Vehicle control device, vehicle control method, and vehicle control computer program

Abstract

A vehicle control device, a vehicle control method, and a computer program for vehicle control, wherein the vehicle control device determines whether there is a possibility that the position of the detected own lane relative to one end of a road on which a vehicle (10) is traveling is different from the actual position, estimates a first required time required for an action required to be performed before the vehicle (10) arrives at a predetermined location when the detected position of the own lane relative to one end of the road is correct, and estimates a second required time required for the action when the detected position of the own lane is assumed to deviate from the actual position by a predetermined number of lanes, and sets a timing for starting the action based on the second required time so that the action ends before the vehicle (10) arrives at the predetermined location if there is a possibility that the detected position of the own lane is different from the actual position and the second required time is longer than the first required time.

Description

Technical Field

[0001] The present invention relates to a vehicle control device, a vehicle control method, and a computer program for vehicle control. Background Art

[0002] When a vehicle is under autonomous driving control, the vehicle's control device performs positioning processing to detect the lane in which the vehicle is traveling (hereinafter sometimes referred to as the host lane). Based on the results of the positioning processing, the vehicle controls the vehicle, such as changing lanes, as needed. Therefore, in order to properly perform autonomous driving control of the vehicle, a technology for highly accurate host lane detection has been proposed (see Japanese Patent Application Laid-Open No. 2019-117059).

[0003] The electronic device disclosed in Japanese Patent Application Laid-Open No. 2019-117059 determines the lane in which the vehicle is traveling by referring to the detected vehicle position and lane information contained in map data. The electronic device then determines whether an error has occurred in determining the vehicle's lane position based on the lane information in the map data. If an error has occurred, the electronic device reduces the reliability of the determined lane position.

[0004] In addition, a technology has been proposed for executing lane change control based on another vehicle traveling in a lane adjacent to the host lane (hereinafter sometimes simply referred to as an adjacent lane) (see Japanese Patent Application Laid-Open No. 2020-45038).

[0005] In the vehicle control method disclosed in Japanese Patent Application Laid-Open No. 2020-45038, sensors detecting the surrounding conditions of a host vehicle acquire detection data of multiple other vehicles traveling in an adjacent lane. A first other vehicle and a second other vehicle located behind the first other vehicle are identified from the multiple other vehicles. In this vehicle control method, a lane-changeable area is defined in the adjacent lane between the first and second other vehicles, where the host vehicle can perform a lane change. A lane change starting point, where the lane change is initiated, is set to a position in the driving lane and forward of the center of the lane-changeable area along the driving lane. Furthermore, in this vehicle control method, from the time the host vehicle's position reaches the lane change starting point, lane change control is executed to cause the host vehicle to change lanes from the driving lane to the adjacent lane, decelerating the vehicle from the lane change starting point toward the lane-changeable area. Summary of the Invention

[0006] Depending on the circumstances, the vehicle control device may also have difficulty accurately detecting the vehicle's lane in the technology disclosed in Japanese Patent Application Laid-Open No. 2019-117059. For example, if lane markings are unclear, accurate lane detection is difficult. Consequently, if lane detection is difficult, it may be impossible to properly execute the intended vehicle control, such as the lane change disclosed in Japanese Patent Application Laid-Open No. 2020-45038.

[0007] Therefore, an object of the present invention is to provide a vehicle control device that can terminate a predetermined vehicle behavior before the vehicle reaches a predetermined position even if a detection result of a host lane is inaccurate.

[0008] According to one embodiment, a vehicle control device is provided. The vehicle control device comprises: a lane detection unit for detecting a lane in which the vehicle is traveling among a plurality of lanes by comparing sensor signals indicating the surroundings of the vehicle obtained by sensors mounted on the vehicle with a map containing information related to a plurality of lanes set on the road on which the vehicle is traveling; a false detection determination unit for determining whether there is a possibility that the position of the detected lane relative to one end of the road on which the vehicle is traveling is different from the actual position; a determination unit for determining, based on the vehicle's driving route, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location a predetermined distance forward from the vehicle's current location; an estimation unit for estimating a first required time or a first required distance required for the action determined when the position of the detected lane relative to one end of the road on which the vehicle is traveling is assumed to be correct, and a first required time or a first required distance required for the action determined when the position of the detected lane relative to one end of the road on which the vehicle is traveling is assumed to be correct, and a first required time or a first required distance required for the action determined when the position of the detected lane relative to one end of the road on which the vehicle is traveling is assumed to be different from the actual position. a second required time or a second required distance required for an action determined when the vehicle is away from a predetermined number of lanes; and a control unit, when there is a possibility that the position of the detected own lane relative to one end of the road on which the vehicle is traveling is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, setting the timing for starting the action based on the second required time or the second required distance in such a manner that the action determined before the vehicle arrives at the predetermined location is completed. On the other hand, when there is no possibility that the position of the detected own lane relative to one end of the road on which the vehicle is traveling is different from the actual position, or when the first required time or the first required distance is longer than the second required time or the second required distance, setting the timing for starting the action based on the first required time or the first required distance in such a manner that the action determined before the vehicle arrives at the predetermined location is completed.

[0009] Preferably, in the vehicle control device, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when construction information received from other equipment indicates that road construction is being carried out within a predetermined period including the current moment in an interval from the current position of the vehicle to a predetermined distance forward.

[0010] Alternatively, preferably, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the last update date and time of the map is more than a predetermined period before the current time, or when the structure of the road in the section from the current position of the vehicle to a predetermined distance ahead is different between the map and a route search map used by the navigation device in searching for the vehicle's traveling route.

[0011] Alternatively, it is also preferred that the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the driving trajectory of other vehicles in the interval from the current position of the vehicle to a predetermined distance forward passes through an area in the map where it is impossible for the vehicle to travel.

[0012] Alternatively, it is also preferred that the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position before a certain time has passed since the start of detection of the own lane, or until the vehicle has traveled a certain distance since the start of detection of the own lane.

[0013] Moreover, preferably, the false detection determination unit determines that there is a possibility that the position of the detected own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the number of lanes indicated by the map or sensor signal at the current position of the vehicle is greater than a predetermined number and the detected own lane is within a predetermined range from the centers of the multiple lanes at the current position.

[0014] In addition, preferably, the false detection determination unit determines that there is a possibility that the position of the detected lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the number of lanes at the current position has changed by more than a predetermined number relative to the number of lanes at the position of the vehicle a predetermined time ago.

[0015] In addition, preferably, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the position of a predetermined feature (object on the ground) detected based on the sensor signal is different from the position of the corresponding feature on the map based on the detected position of the own lane.

[0016] In addition, preferably, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the degree of certainty of the predetermined feature detected based on the sensor signal is below a predetermined confidence threshold.

[0017] In addition, preferably, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when it is unable to detect other vehicles traveling within the following range within a certain period of time. The range is a range represented by the sensor signal that is equivalent to an area on a map where the vehicle can travel based on the detected position of the own lane.

[0018] In addition, preferably, the false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road on which the vehicle is traveling is different from the actual position when the current position of the vehicle is within a predetermined range from a pre-designated false detection danger point in the map.

