Parking assistance device and parking assistance method

Abstract

The present invention relates to a parking assistance device and a parking assistance method. In a path generation unit (52), based on the recognition of three-dimensional objects in a recognition processing unit (51), area determination is performed on an area around a vehicle including a parking path, and this area determination determines which of an area with obstacles when the vehicle moves, an area without obstacles, and an unclear area where the presence or absence of obstacles is unclear. Further, in the path generation unit, based on the result of the area determination, the target vehicle speed is set such that the target vehicle speed in the unclear area is lower than the target vehicle speed in the area without obstacles.

Description

[0001] Cross - reference to related applications

[0002] This application is based on Japanese Patent Application No. 2020 - 030738 filed on February 26, 2020, the contents of which are incorporated herein by reference. Technical field

[0003] The present disclosure relates to a parking assistance device and a parking assistance method that generate a path to a parking reservation position during automatic parking and move the vehicle to the parking reservation position following the path. Background art

[0004] In an automatic parking system, parking assistance is performed. That is, based on the positional relationship between the current position of the vehicle and the parking reservation position, a movement path from the current position to the parking reservation position (hereinafter referred to as a parking path) is generated, and the vehicle is moved along the parking path to perform automatic parking. At this time, when moving the vehicle along the parking path, the presence or absence of obstacles in the vehicle periphery is sensed by a peripheral monitoring sensor, and the target vehicle speed is set based on the result of the presence or absence. For example, in Patent Document 1, when the distance to an obstacle is short, the upper limit vehicle speed is reduced, thereby reducing the vehicle speed to suppress an inadvertent approach to the obstacle.

[0005] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2006 - 335239

[0006] There is an event where, in the case of actually performing automatic parking following a parking path, due to the performance of the peripheral monitoring sensor, there is an area where the presence or absence of an obstacle is unknown, and when there is an obstacle in this area, the obstacle is not detected when approaching it. In this case, it becomes a state where an obstacle that could not be detected in the positional relationship with the vehicle at the start of parking assistance is first detected by approaching based on parking assistance.

[0007] To avoid a collision between the vehicle and the obstacle caused thereby, it is necessary to take measures such as reducing the moving speed of the vehicle during parking assistance or increasing the deceleration based on the brake. However, in the case of taking the former measure, the time required for parking assistance becomes longer, and in the case of taking the latter measure, it becomes an emergency brake, resulting in deterioration of riding comfort. Summary of the invention

[0008] An object of the present disclosure is to provide a parking assistance device and a parking assistance method that can suppress an increase in the time required for parking assistance and can suppress deterioration of riding comfort caused by emergency braking.

[0009] In one aspect of the present disclosure, there is provided a parking assistance device that generates a parking path when the vehicle moves from the current position to a parking target position for parking and moves the vehicle along the parking path to the parking target position. The parking assistance device includes: an identification processing unit that performs: solid object identification for identifying solid objects in the space around the vehicle, and free space identification for identifying a free space for parking the vehicle from a parking lot based on the result of the solid object identification; a path generation unit that sets the free space identified by the free space identification as the parking target position, generates a parking path when the vehicle moves from the current position to the parking target position for parking, and generates a target vehicle speed when the vehicle moves along the parking path; and a path following control unit that performs path following control in which the vehicle is moved to the parking target position for parking while following the parking path and the target vehicle speed generated by the path generation unit. Further, the path generation unit determines, based on the solid object identification in the identification processing unit, which of the following regions exists in the area around the vehicle including the parking path: a region with an obstacle when the vehicle moves (a "with-region"), a region without an obstacle (a "without-region"), and a region with an unknown presence or absence of an obstacle (an "unknown-region"). Based on the result of the region determination, the target vehicle speed is set such that the target vehicle speed in the unknown region is lower than the target vehicle speed in the without-region.

[0010] In this way, it is determined which of the three states, i.e., "with-region", "without-region", and "unknown-region", the area of the host vehicle including the parking path belongs to. Moreover, the three states are clearly separated and used, and the target vehicle speed corresponding to the attribute of the determined region is set. The target vehicle speed in the "unknown-region" is lower than that in the "without-region". In this way, for the "unknown-region", since the presence or absence of an obstacle is unknown, the target vehicle speed is set lower than that in the "without-region". Therefore, even when an obstacle is suddenly detected while the host vehicle moves into the "unknown-region", since the vehicle speed has already been reduced, the vehicle can stop at a location where it does not contact the obstacle without performing emergency braking, and deterioration of riding comfort caused by emergency braking can be suppressed.

[0011] Conversely, for the "without-region", since the target vehicle speed is set higher than that in the "unknown-region", the vehicle speed can be increased as much as possible in the case where it is determined that there is no obstacle, and an increase in the time required for parking assistance can be suppressed.

[0012] Therefore, it is possible to provide a parking assistance device that can suppress an increase in the time required for parking assistance and deterioration of riding comfort caused by emergency braking.

[0013] In another aspect of the present disclosure, there is a parking assistance method of generating a parking path when moving a vehicle from a current position to a parking reservation position and moving the vehicle along the parking path to the parking reservation position, including: performing a solid object recognition for recognizing solid objects in the space around the vehicle, and a free space recognition for recognizing a free space for parking the vehicle from a parking lot based on the result of the solid object recognition; setting the free space recognized by the free space recognition as the parking reservation position, generating a parking path when moving the vehicle from the current position to the parking reservation position, and generating a target vehicle speed when moving the vehicle along the parking path; and performing path following control, in which the vehicle is moved to the parking reservation position for parking while following the generated parking path and the target vehicle speed. Moreover, when generating the parking path and the target vehicle speed, based on the solid object recognition, it is determined for the area around the vehicle including the parking path which area among an area with obstacles when the vehicle moves, an area without obstacles, and an area with unknown presence or absence of obstacles is present, and based on the result of the area determination, the target vehicle speed is set such that the target vehicle speed in the unknown area is lower than the target vehicle speed in the area without obstacles.

