Vehicle travel control method and vehicle travel control device

Abstract

A vehicle travel control method executed by a processor provided in a vehicle travel control device (100) mounted on a vehicle, wherein the processor: sets, when the vehicle passes through a crosswalk, a boundary line to separate the roadway and a pedestrian space, which is a space combining the sidewalks and the crosswalk, on the basis of the shape of the roadway and the shape of the crosswalk; generates a path of movement of a fluid virtually filled in the pedestrian space; predicts, on the basis of the path of movement of the fluid, the walking paths of the pedestrians on the sidewalks heading toward the crosswalk assuming that all pedestrians on the sidewalks enter the crosswalk; predicts future positions of the pedestrians on the basis of the walking paths of the pedestrians; and controls the travel of the vehicle on the basis of the future positions of the pedestrians.

Description

Vehicle Travel Control Method and Vehicle Travel Control Device

[0001] The present invention relates to a vehicle travel control method and a vehicle travel control device.

[0002] Technological development related to the automatic driving of vehicles is underway. For example, Patent Document 1 discloses that when the traveling direction of a vehicle intersects with the traveling direction of a pedestrian or a bicycle, it determines whether there is a crosswalk in the traveling direction of the pedestrian or the bicycle. When it is determined that there is a crosswalk, compared with the case where it is determined that there is no crosswalk, by making it easier to activate the driving support control, it discloses a driving support device that performs driving support control for collision avoidance.

[0003] Japanese Patent Application Laid-Open No. 2018-122633

[0004] For example, even if a pedestrian or the like traveling in the same direction as the vehicle intends to cross a crosswalk ahead in the traveling direction, as shown in FIG. 11, the driving support control is highly likely not to be activated until the traveling direction is changed toward the road side, and it cannot be said that appropriate collision avoidance is being performed.

[0005] The present invention has been made in view of the above circumstances, and an object thereof is to appropriately support collision avoidance between a vehicle and a pedestrian crossing a crosswalk.

[0006] To achieve the above object, a vehicle travel control method according to the present invention is a vehicle travel control method executed by a processor included in a vehicle travel control device mounted on a vehicle. When the vehicle passes through a crosswalk, the processor sets a boundary line that divides the road and a walking space, which is a space combining the road, the sidewalk, and the crosswalk, based on the road shape and the crosswalk shape. It generates a movement path of a fluid virtually filled in the walking space, predicts the walking path of a pedestrian assuming that all pedestrians on the sidewalk heading toward the crosswalk enter the crosswalk based on the movement path of the fluid, predicts the future position of the pedestrian according to the walking path of the pedestrian, and controls the travel of the vehicle based on the future position of the pedestrian.

[0007] According to the present invention, a vehicle travel control method and a vehicle travel control device can appropriately support collision avoidance between a vehicle and a pedestrian crossing a crosswalk.

[0008] This is a block diagram illustrating an example of the configuration of a vehicle according to an embodiment. This is a block diagram illustrating an example of the hardware configuration of a vehicle driving control device. This is a block diagram illustrating an example of the functional configuration of a vehicle driving control device. This is a diagram illustrating the step of setting the intersection point between the outline of the roadway shape and the outline of the pedestrian crossing shape when setting a boundary line. This is a diagram illustrating the step of connecting the outline of the roadway shape and the outline of the pedestrian crossing shape when setting a boundary line. This is a diagram illustrating the step of smoothly shaping the connection between the upper boundary line and the lower boundary line when setting a boundary line. This is a diagram illustrating the movement path of a liquid virtually filled in a pedestrian space. This is a diagram illustrating the method of setting a linear charge in the substitute charge method. This is a diagram illustrating the method of obtaining a pedestrian's walking path assuming that the pedestrian will always enter the pedestrian crossing. This is a diagram illustrating the method of calculating the effective speed of a pedestrian and the method of generating a speed plan. This is a diagram illustrating the method of predicting the future position of a pedestrian. This is a diagram showing a section of the walking path with a large curvature. This is a flowchart illustrating the flow of the vehicle driving control processing. This is a flowchart illustrating the flow of the pedestrian future position prediction processing. This is a diagram showing a risk area extended in the direction of the pedestrian's movement. This is a diagram showing a risk area deformed to match the pedestrian's walking path. This is a diagram illustrating a conventional example.

[0009] The vehicle driving control method and vehicle driving control device according to embodiments of the present invention will be described in detail below with reference to the drawings. The vehicle driving control method and vehicle driving control device according to embodiments of the present invention predict the movement path of pedestrians near a crosswalk and appropriately avoid collisions between vehicles and pedestrians. In the following description, all moving objects that pass through a crosswalk, such as pedestrians and cyclists, will be referred to as "pedestrians."

[0010] As shown in Figure 1, the vehicle 1 according to an embodiment of the present invention is a mobile body comprising a vehicle driving control device 100 for controlling the driving of the vehicle 1, a display device 200 for displaying various information, a sensor unit 300 for detecting various information, and a driving mechanism 400 for driving the vehicle 1.

[0011] The vehicle driving control device 100 is an information processing device that can be mounted on the vehicle 1. For example, the vehicle driving control device 100 performs driving control to avoid collisions between the vehicle 1 and pedestrians based on detection information from the sensor unit 300.

[0012] The display device 200 is an image display device such as an LCD (Liquid Crystal Display), PDP (Plasma Display Panel), or organic EL (Electro-Luminescence) display, and displays various types of information. For example, the display device 200 displays map information, images of the area around the vehicle 1, etc., on its display screen.

