Vehicle control device

Abstract

Provided is a vehicle control device capable of performing collision avoidance control while ensuring a control remaining time required for avoidance control. The invention provides a component and a device which store identification information of a two-wheeled vehicle when the two-wheeled vehicle is exceeded, read identification information of the two-wheeled vehicle recognized in the past when the vehicle turns left at an intersection, determine the possibility of collision with the vehicle, and calculate predicted identification information. When an obstacle is detected in real time when the own vehicle turns left, avoidance control is performed by using predicted identification information calculated in advance to determine that the detected obstacle is a re-detection of a moving body identified in the past. As a result, it is possible to secure a control remaining time for collision avoidance for an obstacle coming from the rear of the own vehicle when turning left, which is difficult to avoid, and to prevent collision.

Description

Technical Field

[0001] The invention relates to a vehicle control device. Background Art

[0002] Conventionally, there have been proposed vehicle control devices and vehicle control methods that perform collision avoidance control to avoid a collision between an object in front of the own vehicle and the own vehicle.

[0003] As a previous vehicle control device, the following vehicle control device is known: the detection result of an object in front of the own vehicle is used to determine the possibility of the object colliding with the own vehicle, and when it is determined that the possibility of the object colliding with the own vehicle in front of the own vehicle is high, collision avoidance control is implemented to avoid the collision with the object.

[0004] In addition, when the own vehicle turns left or right at an intersection, the detection area changes with the direction of the own vehicle. When an object originally located in a blind spot enters the detection area, the object is close to the position of the own vehicle at the time when the object is detected, so the timing of the collision avoidance control will be delayed. To address this problem, the following technology is proposed: when it is determined that the own vehicle will turn right or left after it is determined that the own vehicle has overtaken the object, the timing of the collision avoidance control is advanced compared to a case where it is not determined that the own vehicle has overtaken the object and a case where it is determined that the own vehicle will not turn right or left after it is determined that the own vehicle has overtaken the object (refer to patent document 1). Prior art literature Patent Literature

[0005] Patent Document 1: Japanese Patent No. 6617696 Summary of the invention Problem that the invention aims to solve

[0006] In addition, in order to perform avoidance control, the vehicle usually uses external recognition sensors such as camera sensors to detect objects and track them for a certain period of time to determine whether the object is an avoidance target. Then, the avoidance control process for the avoidance target is entered. In order to determine that it is an avoidance target, a certain period of tracking processing time is usually required to prevent misidentification.

[0007] Furthermore, consider the case where evasion control is performed on an object (a bicycle, two-wheeled vehicle, etc.) coming from a crossing direction or from behind when the own vehicle is turning left or right at an intersection. In the case of evasion control for an object coming from a crossing direction, the position where the evasion control is started is far from the own vehicle, so it is easy to ensure the remaining control time for evasion. On the other hand, in the case of evasion control for an object coming from behind, the position where the evasion control is started is close to the own vehicle, so it is difficult to ensure the remaining control time for evasion. In fact, the own vehicle side is also moving, so it is even more difficult to ensure the remaining control time for evasion control for an object coming from behind. In this way, whether or not evasion control is performed in time varies depending on the direction in which the object is coming from the own vehicle (refer to Figure 1 ).

[0008] Therefore, when a fast moving obstacle such as a bicycle approaches the vehicle from behind, time is also required to determine it as an avoidance object. However, even if avoidance control is switched after it is determined as an avoidance object, the possibility of not being able to avoid it in time increases.

[0009] The collision avoidance control proposed in the method of Patent Document 1 is to advance the operation timing according to the additional conditions after determining that the object is an avoidance object, thereby dealing with the object originally located in the blind spot, but does not consider the time until it is determined as an avoidance object. Therefore, even if Patent Document 1 is used, when a fast moving object such as a bicycle is coming from behind, it is still necessary to determine the time for avoiding the object, so there is a high possibility that the collision cannot be avoided even if the operation timing is advanced.

[0010] The present invention has been made in view of the above-mentioned reasons, and an object of the present invention is to provide a vehicle control device capable of performing collision avoidance control while ensuring a control surplus time required for avoidance control. Technical means of solving problems

[0011] The vehicle control device of the present invention is characterized in that it comprises: an outside recognition unit, which obtains outside information at least in front of the own vehicle; a control unit, which controls the own vehicle based on the outside information obtained by the outside recognition unit; and a storage unit, which stores detection information of the moving body obtained by the outside recognition unit when the own vehicle exceeds the moving body, and when it is predicted that the forward route of the own vehicle will change, the outside recognition unit reads out the past detection information of the moving body stored in the storage unit, judges the possibility of collision between the own vehicle and the moving body, and sets the moving body with a high possibility of collision with the own vehicle when the forward route of the own vehicle changes as the predicted identification information based on the collision possibility, and judges whether the detection information of the moving body detected by the outside recognition unit when the forward route of the own vehicle changes is the same as the past detection information of the moving body included in the predicted identification information. Effects of the Invention

[0012] According to the present invention, for example, when turning left or right at an intersection, it is possible to perform collision avoidance control while ensuring the control surplus time for avoidance control against an obstacle coming from behind the own vehicle.

[0013] According to the present invention, for example, it is possible to prevent a two-wheeled vehicle from being involved in an accident when turning left or right at an intersection, which was difficult to handle conventionally.

[0014] Other problems, structures, and effects than those described above will become clear from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a problem outline diagram for explaining a problem to be solved by the present invention. Figure 2 A schematic diagram of an embodiment of the present invention. Figure 3 It is a schematic block diagram of a vehicle control device in an embodiment of the present invention. Figure 4 It is an overall flow chart of an embodiment of the present invention. Figure 5 This is a flowchart related to collision possibility determination in an embodiment of the present invention. Figure 6 A flowchart related to setting the prediction area in an embodiment of the present invention. Figure 7 A flowchart related to the shortened tracking process using the predicted identification information in the embodiment of the present invention. Figure 8 This is a flowchart related to left and right turn determination in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0017] Before describing the embodiments of the present invention, Figure 1 The problems to be solved by the present invention are described. Figure 1 The diagram shows a situation where a vehicle turns left at an intersection, and includes a vehicle 50 with an avoidance control function traveling on a normal road 55 with an intersection, a sensor detection area 51 for the avoidance control function, a period 54 required to determine an avoidance object when performing avoidance control, a bicycle 53 coming toward the vehicle 50 from the rear, and a bicycle 52 coming toward the vehicle 50 from a crossing direction.

