WO2019239471A1

WO2019239471A1 - Driving assistance device, driving assistance system, and driving assistance method

Info

Publication number: WO2019239471A1
Application number: PCT/JP2018/022282
Authority: WO
Inventors: 健一名倉; 雄末廣; 響子細井
Original assignee: 三菱電機株式会社
Priority date: 2018-06-11
Filing date: 2018-06-11
Publication date: 2019-12-19
Also published as: JPWO2019239471A1

Abstract

The present invention comprises: a receiving unit (110) that receives first sensor information from a first sensor, and second sensor information from a second sensor; a movement information extraction unit (111) that uses the first sensor information to calculate the position of an object detected by the first sensor; a time-series determination unit (115) that manages position information by deeming a plurality of objects derived from the same object to be one object; an object attribute detection unit (112) that uses the second sensor information to generate attribute information; an information integration unit (113) that integrates and manages, in object units, the position information and the attribute information of objects managed by the time-series determination unit (115); and an object position estimation unit (114) that uses the integrated information in object units to estimate an arrival range of the object at the time the next first sensor information is received. The time-series determination unit (115) manages position information for an object within the arrival range, from among objects for which a first position was calculated by the movement information extraction unit (111).

Description

Driving support device, driving support system, and driving support method

The present invention relates to a driving support device, a driving support system, and a driving support method that perform driving support for an autonomous vehicle.

In the advanced driving support system and the automatic driving system that have been attracting attention recently, the autonomous driving vehicle controls the driving of the own vehicle by grasping the surrounding situation from the information acquired from the sensor provided in the own vehicle. In addition, the autonomous driving vehicle can acquire information on surrounding vehicles from sensors installed on the side of the road or the like, and can control the driving of the vehicle by grasping the surrounding situation. In either case of acquiring information from sensors provided in the own vehicle or acquiring information from sensors installed on the roadside, an autonomous driving vehicle uses the detection results of a plurality of sensors to detect obstacles. False detection can be reduced and unnecessary sudden stops can be avoided. On the other hand, even if the autonomous driving vehicle acquires the detection results of a plurality of sensors individually, it takes a processing load to accurately grasp the surrounding situation by using the detection results of each sensor in combination.

For such a problem, the object detection device described in Patent Document 1 determines whether the radar detection object and the image detection object are the same object using information acquired from the millimeter wave radar and the stereo camera. In addition, a technique is disclosed that uses information on an object determined to be the same object for vehicle travel control. When the object detection device described in Patent Document 1 determines whether or not the radar detection object and the image detection object are the same object, the object detection object is based on the received intensity of the reflected wave of the millimeter wave radar. By setting the range for searching for the image detection object corresponding to the radar detection object according to the type of the object, the object is detected with high accuracy while reducing the processing load.

JP 2007-132748 A

However, the object detection device described in Patent Document 1 defines a search range in advance so as to include the object for each object type, and the object type estimated based on the reception intensity of the reflected wave of the millimeter wave radar The search range corresponding to is selected. Therefore, when the search range is significantly different from the actual size of the object, there is a problem that the search area may not be set properly and the detection accuracy may deteriorate. This problem applies to both cases where the object detection device is mounted on an autonomous vehicle and installed on the roadside.

The present invention has been made in view of the above, and an object of the present invention is to obtain a driving support device capable of improving the detection accuracy of an object while suppressing a processing load.

In order to solve the above-described problems and achieve the object, the driving support apparatus of the present invention receives first sensor information from a first sensor using a radar and obtains image information from a second sensor. A receiving unit that receives the second sensor information; a movement information extracting unit that calculates a first position that is the position of the object detected by the first sensor using the first sensor information; A time-series determination unit that manages position information using a plurality of objects derived from the same object as one object among the objects whose first position is calculated by the unit. Further, the driving support device determines the size of the object detected by the second sensor, the second position that is the position of the object, and the type of the object using the second sensor information, and includes the determination result. Integrated by the information integration unit, which integrates and manages the object attribute detection unit that generates attribute information, the position information of the object managed by the time series determination unit, and the attribute information in units of objects And an object position estimation unit that estimates the reach of the object at the reception time at which the first sensor information is received next using the information in units of objects. The time-series determination unit is characterized in that the position information is managed for an object within the reachable range among the objects whose first position is calculated by the movement information extraction unit.

The driving support device according to the present invention has an effect of improving the object detection accuracy while suppressing the processing load.