[0013] The sensor unit 300 is composed of a combination of various sensors that detect various information about the vehicle 1 and its surroundings (e.g., vehicle position, road conditions, pedestrians, etc.). The sensor unit 300 includes, for example, a GPS (Global Positioning System) device, an imaging device, a Lidar (Laser Imaging Detection And Ranging), etc.

[0014] The driving mechanism 400 includes an engine or motor that supplies driving power to the vehicle 1, brakes that stop or suppress the movement of the vehicle 1, and a steering mechanism that steers the vehicle 1.

[0015] Next, an example of the hardware configuration of the vehicle driving control device 100 will be described. As shown in Figure 2, the vehicle driving control device 100 physically comprises a processor 101, memory 102, storage 103, a communication interface (communication I / F (InterFace)) 104, and an input / output interface (input / output I / F) 105. Each of these components is electrically connected to the others via a bus line 106.

[0016] The processor 101 is a computing device that controls the operation of the entire vehicle driving control device 100. The processor 101 is a general-purpose processor such as a CPU (Central Processing Unit), MPU (Micro Processing Unit), or GPU (Graphics Processing Unit). However, the processor 101 is not limited to a general-purpose processor; it may also be a dedicated processor composed of an ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), etc.

[0017] Memory 102 is the main memory and includes ROM (Read Only Memory), RAM (Random Access Memory), etc. The processor 101 reads programs and various data from ROM or storage 103 onto RAM and executes processing to realize the overall control and functions of the vehicle driving control device 100.

[0018] Storage 103 is an auxiliary storage device that stores programs and various data necessary for program execution, and includes non-volatile storage devices such as HDDs (Hard Disk Drives) and SSDs (Solid State Drives). Storage 103 stores, for example, the OS (Operating System), which is the basic software that controls the entire vehicle driving control device 100, and applications that run on the OS and provide various functions. Storage 103 also stores, for example, map information including road information representing the road network. Note that some of the programs and various data necessary for program execution may be stored in ROM.

[0019] The communication interface 104 connects to a network (not shown) and is an interface for the vehicle driving control device 100 to communicate data with external devices. The network is composed of, for example, an internet line network, a LAN (Local Area Network), a WAN (Wide Area Network), a VPN (Virtual Private Network), a dedicated communication network, etc., and may involve various network relay devices such as antennas, gateways, routers, and hubs. The communication interface 104 includes, for example, a network board, a wireless communication module, etc.

[0020] The input / output interface 105 is an interface for connecting to input devices such as keyboards and mice that input operation signals, and output devices such as speakers that output audio data. The input / output interface 106, for example, takes in operation data input by the operator via the input device and outputs data such as control signals, moving images, and audio to the output device.

[0021] Next, an example of the functional configuration of the vehicle driving control device 100 will be described. As shown in Figure 3, the vehicle driving control device 100 functionally comprises a control unit 110, a storage unit 120, a communication unit 130, and an input / output unit 140.

[0022] The control unit 110 is implemented, for example, by the processor 101 and memory 102 shown in Figure 2. The storage unit 120 is implemented, for example, by the memory 102 and storage 103 shown in Figure 2. The communication unit 130 is implemented by the communication interface 104 shown in Figure 2. The input / output unit 140 is implemented, for example, by the input / output interface 105 shown in Figure 2.

[0023] The control unit 110 controls the overall functions of the vehicle driving control device 100. The control unit 110 includes a road environment acquisition unit 111, a driving route planning unit 112, a pedestrian crossing information acquisition unit 113, a boundary line setting unit 114, a fluid movement path generation unit 115, a pedestrian detection unit 116, a pedestrian future position prediction unit 117, a driving mode determination unit 118, and a vehicle control unit 119.

[0024] The road environment acquisition unit 111 acquires road environment information around the vehicle 1. The road environment acquisition unit 111 acquires road environment information, which is information indicating the road structure around the vehicle 1, road traffic rules, etc., based on map information stored in the storage unit 120 and images captured by the imaging device equipped on the vehicle 1. The map information includes road information, which is composed of nodes representing intersections and junctions on the road network representation, and links representing road sections connecting each node. The road information may also include the names of intersections, etc., and attribute information of road shapes. The map information also includes feature information indicating traffic lights, road signs, road markings, facilities, etc. Furthermore, the road environment acquisition unit 111 acquires road environment information, which is information indicating the road environment around the vehicle 1, by analyzing images captured by the imaging device equipped on the vehicle 1 using well-known image analysis techniques. Furthermore, the road environment information includes traffic conditions (traffic volume, congestion, etc.), accident and construction conditions, etc., acquired by the road environment acquisition unit 111 from an external source via a network not shown.

[0025] The route planning unit 112 plans the route the vehicle 1 will take to travel from its current location to its destination. The route planning unit 112 takes into account the location information (current location) from the GPS device included in the sensor unit 300 of the vehicle 1, the destination entered by the user via the input / output unit 140, and the road environment information acquired by the road environment acquisition unit 111 to generate route planning information that shows the route the vehicle 1 will take.

[0026] The pedestrian crossing information acquisition unit 113 acquires pedestrian crossing information relating to pedestrian crossings provided on the driving route planned by the driving route planning unit 112 from map information acquired by the road environment acquisition unit 111 and images captured by the imaging device installed in the vehicle 1. A pedestrian crossing is a road marking provided for pedestrians to cross the roadway. Pedestrian crossings are usually drawn in stripes on the paved surface of the roadway with white paint. Pedestrian crossing information includes information such as the pedestrian crossing location which can identify the location of the pedestrian crossing, and the pedestrian crossing size which can identify the size of the pedestrian crossing. The pedestrian crossing location is represented by three-dimensional coordinates. The pedestrian crossing size is represented by the position coordinates of the center of the pedestrian crossing, the position coordinates of the endpoints, and the dimensions in the width direction (roadway width direction) and depth direction (lane direction).