[0018] Consider the case where the vehicle 50 performs avoidance control on a bicycle 52 coming from the opposite direction and a bicycle 53 coming from behind when turning left at an intersection. The avoidance control function detects an obstacle in the sensor detection area 51 and determines it as an avoidance target and then performs avoidance control to prevent a collision.

[0019] First, in the case of avoidance control for a bicycle 52 coming from a crossing direction, avoidance control actually starts after the bicycle 52 enters the sensor detection area 51 and after a period 54 has passed since it was determined as an avoidance object. However, the position where avoidance control starts is far from the own vehicle 50, so the remaining control time for avoidance can be ensured.

[0020] In contrast, in the case of the avoidance control for the bicycle 53 coming from behind, the avoidance control actually starts after the bicycle 53 enters the sensor detection area 51 and after the period 54 for determination as the avoidance target has passed, but the position where the avoidance control starts is close to the own vehicle 50, so the remaining control time for avoidance cannot be ensured. In fact, the own vehicle 50 is also moving, so it is more difficult to ensure the remaining control time for the avoidance control for the obstacle coming from behind.

[0021] In this way, whether or not the avoidance control is performed in time varies due to the different directions of the bicycles 52 and 53 coming toward the vehicle 50. In view of such a problem, even when the above-mentioned prior art (Patent Document 1) is used, since the period 54 determined as the avoidance object is not clearly described, it is difficult to deal with a situation such as a bicycle 53 coming from behind.

[0022] Furthermore, the idea of ​​simply starting avoidance control at the stage of entering sensor detection area 51 is difficult to prevent erroneous operation, and it is unrealistic to simply shorten the period 54 during which the detected object is determined as the avoidance target.

[0023] An object of the present invention is to shorten the period for specifying an avoidance target for avoidance control, while preventing erroneous operation not only for obstacles coming from the lateral direction or the front but also for obstacles coming from the rear.

[0024] In an embodiment of the present invention, a component and device are provided for this purpose, which have the following three actions: (1) storing identification information of a mobile object when overtaking a mobile object such as a bicycle while driving near an intersection, thereby reading out the identification information stored in the past when predicting that the own vehicle will turn left at the intersection to determine the possibility of collision between the own vehicle and the mobile object stored in the past; (2) determining the possibility of collision, using the information of the mobile object determined as having a high possibility of collision to calculate the detection time and detection area predicted to be detected within the sensor detection range of the own vehicle when the own vehicle turns left as predicted identification information, and setting the area predicted to be detected within the sensor detection area at least in front of the own vehicle; (3) determining that the mobile object stored in the past has been detected again when an object is detected at the predicted detection time and detection area when the own vehicle turns left at the intersection, thereby shortening the period determined as the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control.

[0025] First, use Figure 2 , the basic concept of the present invention is explained. The vehicle control device 100 ( Figure 3 ) is mounted in the own vehicle 3, and performs collision avoidance control for avoiding a collision between an object in front of the own vehicle and the own vehicle. This embodiment assumes that the collision avoidance control is implemented on an ordinary road 1 including a traffic light 2 and an intersection.

[0026] Figure 2 This is a conceptual diagram of a situation in which the vehicle 3 performs collision avoidance control on a two-wheeled vehicle 6 traveling from behind when turning left at an intersection, and the action is described by steps 1, 2, and 3 showing the action flow of this embodiment.

[0027] Step 1 and Step 2 are used to illustrate action (1) among the three actions of an embodiment of the present invention. Action (1) is to store identification information of a moving object when overtaking a moving object such as a bicycle while driving near an intersection. Thus, when it is predicted that the own vehicle will turn left at the intersection, the identification information stored in the past is read out to determine the possibility of collision between the own vehicle and the moving object stored in the past.

[0028] First, if Figure 2 As shown in step 1, the own vehicle 3 and the two-wheeled vehicle 6 are traveling on the ordinary road 1. When the two-wheeled vehicle 6 is detected in the sensor detection area 4 of the own vehicle 3, the speed, relative position relative to the own vehicle, detection time, etc. are obtained as identification information of the two-wheeled vehicle 6, and when the own vehicle 3 exceeds the two-wheeled vehicle 6 and the two-wheeled vehicle 6 leaves the sensor detection area 4, it is stored in the storage unit 5 as reusable information.

[0029] Next, as shown in step 2, the driving path of the own vehicle 3 is obtained from the map information of the navigation system, etc., and it is determined whether the own vehicle 3 is likely to pass through the intersection. When it is determined that the own vehicle 3 will pass through the intersection, the identification information of the two-wheeled vehicle 6 stored in the storage unit 5 is read, and the possibility of collision between the two-wheeled vehicle 6 and the own vehicle 3 when turning left at the intersection is determined based on the read past identification information. Specifically, the distance from the own vehicle 3 to the intersection is calculated and the intersection arrival time is calculated. Next, based on the identification information of the two-wheeled vehicle 6 read from the storage unit 5, the intersection arrival time of the two-wheeled vehicle 6 is calculated assuming that it is moving in a straight line at a constant speed from the detection time point. The calculated intersection arrival times of the own vehicle 3 and the two-wheeled vehicle 6 are compared, and the possibility of the two-wheeled vehicle 6 coming from behind when the own vehicle 3 turns left at the intersection is determined as the collision possibility. When it is determined that the collision possibility is high, the identification information (past identification information) stored in the storage unit 5 as the predicted identification information is determined as an obstacle with a high possibility of being detected again when turning left at the intersection. Thus, when the vehicle passes a moving object such as a bicycle while approaching an intersection, the identification information of the moving object is stored, and when it is predicted that the vehicle will turn left at the intersection, the identification information stored in the past is read to determine the possibility of collision between the vehicle and the moving object stored in the past.