[0019] According to another embodiment, a vehicle control method is provided. The vehicle control method includes: detecting a lane in which the vehicle is traveling among the plurality of lanes by comparing sensor signals indicating the surroundings of the vehicle obtained by sensors mounted on the vehicle with a map containing information related to a plurality of lanes provided on a road on which the vehicle is traveling; determining whether there is a possibility that the position of the detected lane relative to one end of the road on which the vehicle is traveling is different from the actual position; determining, based on the vehicle's travel route, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location a predetermined distance forward from the vehicle's current location; estimating a first required time or a first required distance required for the action determined assuming that the position of the detected lane relative to one end of the road on which the vehicle is traveling is correct; and estimating a first required time or a first required distance required for the action determined assuming that the position of the detected lane relative to one end of the road on which the vehicle is traveling deviates by a predetermined number of lanes from the actual position. The method further comprises setting a second required time or a second required distance required for the action determined in the case of a quantity; if there is a possibility that the position of the detected own lane relative to one end of the road traveling on the vehicle is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, setting the timing for starting the action based on the second required time or the second required distance in such a manner that the action determined is completed before the vehicle arrives at the predetermined location; and if there is no possibility that the position of the detected own lane relative to one end of the road traveling on the vehicle is different from the actual position, or if the first required time or the first required distance is longer than the second required time or the second required distance, setting the timing for starting the action based on the first required time or the first required distance in such a manner that the action determined is completed before the vehicle arrives at the predetermined location.

[0020] According to another embodiment, a computer program for controlling a vehicle is provided. The computer program includes instructions for causing a processor mounted on the vehicle to execute the following processing, the processing comprising: detecting a lane in which the vehicle is traveling among a plurality of lanes by comparing sensor signals representing the vehicle's surroundings obtained by sensors mounted on the vehicle with a map including information related to a plurality of lanes provided on a road on which the vehicle is traveling; determining whether there is a possibility that the position of the detected lane relative to one end of the road on which the vehicle is traveling differs from the actual position; determining, based on the vehicle's travel route, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location a predetermined distance forward from the vehicle's current location; estimating a first required time or a first required distance required for the action determined assuming that the position of the detected lane relative to one end of the road on which the vehicle is traveling is correct, and estimating a first required time or a first required distance required for the action determined assuming that the position of the detected lane relative to one end of the road on which the vehicle is traveling is correct. A second required time or a second required distance required for an action determined when the actual position deviates from the predetermined lane number; when there is a possibility that the position of the lane detected relative to one end of the road where the vehicle is traveling is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, the timing for starting the action is set based on the second required time or the second required distance so that the action determined is completed before the vehicle reaches the predetermined location; when there is no possibility that the position of the lane detected relative to one end of the road where the vehicle is traveling is different from the actual position, or when the first required time or the first required distance is longer than the second required time or the second required distance, the timing for starting the action is set based on the first required time or the first required distance so that the action determined is completed before the vehicle reaches the predetermined location.

[0021] The vehicle control device of the present invention can achieve the effect of "ending the planned vehicle action before the vehicle reaches the planned position even if the detection result of the own lane is inaccurate." BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a diagram showing an example of the relationship between the detection result of the own lane and the start timing of a predetermined action.

[0023] Figure 2 This is a schematic diagram of the vehicle control system in which the vehicle control device is installed.

[0024] Figure 3 This is a hardware configuration diagram of an electronic control device as one embodiment of a vehicle control device.

[0025] Figure 4 This is a functional block diagram of a processor in an electronic control device related to vehicle control processing.

[0026] Figure 5A This is a diagram showing an example in which the detected position of the own lane relative to one end of the road is different from the actual position.

[0027] Figure 5B This is a diagram showing an example in which the detected position of the own lane relative to one end of the road is different from the actual position.

[0028] Figure 6 It is a workflow diagram of vehicle control processing. DETAILED DESCRIPTION

[0029] A vehicle control device, a vehicle control method executed by the vehicle control device, and a vehicle control computer program will be described below with reference to the accompanying drawings. The vehicle control device detects the vehicle's current lane among multiple lanes by comparing sensor signals representing the vehicle's surroundings obtained by sensors mounted on the vehicle with a map containing information regarding multiple lanes on the road on which the vehicle is traveling. The vehicle control device also determines whether there is a possibility that the detected position of the vehicle's current lane relative to one end of the road on which the vehicle is traveling differs from its actual position. Furthermore, based on at least one of the vehicle's current position and the vehicle's destination, the vehicle control device determines a vehicle action that is required to be performed before the vehicle reaches a predetermined location a predetermined distance forward of the vehicle's current position. Furthermore, the vehicle control device estimates a first required time or a first required distance to complete the determined action, assuming the detected position of the vehicle's current lane is correct. Furthermore, the vehicle control device estimates a second required time or a second required distance to complete the determined action, assuming the detected position of the vehicle's current lane actually deviates by a predetermined number of lanes. Furthermore, when there is a possibility that the detected position of the own lane differs from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, the vehicle control device determines the timing for starting the determined action based on the second required time or the second required distance. Thus, even if the detected position of the own lane is inaccurate, the vehicle control device can complete the determined vehicle action before the target position or timing.

[0030] Figure 1 : is a diagram showing an example of the relationship between the detection result of the own lane and the start timing of the scheduled action. Figure 1In the example shown, vehicle 10 is traveling on a six-lane road 100. The planned action is to "change lanes to lane 111 at the left end of road 100 before point P1, where road 101 branches from the left end of road 100 toward the destination." Vehicle 10 is traveling in lane 112, the second lane from the right end of road 100. In this case, vehicle 10 needs to make four lane changes before reaching lane 111. However, assume that lane 113, adjacent to the left of lane 112, is mistakenly detected as the vehicle's own lane. In this case, it is sufficient to recognize that vehicle 10 has made three lane changes before reaching lane 111. Therefore, if vehicle 10 starts lane changes at point P2, assuming it is traveling in lane 113, in order to reach lane 111 before branching point P1, there is a possibility that vehicle 10 may not have reached lane 111 by the time it reaches branching point P1.

[0031] Therefore, in this embodiment, when lane 113 is detected as the vehicle's lane, the required time l1 or required distance d1 required to change lanes from lane 113 to lane 111 is estimated. Furthermore, assuming that the vehicle's lane position is actually one lane off, and vehicle 10 is traveling in lane 112, the required time l2 or required distance d2 required to change lanes from lane 112 to lane 111 is estimated. The timing for starting the lane change is then set based on the longer of the required time l1 or required distance d1 and the required time l2 or required distance d2. In this example, the number of lane changes is one more, and the required time l2 or required distance d2 is longer than the required time l1 or required distance d1. Therefore, to ensure that vehicle 10 moves to lane 111 at the left end before point P1, the lane change is initiated at point P3, which is immediately before point P2. Therefore, even if the detected position of the vehicle's lane relative to one end of road 100 differs from the actual position, the lane change to the target lane 111 is completed before vehicle 10 reaches point P1.