[0027] The boundary setting unit 114 sets a boundary line that separates the roadway from the pedestrian space. Here, the pedestrian space is a road area where pedestrians can walk in accordance with traffic rules, and is basically a space (area) that combines the sidewalk and the pedestrian crossing provided on the roadway.

[0028] The boundary line setting unit 114 sets a boundary line that separates the roadway and the pedestrian space based on the roadway shape and the pedestrian crossing shape. Specifically, as shown in Figure 4A, the boundary line setting unit 114 sets four intersection points PI1 to PI4 where the outline lines ORL and ORR of the roadway shape intersect with the outline line OPC of the pedestrian crossing shape. In the following explanation, the upward direction is considered up in the plane of the paper in Figures 4A, 4B, and 4C, and the opposite direction is considered down.

[0029] Next, as shown in Figure 4B, the boundary line setting unit 114 sets an upper boundary line BU by connecting the line segment connecting intersection PI1 and intersection PI2, which are aligned in the width direction (length direction, or the direction in which pedestrians cross the crosswalk) of the four intersections PI1 to PI4 (a part of the upper side of the outer outline line OPC of the crosswalk shape) and the parts of the outer outline lines ORL and ORR of the roadway shape, which are separated by the outer outline line OPC of the crosswalk shape, that are located above intersections PI1 and PI2. Similarly, the boundary line setting unit 114 sets a lower boundary line BL by connecting the line segment connecting intersection PI3 and intersection PI4, which are aligned in the width direction of the crosswalk (a part of the lower side of the outer outline line OPC of the crosswalk shape) and the parts of the outer outline lines ORL and ORR of the roadway shape, which are separated by the outer outline line OPC of the crosswalk shape, that are located below intersections PI3 and PI4. In Figure 4B, to make the lines easier to see, the overlapping portions of the upper boundary line BU and the lower boundary line BL with the outline of the pedestrian crossing OPC are slightly offset in the illustration.

[0030] Next, as shown in Figure 4C, the boundary line setting section 114 shapes the corners (connecting parts) of the upper boundary line BU and the lower boundary line BL into a rounded, smooth shape. The boundary line setting section 114 forms the connecting part (corner) of the upper boundary line BU between the line segment connecting intersection point PI1 and intersection point PI2 and the portion above intersection points PI1 and PI2 of the outer contour lines ORL and ORR of the roadway shape, which are separated by the outer contour line OPC of the pedestrian crossing shape, into a rounded, curved shape. The boundary line setting section 114 also forms the connecting part (corner) of the lower boundary line BL between the line segment connecting intersection point PI3 and intersection point PI4 and the portion below intersection points PI3 and PI4 of the outer contour lines ORL and ORR of the roadway shape, which are separated by the outer contour line OPC of the pedestrian crossing shape, into a rounded, curved shape.

[0031] The fluid movement path generation unit 115 generates a movement path for a fluid virtually filling the walking space. More specifically, the fluid movement path is the path taken by the fluid virtually filling the walking space as it enters the crosswalk from the sidewalk. The estimation of the flow field, which is a type of vector field in which the flow velocity is defined at any point in space, is calculated under the assumption that the flow field follows the fundamental equations of an inviscid, incompressible fluid (ideal fluid). The governing equations for an ideal fluid are known to be, for example, the Laplace equation, which is a partial differential equation. In this embodiment, the approximate solution (potential value) of the Laplace equation is calculated using the substitute charge method (charge superposition method), which is one of the approximate solution methods for partial differential equations. In this way, the walking path of a pedestrian is predicted by calculating the potential value, which is the solution to the Laplace equation, using the substitute charge method.

[0032] As shown in Figure 5A, a potential field is formed by virtually filling the walking space with liquid and applying different potential values ​​(for example, +1V and -1V) to the upper boundary line BU and lower boundary line BL set by the boundary line setting unit 114. The fluid filling the walking space is assumed to be an inviscid, incompressible fluid (ideal fluid) and to be a steady flow that does not change over time. Equicharge lines (equipotential lines) are obtained by searching for a reference position where the charges (potential values) are equal. In Figure 5A, an electric field was generated as a potential field by applying a voltage of +1V to the upper boundary line BU and -1V to the lower boundary line BL, so multiple equicharge lines are obtained in the walking region formed by the upper boundary line BU and the lower boundary line BL. These equicharge lines can be considered as fluid movement paths.

[0033] As shown in Figure 5B, the fluid's movement path can be adjusted by setting linear charges in addition to the upper boundary line BU and the lower boundary line BL. For example, a linear charge LC1 is set by offsetting the line segment (drawn as a V-shaped solid line) that connects each endpoint of the upper boundary line BU to the midpoint of the upper boundary line BU inward. Similarly, a linear charge LC2 is set by offsetting the line segment (drawn as a solid line) that connects each endpoint of the lower boundary line BL inward. By setting linear charges LC1 and LC2 in this way, equicharge lines can be obtained inside the upper boundary line BU and below the lower boundary line BL.