[0030] Next, step 3 is used to explain the remaining two actions, namely action (2) and action (3). Action (2) is to judge the possibility of collision, use the information of the mobile body determined as having a high possibility of collision to calculate the detection time and detection area when the mobile body will be detected within the sensor detection range of the own vehicle when the own vehicle turns left as predicted identification information, and set the area where the mobile body is predicted to be detected within the sensor detection area at least in front of the own vehicle. Action (3) is to judge that the mobile body stored in the past has been detected again when the own vehicle detects an object at the predicted detection time and the detection area when turning left at the intersection, thereby shortening the period determined as the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control.

[0031] Step 3 is a situation where a two-wheeled vehicle 6 is coming from behind when the own vehicle 3 actually turns left at the intersection. It is determined based on the control information of the turn signal lamp, etc. that the own vehicle 3 will turn left at the intersection. It is determined whether the predicted recognition information calculated in step 2 exists, using the fact that the own vehicle 3 will turn left at the intersection as a trigger. In the case where the predicted recognition information exists, the obstacle detection timing (also called predicted detection timing) and the detection area (also called predicted detection area) when turning left at the intersection are predicted based on the predicted recognition information, and the predicted area 8 corresponding to the predicted detection area is set for the image 7 generated by the image processing unit. Thus, the information of the mobile body determined to have a high possibility of collision can be used to calculate the detection timing and detection area that predicts that the mobile body will be detected within the sensor detection range of the own vehicle when the own vehicle turns left, and set the area that predicts that the mobile body will be detected in the sensor detection area at least in front of the own vehicle.

[0032] When the self-vehicle 3 detects an object in the sensor detection area 4 when turning left at the intersection, the predicted identification information is combined to perform conditional judgment, thereby being able to immediately determine whether it is a previously predicted obstacle, thereby shortening the period required to determine it as an avoidance object required in the usual collision avoidance control and switching to avoidance control (or alarm). Specifically, when an object is detected in the sensor detection area 4, it is immediately determined to be the stored two-wheeled vehicle 6 based on conditions such as whether the timing of detecting the object coincides with the predicted timing and whether the detection position is detected in the predicted area 8 (corresponding to the predicted detection area), thereby shortening the time required to determine it as an avoidance object in the collision avoidance control.

[0033] According to the present invention, the time required for detecting an object and performing object recognition for a certain period of time to determine that the object is an avoidance target, which is difficult to simply shorten, can be shortened by adding predicted recognition information. By shortening the time, the timing of starting avoidance control can be advanced, so the remaining control time for avoiding obstacles can be ensured.

[0034] The configuration of the vehicle control device in the embodiment of the present invention is shown in FIG. Figure 3 The vehicle control device 100 of this embodiment is a vehicle control device that implements collision avoidance control and includes an external recognition unit 101, a control unit 102, and a storage unit 106. The external recognition unit 101 performs obstacle recognition and traffic light recognition, the control unit 102 actually instructs avoidance control, and the storage unit 106 stores (saves) recognition information given by the external recognition unit 101.

[0035] The external recognition unit 101 obtains external information in front of the own vehicle, and has an image processing unit 103, a recognition unit 104, and an obstacle prediction recognition unit 105. The image processing unit 103 is used to generate an image used in the recognition process. The recognition unit 104 detects vehicles or lanes and tracks (traces) the time series of the detection objects. The obstacle prediction recognition unit 105 determines the possibility of a collision when the own vehicle turns left at an intersection based on past recognition information stored in a storage unit 106 that stores the information recognized by the recognition unit 104, and generates prediction recognition information.

[0036] The obstacle prediction and recognition unit 105 includes an intersection distance calculation unit 114, an intersection arrival time calculation unit 115, a collision possibility judgment unit 116, and a prediction reliability setting unit 117. The intersection distance calculation unit 114 obtains the position information of the vehicle from an MPU having high-precision map information such as intersection width information, intersection stop line position, number of lanes and other road structure-related data and road signs, traffic light positions and other three-dimensional information on the road, or a GNSS (GLOBAL NAVIGATION SATELLITE SYSTEM, global navigation satellite system) etc., calculates the distance from the own vehicle to the intersection, the intersection arrival time calculation unit 115 calculates the intersection arrival time when the obstacle travels to the intersection in a straight line at a constant speed and the intersection arrival time of the own vehicle at the intersection according to the obstacle information (past identification information) stored in the storage unit 106, the collision possibility judgment unit 116 judges the collision possibility by comparing the obstacle calculated by the intersection arrival time calculation unit 115 with the intersection arrival time of the own vehicle and calculates the predicted identification information, the predicted reliability setting unit 117 adds reliability to the predicted identification information calculated by the collision possibility judgment unit 116 according to external factors that reduce the detection accuracy, such as the elapsed time after the obstacle is detected and the weather conditions when the obstacle is detected. Furthermore, the reliability set by the prediction reliability setting unit 117 is used by the prediction area detection judgment unit 112 described later to judge the possibility that the obstacle detected (when the own vehicle turns left at the intersection) is an obstacle detected in the past and stored in the storage unit 106 (judgment of the consistency or identity of the obstacle detection information with the past detection information).