[0032] Figure 2 This is a schematic diagram of the vehicle control system in which the vehicle control device is installed. Figure 3This is a hardware configuration diagram of an electronic control device as an embodiment of a vehicle control device. In this embodiment, a vehicle control system 1 that is mounted on a vehicle 10 and controls the vehicle 10 includes a camera 2, a GPS receiver 3, a navigation device 4, a wireless communicator 5, a storage device 6, and an electronic control unit (ECU) 7 as an example of a vehicle control device. The camera 2, the GPS receiver 3, the navigation device 4, the wireless communicator 5, and the storage device 6 are connected to the ECU 7 in a communicative manner via an in-vehicle network that complies with a standard such as a controller area network. In addition, the vehicle control system 1 may also include a distance sensor (not shown) such as a laser radar (LiDAR) or a radar that measures the distance from the vehicle 10 to objects existing around the vehicle 10.

[0033] The camera 2 is an example of a sensor that generates a sensor signal representing the surroundings of the vehicle 10. It has a two-dimensional detector composed of an array of photoelectric conversion elements sensitive to visible light, such as CCD or C-MOS, and an imaging optical system that forms an image of the area to be photographed on the two-dimensional detector. Furthermore, the camera 2 is installed, for example, in the interior of the vehicle 10 so as to face the front of the vehicle 10. The camera 2 photographs the area in front of the vehicle 10 at a predetermined shooting cycle (for example, 1 / 30 second to 1 / 10 second) and generates an image showing the area in front. The image obtained by the camera 2 is an example of a sensor signal. In addition, a plurality of cameras with different shooting directions or focal lengths may be provided in the vehicle 10.

[0034] Whenever the camera 2 generates an image, it outputs the generated image to the ECU 7 via the in-vehicle network.

[0035] The GPS receiver 3 receives GPS signals from GPS satellites at predetermined intervals and measures the vehicle 10's own position based on the received GPS signals. Furthermore, the GPS receiver 3 outputs positioning information indicating the vehicle 10's own position measurement results based on the GPS signals to the navigation device 4 and the ECU 7 via the in-vehicle network at predetermined intervals. Alternatively, the vehicle 10 may include a receiver that receives positioning signals from satellites of other satellite positioning systems to measure the vehicle 10's own position, instead of the GPS receiver.

[0036] The navigation device 4 performs navigation processing for the vehicle 10 according to a navigation program running on the device. For example, the navigation device 4 initiates the navigation program in response to a driver's operation. When the vehicle 10's destination is input, the navigation device 4 searches for a route for the vehicle 10 from the vehicle 10's current location to the destination. In this process, the navigation device 4 searches for the route using a predetermined path search method, such as the Dykstra algorithm, by referring to a route search map (hereinafter sometimes referred to as a road map) stored on the device, which displays various road sections and their connections. The route includes, for example, information indicating the roads to the destination, the direction of travel at branch points on the route, and the locations of intersections for right and left turns. Furthermore, the navigation device 4 may utilize the vehicle 10's own position, based on the latest positioning results received from the GPS receiver 3, as the vehicle 10's current location.

[0037] When the navigation device 4 obtains the travel route of the vehicle 10 , it outputs information indicating the travel route to the ECU 7 via the in-vehicle network.

[0038] The wireless communicator 5 follows a predetermined mobile communication standard and performs wireless communication between it and the wireless base station. The wireless communicator 5 receives traffic information indicating the traffic conditions of the road on which the vehicle 10 is traveling or its surroundings, or construction information indicating the implementation status of construction (for example, information based on a road traffic information communication system) from other devices via the wireless base station. Furthermore, the wireless communicator 5 outputs the received traffic information to the ECU 7 via the in-vehicle network. In addition, the construction information includes, for example, information related to the location and time period for implementing road construction. In addition, the wireless communicator 5 can receive a high-precision map of a predetermined area around the current position of the vehicle 10 from a map server via the wireless base station, which is used in automatic driving control, and output the received high-precision map to the storage device 6.

[0039] The storage device 6 includes, for example, a hard disk drive, a nonvolatile semiconductor memory, or an optical recording medium and its access device. Furthermore, the storage device 6 stores a high-precision map. A high-precision map is an example of a map that includes information related to multiple lanes on a road. For example, a high-precision map includes information indicating the number of lanes, road markings such as lane dividing lines or stop lines, and road signs for each road within a predetermined area represented on the high-precision map.

[0040] Furthermore, the storage device 6 may also include a processor for executing high-precision map update processing and processing related to the high-precision map reading request from the ECU 7. Furthermore, the storage device 6, for example, transmits a high-precision map acquisition request and the current position of the vehicle 10 to the map server via the wireless communicator 5 whenever the vehicle 10 moves a predetermined distance. In addition, the storage device 6 may receive a high-precision map of a predetermined area around the current position of the vehicle 10 from the map server via the wireless communicator 5. In addition, when the storage device 6 receives a high-precision map reading request from the ECU 7, it cuts out a range that includes the current position of the vehicle 10 and is relatively narrower than the above-mentioned predetermined area from the stored high-precision map, and outputs it to the ECU 7 via the in-vehicle network.

[0041] ECU 7 performs automatic driving control of vehicle 10. In this embodiment, ECU 7 compares the image obtained by camera 2 with a high-precision map to detect the lane in which vehicle 10 is traveling and determines whether there is a possibility that the detected position of the lane relative to one end of the road on which vehicle 10 is traveling differs from the actual position. Based on the detected lane, ECU 7 then determines the timing for initiating "an action of vehicle 10 required to be performed before vehicle 10 reaches a predetermined location a predetermined distance forward from the current position of vehicle 10." In this case, if there is a possibility that the detected position of the lane differs from the actual position, ECU 7 determines the timing for initiating the determined action, taking into account the possibility that the detected position of the lane has actually deviated.

[0042] like Figure 3 As shown, the ECU 7 includes a communication interface 21, a memory 22, and a processor 23. The communication interface 21, the memory 22, and the processor 23 may be configured as separate circuits, or may be integrally configured as a single integrated circuit.

[0043] The communication interface 21 includes an interface circuit for connecting the ECU 7 to the in-vehicle network. Furthermore, whenever the communication interface 21 receives an image from the camera 2, it transmits the received image to the processor 23. Furthermore, whenever the communication interface 21 receives positioning information from the GPS receiver 3, it transmits the positioning information to the processor 23. Furthermore, when the communication interface 21 receives a driving route from the navigation device 4, it transmits the driving route to the processor 23. Furthermore, when the communication interface 21 obtains information such as traffic information received by the wireless communicator 5 from other devices, it transmits the information to the processor 23. Furthermore, the communication interface 21 transmits a high-precision map read from the storage device 6 to the processor 23.

[0044] The memory 22 includes, for example, volatile semiconductor memory and non-volatile semiconductor memory. Furthermore, the memory 22 stores various data used in the vehicle control processing executed by the processor 23. For example, the memory 22 stores parameters such as the focal length, shooting direction, and installation position of the camera 2, as well as various parameters used to determine the identifiers used for object detection, such as those used in detecting terrain features. Furthermore, the memory 22 stores the driving route, positioning information of the vehicle 10, images of the surrounding area of ​​the vehicle 10, and high-precision maps. Furthermore, the memory 22 temporarily stores various data generated during the vehicle control processing.

[0045] The processor 23 includes one or more CPUs (Central Processing Units) and their peripheral circuits. The processor 23 may also include other arithmetic circuits such as a logical operation unit, a numerical operation unit, or a graphics processing unit. Furthermore, the processor 23 executes vehicle control processing for the vehicle 10 at predetermined intervals.