[0034] The pedestrian detection unit 116 detects pedestrians walking around the crosswalk. The pedestrian detection unit 116 analyzes images captured by the imaging device of the sensor unit 300 of the vehicle 1 using, for example, well-known image analysis techniques, to detect pedestrians around the crosswalk. The pedestrian detection unit 116 generates pedestrian observation information, such as the pedestrian's position, walking direction, and walking speed, and outputs it to the pedestrian future position prediction unit 117. In the following description, the walking direction and walking speed of the pedestrian indicated by the pedestrian observation information may be referred to as the "movement vector".

[0035] The pedestrian future position prediction unit 117 predicts the future position of a pedestrian based on pedestrian observation information acquired from the pedestrian detection unit. The pedestrian future position prediction unit 117 includes a walking path prediction unit 117a, a walking speed plan generation unit 117b, and a future position calculation unit 117c.

[0036] The walking path prediction unit 117a predicts the walking path of a pedestrian based on pedestrian observation information acquired from the pedestrian detection unit 116 and the fluid movement path generated by the fluid movement path generation unit 115. For example, as shown in Figure 6A, the walking path prediction unit 117a can uniquely identify the walking path of a pedestrian P by continuously connecting the unit vectors on the fluid movement path at the current position of the pedestrian P indicated by the pedestrian observation information, assuming that all pedestrians P on the sidewalk heading towards the crosswalk will enter the crosswalk.

[0037] The walking speed plan generation unit 117b calculates the effective speed and generates a walking speed plan for a predetermined time. The walking speed plan generation unit 117b calculates the effective speed based on the angle between the direction of the unit vector in the walking path predicted by the walking path prediction unit 117a and the observed direction of movement. For example, as shown in Figure 6B, the walking speed plan generation unit 117b calculates the effective speed using the angle between the direction of the unit vector r in the walking path and the direction of the movement vector Vped of pedestrian P indicated by the pedestrian observation information. For example, the walking speed plan generation unit 117b calculates the effective speed by calculating the dot product of the unit vector r and the movement vector Vped. The walking speed plan generation unit 117b also generates a walking speed plan for a predetermined time with the calculated effective speed as the maximum value.

[0038] Furthermore, as shown in Figure 7, the walking speed plan generation unit 117b sets the speed in the walking speed plan lower than usual, taking into consideration that pedestrians will decelerate in sections TRa on the walking path TR where the curvature is greater than a predetermined reference value. The predetermined reference value is set, for example, to a curvature of 30 degrees or more. In addition, the walking speed plan generation unit 117b sets an elliptical risk area RA around the pedestrian P. The risk area is the area of ​​risk that is potentially distributed or present around the pedestrian. Risk refers to the risk that the pedestrian poses to the vehicle 1, for example, the risk that the sudden movement of the pedestrian will force the vehicle 1 to brake or steer suddenly. Risk may also refer to the risk that the vehicle 1 poses to the pedestrian. A quantitative index value representing the level of such risk can be called the risk potential, and the area in which this risk potential is greater than or equal to a predetermined threshold can be set as the risk area. Furthermore, the derivation of the risk potential and risk area can be done using well-known techniques.

[0039] Furthermore, the walking speed plan generation unit 117b generates a walking speed plan based on different assumptions depending on the surrounding environment and the pedestrian's behavior. For example, if the pedestrian signal is green, the walking speed plan generation unit 117b generates a walking speed plan based on the assumption that the pedestrian will cross the crosswalk at their current walking speed or accelerate (progress assumption). On the other hand, if the pedestrian signal is red, the walking speed plan is generated based on the assumption that the pedestrian will stop before the crosswalk (stop assumption). If the pedestrian does not stop before the crosswalk despite the pedestrian signal being red, the unit predicts that the pedestrian will forcefully cross the crosswalk and generates a walking speed plan based on the aforementioned progress assumption.

[0040] The future position calculation unit 117c calculates the pedestrian's future position based on the effective speed generated by the walking speed plan generation unit, the amount of movement per predetermined time based on the walking speed plan, and the amount of movement per unit time at the effective speed along the walking path. The future position calculation unit 117c calculates the pedestrian's future position using, for example, a well-known discrete solution method such as the Euler method or the Runge-Kutta method. As shown in Figure 6C, the future position calculation unit 117c determines the future position of pedestrian P at times T0 to Tn (where n is a natural number) along the movement path. The future position calculation unit 117c outputs the calculated future position information of pedestrian P to the running mode determination unit 118.

[0041] The driving mode determination unit 118 determines the driving mode that vehicle 1 should take at a crosswalk based on the pedestrian's future position. For example, if the pedestrian's future position is before the crosswalk, vehicle 1 will pass through the crosswalk; if the pedestrian's future position is entering the crosswalk, vehicle 1 will stop before the crosswalk.

[0042] The vehicle control unit 119 generates a vehicle driving control command to execute the driving mode determined by the driving mode determination unit 118 when passing through a pedestrian crossing, and transmits it to the driving mechanism 400 of the vehicle 1.

[0043] Next, the operation of the vehicle travel control device 100 having the above configuration will be described. The flowchart shown in FIG. 8 is a flowchart regarding the vehicle travel control process executed by the vehicle travel control device 100, which is an example of the operation of the vehicle travel control device 100. This operation corresponds to the vehicle travel control method of the vehicle travel control device 100 according to the present embodiment.

[0044] The control unit 110 of the vehicle travel control device 100 starts the vehicle travel control process, for example, in response to an operation input of a start instruction by an operator (such as the driver of the vehicle 1) of the vehicle travel control device 100.