[0037] The recognition unit 104 includes a moving body detection unit 111, a prediction area detection and judgment unit 112, and a tracking processing unit 113. The moving body detection unit 111 recognizes a moving body such as a two-wheeled vehicle traveling on an ordinary road. The prediction area detection and judgment unit 112 receives the prediction identification information from the obstacle prediction recognition unit 105, calculates the area (prediction area) in which the obstacle included in the prediction identification information is predicted to be detected on the image captured by the camera unit 126, and hands the prediction area information to the prediction area setting unit 110 of the image processing unit 103. The prediction area setting unit 110 sets the prediction area (reference area) for the image captured by the camera unit 126 (more specifically, at least a portion of the image). Figure 2 ). When the moving body detection unit 111 actually detects a moving body, the predicted area detection judgment unit 112 determines whether it is detected in the predicted area. The predicted area detection judgment unit 112 determines the possibility that the detected obstacle is an obstacle detected in the past based on the condition of whether the moving body is detected in the predicted area and whether the time when the moving body is detected is the predicted time, and hands it over to the tracking processing unit 113. In other words, the predicted area detection judgment unit 112 determines whether the detection information of the detected moving body is the same as the past detection information of the moving body based on the predicted detection timing and the predicted detection area, and hands it over to the tracking processing unit 113. The tracking processing unit 113 usually performs vehicle or lane detection and time series tracking (tracking) of the detection object, etc., but in the case of the collision avoidance control of this embodiment, in order to determine the avoidance object for avoidance control, it is necessary to perform tracking processing for a certain period of time after the initial detection of the obstacle to determine whether it is an obstacle that may collide. In addition to the normal tracking process, the tracking processing unit 113 of this embodiment adds the previously recognized information compared to the normal tracking process when receiving the predicted identification information that the detected obstacle is likely to be an obstacle detected in the past (in other words, when it is determined that the detection information of the detected moving body is the same as the past detection information of the moving body included in the predicted identification information), thereby shortening the tracking processing time for determining the detected obstacle as an avoidance object. As a result, the period for determining the avoidance object for implementing collision avoidance control can be shortened.

[0038] The control unit 102 controls the vehicle according to the external information acquired by the external recognition unit 101, and includes a left-right turn judgment unit 120, a collision risk calculation unit 121, a path planning unit 122, and a control instruction unit 123. The left-right turn judgment unit 120 judges whether the vehicle is in a left-right turn state according to the information of the yaw rate sensor 125, the steering angle sensor 127, and the turn signal lamp 130. The collision risk calculation unit 121 calculates the remaining time for collision (TTC) for obstacles with a high probability of collision (in other words, obstacles determined as avoidance objects of avoidance control) determined by the external recognition unit 101 to judge the collision risk, thereby judging whether avoidance control needs to be executed. The path planning unit 122 calculates the driving path for collision avoidance. The control instruction unit 123 calculates the control value of the collision avoidance control and instructs the braking force / driving force control ECU 124.

[0039] The braking force / driving force control ECU 124 issues appropriate control instructions to various actuators of the vehicle based on the control instruction value of the control instruction unit 123 .

[0040] The warning indicator 128 serves to warn the driver of the imminent danger of collision based on the information from the collision risk calculation unit 121, or to inform the driver of the fact that the evasion control is being performed by displaying or sounding a sound during the collision evasion control.

[0041] Regarding the external recognition unit 101, in this embodiment, a camera unit (camera sensor) mounted in front of the vehicle for photographing at least the front of the own vehicle is taken as an example as a component for obtaining external information at least in front of the own vehicle, but not only a single camera sensor can be used, but also a combination of a camera sensor and a radar sensor or a combination of other sensors can be used.

[0042] In the present embodiment, the external recognition unit 101 and the control unit 102 operate independently, but the external recognition unit 101 and the control unit 102 can also be used as an integrated structure.

[0043] The obstacle prediction and recognition unit 105 obtains detection information from the storage unit 106 when determining the possibility of collision. In order to reduce the data calculation load, the obstacle prediction and recognition unit 105 may also use the width information of the detected obstacle contained in the detection information to filter whether it is a two-wheeled vehicle to limit the acquired data.

[0044] The storage unit 106 may be shared with a common storage unit used in the external recognition unit 101, or may be independently configured as a storage unit dedicated to this embodiment.

[0045] The prediction region setting unit 110 is not limited to setting one prediction region, but may set a plurality of prediction regions. In this case, the setting method of the prediction region may be changed according to the reliability included in the prediction identification information.

[0046] Regarding the prediction area set by the prediction area setting unit 110, when the setting information is not detected even after the predicted detection time has passed from the detection timing included in the prediction identification information, the setting information may be deleted from the storage unit 106. In other words, when the moving object included in the prediction identification information is not detected even after the predicted detection timing has passed in the prediction area set by the prediction area setting unit 110, the prediction identification information may be deleted from the storage unit 106.

[0047] As an example of determining the possibility of collision in the collision possibility determination unit 116, a method of simply calculating the possibility of collision based on a comparison of the arrival times at the intersection of the own vehicle 3 and the two-wheeled vehicle 6 is described, but the method of determining the possibility of collision is not limited to this. For example, a method of improving the accuracy of determining the possibility of collision by predicting the future path of the two-wheeled vehicle 6 is not limited to this embodiment.

[0048] From the time when the collision possibility judgment unit 116 makes a collision possibility judgment until the vehicle arrives at the intersection, the time when the vehicle 3 arrives at the intersection will change due to the driver's control, so the collision possibility judgment can be implemented at any time (in cycle time) before the vehicle 3 arrives at the intersection.

[0049] The left-right turn determination unit 120 may also obtain the steering angle control value to determine the left-right turn.

[0050] In this embodiment, the intersection distance calculation unit 114 calculates the intersection distance based on the information in the map information 129, but it can also be calculated based on the recognition result of the sensor (camera sensor, etc.) installed on the own vehicle, or based on the road-to-vehicle communication information received from the roadside communication device. In other words, as a trigger for reading the obstacle information (past detection information) from the storage unit 106 to calculate the intersection distance, the past detection information can be read by calculating the driving path of the own vehicle based on the map information 129 and judging that the possibility of turning left or right at the intersection is high, or by judging that the possibility of turning left or right at the intersection is high based on the intersection recognition result given by the external recognition unit 101, or by judging that the possibility of turning left or right at the intersection is high based on the intersection information obtained by road-to-vehicle communication.

[0051] In this embodiment, an example of performing collision avoidance control on an obstacle coming from behind when the vehicle turns left is described, but the control is not limited to the case of a left turn and can be applied to all cases of changing the route such as a right turn.