[0046] Figure 4 This is a functional block diagram of the processor 23 related to vehicle control processing. The processor 23 includes a lane detection unit 31, a behavior determination unit 32, a false detection determination unit 33, an estimation unit 34, and a control unit 35. Each of the aforementioned components of the processor 23 is, for example, a functional module implemented by a computer program running on the processor 23. Alternatively, each of the aforementioned components of the processor 23 may be a dedicated computing circuit provided in the processor 23.

[0047] The own lane detection unit 31 detects the own lane in which the vehicle 10 is traveling by comparing an image (hereinafter sometimes referred to as an image) generated by the camera 2 and representing the surroundings of the vehicle 10 with a high-precision map. For example, the own lane detection unit 31 assumes the position and posture of the vehicle 10 and projects features on or around the road detected from the image onto the high-precision map, or projects features on or around the road around the vehicle 10 represented in the high-precision map onto the image. In addition, features on or around the road can be, for example, road markings such as lane demarcation lines or stop lines, or curbs. Furthermore, the own lane detection unit 31 infers the position and posture of the vehicle 10 when the features detected from the image are most consistent with the features represented on the high-precision map as the own position of the vehicle 10.

[0048] The own lane detection unit 31 determines the position at which features are projected on the high-precision map or image using the assumed initial values ​​for the position and posture of the vehicle 10 and parameters of the camera 2 such as focal length, installation height, and shooting direction. Furthermore, the initial values ​​for the position and posture of the vehicle 10 are the vehicle 10 position measured by the GPS receiver 3 or the position and posture of the vehicle 10 estimated during the previous own lane detection, corrected using odometry information. Furthermore, the own lane detection unit 31 calculates the degree of consistency (e.g., normalized cross-correlation value) between features on or around the road detected from the image and features on or around the road shown on the high-precision map.

[0049] The host lane detection unit 31 repeatedly performs the above-described process while changing the assumed position and posture of the vehicle 10. Furthermore, the host lane detection unit 31 may estimate the assumed position and posture at which the degree of agreement is maximized as the actual host position of the vehicle 10. Furthermore, the host lane detection unit 31 may refer to the high-precision map and identify the lane containing the vehicle 10's own position as the host lane in which the vehicle 10 is currently traveling.

[0050] In addition, the lane detection unit 31 can detect the feature by inputting the image into a recognizer that has been pre-learned to detect the feature to be detected from the image. In this case, the lane detection unit 31 can use, for example, a single shot multibox detector (SSD) or a deep neural network (DNN) with a convolutional neural network (CNN) architecture such as a fast regional convolutional neural network (FasterR-CNN) as a recognizer used in the detection of the feature. The above-mentioned recognizer can also output a confidence level indicating the degree of certainty of the feature for each detected feature. In addition, the recognizer detects an area on the image where the confidence level calculated for the predetermined feature is higher than a predetermined detection threshold as an object area representing the predetermined feature.

[0051] The own lane detection unit 31 notifies the behavior determination unit 32 , the false detection determination unit 33 , the estimation unit 34 , and the control unit 35 of information indicating the detected own lane.

[0052] The action determination unit 32 determines, based on the driving route of the vehicle 10, an action that the vehicle 10 is required to perform before the vehicle 10 reaches a predetermined location (hereinafter sometimes referred to as a target location) a predetermined distance forward from the current location of the vehicle 10. The determined action of the vehicle 10 includes an action that changes to a driving state different from the current driving state of the vehicle 10. For example, the action of the vehicle 10 includes a lane change associated with merging or diverging into a lane for traveling to the destination along the driving route, and a lane change for moving to a lane heading in a specific direction (e.g., a right-turn lane or a left-turn lane).

[0053] For example, the action determination unit 32 refers to a high-precision map to determine whether the road on which the vehicle 10 is traveling includes a branch point while the vehicle 10 is traveling a predetermined distance (e.g., 1 km to 5 km) from its current position to reach its destination along the driving route. Alternatively, the action determination unit 32 may determine the presence of such a branch point by referring to a road map similar to the road map used by the navigation device to search for the driving route, instead of a high-precision map. In this case, the road map may be pre-stored in the memory 22. If such a branch point exists, the action determination unit 32 determines whether the lane in which the vehicle 10 is currently traveling is different from the lane leading to the destination from the branch point (hereinafter sometimes referred to as the target lane). Furthermore, if the vehicle 10 is traveling in a different lane from the target lane, the action determination unit 32 determines a lane change from the vehicle 10 to the target lane as the requested action of the vehicle 10. Furthermore, the action determination unit 32 sets the branch point, or a location that is closer to the front of the branch point at a predetermined offset distance from the branch point when viewed from the current position of the vehicle 10, as the target point. Note that the current position of the vehicle 10 may be the position of the vehicle 10 estimated by the own lane detection unit 31 .

[0054] Similarly, the action determination unit 32 refers to the high-precision map or the road map to determine whether there is an intersection where the vehicle 10 turns right or left to reach the destination during the period in which the vehicle 10 travels a predetermined distance from the current position to reach the destination along the driving route. If such an intersection exists, the action determination unit 32 determines whether the lane in which the vehicle 10 is currently traveling is different from the target lane in which the vehicle 10 turns right or left to reach the destination at the intersection. Furthermore, if the lane is different from the target lane, the action determination unit 32 determines a lane change from the lane to the target lane as the required action of the vehicle 10. In addition, the action determination unit 32 sets the intersection, or a location that is closer to the front side of the intersection at a predetermined offset distance when viewed from the current position of the vehicle 10, as the target location.

[0055] The action identification unit 32 generates information indicating the identified action. The information indicating the identified action includes information indicating the type of the action and information indicating the target location. The action identification unit 32 then notifies the estimation unit 34 and the control unit 35 of the information indicating the identified action.

[0056] The false detection determination unit 33 determines whether there is a possibility that the position of the own lane detected by the own lane detection unit 31 with respect to one end of the road on which the vehicle 10 is traveling is different from the actual position.

[0057] Figure 5A and Figure 5B Each of the figures shows an example where the detected position of the own lane is different from the actual position. Figure 5A In the example shown, the vehicle 10 is traveling in the lane 501 that is the second from the right relative to the direction of travel of the vehicle 10, among the multiple lanes provided on the road 500. However, the lane 502 adjacent to the lane 501 is mistakenly detected as the own lane. Therefore, the position of the detected own lane is different from the actual position of the own lane. Such a false detection of the own lane may occur, for example, when there are many lanes provided on the road on which the vehicle 10 is traveling, or when features such as lane markings provided on the road are erased and the discrimination of the features becomes difficult. In addition, when the time elapsed since the detection of the own lane is started is short, such as when the vehicle 10 has just entered an area covered by a high-precision map, a false detection of the own lane may also occur.

[0058] exist Figure 5B In the example shown, vehicle 10 is traveling in lane 511, the rightmost lane relative to the vehicle 10's direction of travel, among multiple lanes on road 510. Furthermore, in this example, lane 511 itself is detected as the vehicle's own lane. However, lane 512 at the left end of road 510 was added after the high-precision map used by vehicle 10 to detect its own lane was generated. Therefore, lane 512 is not included in the high-precision map of road 510. As a result, the number of lanes identified by the vehicle's own lane detection unit 31 between lane 511 and the left end of road 510 differs from the actual number of lanes between lane 511 and the actual left end of road 510. Consequently, the detected position of the vehicle's own lane relative to the left end of road 510 differs from its actual position. This discrepancy between the number of lanes detected to the left and right of the vehicle's own lane on the high-precision map and the actual environment can occur when the high-precision map does not reflect the latest road information or when road construction is underway.