A diagram showing a configuration example of a driving support system Block diagram showing a configuration example of a driving support device Flowchart showing operation of driving support device delivering position information The figure which shows the example of the position of the object which the object position estimation part estimated The figure which shows the 1st example of the reachable range of the object which the object position estimation part estimated The figure which shows the 2nd example of the reachable range of the object which the object position estimation part estimated The figure which shows the 3rd example of the reachable range of the object which the object position estimation part estimated The figure which shows the example of the hardware constitutions of a driving assistance device

Hereinafter, a driving support device, a driving support system, and a driving support method according to an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment.
FIG. 1 is a diagram illustrating a configuration example of a driving support system 400 according to Embodiment 1 of the present invention. The driving support system 400 includes a driving support device 100, a sensor 101, a sensor 102, and a wireless communication device 103. In the present embodiment, a case where the driving support system 400 is applied to a traffic system installed around an intersection will be described. It is assumed that

vehicles

201, 202, 203, bicycle 204, and

pedestrians

205, 206, 207 exist around the intersection. Of the vehicles 201 to 203, at least the vehicle 201 is an autonomous driving vehicle. For the vehicle 201, a vehicle 203 turning right at the intersection, a bicycle 204 crossing the intersection, and a pedestrian 205 can be obstacles. Note that the bicycle 204 includes a person riding the bicycle 204.

The driving support device 100 acquires sensor information from the sensor 101 and the sensor 102. Based on the sensor information acquired from the

sensors

101 and 102, the driving support device 100 generates position information indicating the position of an obstacle around the intersection, that is, an object, and transmits the position information to the wireless communication device 103. The wireless communication device 103 distributes the position information received from the driving support device 100 to the vehicles 201 to 203.

The sensor 101 is, for example, a sensor using a radar. The sensor 101 has a function of periodically detecting a distance from the sensor 101 to the vehicles 201 to 203 traveling on the main line, an angle indicating the direction of the vehicles 201 to 203 with respect to the direction of the sensor 101, and the like. The sensor 101 may further detect at least one of the speed of the vehicles 201 to 203 and the traveling direction of the vehicles 201 to 203. The sensor 101 transmits information on the detected object to the driving support apparatus 100 as sensor information. The sensor 102 is, for example, a camera. The sensor 102 has a function of periodically acquiring image information of an object. The image information is, for example, a still image or a moving image. The sensor 102 may acquire image information of an object only when it detects an object that moves within a shootable range. The sensor 102 transmits image information of the detected object to the driving support apparatus 100 as sensor information. In the following description, the sensor 101 may be referred to as a first sensor and the sensor 102 may be referred to as a second sensor. In addition, the sensor information transmitted from the sensor 101 may be referred to as first sensor information, and the sensor information transmitted from the sensor 102 may be referred to as second sensor information.

The configuration of the driving support device 100 will be described. FIG. 2 is a block diagram illustrating a configuration example of the driving support apparatus 100 according to the present embodiment. The driving support apparatus 100 includes a reception unit 110, a movement information extraction unit 111, an object attribute detection unit 112, an information integration unit 113, an object position estimation unit 114, a time series determination unit 115, and an information distribution unit 116. .

The receiving unit 110 receives sensor information, which is information about an object detected by the sensor 101, from the sensor 101. The sensor information that the receiving unit 110 acquires from the sensor 101 includes, for example, a distance to an object, an angle, a speed, a traveling direction, and the like. In addition, the receiving unit 110 receives sensor information that is information related to an object detected by the sensor 102 from the sensor 102. The sensor information that the receiving unit 110 acquires from the sensor 102 is, for example, image information of the vehicle 202, the bicycle 204, and the pedestrian 205 that are present in the imageable range of the sensor 102. In the present embodiment, the receiving unit 110 receives the first sensor information from the sensor 101 at a period of several tens of ms, and has a longer period or longer time than the period at which the sensor 102 receives the first sensor information. Although it is assumed that the second sensor information is received at intervals, the frequency with which the receiving unit 110 receives the sensor information is not limited to this.

