JP6558282B2 - Automated driving system - Google Patents

Description

  The present invention relates to an automatic driving system for a vehicle.

  A system for performing automatic driving control of a vehicle is described in Patent Document 1, for example. When the system described in Patent Document 1 determines that the conditions for performing the automatic driving control are not satisfied when the automatic driving control of the vehicle is performed, the driver is prompted to cancel the automatic driving control. I do. In addition, this system calculates a stop point where a vehicle such as a roadside belt or an emergency parking belt having a sufficient width can be stopped. If the driver does not cancel the automatic driving control even if the driver prompts the driver to cancel the automatic driving control, the system automatically drives the vehicle to the calculated stop point and stops the vehicle.

JP 2014-106854 A

  Here, when the vehicle automatically travels to the stop point because the driver does not cancel the automatic driving control, it may not be easy to automatically drive the vehicle to the stop point. In this case, since it is difficult to continue the automatic driving control, it is desirable to take over the driving operation to the driver. When handing over the driving operation to the driver, it is desirable to continue the automatic driving control as long as possible and to secure a long time for taking over the driving operation to the driver. However, in such a case, it is switched to manual operation by the driver, and the frequency with which the driver performs manual operation may increase.

  Then, an object of this invention is to provide the automatic driving | operation system which can suppress the increase in the frequency of the manual driving | operation by a driver.

The present invention is an automatic driving system that performs automatic driving control of a vehicle, and a position recognition unit that recognizes the position of the vehicle based on the measurement result of the position measuring unit of the vehicle, and an external sensor that detects an external situation of the vehicle An environment recognition unit that recognizes the surrounding environment of the vehicle based on the detection result of the vehicle, a vehicle state recognition unit that recognizes the state of the vehicle based on the detection result of the internal sensor that detects the state of the vehicle, and the recognized vehicle A first planning unit for generating a first travel plan of the vehicle based on the position of the vehicle, the recognized surrounding environment of the vehicle, and the recognized vehicle state, and the reliability of the recognized vehicle position, First reliability for calculating the reliability of the first travel plan based on at least one of the reliability of recognition of the surrounding environment of the vehicle, the reliability of recognition of the recognized state of the vehicle, and the generated first travel plan. Reliable calculator and recognized Position of both, recognized vehicle surrounding environment, and recognized the states of the vehicle, one or two on the basis of the vehicle that does not significantly change the state of the vehicle per unit time than the first travel plan Of the second planning unit that generates the second travel plan, the recognized vehicle position, the recognized surrounding environment of the vehicle, and the recognized vehicle state, the second planning unit generates the second travel plan. A second reliability calculation unit that calculates the reliability of the second travel plan based on at least one of the reliability of the recognition result used at the time and the generated second travel plan, the first travel plan, A selection unit that selects one of the two travel plans, a presentation unit that presents the first travel plan to the driver of the vehicle when the second travel plan is selected by the selection unit, and a first by the driver Accepts input operation for selection of travel plan An input unit, and a travel control unit that performs automatic driving control of the vehicle based on the travel plan selected by the selection unit, and the selection unit includes the reliability of the first travel plan and the reliability of the second travel plan. When the travel plan having a high reliability of the travel plan is selected from the first travel plan and the second travel plan and the second travel plan is selected based on the reliability of the travel plan, If an input operation for selecting the first travel plan is accepted by the input unit when the first travel plan is presented, the first travel plan is selected instead of the second travel plan.

