Driving control device

Abstract

To provide a travel control device capable of controlling travel of a vehicle so as not to give a sense of discomfort to a driver and peripheral traffic participants.SOLUTION: A travel control device 1 calculates reliability degree of detection of a target while detecting the target from peripheral data of a vehicle, determines whether or not to change the travel control from first control to second control based on the detected target, determines to continue the travel control as the second control, if the reliability degree of the target detection from the peripheral data of a first time is higher than a reliability threshold and the target is not detected from the peripheral data at a second time, after the first time when changing the first control to the second control based on the peripheral data at the first time, determines to change the travel control from the second control to the first control if the reliability degree at the first time is lower than the reliability threshold and the target is not detected from the peripheral data corresponding to the second time, and controls the travel of the vehicle based on the determined control.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to a driving control device that controls driving of a vehicle. [Background technology]

[0002] A vehicle whose driving is controlled by a driving control device is required to drive in accordance with the behavior of other vehicles driving in the vicinity. The driving control device described in Patent Document 1 uses the output of a sensor that detects an object ahead of the vehicle to control the vehicle so that the distance to the object ahead of the vehicle is within a predetermined range. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-137316 Summary of the Invention [Problem to be solved by the invention]

[0004] If the sensor can no longer properly detect information about objects ahead, the control that was initiated when the object was detected will be discontinued, which may change the vehicle's behavior and cause confusion for the vehicle driver or surrounding traffic participants.

[0005] An object of the present disclosure is to provide a driving control device that can control the driving of a vehicle so as not to cause discomfort to the driver or surrounding traffic participants. [Means for solving the problem]

[0006] The gist of the present disclosure is as follows.

[0007] (1) A detection unit that detects a target and calculates a reliability indicating the likelihood of detecting the target by inputting each piece of time-series surrounding data representing the situation around the vehicle into a classifier that has been trained in advance to detect the target from the surrounding data; a control decision unit that decides, based on the detected target, whether to continue a first control used for current travel control of the vehicle or to change the travel control of the vehicle to a second control different from the first control; a continuation determination unit that, when control of traveling of the vehicle is changed from the first control to the second control based on a target detected in surrounding data corresponding to a first time, determines to continue control of traveling of the vehicle as the second control from the second time onwards, if the reliability of the detected target calculated from the surrounding data corresponding to the first time is higher than a reliability threshold and the target is not detected in surrounding data corresponding to a second time later than the first time, and determines to change control of traveling of the vehicle from the second control to the first control from the second time onwards, if the reliability at the first time is lower than the reliability threshold and the target is not detected in the surrounding data corresponding to the second time; a travel control unit that controls travel of the vehicle according to the determined control; Equipped with driving control device.

[0008] According to the present disclosure, it is possible to control the driving of a vehicle so as not to cause discomfort to the driver or other traffic participants in the vicinity. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 2 is a hardware schematic diagram of a driving control device. [Figure 2] FIG. 2 is a functional block diagram of a processor included in the driving control device. [Figure 3] 4 is a flowchart of a driving control process. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, with reference to the drawings, a cruise control device capable of controlling vehicle travel so as not to cause discomfort to the driver or surrounding traffic participants will be described in detail. The cruise control device detects targets by inputting time-series peripheral data representing the conditions around the vehicle to a classifier that has been trained in advance to detect targets from the peripheral data. In detecting the targets, the cruise control device also calculates a reliability indicating the likelihood of detecting the targets. Next, based on the detected targets, the cruise control device determines whether to continue the first control used for current vehicle travel control or to change the vehicle travel control to a second control different from the first control. When the cruise control device changes the vehicle travel control from the first control to the second control based on a target detected from peripheral data corresponding to a first time, if the reliability of the detected target calculated from the peripheral data corresponding to the first time is higher than a reliability threshold and no target is detected from peripheral data corresponding to a second time later than the first time, the cruise control device determines to continue the vehicle travel control with the second control from the second time onwards. On the other hand, in such a case, if the reliability at the first time is lower than the reliability threshold and no target is detected from the surrounding data corresponding to the second time, the cruise control device determines to change the control of the vehicle's travel from the second control to the first control after the second time, and then controls the vehicle's travel in accordance with the determined control.