[0014] Thus, for the parking assistance method, area determination is performed in the same manner as in one aspect of the present disclosure described above, and based on the result of the area determination, the target vehicle speed is set such that the target vehicle speed in the unknown area is lower than the target vehicle speed in the area without obstacles. Thereby, the same effect as in one aspect of the present disclosure described above can be obtained.

[0015] In addition, the reference numerals in parentheses attached to each component, etc. indicate an example of the correspondence relationship between the component, etc. and the specific components, etc. described in the embodiments described later. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a diagram showing the block structure of the automatic parking system according to the first embodiment.

[0017] Figure 2 is a diagram showing an example of a parking path in a parking scenario.

[0018] Figure 3 is a diagram showing an example of a parking path and area determination.

[0019] Figure 4 is a diagram showing an example of setting the target vehicle speed.

[0020] Figure 5 is a flowchart of parking assistance control.

[0021] Figure 6This is a diagram showing an example of setting the target vehicle speed in the automatic parking system according to the second embodiment. Detailed Embodiment

[0022] Hereinafter, embodiments of the present invention will be described based on the accompanying drawings. In addition, in each of the following embodiments, the same or equivalent parts are denoted by the same reference numerals for explanation.

[0023] (First Embodiment)

[0024] Hereinafter, an automatic parking system to which the parking assistance device and the parking assistance method according to this embodiment are applied will be described. As Figure 1 shown, the automatic parking system 1 includes: a surrounding monitoring sensor 3, various actuators 4, and a parking assistance device 5. The parking assistance device 5 is directly or communicably connected to the surrounding monitoring sensor 3 and various actuators 4 via an in-vehicle LAN (Local Area Network), and the automatic parking system 1 performs automatic parking by controlling these components as parking assistance. In addition, there are various types of assistance such as assistance that displays a parking path for indication and assistance that broadcasts during parking, but here, the assistance for performing automatic parking is referred to as parking assistance.

[0025] The surrounding monitoring sensor 3 is an autonomous sensor that monitors the surrounding environment of the own vehicle (hereinafter referred to as the host vehicle). For example, the surrounding monitoring sensor 3 detects three-dimensional objects around the host vehicle, such as moving dynamic targets such as pedestrians and other vehicles and stationary static targets such as road structures, as detection target objects. Here, as the surrounding monitoring sensor 3, there are provided a surrounding monitoring camera 31 that captures an image of a predetermined range around the host vehicle, a sonar 32 that sends a probing wave to a predetermined range around the host vehicle, a millimeter-wave radar 33, a LIDAR (Light Detection and Ranging) 34, and other probing wave sensors. Each surrounding monitoring sensor 3, for example, detects three-dimensional objects in each control cycle determined for each when performing parking assistance.

[0026] The surrounding monitoring camera 31 corresponds to an imaging device, captures an image of the surrounding of the host vehicle, and outputs the captured data as sensing information to the parking assistance device 5. Here, as the surrounding monitoring camera 31, examples include a front camera 31a, a rear camera 31b, a left-side camera 31c, and a right-side camera 31d that capture images of the front, rear, left, and right sides of the vehicle, but are not limited thereto. Three-dimensional objects can be detected by analyzing the captured data of the surrounding monitoring camera 31. However, there are cases where, even when analyzing the captured data, the presence or absence of an obstacle composed of a three-dimensional object is unclear due to various factors. In this case, it becomes an area where the presence or absence of the obstacle is unclear (hereinafter referred to as an unclear area).

[0027] In addition, the so-called "three-dimensional object" refers to an object such as a three-dimensional structure, a person, or a bicycle that is spatially extended in three dimensions and is detected by the surrounding monitoring sensor 3. The so-called "obstacle" means an object among the "three-dimensional objects" that becomes an obstacle to the movement of the host vehicle during parking assist control. Even for a "three-dimensional object", an object such as a wall located at a position higher than the host vehicle or a step difference with a height that can be crossed and does not become an obstacle to the movement of the host vehicle may not be included in the "obstacle".

[0028] The detection wave sensor sequentially outputs the measurement results such as the relative speed, relative distance, and azimuth angle of the target obtained by outputting a detection wave and acquiring its reflected wave to the parking assist device 5 as sensing information. The sonar 32 performs measurements using ultrasonic waves as the detection wave and is arranged at multiple positions on the vehicle. For example, a plurality of them are arranged along the left and right directions of the vehicle on the front and rear bumpers, and measurements are performed by outputting a detection wave to the surrounding of the vehicle. The millimeter-wave radar 33 performs measurements using millimeter waves as the detection wave, and the LIDAR 34 performs measurements using laser as the detection wave. For example, both output a detection wave within a specified range in front of the vehicle and perform measurements within the output range. The detection range of these detection wave sensors for the detection target object is determined based on the sensor performance, and three-dimensional objects existing within this detection range can be detected by the detection wave sensors.

[0029] In addition, in the present embodiment, an example is given in which the surrounding monitoring camera 31, the sonar 32, the millimeter-wave radar 33, and the LIDAR 34 are provided as the surrounding monitoring sensor 3. However, as long as the surrounding monitoring is performed by a combination of one or more of them, it is not necessary to provide all of them. When only a part of the above-mentioned surrounding monitoring camera 31, sonar 32, millimeter-wave radar 33, and LIDAR 34 is provided as the surrounding monitoring sensor 3, the detection range is determined according to the provided surrounding monitoring sensor 3.