[0052] According to the above configuration, use Figure 4 The overall process of this embodiment is described by using the flowchart of Figure 5 The flowchart of action (1) is used to illustrate the Figure 6 The flowchart of action (2) is used to illustrate the Figure 7 The flowchart of action (3) is used to illustrate the Figure 8 The flowchart of the left turn judgment method of the own vehicle performed by the left turn judgment unit 120 is explained, the action (1) is to store the identification information of the mobile body when overtaking a mobile body such as a bicycle when driving near an intersection, thereby reading the identification information stored in the past when predicting that the own vehicle will turn left at the intersection to judge the possibility of collision between the own vehicle and the mobile body stored in the past, the action (2) is to judge the collision possibility, use the information of the mobile body determined as having a high collision possibility to calculate the detection time and detection area predicted to be detected within the sensor detection range of the own vehicle when the own vehicle turns left and set the area predicted to be detected within the sensor detection area at least in front of the own vehicle, and the action (3) is to judge that the mobile body stored in the past is detected again when the own vehicle detects an object at the predicted detection time and the predicted detection area when the own vehicle turns left at the intersection, thereby shortening the period determined as the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control.

[0053] First of all Figure 4 Furthermore, Figure 4 Steps 1 to 3 correspond to Figure 2 Step 1 to step 3.

[0054] First, the external recognition unit 101 installed on the own vehicle obtains external information in front of the own vehicle (S300). It is determined whether the obtained external information includes a two-wheeled vehicle (S301). If not, the process returns to S300, and if yes, the process proceeds to S302. The detection information and detection time of the two-wheeled vehicle recognized by the recognition unit 104 are stored in the storage unit 106 (S302).

[0055] The obstacle prediction and recognition unit 105 obtains the travel plan information of the own vehicle based on the map information 129 and the position information of the own vehicle, etc. (S303). The obstacle prediction and recognition unit 105 determines the possibility of collision when the own vehicle turns left or right based on the moving body information in the obstacle information (past recognition information) stored in the storage unit 106, and extracts obstacles that are determined to have a high possibility of collision (S304).

[0056] Thereafter, the left-right turn judgment unit 120 judges whether the vehicle is turning left or right (S305). If not, the process returns to S300, and if yes, the process proceeds to S306. Determine the moving body with a high possibility of collision when the vehicle turns left or right, and obtain the detection timing (predicted detection timing) and detection area (predicted detection area) of the moving body predicted to be detected in the detection area of ​​the vehicle when turning left or right as the predicted identification information (S306). Determine whether an obstacle is detected in the detection area of ​​the vehicle (S307). If not, the process returns to S306, and if yes, the process proceeds to S308. Use the predicted identification information obtained in S306 to compare with the detection timing and detection area of ​​the obstacle actually detected in S307, and determine the possibility that the obstacle actually detected in S307 is the same as the obstacle determined using the predicted identification information (S308). In other words, determine whether the detection information of the moving body actually detected in S307 when the vehicle turns left or right is the same as the past detection information of the moving body contained in the predicted identification information. If the possibility is high, proceed to S309; ​​if the possibility is low, proceed to S310. If it is judged in S308 that the obstacle detected again is likely to be the same as the predicted identification information detected in the past, the period (number of detections) required for determining it as an avoidance object in the usual collision avoidance control is reduced, and the obstacle whose predicted identification information is detected is immediately determined (S309). If it is judged in S308 that the possibility is low that the obstacle detected again is the same as the predicted identification information detected in the past, tracking processing is performed for a certain period of time (that is, the usual number of detections) in order to determine it as an avoidance object, as in the usual collision avoidance control, and the obstacle is determined as an avoidance object (S310). Avoidance control is implemented or an alarm is issued for the avoidance object determined in the above order (S311).

[0057] Next, use Figure 5 The action of reading the identification information stored in the past to determine the possibility of collision between the own vehicle and the moving object stored in the past in action (1) is explained. The action (1) is to store the identification information of the moving object when overtaking a moving object such as a bicycle when driving near an intersection, thereby reading the identification information stored in the past to determine the possibility of collision between the own vehicle and the moving object stored in the past when it is predicted that the own vehicle will turn left at the intersection.

[0058] The obstacle prediction and recognition unit 105 obtains the driving plan information of the own vehicle based on the map information 129 or the position information of the own vehicle (S400). The obstacle prediction and recognition unit 105 determines whether the own vehicle is scheduled to pass through the intersection based on the map information 129 or the position information of the own vehicle (S401). In the case of no, it returns to S400, and in the case of yes, it proceeds to S402. The moving body information in the past identification information stored in the storage unit 106 is read out (S402). The prediction reliability setting unit 117 calculates and adds reliability to the moving body information read out from the storage unit 106 based on the time when the moving body is detected and the detection accuracy (S403). The intersection distance calculation unit 114 calculates the distance to the intersection based on the current location of the own vehicle and the information of the map information 129 (S404). The intersection arrival time calculation unit 115 calculates the intersection arrival time predicted when the own vehicle arrives at the intersection and turns left or right based on the intersection distance information calculated in S404 (S405). The collision possibility judgment unit 116 calculates the intersection turn time (intersection arrival time) for each moving body information read out in S402, and judges the collision possibility corresponding to each moving body information when the own vehicle turns left or right for each moving body information (S406). The prediction reliability setting unit 117 evaluates and sets the reliability for the determined collision possibility (S407). Based on the collision possibility evaluated in S406 and the reliability evaluated in S407, the moving body information with a high possibility of collision when the own vehicle turns left or right at the intersection is determined, and is handed over to the recognition unit 104 as prediction recognition information (S408).

[0059] use Figure 6 Action (2) is explained, and the action (2) is to judge the possibility of the collision, use the information of the mobile body determined as having a high possibility of collision to calculate the detection timing and detection area of ​​the mobile body that will be detected within the sensor detection range of the own vehicle when the own vehicle turns left as predicted identification information, and set the area in which the mobile body is predicted to be detected within the sensor detection area at least in front of the own vehicle.