[0011] The vehicle includes a surrounding sensor, a Global Navigation Satellite System (GNSS) receiver, a storage device, and a driving control device.

[0012] The surrounding sensor outputs surrounding data that indicates the surrounding situation. The surrounding sensor is, for example, a surrounding camera that has a two-dimensional detector composed of an array of photoelectric conversion elements sensitive to visible light and an imaging optical system that forms an image of the area to be photographed on the two-dimensional detector, and outputs a surrounding image. The GNSS receiver outputs a positioning signal that indicates the vehicle's own position based on the GNSS signal received from the GNSS satellite. The storage device is a hard disk drive or a storage device with non-volatile semiconductor memory, and stores map data including lane information.

[0013] The driving control device 1 is an ECU (Electronic Control Unit) that includes a communication interface, a memory, and a processor. The driving control device 1 controls the driving of the vehicle using surrounding data that indicates the conditions around the vehicle.

[0014] 1 is a hardware schematic diagram of a driving control device 1. The driving control device 1 includes a communication interface 11, a memory 12, and a processor 13.

[0015] The communication interface 11 is an example of a communication unit, and has a communication interface circuit for connecting the driving control device 1 to an in-vehicle network. The communication interface 11 supplies received data to the processor 13. The communication interface 11 also outputs data supplied from the processor 13 to the outside.

[0016] The memory 12 includes a volatile semiconductor memory and a non-volatile semiconductor memory. The memory 12 stores various data used in processing by the processor 13, such as parameters for defining a neural network used as a classifier for detecting a target from peripheral data, a reliability threshold for determining the likelihood of target detection from peripheral data, etc. The memory 12 also stores various application programs, such as a cruise control program for executing cruise control processing.

[0017] The processor 13 is an example of a control unit and includes one or more processors and their peripheral circuits. The processor 13 may further include other arithmetic circuits such as a logic unit, a numerical calculation unit, or a graphics processing unit.

[0018] FIG. 2 is a functional block diagram of the processor 13 included in the driving control device 1. As shown in FIG.

[0019] The processor 13 of the driving control device 1 has, as functional blocks, a detection unit 131, a control decision unit 132, a continuation determination unit 133, and a driving control unit 134. Each of these units of the processor 13 is a functional module implemented by a computer program stored in the memory 12 and executed on the processor 13. Each of these units of the processor 13 may be implemented in the driving control device 1 as an independent integrated circuit, a microprocessor, or firmware.

[0020] The detection unit 131 detects targets and calculates a reliability indicating the likelihood of detecting the targets by inputting each piece of time-series surrounding data representing the situation around the vehicle into a classifier that has been trained in advance to detect targets (e.g., other vehicles) from the surrounding data.

[0021] The classifier can be, for example, a convolutional neural network (CNN) with multiple convolution layers connected in series from the input side to the output side, such as Single Shot MultiBox Detector or Faster R-CNN. Images containing detection targets, such as other vehicles, are used as training data to train the CNN in advance according to a predetermined learning method, such as backpropagation. The CNN operates as a classifier that detects targets from surrounding data. When detecting targets, the CNN calculates the reliability of the detection.

[0022] The control decision unit 132 decides, based on the target detected by the detection unit 131, whether to continue the first control used to control the current driving of the vehicle, or to change the control of the driving of the vehicle to a second control different from the first control.

[0023] The first control is, for example, a distance maintenance control that maintains a distance to a preceding vehicle. The second control is a yield control that widens the space when a merging vehicle traveling in the merging lane changes lanes into the merging lane, for example, when the lane in which the vehicle is traveling is a merging lane among the main lanes that merges into the merging lane.