The movement information extraction unit 111 extracts the first sensor information acquired from the sensor 101 from the sensor information received by the reception unit 110. Each time the movement information extraction unit 111 extracts the first sensor information, the movement information extraction unit 111 calculates the position of the object detected by the sensor 101 using the distance, the angle, and the like to the object included in the first sensor information. The position information of the object is generated. The position information may be expressed by latitude and longitude, may be expressed by the distance from the starting point of the road, and the expression method is not limited as long as the position of the object can be uniquely identified. The position of the object calculated by the movement information extraction unit 111 may be referred to as a first position. The movement information extraction unit 111 associates the position information with the reception time of the first sensor information and assigns and manages a reflection point ID (IDentification) for identifying the position information. When the first sensor information includes speed information, the movement information extraction unit 111 manages the position information by including the speed information of the object.

The time series determination unit 115 acquires the position information to which the reflection point ID is assigned from the movement information extraction unit 111, and arranges the position information in order of reception time and manages them in time series. The time-series determination unit 115 integrates and manages position information of objects that can be determined as the same object based on the reception time, the position of the object, the speed of the object, and the like. That is, the time-series determination unit 115 manages position information with a plurality of objects derived from the same object among the objects whose positions are calculated by the movement information extraction unit 111 as one object. For example, the time-series determination unit 115 uses the position information with the reflection point ID that has the same reception time and whose position is close and moving at the same speed as the position information derived from the same object. Alternatively, the time-series determination unit 115 uses, as position information derived from the same object, position information with different reflection times among the position information assigned with the reflection point ID and moving at a constant speed in a certain direction. . The time-series determination unit 115 assigns a trajectory ID to an object regarded as the same object, and associates the position of the object to which the trajectory ID is assigned with the position information to which the reflection point ID related to the object is assigned. to manage. In addition, when the reflection point ID does not include speed information, the time series determination unit 115 determines the difference between the position indicated by the position information at different times and the time among the position information of the object to which the trajectory ID is assigned. The speed of the object is calculated from the difference between the two.

The object attribute detection unit 112 extracts second sensor information acquired from the sensor 102 from the sensor information received by the reception unit 110. The object attribute detection unit 112 generates attribute information indicating the attribute of the object using the second sensor information, that is, image information. Specifically, the object attribute detection unit 112 performs image analysis on the image information to recognize the object, and determines the time when the object is detected, the full width of the object, the full length of the object, the position of the object, and the type of the object. Determine, and generate attribute information including the determination result. The position of the object determined by the object attribute detection unit 112 may be referred to as a second position. As the type of object, a vehicle, a bicycle, a pedestrian, or the like is assumed. The total width of the object and the total length of the object may be combined to determine the size of the object. The object attribute detection unit 112 generates object attribute information by combining information on the time when the object is detected, the size of the object, the position of the object, and the type of the object, and assigns an object attribute ID for identifying the attribute information And manage. As for the time at which the object is detected, the time at which the object overlaps a virtual line whose position set on the lane is known may be obtained from image analysis, or the second sensor information It may be a reception time.

The information integration unit 113 integrates and manages the object position information managed by the time series determination unit 115 and the attribute information managed by the object attribute detection unit 112 in units of objects. Specifically, the information integration unit 113 uses the attribute information indicated by the object attribute ID based on the time and the position of the object included in the attribute information and the time-series position information of the object managed by the trajectory ID. It is determined whether or not the object indicated by orbit ID is the same object. The information integration unit 113 assigns an object ID to the objects determined to be the same object, links the trajectory ID related to the object and the object attribute ID, and integrates and manages information related to the object. When the object attribute detection unit 112 newly detects the object attribute ID, the information integration unit 113 compares the position of the object included in the attribute information with the position of the object managed by the time series determination unit 115, for example. When the position difference is equal to or less than the prescribed threshold value, the objects whose positions are compared are determined to be the same object. If the time indicated by the attribute information is different from the time stored in time series by the trajectory ID, the information integration unit 113 averages the positions of the objects at different times managed by the trajectory ID. By this method, the position corresponding to the time of the object attribute ID is calculated. Further, the information integration unit 113 generates position information to be distributed from the information distribution unit 116 to the wireless communication device 103 using the position information of the object managed by the time series determination unit 115 or the information of the integrated object unit. . For example, based on the position information of the object managed by the time-series determination unit 115, the information integration unit 113 calculates the position of the object at the time when the information distribution unit 116 next outputs the position information to the wireless communication device 103. . The information integration unit 113 sets the calculated position of the object as position information to be distributed to the vehicles 201 to 203 and outputs the position information to the information distribution unit 116. The position information to be distributed may be at the center of the object or at the head center in the traveling direction of the object.