[0045] When starting the vehicle travel control process, the control unit 110 first acquires road environment information (step S101). The road environment acquisition unit 111 of the control unit 110 acquires the road environment information around the vehicle 1 based on the map information stored in the storage unit 120, the captured image by the imaging device provided in the vehicle 1, the traffic situation acquired from the outside via the communication unit 130, and the like.

[0046] Next, the control unit 110 generates a travel route plan (step S1). The travel route planning unit 112 of the control unit 110 plans a travel route from the current location of the vehicle 1 to the destination according to the position information (current location) by the GPS device provided in the vehicle 1 and the destination acquired by the input / output unit 140. The travel route planning unit 112 explores an appropriate travel route in consideration of the road environment information acquired by the road environment acquisition unit 111 and generates a travel route plan.

[0047] Next, the control unit 110 determines whether there is a crosswalk on the travel route (step S103). The crosswalk information acquisition unit 113 of the control unit 110 determines whether there is a crosswalk within a predetermined distance (such as a distance where the vehicle can decelerate and stop) in front of the vehicle 1 traveling on the travel route planned by the travel route planning unit 112 based on the image analysis result of the captured image by the imaging device provided in the vehicle 1. <When it is determined that there is a crosswalk on the travel route (step S103: YES), the control unit 110 sets a boundary line that divides the roadway and the pedestrian space. For example, the boundary line setting unit 114 of the control unit 110 sets a boundary line that divides the roadway on which the vehicle 1 is traveling and the pedestrian space where pedestrians can walk by the above-described method.

[0049] Subsequently, the control unit 110 generates a fluid movement path (step S104). The fluid movement path generation unit 115 of the control unit 110 generates a movement path of a liquid virtually filled in the pedestrian space.

[0050] Next, the control unit 110 determines whether there are pedestrians around the crosswalk (step S105). The pedestrian detection unit 116 of the control unit 110 determines whether there are pedestrians around the crosswalk based on the image analysis result of the captured image of the imaging device provided in the vehicle 1.

[0051] When it is determined that there are pedestrians around the crosswalk (step S106: YES), the control unit 110 executes pedestrian future position prediction processing (step S107). When it is determined that there are pedestrians around the crosswalk, the pedestrian detection unit 116 generates pedestrian observation information based on the image analysis result of the captured image of the pedestrians and outputs it to the pedestrian future position prediction unit 117 that executes the pedestrian future position prediction processing. The pedestrian observation information includes, for example, the position, walking speed, walking direction, etc. of the pedestrians.

[0052] Here, referring to FIG. 9, the pedestrian future position prediction processing (step S107), which is a subroutine of the vehicle travel control processing, will be described.

[0053] When the pedestrian future position prediction process is started, the control unit 110 predicts the pedestrian's walking path (step S201). The walking path prediction unit 117a of the control unit 110 predicts the pedestrian's walking path based on pedestrian observation information obtained from the pedestrian detection unit 116 and the fluid movement path generated by the fluid movement path generation unit 115. More specifically, the walking path prediction unit 117a can uniquely predict the pedestrian's walking path assuming that the pedestrian will always enter the crosswalk by continuously connecting the unit vectors on the fluid movement path at the pedestrian's current position indicated by the pedestrian observation information.

[0054] Next, the control unit 110 generates a walking speed plan for the pedestrian (step S202). The walking speed plan generation unit 117b of the control unit 110 obtains the effective speed based on the angle between the unit vector in the pedestrian's walking path predicted by the walking path prediction unit 117a and the pedestrian's walking vector indicated by the pedestrian observation information. The walking speed plan generation unit 117b obtains the speed obtained from the dot product of the unit vector and the walking vector as the effective speed, for example. Next, the walking speed plan generation unit 117b generates a speed plan for a predetermined time period with the effective speed as the maximum value.

[0055] When generating a speed plan, the pedestrian speed plan generation unit 117b sets the speed in the pedestrian speed plan lower than usual, taking into account that pedestrians will slow down in sections of the walking path with a large curvature. Furthermore, the pedestrian speed plan generation unit 117b generates different pedestrian speed plans based on different assumptions about the actions that pedestrians may take depending on the color of the pedestrian signal. Specifically, when the pedestrian signal is green, it is considered that pedestrians are more likely to cross the crosswalk. Therefore, the pedestrian speed plan generation unit 117b generates a pedestrian speed plan assuming that pedestrians will proceed to the crosswalk. The pedestrian speed plan assuming that pedestrians will proceed is a normal pedestrian speed plan with the effective speed as the maximum value. On the other hand, when the pedestrian signal is red, it is considered that pedestrians are less likely to cross the crosswalk. Therefore, the pedestrian speed plan generation unit 117b generates a pedestrian speed plan assuming that pedestrians will stop before the crosswalk. The pedestrian speed plan assuming that pedestrians will stop is a pedestrian speed plan in which pedestrians decelerate at a constant rate. Furthermore, for example, if the pedestrian signal has just turned from flashing green to red (and the road signal is also red, etc.), a pedestrian speed plan may be provided that assumes the pedestrian may accelerate. The following describes the process of calculating the pedestrian's future position using the pedestrian speed plan assuming forward movement and the pedestrian speed plan assuming stopping, which are generated by the pedestrian speed plan generation unit 117b.

[0056] The control unit 110 determines whether the pedestrian signal is red or not (step S203). The pedestrian future position prediction unit 117 of the control unit 110 determines whether the pedestrian signal installed at the crosswalk is red (red light on) or not, based on the image analysis results of the image captured by the imaging device installed on the vehicle 1. The pedestrian signal is assumed to be green (blue light on, blue flashing) or red (red light on). If there is no pedestrian signal installed at the crosswalk, the pedestrian future position prediction unit 117 determines that the pedestrian signal is green.