[0059] Therefore, the false detection determination unit 33 determines whether there is a possibility that the detected position of the own lane relative to the end of the road on which the vehicle 10 is traveling differs from its actual position based on the structure of the road on which the vehicle 10 is traveling, the driving environment, the timing of generating or updating the high-precision map, and other factors. Specifically, the false detection determination unit 33 may determine whether there is a possibility that the detected position of the own lane differs from its actual position based on any of the determination processes described below. By executing the determination processes described below, the false detection determination unit 33 can accurately determine whether there is a possibility that the detected position of the own lane relative to the end of the road on which the vehicle 10 is traveling differs from its actual position. Furthermore, the false detection determination unit 33 does not need to execute all of the determination processes described below; it only needs to execute at least one of these determination processes. In the following, "the detected position of the own lane relative to the end of the road on which the vehicle 10 is traveling differs from its actual position" may be simply expressed as "the detected position of the own lane differs from its actual position."

[0060] For example, the false detection determination unit 33 refers to construction information received from another device via the wireless communicator 5. If the construction information indicates that "road construction is being carried out within a predetermined period, including the current time, between the current position of the vehicle 10 and a predetermined distance ahead," the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane differs from the actual position. This is because there is a possibility that the number of passable lanes has changed due to the road construction, causing the detected position of the own lane to differ from the actual position.

[0061] Furthermore, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane differs from the actual position if the high-precision map was last updated at least a predetermined period before the current time. Alternatively, the false detection determination unit 33 may determine that there is a possibility that the detected position of the own lane differs from the actual position if the road structure between the high-precision map and the road map differs from the current position of the vehicle 10 to a predetermined distance ahead. This is because the high-precision map may not accurately represent the road structure around the current position of the vehicle 10, and the detected position of the own lane may be incorrect.

[0062] Furthermore, the false detection determination unit 33 may determine that there is a possibility that the detected position of the own lane is different from the actual position if the driving trajectory of another vehicle traveling ahead of the vehicle 10 in the section from the current position of the vehicle 10 to a predetermined distance ahead passes through an area in the high-precision map where the vehicle is unlikely to travel. This is because the other vehicle is unlikely to actually travel in the area where the vehicle is unlikely to travel, and therefore the detected position of the own lane is likely to be incorrect.

[0063] In this case, the false detection determination unit 33 detects other vehicles traveling around the vehicle 10 based on a series of time-series images obtained by the camera 2. In this case, the false detection determination unit 33 can detect other vehicles around the vehicle 10 from the series of images by inputting the series of images to a recognizer that has been pre-trained to detect objects to be detected from images. The control unit 35 can use, for example, a DNN having a CNN-type architecture as such a recognizer.

[0064] The false detection determination unit 33 tracks other vehicles detected from a series of time-series images to determine the trajectory of the other vehicles. In this case, the false detection determination unit 33 applies an optical flow-based tracking process, such as the Lucas-Kanade method, to the object region representing the other vehicle of interest in the latest image captured by the camera 2 and the object region in the past image, thereby tracking the other vehicle represented by the object region. Therefore, the false detection determination unit 33 applies a feature point extraction filter, such as the SIFT (Scale-Invariant Feature Transform) or the Harris operator, to the object region of interest, extracting multiple feature points from the object region. Furthermore, the false detection determination unit 33 calculates the optical flow for each of the multiple feature points by determining the corresponding point in the object region in the past image using the applied tracking method. Alternatively, the false detection determination unit 33 may apply another tracking method suitable for tracking moving objects detected from images to the object region of interest in the latest image and the object region in the past image, thereby tracking the other vehicle represented by the object region.

[0065] The false detection determination unit 33 uses information such as the camera 2's installation position on the vehicle 10 to perform viewpoint conversion processing on each of the other vehicles being tracked, converting the other vehicle's in-image coordinates into coordinates on the bird's-eye view image (bird's-eye coordinates). The false detection determination unit 33 can estimate the position of the other vehicle detected at the time each image was acquired based on the vehicle 10's position and posture at the time each image was acquired, the estimated distance from the detected other vehicle, and the direction from the vehicle 10 toward the other vehicle. Alternatively, the false detection determination unit 33 can obtain the vehicle 10's position and posture from the lane detection unit 31. Furthermore, the false detection determination unit 33 can determine the direction from the vehicle 10 toward the other vehicle based on the position of the object area in the image containing the detected other vehicle and the direction of the camera 2's optical axis. Furthermore, the estimated distance from the vehicle 10 to the detected other vehicle can be calculated based on the ratio of the size of the area representing the other vehicle in the image to the reference size of the other vehicle in the image, assuming the distance from the other vehicle is the reference distance, and the actual size of the other vehicle in space. The reference distance, the reference size of the detected other vehicles on the image, and the size of the real space may be pre-stored in memory 22, for example. Furthermore, the position of the lower end of the object region is assumed to indicate the position where the other vehicle represented by the object region contacts the road surface. Therefore, the false detection determination unit 33 may estimate the distance to the other vehicle represented by the object region based on the orientation of the lower end of the object region from the camera 2 and the installation height of the camera 2.

[0066] The false detection determination unit 33 overlays the driving trajectory of another vehicle, calculated as described above, with the high-precision map to determine whether the trajectory crosses an area where the vehicle is unlikely to travel. Furthermore, if at least a portion of the driving trajectory calculated for any other vehicle overlaps with an area where the vehicle is unlikely to travel, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane differs from the actual position.

[0067] Furthermore, the false detection determination unit 33 may determine that there is a possibility that the detected position of the own lane is different from the actual position before a certain time has passed since the start of the own lane detection process, or until the vehicle 10 has traveled a certain distance since the start of the own lane detection process. This is because the detection accuracy of the own lane may be insufficient immediately after the start of the own lane detection process.

[0068] Furthermore, the false detection determination unit 33 determines whether the number of lanes indicated by the high-precision map or the image generated by the camera 2 at the current position of the vehicle 10 is greater than a predetermined number. The predetermined number is set to an arbitrary number greater than or equal to three, for example, 3 to 5. If the number of lanes is greater than or equal to the predetermined number and the detected own lane is within a predetermined range from the center of the multiple lanes at the current position of the vehicle 10, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane is different from the actual position. This is because the detected position of the own lane is likely to be different when the vehicle 10 is traveling near the center of a road with many lanes.

[0069] Furthermore, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane is different from the actual position when the number of lanes at the current position of the vehicle 10 has changed by a predetermined number or more compared to the number of lanes at the position of the vehicle 10 a predetermined time ago. At locations where the number of lanes on the road on which the vehicle 10 is traveling changes dramatically, the detected position of the own lane is likely to be different.