[0024] The control decision unit 132 acquires lane information of roads around the vehicle's own position represented by the positioning signal received from the GNSS receiver from a storage device. Then, the control decision unit 132 identifies the lane in which the vehicle is traveling using lane markings detected from the surrounding data, and determines whether the lane in which the vehicle is traveling is a merging lane.

[0025] The control decision unit 132 uses lane markings detected from the surrounding data to identify which of one or more lanes on the road the vehicle is traveling in. For example, if the lane information acquired from the storage device includes two lanes and one lane marking is detected on the left side of the vehicle and two lane markings are detected on the right side from the surrounding image, the control decision unit 132 can identify the left lane of the two lanes as the vehicle's traveling lane. Then, the control decision unit 132 identifies the position of the traveling lane in the lane information acquired from the storage device and determines whether the traveling lane is a merging lane.

[0026] When the driving lane is determined to be the merging lane, the control decision unit 132 determines whether or not yielding control is necessary to widen the space when a merging vehicle, which is another vehicle traveling in the merging lane, changes lanes into the merging lane. For example, when the merging vehicle reaches a point where it is possible to change lanes from the merging lane to the merging lane, the control decision unit 132 determines that yielding control is necessary if the vehicle is predicted to travel within a predetermined range before the point. When it is determined that yielding control is necessary, the control decision unit 132 changes the first control (distance maintenance control) used for the current driving control to the second control (yielding control), and when it is determined that yielding control is not necessary, it decides to continue the first control.

[0027] The control decision unit 132 may detect lane markings by inputting each piece of time-series peripheral data representing the conditions around the vehicle to a classifier that has been trained in advance to detect lane markings from the peripheral data. The classifier may be, for example, a CNN. Images containing lane markings are used as training data, and the CNN is trained in advance according to a predetermined learning method, such as backpropagation. The CNN operates as a classifier that detects lane markings from the peripheral data. Note that a CNN that has been trained in advance to detect targets such as other vehicles and lane markings from the peripheral data using images containing targets such as other vehicles and lane markings as training data may be used as both the classifier used by the detection unit 131 to detect targets and the classifier used by the control decision unit 132 to detect lane markings.

[0028] When the continuation determination unit 133 changes the vehicle traveling control from the first control to the second control based on a target detected in the surrounding data corresponding to a first time, and the reliability of the detected target calculated from the surrounding data corresponding to the first time is higher than the reliability threshold and no target is detected in the surrounding data corresponding to a second time after the first time, the continuation determination unit 133 determines to continue the vehicle traveling control from the second time onwards as the second control. Furthermore, in such a case, when the reliability at the first time is lower than the reliability threshold and no target is detected in the surrounding data corresponding to the second time, the continuation determination unit 133 determines to change the vehicle traveling control from the second control to the first control from the second time onwards.

[0029] The continuation determination unit 133 can use, as the reliability, the reliability of the detection calculated when the classifier detects a target from surrounding data.

[0030] The driving control unit 134 transmits control signals to the driving mechanisms of the vehicle, such as the engine or motor and the steering mechanism, according to the determined control, and controls the driving of the vehicle.

[0031] Fig. 3 is a flowchart of the driving control process. The driving control device 1 repeatedly executes the driving control process shown in Fig. 3 at predetermined time intervals (for example, every one second) while the vehicle is driving under automatic driving control.

[0032] First, the detection unit 131 of the processor 13 of the cruise control device 1 detects a target from the surrounding data corresponding to a first time, and calculates the reliability of the detection of the target (step S1).

[0033] Next, the control decision unit 132 of the processor 13 of the driving control device 1 decides, based on the detected target, whether to continue the first control used for the current driving control or to change the driving control of the vehicle to a second control different from the first control (step S2). If it is decided to change to the second control, the control decision unit 132 changes the driving control of the vehicle to the second control.