[0030] The parking assist device 5 is an ECU (electronic control unit) that functions as various control units for realizing the parking assist method in the automatic parking system 1, and is composed of a microcomputer including a CPU, a ROM, a RAM, an I / O, etc. In the present embodiment, the parking assist device 5 inputs sensing information that is a detection result of the surrounding monitoring sensor 3 during parking assistance, and performs various controls for parking assistance based on the sensing information. When the driver issues an instruction to perform parking assistance, such as when a parking assist switch (not shown) is pressed when performing parking assistance, parking assistance is performed. When the parking assist instruction is issued, the parking assist device 5 recognizes a free space where parking is possible based on the sensing information of the surrounding monitoring sensor 3, generates a parking path from the current position of the vehicle to the scheduled parking position during automatic parking, and performs a path following control according to the parking path. Specifically, the parking assist device 5 is configured to include a recognition processing unit 51, a path generation unit 52, and a path following control unit 53 as functional units that perform various controls.

[0031] The recognition processing unit 51 receives sensing information from the surrounding monitoring sensor 3 and recognizes the surrounding environment of the vehicle to be parked based on the sensing information, specifically, recognizes three-dimensional objects existing around the vehicle. Here, the recognition processing unit 51 is composed of an image recognition unit 51a, a space recognition unit 51b, and a free space recognition unit 51c.

[0032] The image recognition unit 51a includes a three-dimensional object recognition unit 51aa. The three-dimensional object recognition unit 51aa receives image data from the surrounding monitoring camera 31 as sensing information, and performs image analysis on the image data to recognize a three-dimensional object.

[0033] In three-dimensional object recognition, three-dimensional objects such as dynamic object marks and static object marks that exist around the vehicle are recognized as detection objects. Based on the shapes of obstacles in the three-dimensional objects that become detection objects recognized by the three-dimensional object recognition, preferably the shapes of static object marks, the path generation described below is performed, and the presence or absence of obstacles is determined.

[0034] The image data input from the surrounding monitoring camera 31 is data reflecting the appearance of the surroundings, so if the image is analyzed, it is possible to identify whether there is a three-dimensional object. In addition, it is possible to determine whether the three-dimensional object is a dynamic object or a static object based on the shape of the recognized three-dimensional object or the optical flow of the image, and it is possible to detect the position of the three-dimensional object, that is, the position, distance, and height of the three-dimensional object relative to the vehicle.

[0035] The space recognition unit 51b also includes a three-dimensional object recognition unit 51ba. The three-dimensional object recognition unit 51ba performs three-dimensional object recognition in the space around the host vehicle based on the sensing information from at least one of the sonar 32, millimeter-wave radar 33, and LIDAR 34. The three-dimensional object recognition here is the same as the three-dimensional object recognition performed by the image recognition unit 51a. Therefore, as long as either the image recognition unit 51a or the space recognition unit 51b is provided, three-dimensional recognition can be performed.

[0036] In addition, although three-dimensional object recognition can be performed by either the image recognition unit 51a or the space recognition unit 51b, better-precision three-dimensional object recognition can be performed by using both. For example, by using the three-dimensional object recognition performed by the space recognition unit 51b to supplement the three-dimensional object recognition performed by the image recognition unit 51a, three-dimensional object recognition can be performed with better precision.

[0037] The free space recognition unit 51c performs free space recognition to identify the places in the parking lot that become free spaces. The free space is the place in the parking lot where other vehicles are not parked, meaning the parking space with the area and shape where the host vehicle can park. It is not limited to the case where there are multiple parking spaces in the parking lot, and also includes the case where there is only one. The place recognized as this free space is set as the parking reservation position. The free space recognition unit 51c recognizes the free space in the parking lot based on the recognition results of the three-dimensional object recognition by the image recognition unit 51a and the space recognition unit 51b. For example, the free space recognition unit 51c can grasp the shape of the parking lot and the presence or absence of other vehicles parked based on the results of the three-dimensional object recognition, and thus can recognize the free space from the parking lot based on this.

[0038] The path generation unit 52 generates a path based on the results of the three-dimensional object recognition and the free space recognition, or performs area determination to determine the presence or absence of obstacles in the area around the host vehicle including the parking path, or generates a target vehicle speed corresponding to the parking path. Specifically, the path generation unit 52 is configured to include a path calculation unit 52a, an area determination unit 52b, and a target vehicle speed generation unit 52c.

[0039] The path calculation unit 52a calculates a moving path from the current position of the vehicle to the scheduled parking position identified by free space recognition while avoiding obstacles identified by three-dimensional object recognition, and generates a path represented by the calculation result as a parking path. When there are some constraints when generating the path, the path calculation unit 52a generates the parking path in a manner that satisfies the constraints. For example, the path calculation unit 52a generates a parking path with the minimum number of turns within a specified range. In addition, when there are constraints on the direction when parking, that is, the direction of approach to the scheduled parking position, the parking path is calculated by adding the constraints to the direction. For example, in the case of forward parking where the vehicle moves forward and stops at the scheduled parking position, or in the case of rearward parking where the vehicle moves backward and stops, the direction of the vehicle when parking is set as a constraint. Regarding the direction of the vehicle when parking, when the imaged data of the surrounding monitoring camera 31 contains a sign that writes information such as "park forward" or "park rearward", or contains a mark indicating the direction when parking, the information is included in the constraints. Furthermore, when there is a setting switch or the like for the user to set the direction of the vehicle when parking, the direction of the vehicle when parking can also be included in the constraint conditions according to the setting state of the setting switch.