[0070] Furthermore, if the position of a predetermined feature detected from the image generated by the camera 2 differs from the position of the corresponding feature on the high-precision map based on the detected position of the own lane, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane differs from the actual position. This discrepancy between the position of the feature on the image and the position of the corresponding feature on the high-precision map may be caused by the fact that the estimated position of the vehicle 10 is inaccurate, resulting in the possibility that the estimated position of the own lane is incorrect. Furthermore, the false detection determination unit 33 can simply project the feature detected from the image onto the high-precision map using the same method as that used to estimate the position of the own vehicle described in the own lane detection unit 31, and compare the position of the feature on the image with the position of the corresponding feature on the high-precision map.

[0071] Furthermore, if the confidence level of a predetermined feature detected from an image generated by the camera 2 is below a predetermined confidence threshold, the false detection determination unit 33 determines that there is a possibility that the detected position of the own lane differs from the actual position. In this case, the confidence level can be the confidence level output by the identifier used by the own lane detection unit 31. Furthermore, the predetermined confidence threshold is preferably set to a value higher than the detection threshold used by the own lane detection unit 31 to detect features from the image. This is because the estimated position of the own lane may be erroneous due to insufficient detection accuracy of features surrounding the vehicle 10.

[0072] Furthermore, the false detection determination unit 33 detects other vehicles traveling within a range represented by the image generated by the camera 2, corresponding to the area where the vehicle 10 can travel on the high-precision map based on the detected position of the own lane. If the false detection determination unit 33 fails to detect such other vehicles within a certain period of time, it determines that there is a possibility that the detected position of the own lane differs from the actual position. Consider that other vehicles often travel in areas where the vehicle 10 can travel on the high-precision map. Therefore, if no other vehicles are detected within the range on the image corresponding to this area, there is a possibility that the vehicle 10's position is estimated incorrectly, and consequently, the position of the own lane is incorrect. Furthermore, the false detection determination unit 33 can simply project the area where the vehicle 10 can travel on the high-precision map onto the image using the same method as the vehicle position estimation described in the own lane detection unit 31, and determine the range on the image corresponding to this area.

[0073] Alternatively, the high-precision map may include information indicating locations where detection of the own lane is likely to fail (hereinafter referred to as misdetection risk locations). In this case, the misdetection determination unit 33 determines that there is a possibility that the detected location of the own lane is different from the actual location if the current location of the vehicle 10 is within a predetermined range from the pre-specified misdetection risk location in the high-precision map.

[0074] The false detection determination unit 33 notifies the estimation unit 34 and the control unit 35 of a determination result as to whether or not there is a possibility that the detected position of the own lane is different from the actual position.

[0075] When the action determination unit 32 notifies the driver of the determined action, the estimation unit 34 estimates a first required time, which is the time required for the determined action assuming the vehicle 10 is traveling in the detected own lane. Furthermore, if it is determined that the detected own lane relative to one end of the road on which the vehicle 10 is traveling may differ from the actual own lane position, the estimation unit 34 assumes that the detected own lane position is offset by a predetermined number of lanes relative to the actual position. The estimation unit 34 then estimates a second required time, which is the time required for the determined action assuming this assumption. In this embodiment, the estimation unit 34 estimates the second required time based on the probability of deviation from one lane, but the present invention is not limited to this and may also estimate the second required time based on the probability of deviation from two or more lanes. For example, the more lanes the road on which the vehicle 10 is traveling includes, the greater the number of lanes the estimation unit 34 specifies. Furthermore, if the detected position of the host lane is likely to deviate to either the left or right relative to the direction of travel of the vehicle 10, the estimating unit 34 may estimate the second required time by assuming that the detected position of the host lane has deviated by a predetermined number of lanes in the direction that results in a longer required time for the determined action. Alternatively, if the detected position of the host lane is likely to deviate only to one of the left and right directions relative to the direction of travel of the vehicle 10, the estimating unit 34 may estimate the second required time by assuming that the detected position of the host lane has deviated by a predetermined number of lanes in the direction that is likely to deviate. For example, if a three-dimensional structure such as a guardrail is shown adjacent to either the left or right side of the detected host lane in the image generated by the camera 2, the estimating unit 34 may estimate the second required time by assuming that the detected position of the host lane has deviated by a predetermined number of lanes in the direction opposite to the side where the structure is located.

[0076] For example, the estimation unit 34 estimates the first and second required times by referring to a required time table pre-stored in the memory 22, which indicates the relationship between the determined action and the required time required for the action. For example, if the determined action is one or more lane changes from the host lane to the target lane, as described above, the estimation unit 34 refers to the required time table to determine the required time for each lane change. The estimation unit 34 then estimates the first required time by multiplying the required time for each lane change by the number of lane changes required to move from the detected host lane to the target lane. Furthermore, the estimation unit 34 estimates the second required time by multiplying the required time for each lane change by the number of lane changes required to move from the host lane position at the time of the assumed departure to the target lane.

[0077] Furthermore, multiple required timetables may be prepared for each determined action based on the driving conditions of the vehicle 10. For example, a separate required timetable may be prepared based on the speed of the vehicle 10, the legal speed limit of the road on which the vehicle 10 is traveling, the number of other vehicles traveling in lanes adjacent to the detected lane, and the like. In this case, the estimation unit 34 selects the required timetable to be used by referring to the speed of the vehicle 10 measured by a speed sensor (not shown) installed on the vehicle 10, the legal speed limit at the current position of the vehicle 10 as shown on a high-precision map, or the number of other vehicles detected from images captured by the camera 2.

[0078] Alternatively, the estimating unit 34 may estimate a first required distance, the distance required for the vehicle 10 to travel for the action determined assuming the vehicle 10 is traveling in the detected own lane, instead of the first required time, or in addition to the first required time. Furthermore, the estimating unit 34 may estimate a second required distance, the distance required for the vehicle 10 to travel for the action determined assuming the position of the detected own lane deviates by a predetermined number of lanes, instead of the second required time, or in addition to the second required time. In this case, as in the above-described embodiment, the estimating unit 34 may estimate the first and second required distances by referring to a required distance table pre-stored in the memory 22 that indicates the relationship between the determined action and the required distance required for that action.

[0079] For example, if the determined action involves one or more lane changes from the host lane to the target lane, the estimation unit 34 refers to the required distance table to determine the required distance for each lane change. Furthermore, the estimation unit 34 determines the first required distance by multiplying the number of lane changes required to move from the host lane position to the target lane when the vehicle deviates from the lane. Furthermore, the estimation unit 34 may estimate the second required distance by multiplying the required distance for each lane change by the number of lane changes required to move from the host lane position to the target lane when the vehicle deviates from the lane. Furthermore, for two consecutive lane changes, the sections in which each lane change is performed may partially overlap. Therefore, the estimation unit 34 may correct the first and second required distances calculated above by subtracting the distance obtained by multiplying the overlapping distance between the sections in which each lane change is performed by the total number of lane changes performed, minus one. This overlap distance may be pre-stored in the memory 22.

[0080] Furthermore, multiple required distance tables may be prepared for each determined behavior based on the driving conditions of the vehicle 10. For example, separate required distance tables may be prepared based on the speed of the vehicle 10, the legal speed limit for the road on which the vehicle 10 is traveling, the number of other vehicles traveling in lanes adjacent to the detected lane, and the like. In this case, the estimation unit 34 selects the required distance table to be used by referring to the speed of the vehicle 10 measured by a speed sensor (not shown), the legal speed limit at the current position of the vehicle 10 as shown on a high-precision map, or the number of other vehicles detected from images captured by the camera 2.