[0060] First, the left-right turn determination unit 120 detects that the vehicle is turning left-right (S500). If no, the process returns to S500, and if yes, the process proceeds to S501. The recognition unit 104 determines whether it has received a signal from the obstacle prediction recognition unit 105. Figure 5The predicted identification information (determined by the collision possibility judgment unit 116) described in (S501). If no, the processing is terminated, and if yes, the process proceeds to S502. The recognition unit 104 calculates and predicts the timing of a collision between the own vehicle and a moving body with a high possibility of collision when the own vehicle turns left or right, which is contained in the received predicted identification information (S502). In addition, the recognition unit 104 calculates the sensor detection area that predicts that the moving body will be detected when the own vehicle turns left or right based on the relative position information of the moving body relative to the own vehicle contained in the predicted identification information (S503). The predicted area setting unit 110 of the image processing unit 103 sets the sensor detection area for (at least a part of) the sensor detection area at least in front of the own vehicle as the predicted area based on the calculated collision timing and the predicted identification information containing the sensor detection area information (reference Figure 2 )(S504).

[0061] use Figure 7 Action (3) is explained, which is to determine that a moving object stored in the past has been detected again when an object is detected within the predicted detection timing and the detection area when the own vehicle turns left at the intersection, thereby shortening the period used to determine the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control.

[0062] The left-right turn judgment unit 120 judges whether the vehicle is in the right-left turn control (S600). If not, the process ends; if yes, the process proceeds to S601. The recognition unit 104 judges whether a moving object is detected in the sensor detection area at least in front of the vehicle (S601). If not, the process returns to S600; if yes, the process proceeds to S602. The prediction area detection judgment unit 112 of the recognition unit 104 performs a prediction on the moving object detected in S601. Figure 6 The predicted collision timing included in the identification information calculated in the action is compared with the timing of the actual detection of the moving body (S602). In other words, it is determined whether the moving body is detected at the predicted collision timing. If not, it proceeds to S604, and if yes, it proceeds to S603. In addition, the predicted area detection judgment unit 112 determines whether it is through Figure 6 The moving object detected in S601 is detected in the predicted area set by the action of the vehicle (S603). In other words, it is determined whether the moving object is detected in the predicted area. If not, proceed to S604, and if yes, proceed to S606. That is, for the detection information of the moving object detected in S601 when the vehicle turns left or right, it is determined whether it is the same as the past detection information of the moving object included in the predicted identification information based on the predicted detection timing and the predicted detection area.

[0063] In S604, the prediction region detection judgment unit 112 judges that the possibility that the moving object detected in S601 is the same as the moving object included in the prediction recognition information that was recognized in the past and judged to have a high probability of collision is low, and cancels the prediction region detection judgment unit 112. Figure 6 The prediction area set in the process is set (S604). Then, the tracking process implemented in the normal collision avoidance control is performed (S605).

[0064] If the answer is yes in S603, the prediction area detection judgment unit 112 judges that there is a high possibility that the moving body detected in S601 is the same as the moving body included in the prediction identification information, which was identified in the past and judged to have a high possibility of collision. That is, it is judged that the detection information of the moving body detected in S601 when the own vehicle turns left or right is the same as the past detection information of the moving body included in the prediction identification information. In this case, the tracking process can be limited to the moving body detected in the prediction area and the number of detections required for the tracking process (tracking process time) can be reduced using the prediction identification information to determine the avoidance object, thereby simplifying the tracking process method while maintaining the detection accuracy. As a result, the period required for the usual collision avoidance control to determine the avoidance object can be shortened (S606). The control unit 102 implements collision avoidance control or an alarm (S607) on the avoidance object determined by the recognition process of S605 or S606.

[0065] Finally, use Figure 8 How the left-right turn determination unit 120 of the control unit 102 determines whether the vehicle is turning left-right will be described.

[0066] The left-right turn determination unit 120 obtains control information of the turn signal lamp 130 (S700). The left-right turn determination unit 120 obtains information of the yaw rate sensor 125 and obtains information of the position of the own vehicle based on the map information 129, etc. (S701). Based on the information obtained in S700, it is determined whether the turn signal lamp operation of a left turn or a right turn is being performed (S702). Based on the yaw rate sensor information obtained in S701, it is determined whether the own vehicle is changing in the same direction as the indicated turn signal lamp direction (S703). When the judgment results of the above S702 and S703 are yes, the left-right turn determination unit 120 determines that it is in a left turn or a right turn state (S704).

[0067] In this embodiment, with the help of the three methods (1), (2) and (3), the period from the time when an object is detected and the object is identified as an avoidance object for a certain period of time, which is difficult to shorten simply and is required to execute collision avoidance control, can be shortened by adding predicted recognition information. The method (1) stores the recognition information of a moving object when passing a moving object such as a bicycle when driving near an intersection, thereby reading the recognition information stored in the past when predicting that the own vehicle will turn left at the intersection to determine the possibility of a collision between the own vehicle and the moving object stored in the past. The method (2) determines the possibility of the collision and uses the result to determine the collision possibility. The method (3) is to calculate the detection timing and detection area of ​​the moving body that will be detected within the sensor detection range of the own vehicle when the own vehicle turns left based on the information of the moving body that is determined to have a high collision probability as the prediction identification information, and set the area in which the moving body is predicted to be detected within the sensor detection area at least in front of the own vehicle, and judge that the moving body stored in the past is detected again when the own vehicle detects an object at the predicted detection timing and the detection area when the own vehicle turns left at the intersection, thereby shortening the period for determining the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control. By shortening the period, the timing from the first detection of the obstacle to the start of avoidance control can be brought forward, so that the remaining control time for avoiding the obstacle can be ensured.

[0068] Regarding the ensuring of the remaining control time for avoidance control of obstacles coming from behind the vehicle when turning left or right at an intersection, which is defined as the subject at the beginning of this specification, collision avoidance control can be performed by shortening the period determined as the avoidance object using predictive recognition information to ensure the remaining control time for avoidance control for obstacles coming from behind.