[0040] This will be described using an example. In a normal parking scenario, an arbitrary parking space is selected as a planned parking position from a plurality of parking spaces in a flat parking lot based on a relationship with the current position of the vehicle. Figure 2 The figure shows a case where, when the parking assist instruction is given, the parking space in the left front relative to the current position Pa of the host vehicle V is selected as the planned parking position Pb. At the current position Pa, the host vehicle V is located diagonally in front of the planned parking position Pb with the moving direction facing the left. In this case, if the constraint condition of the direction of the host vehicle V during parking is "rearward parking", a parking path from the current position Pa to the planned parking position Pb is generated so that the moving direction to the planned parking position Pb is "rearward parking".

[0041] by Figure 2 For example, a parking path indicated by a dotted line in the figure is generated. Specifically, a path indicated by a dotted line in the figure is generated, that is, a path is generated in which the vehicle V first turns right and moves forward as indicated by arrow A1, then switches from the forward direction to the reverse direction, and turns left and moves backward as indicated by arrow A2. Thus, a parking path is generated for parking at the planned parking position Pb by "parking backward".

[0042] In addition, when generating a parking path, obstacles formed by the three-dimensional objects identified by three-dimensional object recognition are avoided, but only the static landmarks among them are avoided to generate the parking path. Since the dynamic landmarks move, the host vehicle V can be moved after the risk of collision with the dynamic landmarks disappears. In this case, it is sufficient to generate a parking path that only considers static landmarks.

[0043] The area determination unit 52b determines which area among the area where obstacles exist (hereinafter referred to as the "occupied area"), the area where there are no obstacles (hereinafter referred to as the "free area"), and the unknown area is the area around the host vehicle V including the parking path calculated by the path calculation unit 52a.

[0044] As described above, based on the sensing information of the surrounding monitoring sensor 3, it is possible to detect the presence or absence of three-dimensional objects, discriminate whether the three-dimensional object is a dynamic landmark or a static landmark, and determine the position, distance, and height of the three-dimensional object. Therefore, it is possible to perform area determination of the presence or absence of obstacles to discriminate which area among the occupied area, the free area, and the unknown area is the area around the host vehicle V based on the sensing information of the surrounding monitoring sensor 3.

[0045] For area determination, it is also possible to set all areas where three-dimensional objects exist as occupied areas, but it is also possible not to set an area where a three-dimensional object exists as an occupied area even if the three-dimensional object is not an obstacle. For example, when a three-dimensional object exists at a relatively high position or is a relatively low object that can be crossed by the tires, since it does not become an obstacle to the movement of the host vehicle V, it may not be discriminated as an occupied area in such a case. In addition, an area where no three-dimensional object exists is determined as a free area, and if it is neither an occupied area nor a free area, it is determined as an unknown area.

[0046] Here, the details of the unknown area are described. As described above, the unknown area is an area where the presence or absence of an obstacle as a three-dimensional object is unknown, and is discriminated as an unknown area when it is not discriminated as either an occupied area or a free area. The unknown area may be generated due to the following factors (1) to (4), for example.

[0047] (1) Outside the detection range of the surrounding monitoring sensor 3. For example, when the surrounding monitoring sensor 3 is composed of a probing wave sensor, the detection range of the detection target object is determined based on the sensor performance. Therefore, outside this detection range, three-dimensional objects cannot be detected, resulting in an unknown area.

[0048] (2) Within the detection range of the peripheral monitoring sensor 3, there is a region hidden behind a three-dimensional object due to the presence of the object in front. For example, when the peripheral monitoring sensor 3 is the sonar 32, if the detection range of the sonar 32 is a specified angular range of 4 m from the sonar 32, when there is a three-dimensional object at a position 2 m from the sonar 32, the position behind the object, that is, the position farther from the sonar 32, is hidden and cannot be detected. Therefore, even within the detection range, the hidden and undetected region becomes an unknown region.

[0049] (3) In the range where various interference factors occur and the detection of the peripheral monitoring sensor 3 cannot be performed. For example, when the peripheral monitoring sensor 3 is composed of a detection wave sensor, it sometimes interferes with other peripheral monitoring sensors 3 of the own vehicle V or sensors of other vehicles and cannot accurately perform three-dimensional object recognition. In addition, when the sensing unit of the peripheral monitoring sensor 3 is contaminated, if it is a detection wave sensor, sometimes the transmission and reception of the detection wave cannot be accurately performed, and if it is the peripheral monitoring camera 31, sometimes the shooting cannot be accurately performed. In addition, in the case of the peripheral monitoring camera 31, there is a difference in contrast between the sunlight area and the non-sunlight area, and it is difficult to measure the distance to the three-dimensional object only by the shooting data, etc., and three-dimensional object recognition cannot be accurately performed.

[0050] (4) It reaches the performance limit of the peripheral monitoring sensor 3. Due to the existence of the performance limit of the peripheral monitoring sensor 3, although it is possible to detect the presence of something, it is not possible to detect its position, etc., and only low-reliability three-dimensional object recognition can be performed. For example, in the case of the sonar 32, even if it is possible to confirm the presence of a three-dimensional object within the detection range, its height cannot be recognized, and it cannot be determined whether it is a three-dimensional object that can be crossed. In this way, in the case where a three-dimensional object that can be an obstacle is recognized even though the height, etc. cannot be recognized, although this area can also be determined as an area with an object, it can also be determined as an unknown area.