The object position estimation unit 114 uses the object unit information integrated by the information integration unit 113 to detect that there is an object to which the object ID is assigned at the reception time when the first sensor information is next received from the sensor 101. Estimate the expected position. The position where the object is expected to exist is defined as the reachable range. The reach range varies depending on the speed of the object, the attribute of the object, and the like. The information of the object unit integrated by the information integration unit 113 is the position information of the object indicated by the trajectory ID linked to the object ID and the attribute information indicated by the object attribute ID. The object position estimation unit 114 notifies the time series determination unit 115 of information on the reachable range.

The information distribution unit 116 outputs the position information of the object acquired from the information integration unit 113 to the wireless communication device 103. The information distribution unit 116 periodically distributes position information to the vehicles 201 to 203 via the wireless communication device 103. The transmission cycle for distributing the position information is, for example, adjusted to the operation cycle of the automatic driving system, and is typically 100 ms, but is not limited thereto. The information distribution unit 116 includes a storage unit 117 that stores the position information acquired from the information integration unit 113, and a transmission unit 118 that transmits the position information to the vehicles 201 to 203.

Subsequently, an operation in which the driving support device 100 periodically distributes position information will be described. FIG. 3 is a flowchart showing an operation in which the driving support apparatus 100 according to the present embodiment distributes position information. First, in the driving assistance device 100, the receiving unit 110 determines whether sensor information is received from the sensor 101 or the sensor 102 (step S101). If the sensor information is not received (step S101: No), the receiving unit 110 waits until sensor information is received. When the receiving unit 110 receives the sensor information (step S101: Yes) and the sensor information is the first sensor information (step S102: Yes), the movement information extraction unit 111 receives the first sensor information. Using this, the position of the object is calculated, and the position information of the object is generated (step S103). The movement information extraction unit 111 manages the position information by adding a reflection point ID. The time series determination unit 115 acquires position information from the movement information extraction unit 111 and integrates position information of a plurality of objects derived from the same object (step S104). The time series determination unit 115 performs management by assigning a trajectory ID to an object obtained by integrating position information. When the sensor information received by the receiving unit 110 is not the first sensor information (step S102: No), the driving support device 100 omits the operations of step S103 and step S104.

When the sensor information received by the receiving unit 110 is the second sensor information (step S105: Yes), the object attribute detection unit 112 generates object attribute information using the second sensor information (step S105). S106). The object attribute detection unit 112 assigns and manages an object attribute ID to the attribute information. When the sensor information received by the receiving unit 110 is not the second sensor information (step S105: No), the driving support device 100 omits the operation of step S106.

The information integration unit 113 manages the position information of the objects integrated by the time series determination unit 115 and the attribute information is generated by the object attribute detection unit 112 (step S107: Yes), the time series determination unit The position information of the object managed in 115 and the attribute information managed in the object attribute detection unit 112 are integrated (step S108). The information integration unit 113 assigns an object ID to an object estimated as the same object. The information integration unit 113 only needs to generate attribute information in the object attribute detection unit 112, and the attribute information used for integration may be the same as the attribute information used when the previous integration was performed.

Based on the position information of the object indicated by the trajectory ID and the attribute information indicated by the object attribute ID associated with the object ID by the information integration unit 113, the object position estimation unit 114 is next connected to the first sensor from the sensor 101. At a reception time at which information is received, an arrival range that is a position where an object with an object ID is expected to exist is estimated (step S109).

The driving support device 100 returns to the process of step S101 when there is a prescribed period until the information distribution unit 116 distributes the position information next time (step S110: Yes). The prescribed period is longer than the processing period from step S101 to step S109 described above in the driving support device 100. If there is no specified period until the information distribution unit 116 distributes the next position information (step S110: No), the information integration unit 113 next transmits the position information to the wireless communication apparatus 103 by the information distribution unit 116. The position of the object at the distribution time is calculated (step S111). For example, the information integration unit 113 corrects, that is, calculates, the latest position information currently managed by the time-series determination unit 115 to the position of the object at the distribution time, using constant velocity linear motion as a model. The information integration unit 113 outputs the calculated position of the object to the information distribution unit 116 as position information. The information distribution unit 116 distributes the position information acquired from the information integration unit 113 after the distribution time is reached (step S112). Then, the driving assistance device 100 returns to the process of step S101.