  In this automatic driving system, the travel control unit performs automatic driving control of the vehicle based on the travel plan selected by the selection unit among the first travel plan and the second travel plan. Here, a 1st plan part produces | generates a 1st driving | running plan using all the recognition results in a position recognition part, an environment recognition part, and a vehicle state recognition part. For this reason, for example, when the reliability of any of the recognition results of the position recognition unit, the environment recognition unit, and the vehicle state recognition unit decreases, the reliability of the first travel plan generated using all of these results decreases. To do. On the other hand, a 2nd plan part produces | generates a 2nd driving plan based on the recognition result of a kind smaller than a 1st plan part. Since the second travel plan is generated based on fewer types of recognition results than the first travel plan, when the second travel plan is generated without using the recognition result having reduced reliability, the recognition result The second travel plan is generated without being affected by the decrease in reliability. When the second travel plan is generated in this way, since the recognition result with reduced reliability is not used, the reliability of the second travel plan is reduced due to the decrease in the reliability of the recognition result. Absent. For this reason, the reliability of a 1st travel plan falls with the fall of the reliability of the recognition result in any one of a position recognition part, an environment recognition part, and a vehicle state recognition part, and automatic operation control based on a 1st travel plan is carried out. When it becomes difficult and a situation occurs in which the driver takes over the driving operation, the second travel plan is selected by the selection unit. Thus, even if the situation where the automatic driving control based on the first travel plan becomes difficult and the driver takes over the driving operation occurs, the vehicle is not changed based on the second driving plan until the driver takes over the driving operation. Automatic operation control is continued. The driver can take over the driving operation while the vehicle is traveling based on the second travel plan.

  Here, even if the reliability of the first travel plan is reduced and the second travel plan is selected, the driver may determine that travel based on the first travel plan is acceptable. For this reason, the presentation unit presents the first travel plan when the second travel plan is selected by the selection unit. Thereby, the driver can determine whether or not the first travel plan is acceptable. The input unit receives an input operation for selecting the first travel plan performed by the driver. When the input operation for selecting the first travel plan is received, the selection unit selects the first travel plan instead of the second travel plan. Thus, the travel control unit performs automatic driving control of the vehicle based on the first travel plan instead of the second travel plan. As described above, since the first travel plan is acceptable, when the driver performs an input operation for selecting the first travel plan, automatic operation control based on the first travel plan is performed. As described above, since the automatic operation control based on the first travel plan is resumed, it is possible to suppress an increase in the frequency with which the driver performs the manual operation as the automatic operation control based on the first travel plan ends.

  ADVANTAGE OF THE INVENTION According to this invention, the increase in the frequency of the manual operation by a driver can be suppressed.

It is a figure showing the schematic structure of the automatic operation system concerning an embodiment. It is a figure which shows the example of a display image which presents a 1st travel plan when the 2nd travel plan is selected. It is a flowchart which shows the flow of the process performed by ECU. It is a figure which shows schematic structure of the automatic driving system which concerns on a modification.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted.

  As shown in FIG. 1, the automatic driving system 100 is mounted on a vehicle M such as a passenger car, and is a system for performing automatic driving control of the vehicle M. The automatic driving system 100 determines whether or not automatic driving is possible. When the automatic driving system 100 determines that automatic driving is possible, the automatic driving control start operation by the driver (operation for pressing an automatic driving control start button) Etc.) is started, the automatic driving control of the vehicle M is started. The automatic driving control is vehicle control for automatically driving the vehicle M along a preset target route. In the automatic driving control, the driver does not need to perform a driving operation, and the vehicle M travels automatically. The target route is a route on the map on which the vehicle M travels in the automatic driving control.

  The automatic driving system 100 includes an ECU 6 for executing automatic driving control. The ECU 6 is an electronic control unit having a CPU [Central Processing Unit], a ROM [Read Only Memory], a RAM [Random Access Memory], a CAN [Controller Area Network] communication circuit, and the like. The ECU 6 implements various functions by loading a program stored in the ROM into the RAM and executing the program loaded in the RAM with the CPU. The ECU 6 may be composed of a plurality of electronic control units. The ECU 6 is connected to the GPS receiver 1, external sensor 2, internal sensor 3, map database 4, navigation system 5, actuator 7, HMI [Human Machine Interface] 8, and operation detector 9 via the CAN communication circuit. Has been.

  The GPS receiver 1 is mounted on the vehicle M and functions as a position measuring unit that measures the position of the vehicle M. The GPS receiving unit 1 measures the position of the vehicle M (for example, the latitude and longitude of the vehicle M) by receiving signals from three or more GPS satellites. The GPS receiver 1 transmits information on the measured position of the vehicle M to the ECU 6.