[0051] In this way, the area determination unit 52b determines which area among the area with an object, the area without an object, and the unknown area it is. According to each specified control cycle, for example, according to the sampling cycle of each peripheral monitoring sensor 3, for example, every 100 ms, the area determination by the area determination unit 52b is performed, and the result of the area determination can be sequentially updated as the own vehicle V moves. However, due to the above factors, sometimes the "unknown area" is maintained and not updated to the "area with an object" or the "area without an object", and the own vehicle V may also move to the "unknown area".

[0052] In addition, for the area where the area determination is performed, as long as it is the area around the own vehicle V that includes the parking path calculated by the path calculation unit 52a, specifically, the area that includes the moving range considering the vehicle width of the own vehicle V is sufficient. For example, as Figure 3As shown, in addition to the detection range Ra of the surrounding monitoring sensor 3 starting from the current position Pa of the own vehicle V, in this figure, the specified range Rb including the parking path indicated by the dashed line in the figure is also set as the area for area determination. In the example of this figure, within the detection range of the surrounding monitoring sensor 3 starting from the current position Pa of the own vehicle V, the front of the own vehicle V is determined as "having an area", and the other ranges are determined as "having no area". For the areas of the parking path outside the detection range, they are determined as "unknown areas".

[0053] The target vehicle speed generation unit 52c uses the determination result in the area determination unit 52b and sets the target vehicle speed at each position on the path when the own vehicle V moves along the parking path calculated by the path calculation unit 52a. Regarding the method of setting the target vehicle speed, various methods can be considered, but here, the upper limit control vehicle speed for each area is set, and the target vehicle speed is determined based on this upper limit control vehicle speed.

[0054] The so-called upper limit control vehicle speed is the upper limit value of the vehicle speed set for each area. The upper limit control vehicle speed is set for "having an area", "having no area", and "unknown area" respectively. However, for "having an area", the upper limit control vehicle speed may not be determined, and the vehicle speed may be adjusted according to the distance from the obstacle.

[0055] For example, the upper limit control vehicle speed for the "unknown area" is set lower than that for the "having no area". The upper limit control vehicle speed for the "having no area" is set to a certain speed so as to perform automatic parking based on parking assist control as early as possible. For the upper limit control vehicle speed in the "unknown area", since an obstacle may be suddenly detected, the upper limit control vehicle speed is set to a smaller value so as to be able to cope with such a state. When showing an example, the upper limit control vehicle speed for the "unknown area" is set to about 1 / 2 of the upper limit control vehicle speed for the "having no area". When the upper limit control vehicle speed for the "having no area" is set to 5 km / h, the upper limit control vehicle speed for the "unknown area" is set to about 2 km / h. In addition, when the upper limit control vehicle speed for the "having no area" is set to 2 - 3 km / h, the upper limit control vehicle speed for the "unknown area" is set to about 1 km / h.

[0056] For example, in the case of Figure 3 assuming the parking path, follow the parking path that moves from the "having no area" to the "unknown area". Therefore, for example, as Figure 4As shown, after setting the upper limit control vehicle speed for the "No Area", set the upper limit control vehicle speed for the "Unknown Area". Then, at the start of parking assistance, i.e., at the start of driving, set the target vehicle speed that gradually rises to the upper limit control vehicle speed of the "No Area". Next, at the boundary position between the "No Area" and the "Unknown Area", connect each consecutive upper limit control vehicle speed together so that the upper limit control vehicle speed smoothly changes from the "No Area" to the "Unknown Area". For example, decelerate from the upper limit control vehicle speed for the "Unknown Area" from a point at a specified distance in front of the boundary position between the "No Area" and the "Unknown Area" (hereinafter referred to as the preparatory deceleration start position). The preparatory deceleration start position can be set, for example, as a point 1 m in front of the boundary position. The deceleration start position can be set based on the upper limit control vehicle speed for the "No Area", the upper limit control vehicle speed for the "Unknown Area", and the deceleration (specified value) towards the "Unknown Area". Preferably, the deceleration towards the "Unknown Area" is set to a value that does not give discomfort to the user (e.g., 0.01G to 0.1G). Then, near the parking predetermined position Pb, gradually reduce the target vehicle speed from the upper limit control vehicle speed to 0. Set the target vehicle speed in this way. At this time, consider the ride comfort of the passengers, etc., and set the upper and lower limit accelerations to smoothly change the target vehicle speed so that the change in the target vehicle speed converges within the range of the upper and lower limit accelerations.

[0057] In addition, according to the environment around the vehicle V, even at the same vehicle speed, the sense of fear given to the passengers during automatic parking may change. For example, when there is a wall or the like nearby and when there is nothing, even at the same vehicle speed, the sense of fear given to the passengers may change. Therefore, the upper limit control vehicle speed for each area can also be changed according to the presence or absence of three-dimensional objects around the parking path. If there are no three-dimensional objects, it becomes a relatively high first upper limit value, and if there are three-dimensional objects, it becomes a second upper limit value lower than the first upper limit value. Taking the above example, when there are no three-dimensional objects around the parking path, the upper limit control vehicle speed for the "No Area" can be set to 5 km / h, and when there are three-dimensional objects around the parking path, the upper limit control vehicle speed for the "No Area" can be set to 2 - 3 km / h.

[0058] In addition, as described above, for area determination, since it is performed at each specified control cycle and the area determination result is updated sequentially, there may be a case where a part of the "Unknown Area" becomes not the "Unknown Area" as the vehicle V moves. In this case, the target vehicle speed is also updated according to the attribute of the updated area. For example, if it is the "No Area", the target vehicle speed is set based on the upper limit control vehicle speed for the "No Area".