The information integration unit 113 manages at least one of the conditions in which the position information of the object whose position information is integrated by the time-series determination unit 115 or the attribute information is generated by the object attribute detection unit 112. If the condition information is not satisfied (step S107: No), it is determined whether or not the position information of the object with the position information integrated by the time series determination unit 115 is managed (step S113). The driving support apparatus 100 performs the process of step S110 when the position information of the object in which the position information is integrated is managed by the time-series determination unit 115 (step S113: Yes). The subsequent processing is as described above. If the position information of the object in which the position information is integrated is not managed by the time series determination unit 115 (step S113: No), the driving support apparatus 100 returns to the process of step S101.

Here, the range of the object estimated by the object position estimation unit 114 will be described. FIG. 4 is a diagram illustrating an example of an object position estimated by the object position estimation unit 114 according to the present embodiment. The object position estimation unit 114 is a target object whose range is estimated from the information integrated by the information integration unit 113, specifically, the size of the object included in the attribute information managed by the object attribute detection unit 112. The size of can be recognized. As shown in FIG. 4, the total length of the target object whose range is estimated is recognized as L1, and the total width is recognized as W1. Further, the object position estimation unit 114 recognizes the object as the vehicle 300 from the type of the object included in the attribute information. Further, the object position estimation unit 114 uses the information integrated by the information integration unit 113, specifically, the speed of the object managed by the time series determination unit 115, and then the first sensor from the sensor 101. The position of the object at the reception time when information is received is estimated. Note that the position of the vehicle 300 may be the center of the vehicle 300 or the center at the beginning of the traveling direction of the vehicle 300. In FIG. 4, the center at the beginning of the traveling direction of the vehicle 300 is the current vehicle position 301, that is, the reference position. The object position estimation unit 114 estimates, for example, the position 301a of the object at the reception time when the first sensor information is next received from the sensor 101 using a constant velocity linear motion as a model. Note that the object position estimation unit 114 uses information on the object size, that is, the entire width of the object and the total length of the object, when the object position is represented at the center of the object, and represents the object position at the beginning center of the object traveling direction. In this case, information on the entire width of the object among the sizes of the object is used, and information on the total length of the object may not be used.

The object position estimation unit 114 uses a position 301a of the object and the type of the object to determine a trapezoid area around the position 301a from the acceleration change range and the course change angle θ set according to the type of the object. Is set, and the set area is set as the aforementioned reachable range. FIG. 5 is a diagram illustrating a first example of the object reachable range estimated by the object position estimation unit 114 according to the present embodiment. For example, the object position estimation unit 114 sets the reach range width W2 in consideration of the course change range for the entire width W1 of the object, that is, the vehicle 300, and calculates the reach range width W3 by calculating from the maximum acceleration, the course change angle, and the like. Set. In addition, the object position estimation unit 114 calculates the distance L2 from, for example, the type of object, that is, the range of vehicle acceleration.

The object position estimation unit 114 may change the size of the reach range and the shape of the reach range according to the type of the object. FIG. 6 is a diagram illustrating a second example of the object reachable range estimated by the object position estimation unit 114 according to the present embodiment. In FIG. 6, it is assumed that the object type is a bicycle. When the type of the object is a bicycle, it is assumed that the negative acceleration change range is larger than the vehicle and the course change angle is wide. Therefore, in the case of the reference position of the object, that is, the current bicycle position 302, the object position estimation unit 114 estimates the object position 302a at the reception time when the first sensor information is next received from the sensor 101, and the width W21. The trapezoid reach indicated by the width W22 and the distance L20 is estimated.

FIG. 7 is a diagram illustrating a third example of the object reachable range estimated by the object position estimation unit 114 according to the present embodiment. In FIG. 7, it is assumed that the object type is a pedestrian. When the object type is a pedestrian, the acceleration change range is smaller than that of the vehicle and the bicycle, but since the course change angle is larger than that of the vehicle and the bicycle, the object position estimation unit 114 sets a circular reachable range. Therefore, the object position estimation unit 114 estimates the object position 303a at the reception time when the first sensor information is next received from the sensor 101 in the case of the object reference position, that is, the current pedestrian position 303, and the width The circular reachable range indicated by W30 and distance L30 is estimated. The object position estimation unit 114 may set the shape of the reach range in consideration of the current object speed in addition to the object type. As described above, the object position estimation unit 114 can set the size of the reachable range based on the object size, the object type, and the object speed, and can set the reachable range shape based on the object type. can do.

