterized by:

4. 4. A behavior prediction method according to claim 3, The longer the elapsed time, the larger the predetermined threshold value. A behavior prediction method characterized by:

5. A behavior prediction method according to any one of claims 1 to 4, acquire the elapsed time from the timing when the current display state changed from the permission display to the intermediate display; setting a change amount obtained by subtracting the first likelihood from the second likelihood based on the elapsed time; A behavior prediction method characterized by:

6. 6. A behavior prediction method according to claim 5, The longer the elapsed time, the smaller the amount of change. A behavior prediction method characterized by:

7. A behavior prediction method according to any one of claims 1 to 6, When the pedestrian starts accelerating while the current display state is the intermediate display state, the second likelihood is set to be greater than the first likelihood. A behavior prediction method characterized by:

8. A behavior prediction method according to claim 7, When the pedestrian starts accelerating within a predetermined time from the timing when the current display state changes from the permission display to the intermediate display, the second likelihood is set to be larger than the first likelihood. A behavior prediction method characterized by:

9. A behavior prediction method according to any one of claims 1 to 8, When the pedestrian starts to decelerate while the current display state is the intermediate display state, the second likelihood is set to be smaller than the first likelihood. A behavior prediction method characterized by:

10. A behavior prediction method according to any one of claims 1 to 9, acquiring attributes of the pedestrian; setting a difference between the first likelihood and the second likelihood based on the attribute; A behavior prediction method characterized by:

11. The greater the likelihood calculated by the behavior prediction method according to any one of claims 1 to 10, the faster the deceleration of the vehicle is. A driving assistance method characterized by the above.

12. A behavior prediction device including an acquisition unit and a controller, The acquisition unit Obtain information on pedestrian crossings that intersect with the route the vehicle is scheduled to travel, a traffic light associated with the crosswalk, the traffic light providing a warning to pedestrians; A permission sign permitting crossing of the crosswalk; a prohibition sign prohibiting crossing the crosswalk; an intermediate display displayed after the permission display and before the prohibition display; and acquiring a current status of the traffic signal indicating at least one of the following: Acquire the current location and movement information of the pedestrian; The controller calculating a timing at which the pedestrian will arrive at the crosswalk based on the current position and the movement information; In calculating the likelihood that the pedestrian will cross the crosswalk based on the current status and the timing, a likelihood calculated when the current indication state is the permission indication is defined as a first likelihood; a likelihood calculated when the current display state is the intermediate display is defined as a second likelihood, setting the second likelihood to be greater than the first likelihood; A behavior prediction device characterized by the above.

Images