[0059] The path following control unit 53 is a part that performs path following control by performing vehicle motion control such as acceleration / deceleration control and steering control of the host vehicle V. The path following control unit 53 outputs control signals to various actuators 4 so that the host vehicle V can follow the parking path generated by the path generation unit 52 and the target vehicle speed and move and stop at the parking position Pb. Here, it is assumed that the parking assist device 5 is composed of one ECU and the path following control unit 53 is provided in this ECU, but the parking assist device 5 may also be composed of a combination of multiple ECUs, and the path following control unit 53 may also be composed of these ECUs. As the multiple ECUs, for example, a steering ECU that performs steering control, a power unit control ECU that performs acceleration / deceleration control, and a brake ECU are listed.

[0060] Specifically, the path following control unit 53 acquires detection signals output from various sensors such as an accelerator position sensor, a brake pedal force sensor, a steering angle sensor, a wheel speed sensor, and a gear position sensor that are mounted on the host vehicle V but not shown. Moreover, the path following control unit 53 detects the states of each part from the acquired detection signals and outputs control signals to various actuators 4 so that the host vehicle V moves following the parking path and the target vehicle speed.

[0061] The various actuators 4 are various driving control devices related to the running and stopping of the host vehicle V, and include an electronically controlled throttle valve 41, a brake actuator 42, an EPS (Electric Power Steering) motor 43, a transmission 44, and the like. These various actuators 4 are controlled based on control signals from the path following control unit 53, thereby controlling the traveling direction, steering angle, and braking drive torque of the host vehicle V. Thereby, parking assist control including path following control for moving the host vehicle V along the parking path and the target vehicle speed and stopping at the parking position Pb is achieved.

[0062] In addition, when moving the host vehicle V from the current position Pa to the parking position Pb, it is only necessary to move the host vehicle V following this path, but it is also possible that people or other vehicles approach during the movement of the host vehicle V. In this case, the movement of the host vehicle V is stopped until the dynamic object appears outside the range of the movement predicted trajectory of the host vehicle V inferred from the parking path and the vehicle width, so that the host vehicle V does not collide with the dynamic object. In addition, there may be a case where there are static objects that cannot be recognized when initially calculating the parking path in the "unknown area". Therefore, even in the middle of the movement of the host vehicle V following the parking path, the stereo object recognition by the stereo object recognition units 51aa and 51ba is continued. Then, if a static object exists in a place where it may collide when the host vehicle V moves following the parking path, the parking path is regenerated.

[0063] The automatic parking system 1 according to the present embodiment is configured as described above. Next, with reference to Figure 5 the flowchart of the parking assistance control executed by the parking assistance device 5 shown in FIG., the operation of the thus configured automatic parking system 1 will be described. When the driver gives an instruction for parking assistance by pressing a parking assistance switch (not shown) or the like, the processing shown in this figure is executed at each prescribed control cycle. In addition, each process shown in this flowchart is realized by each functional unit of the parking assistance device 5. Further, each step of realizing this process is also grasped as each step of realizing the parking assistance method.

[0064] First, in step S100, the recognition process is started. The recognition process here means inputting the sensing information of the surrounding monitoring sensor 3 and performing the above-described three-dimensional object recognition and free space recognition based on the input sensing information. In addition, the process shown in step S100 is performed by the recognition processing unit 51 in the parking assistance device 5.

[0065] If the recognition process in step S100 is completed, the process proceeds to step S110 to generate a parking path. The generation of the parking path is performed by the method as described above. In addition, the process of step S110 is performed by the path generation unit 52 in the parking assistance device 5.

[0066] After that, the process proceeds to step S120 to determine whether the parking path generated in step S110 is a "no area". The parking path here does not mean the entire area of the parking path from the current position Pa to the parking destination position Pb, but means the range within a prescribed moving distance from the current position Pa in the parking path. That is, it is determined whether the area of the destination to which the own vehicle V in the parking path will move from now on is a "no area". Then, if an affirmative determination is made here, the process proceeds to step S130 to set the upper limit control vehicle speed for the "no area" and the process proceeds to step S140.

[0067] On the other hand, when a negative determination is made in step S120, the process proceeds to step S150 to determine whether the parking path generated in step S110 is an "unknown area". The parking path here also means the range within a prescribed moving distance from the current position Pa in the parking path in the same manner as in step S120. That is, it is determined whether the area of the destination to which the own vehicle V in the parking path will move from now on is an "unknown area". In other words, it is equivalent to the process of determining whether an "unknown area" is included within a prescribed distance from the current position on the parking path or whether the remaining distance to the "unknown area" is within the prescribed distance. Then, if an affirmative determination is made here, the process proceeds to step S160 to set the upper limit control vehicle speed for the "unknown area" and the process proceeds to step S140.

[0068] And when a negative determination is made in step S150, the process proceeds to step S170. In this case, since it means that the area where the host vehicle V in the parking path will move to from now on is an "occupied area", the upper limit control vehicle speed for the "occupied area" is set in step S170, and the process proceeds to step S140.

[0069] In step S140, the set continuous upper limit control vehicle speeds are connected. At this time, the upper and lower limit accelerations are set in such a way that the connection part of the upper limit control vehicle speeds with different values becomes smooth, so as to satisfy the upper and lower limit accelerations and perform the connection of the continuous upper limit control vehicle speeds. In addition, in the connection of the continuous upper limit control vehicle speeds, when the distance to the obstacle or the parking predetermined position Pb is within a specified distance, for example, 50 cm, the target vehicle speed is set to be able to stop at the position in front of the obstacle or the parking predetermined position Pb. In this way, the target vehicle speed as shown is set. Figure 4 The target vehicle speed as shown.