When the process of the flowchart shown in FIG. 3 is repeatedly performed, the time-series determination unit 115 determines the object position when the object reachability range estimated by the object position estimation unit 114 is estimated (step S109). Using the reach range estimated by the estimation unit 114, processing for integrating position information of a plurality of objects derived from the same object is performed (step S104). That is, the time-series determination unit 115 determines whether or not the object whose position is calculated by the movement information extraction unit 111 is the same as the object whose reach range is estimated in the reach range. Specifically, the time-series determination unit 115 performs determination using the arrival range estimated by the object position estimation unit 114 as a threshold for the position information obtained from the movement information extraction unit 111 and provided with the reflection point ID. Filter the position information used for the. The time series determination unit 115 identifies the position information obtained as a result of the filtering process from the position information of the same object, and performs the position update process in order from the closest distance. In the position update process, the time series determination unit 115 may calculate and obtain an average value of positions, or may estimate the position using a Kalman filter or the like. As described above, the time-series determination unit 115 manages position information for objects within the reach of the objects whose positions are calculated by the movement information extraction unit 111. Thereby, when the object is detected in a wide range by the sensor 101, the time-series determination unit 115 efficiently selects only necessary object position information by limiting the processing target region to the reachable range, Processing load can be reduced. In addition, since the reach range is predicted, the time series determination unit 115 can follow a change in position due to the movement of the object.

When the reach range is not estimated by the object position estimation unit 114, the time series determination unit 115 uses the position, speed, etc. of the object at the previous reception time as a model, for example, using constant velocity linear motion as a model. The position of the object at the time may be estimated, and if the difference between the estimated position of the object and the actual position of the object is within a prescribed threshold value, the same object may be estimated. The time series determination unit 115 may estimate the position of the object at the current reception time using a Kalman filter or the like. When the position difference is larger than the prescribed threshold value, the time series determination unit 115 determines that the objects whose positions are compared are different objects.

In the present embodiment, the information distribution unit 116 distributes position information to the vehicles 201 to 203. However, together with the position information, the reach range estimated by the object position estimation unit 114, the object attribute detection unit The attribute information generated at 112 may be distributed.

Further, in the present embodiment, the time series determination unit 115 may perform processing at the same processing cycle as the information integration unit 113, or may perform processing each time position information is acquired from the movement information extraction unit 111. Good.

In the present embodiment, it is assumed that the driving support system 400 is installed on the roadside, but the driving support device 100 and the

sensors

101 and 102 of the driving support system 400 are mounted on an automatic driving vehicle. May be. The driving assistance device 100 generates position information of the mounted autonomous driving vehicle.

Next, the hardware configuration of the driving support device 100 will be described. FIG. 8 is a diagram illustrating an example of a hardware configuration of the driving support apparatus 100 according to the present embodiment. The driving support apparatus 100 includes a CPU (Central Processing Unit) 40, a ROM (Read Only Memory) 41, a RAM (Random Access Memory) 42, a sensor IF (Interface) 43, a communication IF 44, and a memory 45. Are connected to each other by a bus 46. The CPU 40 controls the entire driving support device 100. The ROM 41 stores programs such as a boot program, a communication program, and a data analysis program. The RAM 42 is used as a work area for the CPU 40. The sensor IF 43 is connected to various sensors such as a radar and a lidar, and functions as an interface between the various sensors and the CPU 40. The communication IF 44 is connected to the autonomous driving vehicle via radio and functions as an interface between the autonomous driving vehicle and the CPU 40. A communication network such as a LAN (Local Area Network) may exist between the sensor IF 43 and the various sensors and between the communication IF 44 and the autonomous driving vehicle. One or both of the sensor IF 43 and the communication IF 44 are also connected to a communication network such as the Internet, and function as an interface between the communication network and the CPU 40.

The functions of the reception unit 110, the movement information extraction unit 111, the object attribute detection unit 112, the information integration unit 113, the object position estimation unit 114, and the time series determination unit 115 are software, firmware, or a combination of software and firmware. Realized. Software or firmware is described as a program and stored in the memory 45. The CPU 40 implements the functions of each unit by reading and executing the program stored in the memory 45.

As described above, according to the present embodiment, the driving support device 100 selects a plurality of objects derived from the same object as one object among the objects whose positions are calculated by the sensor 101 using the radar. The position information is managed, and the position information and the attribute information of the object detected by the sensor 102 that acquires the image information are integrated. The driving support device 100 uses the integrated information to estimate the arrival range of the object at the next reception time when the first sensor information is received from the sensor 101, and processes the object detected by the sensor 101 in the arrival range. I decided to make it. Thereby, the driving assistance apparatus 100 can improve the detection accuracy of the object which exists around the driving assistance apparatus 100, suppressing processing load.