  The external sensor 2 is a detection device for detecting an external situation or the like around the vehicle M. The external sensor 2 includes at least one of a camera, a radar [Radar], and a rider [LIDAR: Laser Imaging Detection and Ranging]. The external sensor 2 is also used for white line recognition of a travel lane in which the vehicle M travels, which will be described later. The external sensor 2 may be used for measuring the position of the vehicle M.

  The camera is an imaging device that captures an external situation of the vehicle. The camera is provided on the back side of the windshield of the vehicle M. The camera may be provided on the left and right side surfaces of the vehicle M and the back surface of the vehicle. The camera transmits imaging information obtained by imaging the front of the vehicle M to the ECU 6. The camera may be a monocular camera or a stereo camera. The stereo camera has two imaging units arranged so as to reproduce binocular parallax. The imaging information of the stereo camera includes information in the depth direction.

  The radar detects obstacles around the vehicle M using radio waves (for example, millimeter waves). The radar detects an obstacle by transmitting a radio wave to the periphery of the vehicle M and receiving the radio wave reflected by the obstacle. The radar transmits the detected obstacle information to the ECU 6. The obstacles include fixed obstacles such as curbs, utility poles, poles, guardrails, walls, buildings, signs and signs provided on the roadside, and dynamic obstacles such as people, bicycles, and other vehicles.

  The rider detects an obstacle outside the vehicle M using light. The rider transmits light to the periphery of the vehicle M, receives the light reflected by the obstacle, measures the distance to the reflection point, and detects the obstacle. The rider transmits the detected obstacle information to the ECU 6. Riders and radars do not necessarily have to be provided in duplicate.

  The internal sensor 3 is a detection device that detects the vehicle state of the vehicle M. The internal sensor 3 includes a vehicle speed sensor, an acceleration sensor, and a yaw rate sensor. The vehicle speed sensor is a detector that detects the vehicle speed of the vehicle M. As the vehicle speed sensor, a wheel speed sensor that is provided for a wheel of the vehicle M or a drive shaft that rotates integrally with the wheel and detects the rotation speed of the wheel is used. The vehicle speed sensor transmits the detected vehicle speed information to the ECU 6.

  The acceleration sensor is a detector that detects the acceleration of the vehicle M. The acceleration sensor includes a longitudinal acceleration sensor that detects the longitudinal acceleration of the vehicle M and a lateral acceleration sensor that detects the lateral acceleration of the vehicle M. The acceleration sensor transmits acceleration information of the vehicle M to the ECU 6. The yaw rate sensor is a detector that detects the yaw rate (rotational angular velocity) around the vertical axis of the center of gravity of the vehicle M. A gyro sensor can be used as the yaw rate sensor. The yaw rate sensor transmits the detected yaw rate information of the vehicle M to the ECU 6.

  The map database 4 is a database that stores map information. The map information may include the position information of the fixed obstacle. The map information may include position information of white lines provided on the road. The map database 4 is formed in an HDD [Hard Disk Drive] mounted on the vehicle M. The map database 4 may be connected to the map information management center server by wireless communication and periodically update the map information using the latest map information stored in the map information management center server. The map database 4 is not necessarily installed in the vehicle M. The map database 4 may be provided in a server or the like that can communicate with the vehicle M.

  Further, the map information may include information in which the frequency of switching from automatic driving control to manual driving is associated with each location. The map information may include information in which the ratio of the time traveled by the manual operation to the time traveled by the automatic operation control is associated with each predetermined area. With this information, the position recognition unit 11 or the like of the ECU 6 can recognize that a certain area has a high ratio of manual driving by a driver. Further, the map database 4 may store a weather map indicating the weather for each place. The map database 4 may store a disaster map indicating a place where a disaster (earthquake, flood damage, etc.) occurs.

  The navigation system 5 is mounted on the vehicle M and sets a target route on which the vehicle M travels by automatic driving control. The navigation system 5 determines a target route from the position of the vehicle M to the destination based on the preset destination, the position of the vehicle M measured by the GPS receiver 1, and the map information in the map database 4. Calculate. The destination of the automatic driving control is set by the passenger of the vehicle M operating the input button (or touch panel) provided in the navigation system 5. The navigation system 5 can set the target route by a known method. The navigation system 5 may have a function of performing guidance along the target route during manual driving of the vehicle M by the driver. The navigation system 5 transmits information on the target route of the vehicle M to the ECU 6. The navigation system 5 may be executed by a server of a facility such as an information processing center capable of communicating with the vehicle M in part of the functions. The function of the navigation system 5 may be executed in the ECU 6.