[0070] After that, the process proceeds to step S180, and the path following control process is executed. This process is performed by the path following control unit 53 in the parking assist device 5. Specifically, control signals are output to various actuators 4 so that the host vehicle V can move from the current position Pa and stop at the parking predetermined position Pb while following the parking path generated in step S110 and the target vehicle speed set in step S140. Thereby, various actuators 4 are driven, and the traveling direction, steering angle, and braking drive torque of the host vehicle V are controlled to make the host vehicle V follow the parking path and the target vehicle speed and move.

[0071] Then, the process proceeds to step S190 to determine whether the host vehicle V has reached the parking predetermined position Pb that is the target position. When the host vehicle V moves to the parking predetermined position Pb, a positive determination is made, and the process ends, and the host vehicle V stops at the parking predetermined position Pb.

[0072] As described above, in the automatic parking system 1 of the present embodiment, based on the sensing information of the surrounding monitoring sensor 3, it is discriminated which of the three states of "occupied area", "unoccupied area", and "unknown area" the area of the host vehicle V including the parking path is. Moreover, the three states are clearly separated and used, and the target vehicle speed corresponding to the attribute of the discriminated area is set. For example, the target vehicle speed is lower for the "unknown area" than for the "unoccupied area". In this way, for the "unknown area", since the situation of the presence of obstacles is unknown, the target vehicle speed is set lower than for the "unoccupied area". Therefore, even when the host vehicle V moves to the "unknown area" and an obstacle is suddenly detected, since the vehicle speed has already been reduced, it is possible to stop at a place where the obstacle is not contacted without performing emergency braking, and it is possible to suppress the deterioration of the riding comfort caused by emergency braking.

[0073] On the contrary, for the "no area", since the target vehicle speed is set higher compared to the "unknown area", it is possible to increase the vehicle speed in the case where it is determined that there is no obstacle as much as possible, and it is possible to suppress the lengthening of the time required for parking assistance.

[0074] Therefore, it is possible to provide a parking assistance device and a parking assistance method that can suppress the lengthening of the time required for parking assistance and can suppress the deterioration of riding comfort caused by emergency braking.

[0075] (Second Embodiment)

[0076] The second embodiment will be described. In this embodiment, the method of setting the target vehicle speed is changed compared to the first embodiment, and other than that, it is the same as the first embodiment. Therefore, only the parts different from the first embodiment will be described.

[0077] In this embodiment, the target vehicle speed generation unit 52c obtains the correlation between the moving distance and the curvature of the parking path, and determines the target vehicle speed based on this correlation and the upper limit control vehicle speed.

[0078] For example, in the case of assuming Figure 3 the parking path, try to assume the correlation between the moving distance and the curvature of the parking path as shown in the upper side figure of Figure 6 . In this example, starting from the current position Pa, the curvature gradually increases in section 1, becomes a constant curvature in section 2, gradually decreases in section 3, and becomes constant at a lower curvature in sections 4 and 5 to reach the parking position Pb. The sections 1 to 5 representing the correlation between the moving distance and the curvature are set as segments, and the target vehicle speed considering the upper limit control vehicle speed is set for each segment.

[0079] Regarding the upper limit control vehicle speed, it is the same as the first embodiment, but the upper limit control vehicle speed is adjusted based on the relationship between the moving distance and the curvature to set the target vehicle speed. Specifically, the target vehicle speed is set such that the higher the curvature, the lower the vehicle speed.

[0080] For example, with Figure 3For the parking path, for the segmentation of section 1, since it is a situation where the curvature is increasing and has not yet exceeded the specified value midway, the upper limit of "no area" is set to control the vehicle speed. Next, for the segmentation of section 2, the curvature becomes greater than the specified value, and a vehicle speed lower than the upper limit control vehicle speed of "no area" is set. Then, for the segmentation of section 3, the curvature becomes smaller than the specified value again midway during the decrease, and the upper limit of "no area" is set to control the vehicle speed. After that, for the segmentation of section 4, since it becomes constant in the state of decreasing curvature, the upper limit of "no area" is set to control the vehicle speed. For the segmentation of section 5, even though the curvature is the same as that of section 4 but it becomes an "unknown area", the upper limit of the "unknown area" is set to control the vehicle speed. Then, by smoothly connecting the upper limit control vehicle speeds set for each segmentation, etc., the target vehicle speed is set.

[0081] In this way, it is also possible to correct the upper limit control vehicle speed according to the curvature of the parking path to set the target vehicle speed, rather than setting the upper limit control vehicle speed as the target vehicle speed without change. In this way, the same effect as the first embodiment can also be obtained.

[0082] (Other Embodiments)

[0083] The present disclosure has been described based on the above embodiments, but is not limited to these embodiments, and also includes various modifications and modifications within an equivalent range. In addition, various combinations, methods, and further combinations and methods that include only one element, more or fewer other elements in these combinations and methods also fall within the scope and spirit of the present disclosure.

[0084] For example, for the method of setting the target vehicle speed described in the above embodiments, only one example is given, and other methods can also be set. For example, in the above embodiments, the upper limit control vehicle speed is set for each area, and the target vehicle speed is set based on this upper limit control vehicle speed. However, it is also possible to set the target vehicle speed according to the curvature of the parking trajectory without setting the upper limit control vehicle speed. In addition, the target vehicle speed can also be set in such a way that the longer the distance to the three-dimensional object existing at the position closest to the parking trajectory, the greater the vehicle speed.