[0057] If the pedestrian signal is not red, that is, if the pedestrian signal is green (step S203: NO), the control unit 110 calculates the pedestrian's future position using a walking speed plan assuming progress (step S204). In this case, the walking speed plan assuming progress is a normal walking speed plan with the effective speed as the maximum value. The future position calculation unit 117c of the control unit 110 calculates the pedestrian's future position using the walking speed plan assuming progress, i.e., a normal walking speed plan.

[0058] Then, the control unit 110 adopts the future position calculated using the assumed walking speed plan as the pedestrian's future position (step S205). The future position calculation unit 117c generates pedestrian future position information indicating the pedestrian's future position calculated using the assumed walking speed plan and outputs it to the running mode determination unit 118.

[0059] On the other hand, if in step S203 the control unit 110 determines that the pedestrian signal is red (step S203: YES), the control unit 110 calculates the pedestrian's future position using a walking speed plan assuming a stop (step S206). In this case, the walking speed plan assuming a stop is a walking speed plan in which the pedestrian decelerates at a constant rate. The future position calculation unit 117c calculates the pedestrian's future position using the walking speed plan assuming a stop.

[0060] Next, the control unit 110 determines whether or not the vehicle was able to stop before the pedestrian crossing (step S207). The future position calculation unit 117c determines whether or not the pedestrian's future position, calculated using a walking speed plan assuming a stop, is able to stop before the pedestrian crossing. If it is determined that the vehicle was able to stop before the pedestrian crossing (step S207: YES), the control unit 110 adopts the pedestrian's future position calculated using a walking speed plan assuming a stop as the pedestrian's future position (step S208). The future position calculation unit 117c generates pedestrian future position information indicating the pedestrian's future position calculated using a walking speed plan assuming a stop, and outputs it to the driving mode determination unit 118.

[0061] On the other hand, if the control unit 110 determines that the pedestrian has not stopped before the crosswalk, that is, that the pedestrian has stopped within the crosswalk (step S207: NO), the control unit 110 calculates the pedestrian's future position using the walking speed plan during the process of travel (step S204). If the pedestrian's future position calculated using the walking speed plan assuming a stop is within the crosswalk, it is considered that the pedestrian is likely to ignore the red light and cross the crosswalk, and therefore it is considered that calculating the pedestrian's future position using the walking speed plan during the process of travel, rather than assuming a stop, will yield a more accurate result. Accordingly, the future position calculation unit 117c calculates the pedestrian's future position using the walking speed plan during the process of travel.

[0062] Then, the control unit 110 adopts the future position calculated using the assumed walking speed plan as the pedestrian's future position (step S205). The future position calculation unit 117c generates pedestrian future position information indicating the pedestrian's future position calculated using the assumed walking speed plan and outputs it to the running mode determination unit 118.

[0063] Returning to Figure 8, after performing the pedestrian future position prediction process, the control unit 110 determines the driving mode of vehicle 1 on the crosswalk (step S108). The driving mode determination unit 118 of the control unit 110 determines the driving mode that vehicle 1 should take on the crosswalk based on the pedestrian's future position.

[0064] Next, the control unit 110 generates a vehicle driving control command and transmits it to the vehicle's driving mechanism 400 (step S109). The vehicle control unit 119 of the control unit 110 generates a vehicle driving control command to execute the driving mode when passing through the pedestrian crossing determined by the driving mode determination unit 118 and transmits it to the vehicle's driving mechanism 400.

[0065] After executing the process in step S109, the control unit 110 returns to step S101 and repeats the series of processes from steps S101 to S109 described above until, for example, a stop command is input by the operator of the vehicle travel control device 100.

[0066] As described above, the vehicle driving control device 100 according to this embodiment sets a boundary line separating the roadway and the pedestrian space based on the roadway shape and the pedestrian crossing shape when the vehicle 1 passes over a pedestrian crossing, and generates a movement path for a fluid virtually filled within the pedestrian space. Furthermore, based on the fluid movement path, the vehicle driving control device 100 predicts the walking path of a pedestrian assuming that the pedestrian will always enter the pedestrian crossing, and predicts the pedestrian's future position according to the pedestrian's walking path. Then, the vehicle driving control device 100 controls the driving of the vehicle 1 based on the pedestrian's future position. In this way, the vehicle driving control device 100 according to this embodiment can appropriately support collision avoidance between the vehicle 1 passing over the pedestrian crossing and the pedestrian crossing the pedestrian crossing by predicting the walking path and future position of a pedestrian walking near the pedestrian crossing.

[0067] Furthermore, the vehicle driving control device 100 according to this embodiment divides the outline of the roadway shape and the outline of the pedestrian crossing shape at the intersection point where they meet, and connects the divided line segments to set a boundary line. In this way, the vehicle driving control device 100 according to this embodiment can set a boundary line that separates the roadway and pedestrian space in a simple and versatile manner by combining map shapes. In addition, the vehicle driving control device 100 according to this embodiment shapes the connecting portion of the boundary line into a rounded and smooth shape. As a result, the vehicle driving control device 100 can assume a pedestrian path in which pedestrians smoothly enter from the sidewalk towards the pedestrian crossing in a curved manner.