The configuration described in the above embodiment shows an example of the contents of the present invention, and can be combined with another known technique, and can be combined with other configurations without departing from the gist of the present invention. It is also possible to omit or change the part.

100 driving support device, 101, 102 sensor, 103 wireless communication device, 110 receiving unit, 111 movement information extracting unit, 112 object attribute detecting unit, 113 information integrating unit, 114 object position estimating unit, 115 time series determining unit, 116 information Distribution unit, 117 storage unit, 118 transmission unit, 201-203,300 vehicle, 204 bicycle, 205-207 pedestrian, 400 driving support system.

Claims

A receiving unit that receives first sensor information from a first sensor using a radar and receives second sensor information from a second sensor that acquires image information;
A movement information extraction unit that calculates a first position, which is the position of the object detected by the first sensor, using the first sensor information;
A time-series determination unit that manages position information with a plurality of objects derived from the same object as one object among the objects whose first position is calculated by the movement information extraction unit;
Attribute information including a determination result by determining the size of the object detected by the second sensor, the second position that is the position of the object, and the type of the object using the second sensor information An object attribute detection unit for generating
An information integration unit that integrates and manages the object position information and the attribute information managed by the time-series determination unit;
An object position estimation unit that estimates the reach of the object at a reception time at which the first sensor information is received next, using the information of the object unit integrated by the information integration unit;
With
The time series determination unit manages position information for an object within the reachable range among objects whose first position is calculated by the movement information extraction unit;
A driving support device characterized by that.
The first sensor information includes a distance from the first sensor to the object, and an angle indicating a direction of the object with respect to a directing direction of the first sensor. At least one of the traveling directions of the object is included,
The driving support device according to claim 1, wherein
The image information included in the second sensor information is a still image or a moving image.
The driving support apparatus according to claim 1 or 2, wherein
The object attribute detection unit determines the time at which the object is detected, the second position, and the type of the object using the still image or the moving image, and further determines the full width of the object as the size of the object. And determining the total length of the object, and generating the attribute information,
The driving support apparatus according to claim 3, wherein
The object position estimation unit sets the size of the reach based on the size of the object, the type of the object, and the speed of the object;
The driving support apparatus according to any one of claims 1 to 4, wherein
The object position estimation unit sets the shape of the reach based on the type of the object,
The driving support device according to any one of claims 1 to 5, wherein
The information integration unit compares the second position included in the attribute information with the position of the object managed by the time series determination unit, and if the difference in position is equal to or less than a prescribed threshold value, The objects whose positions are compared are determined as the same object.
The driving support device according to any one of claims 1 to 6, wherein
The information integration unit generates position information to be distributed using the position information of the object managed by the time series determination unit or the information of the integrated object unit,
The driving support device further includes:
An information distribution unit comprising: a storage unit that stores the position information acquired from the information integration unit; and a transmission unit that transmits the position information to an autonomous vehicle.
The driving support apparatus according to any one of claims 1 to 7, further comprising:
The driving support device according to any one of claims 1 to 8,
A first sensor using a radar;
A second sensor for acquiring image information;
A driving support system comprising:
A receiving unit that receives first sensor information from a first sensor using a radar and receives second sensor information from a second sensor that acquires image information;
A second step in which a movement information extraction unit calculates a first position, which is a position of an object detected by the first sensor, using the first sensor information;
A third step in which the time-series determination unit manages the position information by using, as one object, a plurality of objects derived from the same object among the objects whose first position is calculated by the movement information extraction unit;
The object attribute detection unit determines the size of the object detected by the second sensor, the second position that is the position of the object, and the type of the object, using the second sensor information. A fourth step of generating attribute information including the determination result;
A fifth step in which the information integration unit integrates and manages the position information of the object managed by the time-series determination unit and the attribute information in units of objects;
A sixth step in which the object position estimating unit estimates the reach of the object at the reception time at which the first sensor information is next received using the information of the object unit integrated by the information integration unit;
Including
When the steps from the first step to the sixth step are repeatedly performed, in the third step, the time-series determination unit includes the objects whose first position is calculated by the movement information extraction unit, Managing location information for objects within the reach;
A driving support method characterized by the above.