[0085] In addition, the control unit and its method described in the present disclosure can also be implemented by a dedicated computer, which is provided by a processor and a memory configured to be programmed to execute one or more functions embodied by a computer program. Alternatively, the control unit and its method described in the present disclosure can also be implemented by a dedicated computer provided by a processor constituted by one or more dedicated hardware logic circuits. Alternatively, the control unit and its method described in the present disclosure can also be implemented by one or more dedicated computers constituted by a combination of a processor configured to be programmed to execute one or more functions and a memory and a processor constituted by one or more hardware logic circuits. Additionally, the computer program can also be stored as instructions executable by a computer in a computer-readable non-transitory tangible storage medium.

Claims

1. A parking assistance device that generates a parking path when the vehicle moves from the current position to a parking predetermined position for parking and causes the vehicle to move along the parking path to the parking predetermined position, comprising: An identification processing unit that performs solid object identification for identifying solid objects in the space around the vehicle and free space identification for identifying the free space in the parking lot for parking the vehicle; A path generation unit that sets the free space identified by the free space identification as the parking predetermined position, generates the parking path when the vehicle moves from the current position to the parking predetermined position for parking, and generates a target vehicle speed when the vehicle moves along the parking path; and A path following control unit that performs path following control, in which the vehicle is moved to the parking predetermined position for parking by following the parking path and the target vehicle speed generated by the path generation unit; Based on the solid object identification in the identification processing unit, the path generation unit determines which of the areas around the vehicle including the parking path are areas with obstacles when the vehicle moves, areas without obstacles, and areas with unknown presence or absence of obstacles, and based on the result of the area determination, sets the target vehicle speed in the unknown area to a value lower than the target vehicle speed in the area without obstacles.

2. The parking assistance device according to claim 1, wherein The path generation unit updates the result of the area determination by also performing the area determination when the vehicle moves along the parking path in the path following control, and updates the target vehicle speed based on the updated result of the area determination.

3. The parking assistance device according to claim 1 or 2, wherein The path generation unit respectively sets an upper limit control vehicle speed for the area with obstacles in the area with obstacles, sets an upper limit control vehicle speed for the area without obstacles in the area without obstacles, sets an upper limit control vehicle speed for the unknown area that is lower than the upper limit control vehicle speed for the area without obstacles in the unknown area, and sets the target vehicle speed by connecting the set continuous upper limit control vehicle speeds together so as to converge within a predetermined range of upper and lower limit accelerations.

4. The parking assistance device according to claim 1 or 2, wherein The path generation unit at least has an upper limit control vehicle speed for the area without obstacles and an upper limit control vehicle speed for the unknown area that is lower than the upper limit control vehicle speed for the area without obstacles, obtains the correlation between the moving distance and the curvature of the parking path, and adjusts the upper limit control vehicle speed for the area without obstacles and the upper limit control vehicle speed for the unknown area based on the curvature corresponding to the moving distance, and sets the target vehicle speed in such a way that the higher the curvature, the lower the vehicle speed.

5. The parking assistance device according to any one of claims 1 to 4, wherein The above-mentioned recognition processing unit inputs the detection results from the surrounding monitoring sensors, and recognizes the three-dimensional objects in the space around the vehicle based on the detection results of the surrounding monitoring sensors. The surrounding monitoring sensors monitor the surrounding environment of the vehicle. In the above-mentioned area determination, the path generation unit determines the area outside the detection range of the surrounding monitoring sensors as the above-mentioned unknown area.

6. The parking assistance device according to claim 5, wherein In the above-mentioned area determination, the path generation unit determines the area that is within the detection range of the surrounding monitoring sensors but hidden behind the three-dimensional object as the above-mentioned unknown area.

7. The parking assistance device according to claim 5 or 6, wherein In the above-mentioned area determination, the path generation unit determines the area within the range where the detection of the surrounding monitoring sensors cannot be performed due to interference factors as the above-mentioned unknown area.

8. The parking assistance device according to any one of claims 5 to 7, wherein In the above-mentioned area determination, the path generation unit determines the area where, due to the performance limit of the surrounding monitoring sensors, although the existence of the three-dimensional object is confirmed, the position where the three-dimensional object exists cannot be determined as the above-mentioned unknown area.

9. A parking assistance method is a parking assistance method for generating a parking path when moving a vehicle from a current position to a parking predetermined position and moving the vehicle along the parking path to the parking predetermined position, and includes: Performing three-dimensional object recognition for recognizing three-dimensional objects in the space around the vehicle, and free space recognition for recognizing the free space for parking the vehicle in a parking lot; Setting the free space recognized by the above-mentioned free space recognition as the above-mentioned parking predetermined position, generating the above-mentioned parking path when moving the vehicle from the above-mentioned current position to the above-mentioned parking predetermined position for parking, and generating a target vehicle speed when moving the vehicle along the parking path; and Performing path following control, in which the vehicle is moved to the above-mentioned parking predetermined position for parking by following the generated above-mentioned parking path and the above-mentioned target vehicle speed, When generating the above-mentioned parking path and generating the above-mentioned target vehicle speed, Based on the above-mentioned three-dimensional object recognition, a region determination is performed on the region around the vehicle including the above-mentioned parking path to determine which region among the region with obstacles when the vehicle moves, the region without obstacles, and the unknown region where the presence or absence of the obstacles is unknown, and based on the result of the region determination, the target vehicle speed of the above-mentioned unknown region is set to a value lower than the target vehicle speed of the above-mentioned region without obstacles.

Images