[0068] Furthermore, the vehicle travel control device 100 according to this embodiment observes the position, walking speed, and walking direction of pedestrians walking near a crosswalk, and obtains the pedestrian's walking path by continuously connecting unit vectors on the fluid movement path at the pedestrian's position. This uniquely determines the walking path based on the actually observed pedestrian's position.

[0069] Furthermore, the vehicle driving control device 100 according to this embodiment calculates an effective speed based on the pedestrian's walking speed and direction, and a unit vector on the walking path, and generates a speed plan for a predetermined time period with the calculated effective speed as the maximum value. The vehicle driving control device 100 also predicts the pedestrian's future position along the walking path from the amount of movement per predetermined time period. In this way, by generating a speed plan that is not limited to constant velocity motion, the vehicle driving control device 100 can accurately predict the pedestrian's future position by taking into account changes in the pedestrian's future walking speed.

[0070] Furthermore, the vehicle driving control device 100 according to this embodiment sets the effective speed to a speed lower than the pedestrian's walking speed, the larger the angle between the unit vector on the fluid movement path at the pedestrian's position and the pedestrian's movement vector. In this way, the vehicle driving control device 100 compares the direction vector of the fluid movement path with the pedestrian's movement vector and can take into account that the closer the directions of the two vectors are to parallel, the higher the probability that the pedestrian will cross the crosswalk, and the more different the directions of the two vectors are, the lower the probability that the pedestrian will cross the crosswalk. Moreover, in sections where the curvature of the walking path is greater than a predetermined reference value, the vehicle driving control device 100 according to this embodiment sets the speed in the speed plan lower than usual, taking into account that the pedestrian will slow down. In this way, the vehicle driving control device 100 can accurately predict the future position of the pedestrian by changing the speed plan according to the curvature.

[0071] Furthermore, the vehicle driving control device 100 according to this embodiment detects the color of the pedestrian signal at a crosswalk, and when a pedestrian approaches a crosswalk where the pedestrian signal is red, it sets a speed plan assuming that the pedestrian will stop in the future (stop assumption), predicts the pedestrian's future position based on this speed plan, and if the predicted future position is within the crosswalk, it predicts the pedestrian's future position based on a speed plan assuming that the pedestrian will proceed (proceed assumption). In this way, by setting a speed plan that slows down in anticipation of the pedestrian stopping in the future when the pedestrian signal is red, it is possible to ensure that the vehicle 1 passes through the crosswalk smoothly. On the other hand, since there are also pedestrians who cross the crosswalk even when the light is red without stopping, when it is detected that the pedestrian will not stop before the crosswalk, the use of the stop assumption is discontinued, and the future position of the pedestrian is predicted based on a speed plan assuming proceeding, thereby appropriately assisting in avoiding a collision between the vehicle 1 and the pedestrian.

[0072] It should be noted that the present invention is not limited to the embodiments described above, and various modifications and applications are possible without departing from the spirit of the invention.

[0073] (Modification) The vehicle driving control device 100 may, for example, modify and set the risk area around the pedestrian. Normally, the walking speed plan generation unit 117b sets an elliptical risk area RA1 along the direction of travel of the pedestrian P, for example, as shown in Figure 10A. However, the shape of the risk area is not limited to this, and it may be set to a modified shape, such as curving to match the shape of the walking path TR of the pedestrian P, as shown in the risk area RA2 in Figure 10B, to suit the walking range of the pedestrian P. In this way, by setting a risk area along the predicted walking path of the pedestrian, the risk area can be set in a shape that takes into account the possibility of entering the crosswalk in the future, even for pedestrians whose direction of travel is not facing the crosswalk. The vehicle driving control device 100 should control the driving of the vehicle 1 based on the set risk area.

[0074] The vehicle driving control device 100 may, for example, pre-simulate a virtual fluid movement path for each pedestrian crossing and use the corresponding simulation results to control the vehicle's movement when the vehicle 1 passes over a pedestrian crossing. Although fluid simulation is computationally intensive, it is universal as long as the map shape does not change, so if the simulation is completed in advance and saved in a readable data format, the online processing load can be significantly reduced.

[0075] The vehicle driving control device 100 may, for example, set a boundary line separating the roadway and the pedestrian space, taking into consideration obstacles present on the pedestrian crossing (for example, vehicles stopped on the pedestrian crossing because they cannot pass through an intersection, etc.) and vehicles parked on the shoulder of the road. This makes it possible to predict an appropriate pedestrian walking path according to the road conditions.

[0076] In the above embodiment, a vehicle driving control method for assisting in avoiding collisions with pedestrians when vehicle 1 passes through a crosswalk was described. However, the application of the vehicle driving control method according to this embodiment is not limited to this, and the vehicle driving control method may also be applied, for example, when vehicle 1 crosses a sidewalk when entering a facility adjacent to the road (store, parking lot, etc.).

[0077] In the above embodiment, the vehicle driving control device 100 was described as a device mounted on the vehicle 1. However, the functions of the vehicle driving control device 100 may be configured to be implemented on the vehicle 1.

[0078] In the above embodiment, for example, the control program executed by the processor 101 of the vehicle driving control device 100 was mainly stored in storage 103 beforehand. However, the present invention is not limited thereto, and the control program for executing the above-mentioned various processes may be implemented in an existing general-purpose computer, framework, workstation, etc., thereby enabling it to function as a device equivalent to the vehicle driving control device 100 according to the above embodiment.

[0079] The method of providing such programs is optional. For example, they may be distributed by storing them on a computer-readable storage medium (flexible disk, CD (Compact Disc)-ROM, DVD (Digital Versatile Disc)-ROM), or they may be stored on network storage such as the Internet and provided for download.