[0069] As described above, the vehicle control device 100 of this embodiment includes: an outside recognition unit 101, which obtains outside information at least in front of the own vehicle; a control unit 102, which controls the own vehicle based on the outside information obtained by the outside recognition unit 101; and a storage unit 106, which stores the detection information of the moving body obtained (detected) by the outside recognition unit 101 when the own vehicle exceeds the moving body. When it is predicted that the forward route of the own vehicle will change, the outside recognition unit 101 reads the past detection information of the moving body stored in the storage unit 106, judges the possibility of collision between the own vehicle and the moving body, and sets the mobile body (predicted detection timing and predicted detection area) with a high possibility of collision with the own vehicle when the forward route of the own vehicle changes as the predicted recognition information based on the collision possibility (obstacle prediction recognition unit 105), and judges whether the detection information of the moving body detected by the outside recognition unit when the forward route of the own vehicle changes is the same as the past detection information of the moving body included in the predicted recognition information (recognition unit 104).

[0070] When it is determined that the detection information of the moving body detected by the external identification unit 101 when the own vehicle changes its forward route is the same as the past detection information of the moving body contained in the predicted identification information, the vehicle control device 100 shortens the period of determination as the avoidance object of collision avoidance control and determines the moving body as the avoidance object, and switches to avoidance control or alarm for the moving body (identification unit 104).

[0071] The predicted identification information includes a predicted detection timing and a predicted detection area of ​​the moving body. The vehicle control device 100 sets a predicted detection area for predicting that the moving body will be detected by the external recognition unit 101 when the own vehicle's forward route is changed (predicted detection area), and sets a predicted detection timing for predicting that the moving body will be detected by the external recognition unit 101 when the own vehicle's forward route is changed (predicted detection timing). For the detection information of the moving body detected by the external recognition unit 101 when the own vehicle changes its forward route, the predicted detection timing and the predicted detection area are used as conditions to determine whether it is the same as the past detection information of the moving body (identification unit 104).

[0072] The vehicle control device 100 compares the intersection arrival time of the own vehicle with the intersection arrival time of the moving body calculated based on the past detection information of the moving body, and determines the possibility of the collision (the obstacle prediction and recognition unit 105).

[0073] When it is determined that the detection information of the moving body detected by the external identification unit 101 when the own vehicle changes its forward route is the same as the past detection information of the moving body contained in the predicted identification information, the vehicle control device 100 shortens the tracking processing time for determining the avoidance object as the collision avoidance control and determines the moving body as the avoidance object (identification unit 104).

[0074] The vehicle control device 100 sets the predicted detection area (recognition unit 104 ) for a portion of an image captured by a camera sensor that captures at least the front of the own vehicle.

[0075] The vehicle control device 100 sets reliability for the predicted recognition information according to the time when the moving object is detected and the detection accuracy (weather conditions, etc.), and uses the reliability to determine the consistency (identity) between the detection information and the past detection information.

[0076] In other words, the vehicle control device 100 of the present embodiment comprises: a control unit 102, which performs collision avoidance control for avoiding a collision between an object in front of the own vehicle and the own vehicle; an external recognition unit 101, which performs obstacle recognition; a storage unit 106, which stores obstacle recognition information; an obstacle prediction recognition unit 105, which calculates the distance from the own vehicle to the intersection and the intersection arrival time, determines the possibility of the own vehicle colliding when turning left at the intersection based on each obstacle information stored in the storage unit 106, and generates prediction recognition information; and a prediction area setting unit, which determines an obstacle with a high possibility of collision when the own vehicle turns left at the intersection based on the prediction recognition information, and sets an area in which the obstacle is predicted to be detected. When the own vehicle detects an obstacle in the prediction area set by the prediction area setting unit when turning left at the intersection, the vehicle control device 100 determines that a moving body stored in the past has been detected again, thereby shortening the time used to determine the avoidance object for avoiding control, thereby ensuring the remaining control time required for avoiding control.

[0077] That is, the vehicle control device 100 of this embodiment provides a component and a device, which stores identification information when passing a two-wheeled vehicle, reads the identification information of the two-wheeled vehicle identified in the past when turning left or right at an intersection to determine the possibility of collision with the own vehicle and calculate the predicted identification information, and when an obstacle is actually detected when the own vehicle turns left, it uses the predicted identification information to immediately determine that it is a re-detection of a moving body identified in the past and perform avoidance control.

[0078] According to the present invention, when the own vehicle is traveling on the road near the intersection, with the help of a total of three actions, namely action (1), action (2), and action (3), for the current subject, that is, for example, ensuring the remaining time of avoidance control for obstacles coming from the rear of the own vehicle when turning left or right at the intersection, the collision avoidance control can be performed by shortening the period of determining the collision avoidance object using predicted recognition information to ensure the remaining control time of avoidance control for obstacles coming from the rear. The action (1) is to store the recognition information of a moving object when overtaking a moving object such as a bicycle when traveling near the intersection, thereby reading the recognition information stored in the past when predicting that the own vehicle will turn left at the intersection to determine the collision between the own vehicle and the moving object stored in the past. The action (2) is to judge the collision possibility, use the information of the mobile body determined as having a high collision possibility to calculate the detection time and detection area when the mobile body will be detected within the sensor detection range of the own vehicle when the own vehicle turns left as the prediction identification information, and set the area where the mobile body is predicted to be detected within the sensor detection area at least in front of the own vehicle, and the action (3) is to judge that the mobile body stored in the past has been detected again when the object is detected at the predicted detection time and the detection area when the own vehicle turns left at the intersection, thereby shortening the period determined as the avoidance object for implementing avoidance control, thereby ensuring the remaining control time required for avoidance control.

[0079] This embodiment can prevent a two-wheeled vehicle from being involved in an accident when turning left or right at an intersection, which was difficult to handle conventionally.

[0080] Furthermore, the above-described embodiment is only an example, and the present invention is not limited to these contents as long as the features of the invention are not damaged. In addition, various embodiments have been described above, but the present invention is not limited to these contents. Other forms considered within the scope of the technical idea of ​​the present invention are also included in the scope of the present invention.