  It should be noted that the target route mentioned here is “driving support device” described in Japanese Patent No. 5382218 (WO 2011/158347) or “automatic driving device” described in Japanese Patent Application Laid-Open No. 2011-162132. A target route that is automatically generated based on past destination history and map information when the destination is not explicitly set by the driver, such as a road driving route, is also included.

  The actuator 7 is a device that executes traveling control of the vehicle M. The actuator 7 includes at least an engine actuator, a brake actuator, and a steering actuator. The engine actuator controls the driving force of the vehicle M by controlling the amount of air supplied to the engine (throttle opening) in accordance with a control signal from the ECU 6. When the vehicle M is a hybrid vehicle, in addition to the amount of air supplied to the engine, a control signal from the ECU 6 is input to a motor as a power source to control the driving force. When the vehicle M is an electric vehicle, a control signal from the ECU 6 is input to a motor as a power source to control the driving force.

  The brake actuator controls the brake system according to a control signal from the ECU 6 and controls the braking force applied to the wheels of the vehicle M. A hydraulic brake system can be used as the brake system. The steering actuator controls driving of an assist motor that controls steering torque in the electric power steering system in accordance with a control signal from the ECU 6. Thereby, the steering actuator controls the steering torque of the vehicle M.

  The HMI 8 is an interface for transmitting and inputting information between an occupant (for example, a driver) of the vehicle M and the automatic driving system 100. The HMI 8 includes an input unit 8a and a presentation unit 8b. The input unit 8a is a device that accepts an input operation by a driver. The input unit 8a includes an operation button or a touch panel. The presentation unit 8b is a device that presents information to the driver. The presentation unit 8b includes a display that displays image information to a driver and a passenger of the vehicle M. In addition to the display, the presentation unit 8b may include a speaker that outputs sound.

  The HMI 8 indicates that the driver needs to take over the driving operation through the presentation unit 8b when the second traveling plan is selected in a state where the first traveling plan is selected in the selection unit 16 described later. May be notified. That is, when the automatic driving control based on the first travel plan ends, the HMI 8 may notify the driver so that the driver takes over the driving operation.

  The operation detection unit 9 detects a driving operation performed by the driver. The operation detection unit 9 includes an accelerator pedal sensor, a brake pedal sensor, and a steering angle sensor. The accelerator pedal sensor is provided with respect to the shaft portion of the accelerator pedal, and detects the amount of depression of the accelerator pedal (the position of the accelerator pedal). The accelerator pedal sensor transmits accelerator operation information corresponding to the detected depression amount of the accelerator pedal to the ECU 6. The brake pedal sensor is provided for the shaft portion of the brake pedal, and detects the amount of depression of the brake pedal (the position of the brake pedal). You may detect from the operating force of a brake pedal (stepping force with respect to a brake pedal, the pressure of a master cylinder, etc.). The brake pedal sensor transmits the detected brake operation information to the ECU 6. The steering angle sensor is a sensor that detects the steering angle (actual steering angle) of the vehicle M. The steering angle sensor is provided with respect to the steering shaft of the vehicle M. The steering angle sensor transmits the detected steering angle information to the ECU 6.

  Next, a functional configuration of the ECU 6 will be described. The ECU 6 includes a position recognition unit 11, an environment recognition unit 12, a vehicle state recognition unit 13, a first calculation unit 14, a second calculation unit 15, a selection unit 16, a travel control unit 17, and a presentation control unit 18.