[0081] The estimating unit 34 notifies the control unit 35 of the first and second required times or the first and second required distances.

[0082] The control unit 35 controls the vehicle 10 so that the determined action is completed before the vehicle 10 reaches the target location of the determined action. Therefore, if there is a possibility that the detected position of the vehicle's own lane relative to one end of the road on which the vehicle 10 is traveling differs from the actual position, the control unit 35 refers to the first and second required times or the first and second required distances. Furthermore, the control unit 35 sets the timing for starting the determined action (hereinafter sometimes referred to as the action start timing) based on the longer of the first and second required times or the longer of the first and second required distances.

[0083] Specifically, if the second required time is longer than the first required time, the control unit 35 sets the action start timing at a predetermined time point that is at least the second required time before the expected time for vehicle 10 to arrive at the destination. On the other hand, if the first required time is longer than the second required time, the control unit 35 sets the action start timing at a predetermined time point that is at least the first required time before the expected time for vehicle 10 to arrive at the destination. Furthermore, the control unit 35 calculates the distance from the current position of vehicle 10 to the destination by referring to a high-precision map or road map, and then divides this distance by the average speed of vehicle 10 over the most recent predetermined period to calculate the expected time.

[0084] Similarly, if the second required distance is longer than the first required distance, the control unit 35 sets the time when the distance from the current position of the vehicle 10 to the target location reaches the second required distance or the distance obtained by adding a predetermined offset distance to the second required distance as the action start timing. On the other hand, if the first required distance is longer than the second required distance, the control unit 35 sets the time when the distance from the current position of the vehicle 10 to the target location reaches the first required distance or the distance obtained by adding a predetermined offset distance to the first required distance as the action start timing. Furthermore, the control unit 35 may simply determine the distance from the current position of the vehicle 10 to the target location by referring to a high-precision map or a road map.

[0085] When it is determined that there is no possibility that the detected position of the own lane relative to one end of the road on which the vehicle 10 is traveling is different from the actual position, the control unit 35 may set the action start timing based on the first required time or the first required distance as described above.

[0086] When the action start timing arrives, the control unit 35 controls the various components of the vehicle 10 so that the vehicle 10 executes the determined action. Specifically, the control unit 35 generates one or more planned trajectories for the vehicle 10, extending from the vehicle 10's current position to a predetermined distance ahead (e.g., 500 meters to 1 kilometer), according to the determined action. The planned trajectories represent, for example, a set of target positions of the vehicle 10 at various times while the vehicle 10 travels through the predetermined interval. The control unit 35 then controls the various components of the vehicle 10 so that the vehicle 10 travels along these planned trajectories.

[0087] In the example above, if the determined action involves one or more lane changes from the vehicle's lane to the target lane, the control unit 35 generates a planned driving trajectory to complete the lane change. Therefore, the control unit 35 generates the planned driving trajectory to prevent collisions between objects surrounding the vehicle 10, particularly other vehicles traveling in the vehicle's lane and adjacent lanes to the target lane. To this end, the control unit 35 detects objects surrounding the vehicle 10 from a series of time-series images captured by the camera 2. The control unit 35 then tracks the objects detected from the series of time-series images and estimates the predicted trajectory of each object until a predetermined time has passed based on the trajectory obtained from the tracking results. In this case, the control unit 35 can determine the trajectory of objects surrounding the vehicle 10 by performing the same process as that used by the false detection determination unit 33 to determine the trajectories of other vehicles. Alternatively, the control unit 35 can obtain the trajectories of objects surrounding the vehicle 10 from the false detection determination unit 33. Furthermore, the control unit 35 may estimate the predicted trajectory of the object until a predetermined time later by executing a prediction process using a Kalman filter or a particle filter based on the trajectory of the object being tracked within the most recent predetermined period.

[0088] The control unit 35 generates a planned driving trajectory based on the predicted trajectory of each object being tracked so that the predicted value of the distance between each object and the vehicle 10 during the tracking process until a predetermined time is obtained becomes greater than a predetermined distance and a lane change is completed within the distance required for each lane change.

[0089] After setting a planned driving trajectory, the control unit 35 controls various components of the vehicle 10 to ensure that the vehicle 10 travels along the planned driving trajectory. For example, the control unit 35 calculates the acceleration of the vehicle 10 based on the planned driving trajectory and the current speed of the vehicle 10 measured by a vehicle speed sensor (not shown), and sets the accelerator position or braking force to achieve the desired acceleration. Furthermore, the control unit 35 calculates the fuel injection amount based on the set accelerator position and outputs a control signal corresponding to the fuel injection amount to the fuel injection device of the engine of the vehicle 10. Alternatively, the control unit 35 calculates the amount of power supplied to the motor based on the set accelerator position and controls the motor drive circuit to supply this power to the motor. Furthermore, the control unit 35 outputs a control signal corresponding to the set braking force to the brakes of the vehicle 10. Furthermore, when the vehicle 10's path is changed to ensure that the vehicle 10 travels along the planned driving trajectory, the control unit 35 calculates the steering angle of the vehicle 10 based on the planned driving trajectory. Then, the control unit 35 outputs a control signal corresponding to the steering angle to an actuator (not shown) that controls the steering wheels of the vehicle 10 .

[0090] Figure 6 This is a flowchart of the vehicle control process executed by the processor 23. The processor 23 may execute the vehicle control process according to the following flowchart at every predetermined cycle.

[0091] The own lane detection unit 31 of the processor 23 detects the own lane in which the vehicle 10 is traveling (step S101). In addition, the action determination unit 32 of the processor 23 determines an action of the vehicle 10 that is required to be performed before the vehicle 10 reaches a target location that is a predetermined distance forward from the current position of the vehicle 10 (step S102).

[0092] Furthermore, the false detection determination unit 33 of the processor 23 determines whether there is a possibility that the detected position of the own lane with respect to one end of the road on which the vehicle 10 is traveling is different from the actual position (step S103 ).

[0093] If it is determined that there is no possibility that the detected position of the own lane differs from the actual position (step S103 - No), the estimation unit 34 of the processor 23 assumes that the position of the own lane is correct. Based on this assumption, the estimation unit 34 then estimates the first required time l1 or first required distance d1 required for the determined action (step S104). The control unit 35 of the processor 23 then sets the action start timing based on the first required time l1 or first required distance d1 so that the determined action ends before the target location (step S105).

[0094] On the other hand, if it is determined that there is a possibility that the detected position of the own lane is different from the actual position (step S103 - YES), the estimating unit 34 assumes that the position of the own lane is correct and estimates the first required time l1 or the first required distance d1 (step S106). Furthermore, the estimating unit 34 assumes that the detected position of the own lane is deviated by a predetermined number of lanes and estimates the second required time l2 or the second required distance d2 required for the determined action (step S107).

[0095] The control unit 35 determines whether the second required time l2 is longer than the first required time l1, or whether the second required distance d2 is longer than the first required distance d1 (step S108). Then, it is assumed that "the control unit 35 determines that the second required time l2 is longer than the first required time l1, or the second required distance d2 is longer than the first required distance d1" (step S108-Yes). In this case, the control unit 35 sets the action start timing based on the second required time l2 or the second required distance d2 so that the determined action ends before the target location (step S109).