[0080] Furthermore, when the above processing is performed by a division of labor between the OS (Operating System) and the application program, or by collaboration between the OS and the application program, only the application program may be stored on a recording medium or storage. It is also possible to superimpose the program onto a carrier wave and distribute it over a network. For example, the above program may be posted on a bulletin board system (BBS) on a network and distributed over the network. The program may then be designed to execute the above processing by launching it and running it under the control of the OS, just like any other application program.

[0081] This invention allows for various embodiments and modifications without departing from the broad spirit and scope of the invention. Furthermore, the embodiments described above are for illustrative purposes only and do not limit the scope of the invention. In other words, the scope of the invention is indicated by the claims, not by the embodiments. Various modifications made within the scope of the claims and the equivalent meaning of the invention are considered to be within the scope of the invention.

[0082] 1...Vehicle, 100...Vehicle driving control device, 101...Processor, 102...Memory, 103...Storage, 104...Communication interface, 105...Input / output interface, 106...Bus line, 110...Control unit, 111...Road environment acquisition unit, 112...Driving route planning unit, 113...Pedestrian crossing information acquisition unit, 114...Boundary line setting unit, 115...Fluid movement path generation unit, 116...Pedestrian detection unit, 117...Pedestrian future position prediction unit, 117a...Walking path prediction unit, 117b...Walking speed plan generation unit, 117c...Future position calculation unit, 1 18... Driving mode determination unit, 119... Vehicle control unit, 120... Memory unit, 130... Communication unit, 140... Input / output unit, 200... Display device, 300... Sensor unit, 400... Driving mechanism, BL... Lower boundary line, BU... Upper boundary line, LC1, LC2... Linear charge, ORL, ORR... Outline of road shape, OPC... Outline of pedestrian crossing shape, P... Pedestrian, PI1, PI2, PI3, PI4... Intersection, r... Unit vector, RA, RA1, RA2... Risk area, T0 to Tn... Time, TR... Walking path, TRa... Section, Vped... Movement vector.

Claims

1. A vehicle driving control method executed by a processor in a vehicle driving control device mounted on a vehicle, wherein the processor sets a boundary line that separates the roadway and the pedestrian space, which is a space that combines the sidewalk and the pedestrian crossing, based on the shape of the roadway and the shape of the pedestrian crossing when the vehicle passes over a pedestrian crossing; generates a movement path for a fluid virtually filled in the pedestrian space; predicts the walking path of pedestrians assuming that all pedestrians on the sidewalk heading towards the pedestrian crossing enter the pedestrian crossing based on the movement path of the fluid; predicts the future position of the pedestrians according to the walking path of the pedestrians; and controls the driving of the vehicle based on the future position of the pedestrians.

2. The vehicle driving control method according to claim 1, wherein the processor divides the outline of the roadway shape and the outline of the pedestrian crossing shape at the intersection where the outline of the roadway shape and the outline of the pedestrian crossing shape intersect, and connects the divided line segments to set the boundary line.

3. The vehicle driving control method according to claim 2, wherein the processor shapes the connecting portion of the boundary line into a rounded, smooth shape.

4. The vehicle driving control method according to any one of claims 1 to 3, wherein the processor detects a pedestrian walking near the crosswalk, observes the pedestrian's position, walking speed, and walking direction, and obtains the walking path by continuously connecting unit vectors on the fluid movement path at the pedestrian's position.

5. The vehicle driving control method according to claim 4, wherein the processor calculates an effective speed based on the walking speed and direction of the pedestrian and a unit vector on the walking path, generates a speed plan for a predetermined time period with the calculated effective speed as the maximum value, and predicts the future position of the pedestrian along the walking path from the amount of movement per predetermined time period.

6. The vehicle driving control method according to claim 5, wherein the processor sets the effective speed to a speed lower than the walking speed of the pedestrian, the larger the angle between the unit vector on the fluid movement path at the pedestrian's position and the pedestrian's movement vector.

7. The vehicle driving control method according to claim 5 or 6, wherein the processor sets the speed of the speed plan lower than normal in sections where the curvature of the walking path is greater than a predetermined reference value, taking into consideration that the pedestrian will decelerate.

8. The vehicle driving control method according to any one of claims 5 to 7, wherein the processor detects the color of the pedestrian signal light of the crosswalk, and when the pedestrian approaches the crosswalk where the pedestrian signal light is red, sets a speed plan assuming that the pedestrian will stop in the future, predicts the future position based on this speed plan, and if the predicted future position is within the crosswalk, predicts the future position based on a speed plan assuming that the pedestrian will proceed.

9. The vehicle driving control method according to any one of claims 1 to 8, wherein the processor sets a risk area along the walking path.

10. The vehicle driving control method according to any one of claims 1 to 9, wherein the processor pre-simulates the fluid movement path for each pedestrian crossing and controls the vehicle's driving using the corresponding simulation result when the vehicle passes over the pedestrian crossing.

11. A vehicle driving control device installed in a vehicle, comprising a processor that, when the vehicle passes a crosswalk, sets a boundary line separating the roadway from the pedestrian space, which is a space connecting the sidewalk and the crosswalk, based on the shape of the roadway and the crosswalk; generates a movement path for a fluid virtually filled in the pedestrian space; predicts the walking path of pedestrians assuming that all pedestrians on the sidewalk heading towards the crosswalk enter the crosswalk based on the movement path of the fluid; predicts the future position of the pedestrians according to their walking path; and controls the driving of the vehicle based on the future position of the pedestrians.

Images