[0034] When the vehicle driving control is changed from the first control to the second control (step S2: Y), the continuation judgment unit 133 of the processor 13 of the driving control device 1 judges whether the reliability of the target calculated from the surrounding data corresponding to the first time is higher than the reliability threshold stored in the memory 12 (step S3).

[0035] If it is determined that the reliability at the first time is higher than the reliability threshold (step S3: Y), the continuation determination unit 133 determines whether or not a target has been detected from the peripheral data corresponding to a second time later than the first time (step S4).If it is determined that the reliability is not higher than the reliability threshold (step S3: N), the continuation determination unit 133 determines whether or not a target has been detected from the peripheral data corresponding to the second time later than the first time (step S5).

[0036] If the reliability is determined to be lower than the reliability threshold and no target is detected from the surrounding data corresponding to the second time (step S4: N), the continuation determination unit 133 decides to continue controlling the vehicle's travel using the second control from the second time onwards (step S6), and proceeds to step S9, which will be described later.

[0037] If the reliability is determined to be lower than the reliability threshold and no target is detected from the surrounding data corresponding to the second time (step S5: N), the continuation judgment unit 133 decides to change the control of the vehicle's driving from the second time onwards from the second control to the first control (step S7), and proceeds to step S9 described below.

[0038] If the reliability at the first time is determined to be higher than the reliability threshold and a target is detected from the surrounding data corresponding to the second time (step S4: Y), or if the reliability at the first time is determined not to be higher than the reliability threshold and a target is detected from the surrounding data corresponding to the second time (step S5: Y), the continuation judgment unit 133 determines how to control the vehicle's travel from the second time onwards using the target detected from the surrounding data corresponding to the second time (step S8), and proceeds to step S9, which will be described later.

[0039] The driving control unit 134 controls the driving of the vehicle according to the determined control (step S9), and ends the driving control process.

[0040] After the driving control process is completed, the driving control unit 134 continues to control the driving of the vehicle using the determined control until a new control to be used for driving control is determined.

[0041] By executing the driving control process in this manner, the driving control device 1 can control the driving of the vehicle so as not to cause discomfort to the driver or other traffic participants in the vicinity.

[0042] It should be understood that those skilled in the art can make various changes, substitutions, and alterations thereto without departing from the spirit and scope of the present disclosure. [Explanation of symbols]

[0043] 1. Driving control device 131 Detector 132 Control decision unit 133 Continuation Judgment Section 134 Travel control unit

Claims

[Claim 1] a detection unit that detects other vehicles by inputting each piece of time-series surrounding data that represents the situation around the vehicle into a classifier that has been trained in advance to detect other vehicles from the surrounding data, and calculates a reliability that indicates the likelihood of detecting the other vehicles; a control decision unit that decides, based on the detected other vehicle, whether to continue the vehicle-to-vehicle distance maintenance control that maintains the vehicle-to-vehicle distance to the preceding vehicle, which is currently used for controlling the vehicle's travel, or to change the vehicle's travel control to a yield control that is different from the vehicle-to-vehicle distance maintenance control and that widens the space when the other vehicle changes lanes into the travel lane of the subject vehicle; a continuation determination unit that, when the control of the traveling of the vehicle is changed from the distance maintenance control to the yielding control based on another vehicle detected in the surrounding data corresponding to a first time, determines to continue the control of the traveling of the vehicle as the yielding control from the second time onwards, if the reliability of the detected another vehicle calculated from the surrounding data corresponding to the first time is higher than a reliability threshold and the other vehicle is not detected in the surrounding data corresponding to a second time later than the first time, and determines to change the control of the traveling of the vehicle from the yielding control to the distance maintenance control from the second time onwards, if the reliability at the first time is lower than the reliability threshold and the other vehicle is not detected in the surrounding data corresponding to the second time; a travel control unit that controls travel of the vehicle according to the determined control; Equipped with driving control device.

Images