[0081] In addition, the present invention includes various deformation forms and is not limited to the above-mentioned embodiments. For example, the above-mentioned embodiments are detailed descriptions of the present invention in an easy-to-understand manner, and are not necessarily limited to all the structures described. In addition, the above-mentioned various structures, functions, processing units, processing methods, etc., for example, can be implemented in part or in whole with hardware by designing using integrated circuits. In addition, the above-mentioned various structures, functions, etc. can also be implemented in software by interpreting and executing programs that implement various functions by a processor. Information such as programs, tables, files, etc. that implement various functions can be stored in storage devices such as memories, hard disks, SSDs (Solid State Drives) or recording media such as IC cards, SD cards, and DVDs. Explanation of symbols

[0082] 1…Ordinary road including intersections 2…Traffic lights 3…Own vehicle 4…Sensor detection area 5…Storage 6…Two-wheeled vehicles 7…Image generated by the image processing unit 8…Forecast area 50…Own vehicle 51…Sensor detection area 52…Bicycle 53…Bicycle 54…The period of time designated as the object of avoidance 55…Ordinary road with intersections 100…Vehicle control devices 101… External Identification Department 102…Control Department 103…Image Processing Department 104…Identification Department 105…Obstacle prediction and recognition unit 106…Storage 110…Prediction area setting unit 111…Moving object detection unit 112…Prediction area detection and judgment unit 113…Tracking Processing Department 114…Intersection distance calculation unit 115…Intersection arrival time calculation unit 116…Collision possibility judgment unit 117…Prediction reliability setting unit 120…Left and right turn judgment unit 121…Collision Risk Calculation Department 122…Route Planning Department 123…Control indicator 124…Braking force / driving force control ECU 125…Yaw rate sensor 126…Camera Department 127…Steering angle sensor 128…Alarm indication unit 129…Map information 130…Turn indicator light 131…communication network.

Claims

1. A vehicle control device, characterized in that: have: An external recognition unit that obtains external information at least in front of the own vehicle; a control unit configured to control the vehicle according to the external information acquired by the external recognition unit; as well as a storage unit for storing detection information of the moving object acquired by the outside world recognition unit when the own vehicle exceeds the moving object, When it is predicted that the forward route of the own vehicle changes, the external recognition unit reads out the past detection information of the mobile body stored in the storage unit, determines the possibility of collision between the own vehicle and the mobile body, and sets the mobile body with a high possibility of colliding with the own vehicle when the forward route of the own vehicle changes as the predicted recognition information based on the collision possibility, and determines whether the detection information of the mobile body detected by the external recognition unit when the forward route of the own vehicle changes is the same as the past detection information of the mobile body contained in the predicted recognition information.

2. The vehicle control device according to claim 1, characterized in that: When it is determined that the detection information of the moving body detected by the external recognition unit when the own vehicle changes its forward route is the same as the past detection information of the moving body contained in the predicted recognition information, the vehicle control device shortens the period of determination as the avoidance object of collision avoidance control, determines the moving body as the avoidance object, and switches to avoidance control or alarm for the moving body.

3. The vehicle control device according to claim 1, characterized in that: The predicted identification information includes a predicted detection time and a predicted detection area of ​​the moving object. The vehicle control device sets a predicted detection area for at least a portion of the external information of the own vehicle acquired by the external recognition unit to predict that the moving body will be detected by the external recognition unit when the forward route of the own vehicle is changed, and sets a predicted detection timing at which the moving body will be detected by the external recognition unit when the forward route of the own vehicle is changed. For the detection information of the moving body detected by the external recognition unit when the forward route of the own vehicle is changed, it is judged whether it is the same as the past detection information of the moving body based on the predicted detection timing and the predicted detection area.

4. The vehicle control device according to claim 1, characterized in that: The vehicle control device determines the possibility of collision by comparing the intersection arrival time of the own vehicle with the intersection arrival time of the moving body calculated based on past detection information of the moving body.

5. The vehicle control device according to claim 1, characterized in that: When it is determined that the detection information of the moving body detected by the external recognition unit when the own vehicle changes its forward route is the same as the past detection information of the moving body contained in the predicted recognition information, the vehicle control device shortens the tracking processing time used to determine the avoidance object for collision avoidance control and determines the moving body as the avoidance object.

6. The vehicle control device according to claim 3, characterized in that: The vehicle control device sets the predicted detection area for a portion of an image captured by a camera sensor that captures at least the front of the own vehicle.

7. The vehicle control device according to claim 1, characterized in that: The vehicle control device performs the collision possibility judgment at any time before the own vehicle reaches the intersection.

8. The vehicle control device according to claim 3, characterized in that: When the moving object included in the predicted identification information is not detected in the predicted detection area even after the predicted detection timing, the vehicle control device deletes the predicted identification information from the storage unit.

9. The vehicle control device according to claim 1, characterized in that: The vehicle control device sets reliability for the predicted recognition information according to the time when the moving body is detected and the detection accuracy, and uses the reliability for consistency determination between the detection information and the past detection information.

10. The vehicle control device according to claim 1, characterized in that: As a trigger for reading the past detection information, the vehicle control device calculates the travel path of the own vehicle based on map information and determines that the possibility of turning left or right at an intersection is high, thereby reading the past detection information.

11. The vehicle control device according to claim 1, characterized in that: As a trigger for reading the past detection information, the vehicle control device reads the past detection information by judging that the possibility of turning left or right at the intersection is high based on the intersection recognition result provided by the outside world recognition unit.

12. The vehicle control device according to claim 1, characterized in that: The vehicle control device reads the past detection information by judging that the possibility of turning left or right at the intersection is high using the intersection information acquired by roadside-to-vehicle communication as a trigger for reading the past detection information.

13. The vehicle control device according to claim 3, characterized in that: The vehicle control device cancels the setting of the predicted detection area when it is determined that the detection information of the moving object detected by the outside recognition unit when the own vehicle changes the course is different from the past detection information of the moving object included in the predicted recognition information.

Images