[0096] On the other hand, suppose that "the control unit 35 determines that the first required time l1 is longer than the second required time l2, or the first required distance d1 is longer than the second required distance d2" (step S108-No). In this case, the control unit 35 sets the action start timing based on the first required time l1 or the first required distance d1 so that the determined action ends before the target location (step S105).

[0097] After step S105 or step S109, when the action start timing comes, the control unit 35 controls the vehicle 10 so that the vehicle 10 executes the determined action (step S110). After step S110, the processor 23 ends the vehicle control process.

[0098] As described above, the vehicle control device determines whether there is a possibility that the detected position of the vehicle's lane relative to one end of the road on which the vehicle is traveling differs from its actual position. Furthermore, the vehicle control device estimates a first required time or a first required distance to complete the determined action, assuming that the detected position of the vehicle's lane is correct. Furthermore, the vehicle control device estimates a second required time or a second required distance to complete the determined action, assuming that the detected position of the vehicle's lane actually deviates by a predetermined number of lanes. Furthermore, if there is a possibility that the detected position of the vehicle's lane differs from its actual position and the second required time or distance is longer than the first required time or distance, the vehicle control device sets the action start timing based on the second required time or distance. Thus, even if the detected position of the vehicle's lane is inaccurate, the vehicle control device can complete the determined vehicle action before the target position or timing.

[0099] Furthermore, the computer program that realizes the functions of the processor 23 of the ECU 7 in the above-described embodiment or modification may be provided in a form recorded in a computer-readable portable recording medium such as a semiconductor memory, a magnetic recording medium, or an optical recording medium.

[0100] As described above, those skilled in the art can make various changes according to the embodiment within the scope of the present invention.

Claims

1. A vehicle control device comprising: a host lane detection unit for detecting a host lane in which the vehicle is traveling, among the plurality of lanes, by comparing sensor signals indicating the surroundings of the vehicle obtained by sensors mounted on the vehicle with a map including information related to a plurality of lanes provided on the road on which the vehicle is traveling; a false detection determination unit that determines whether there is a possibility that the detected position of the own lane relative to the one end of the road is different from the actual position; a determination unit that determines, based on a travel route of the vehicle, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location that is a predetermined distance forward from a current position of the vehicle; an estimating unit that estimates a first required time or a first required distance required for the action when it is assumed that the position of the detected own lane relative to the one end of the road is correct, and a second required time or a second required distance required for the action when it is assumed that the position of the detected own lane relative to the one end of the road deviates from the actual position by a predetermined number of lanes; and The control unit sets the timing for starting the action based on the second required time or the second required distance so that the action ends before the vehicle reaches the predetermined location, when there is a possibility that the position of the lane detected relative to one end of the road is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance. On the other hand, when there is no possibility that the position of the lane detected relative to one end of the road is different from the actual position, or when the first required time or the first required distance is longer than the second required time or the second required distance, the control unit sets the timing for starting the action based on the first required time or the first required distance so that the action ends before the vehicle reaches the predetermined location.

2. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the lane relative to one end of the road is different from the actual position when the construction information received from other equipment indicates that road construction is being carried out within a predetermined period including the current moment in the interval from the current position of the vehicle to the predetermined distance forward.

3. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the last update date of the map is more than a predetermined period before the current time, or when the structure of the road in the section from the current position of the vehicle to the predetermined distance ahead is different between the map and the route search map used by the navigation device in searching the driving route of the vehicle.

4. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the driving trajectory of other vehicles in the section from the current position of the vehicle to the predetermined distance ahead passes through an area in the map where it is impossible for the vehicle to travel.

5. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position before a certain time has passed since the start of the detection of the own lane, or until the vehicle has traveled a certain distance since the start of the detection of the own lane.

6. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the position of the detected own lane relative to one end of the road is different from the actual position when the number of lanes indicated by the map or the sensor signal at the current position of the vehicle is greater than a predetermined number and the detected own lane is within a predetermined range from the center of a plurality of lanes of the road set at the current position of the vehicle.

7. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the number of lanes at the current position has changed by more than a predetermined number compared to the number of lanes at the position of the vehicle a predetermined time ago.

8. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the position of a predetermined feature detected based on the sensor signal is different from the position of the corresponding feature on the map based on the detected position of the own lane.

9. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the degree of certainty of the predetermined feature detected based on the sensor signal is less than or equal to a predetermined confidence threshold.

10. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when it is unable to detect other vehicles traveling within the following range within a certain period of time. The range is a range represented by the sensor signal that is equivalent to the area in which the vehicle can travel on the map based on the detected position of the own lane.

11. The vehicle control device according to claim 1, The false detection determination unit determines that there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position when the current position of the vehicle is within a predetermined range from a pre-specified false detection risk point in the map.

12. A vehicle control method, comprising: detecting a lane in which the vehicle is traveling among the plurality of lanes by comparing sensor signals indicating the surroundings of the vehicle obtained by sensors mounted on the vehicle with a map including information related to a plurality of lanes provided on the road on which the vehicle is traveling; determining whether there is a possibility that the detected position of the own lane relative to the one end of the road is different from the actual position; determining, based on the travel route of the vehicle, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location that is a predetermined distance forward from a current position of the vehicle; estimating a first required time or a first required distance required for the action if the detected position of the own lane relative to the one end of the road is assumed to be correct, and a second required time or a second required distance required for the action if the detected position of the own lane relative to the one end of the road is assumed to be deviated from the actual position by a predetermined number of lanes; If there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, setting a timing for starting the action based on the second required time or the second required distance so that the action is completed before the vehicle reaches the predetermined location; and In the absence of a possibility that the detected position of the own lane relative to one end of the road is different from the actual position, or that the first required time or the first required distance is longer than the second required time or the second required distance, the timing for starting the action is set based on the first required time or the first required distance so that the action is completed before the vehicle reaches the predetermined location.

13. A computer program product for vehicle control, comprising a vehicle control computer program, wherein the vehicle control computer program causes a processor mounted on a vehicle to execute the following processing: detecting a lane in which the vehicle is traveling among the plurality of lanes by comparing sensor signals indicating the surroundings of the vehicle obtained by sensors mounted on the vehicle with a map including information related to a plurality of lanes provided on a road on which the vehicle is traveling; determining whether there is a possibility that the detected position of the own lane relative to the one end of the road is different from the actual position; determining, based on the travel route of the vehicle, an action of the vehicle that is required to be performed before the vehicle reaches a predetermined location that is a predetermined distance forward from a current position of the vehicle; estimating a first required time or a first required distance required for the action if the detected position of the own lane relative to the one end of the road is assumed to be correct, and a second required time or a second required distance required for the action if the detected position of the own lane relative to the one end of the road is assumed to be deviated from the actual position by a predetermined number of lanes; If there is a possibility that the detected position of the own lane relative to one end of the road is different from the actual position and the second required time or the second required distance is longer than the first required time or the first required distance, setting a timing for starting the action based on the second required time or the second required distance so that the action is completed before the vehicle reaches the predetermined location; and In the absence of a possibility that the detected position of the own lane relative to one end of the road is different from the actual position, or that the first required time or the first required distance is longer than the second required time or the second required distance, the timing for starting the action is set based on the first required time or the first required distance so that the action is completed before the vehicle reaches the predetermined location.

Images