  The position recognition unit 11 recognizes the position of the vehicle M on the map based on the position information of the GPS receiving unit 1 and the map information of the map database 4. The position recognition unit 11 recognizes the position of the vehicle M as a combination of the x coordinate and the y coordinate in the xy orthogonal coordinate system. Further, the position recognition unit 11 recognizes the position of the vehicle M based on the vehicle speed of the vehicle M recognized by the vehicle state recognition unit 13 in addition to the position information of the GPS reception unit 1 and the map information of the map database 4. Correction may be performed. The position recognition unit 11 recognizes the position of the vehicle M by the SLAM technology using the position information of the fixed obstacle such as a curb and the detection result of the external sensor 2 included in the map information of the map database 4. Good. In this case, the external sensor 2 functions as a position measuring unit instead of the GPS receiving unit 1. Further, the position recognition unit 11 recognizes the position of the vehicle M or corrects the position by existing image processing or the like using the position information of the white line included in the map information and the detection result of the white line in the external sensor 2. You may go. In this case, the external sensor 2 also functions as a position measurement unit.

  The position of the vehicle M can be based on the center position of the vehicle M when viewed from the vertical direction (in the plan view). The center position of the vehicle M is the center of the vehicle M in the vehicle width direction and the center of the vehicle M in the front-rear direction.

  Further, the position recognition unit 11 calculates the reliability of the position of the recognized vehicle M. This reliability represents the certainty of the position of the recognized vehicle M. For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the calculation cycle of the position of the vehicle M. In this case, when the calculation cycle is longer than the predetermined cycle, the position recognition unit 11 may lower the reliability of the position of the vehicle M than when the calculation cycle is shorter than the predetermined cycle.

  For example, the position recognizing unit 11 may calculate the reliability of the position of the vehicle M based on the acquisition cycle of the position information of the GPS receiving unit 1 used for recognizing the position of the vehicle M. In this case, the position recognizing unit 11 lowers the reliability of the position of the vehicle M when the acquisition period of the position information of the GPS receiving unit 1 is longer than a predetermined period compared to when the acquisition period is shorter than the predetermined period. May be. For example, the position recognizing unit 11 may calculate the reliability of the position of the vehicle M based on temporal changes such as the position information of the GPS receiving unit 1 used for recognizing the position of the vehicle M. In this case, the position recognizing unit 11 has no contradiction in the case where there is a contradiction in the temporal change in the position information of the GPS receiving unit 1 (when the change is discontinuous), and there is no contradiction (when the change is not discontinuous). Compared to, the reliability of the position of the vehicle M may be lowered.

  For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the recognition error when recognizing the position of the vehicle M. For example, the error is small when the position of the vehicle M is narrowed down to one point. For example, the position of the vehicle M is not narrowed down, and the vehicle M exists in a certain area. If it is recognized, the error is large. In this case, when the error in recognizing the position of the vehicle M is large, the position recognizing unit 11 may lower the reliability of the position of the vehicle M than when the error is small. For example, when the position recognition unit 11 recognizes the position of the vehicle M using the detection results of the fixed structures and white lines around the vehicle M, the position recognition unit 11 can detect the fixed structures and white lines used for the recognition. The reliability of the position of the vehicle M may be calculated based on the ratio. In this case, the position recognizing unit 11 may reduce the reliability of the position of the vehicle M when the ratio of detection of the fixed structure and the white line is low as compared with the case where the ratio of detection is high.

  For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the frequency of switching from automatic driving control associated with each place to manual driving. The position recognition unit 11 can acquire the frequency of switching from automatic driving control associated with each place to manual driving based on the map information. In this case, when the position of the recognized vehicle M is a place where the frequency of switching from automatic driving control to manual driving is high, the position recognizing unit 11 determines the position of the vehicle M compared to the case where the frequency is low. The reliability may be lowered. For example, the position recognizing unit 11 determines the reliability of the position of the vehicle M based on the ratio of the time during which the vehicle travels in the manual operation to the time during which the vehicle travels by the automatic operation control associated with each predetermined area. May be calculated. Based on the map information, the position recognizing unit 11 can acquire the ratio of the time during which the vehicle is traveling in the manual operation to the time during which the vehicle is traveling under the automatic operation control associated with each predetermined area. In this case, the position recognizing unit 11 reduces the reliability of the position of the vehicle M when the recognized position of the vehicle M is an area where the ratio of time for manual driving is high compared to the area where the ratio is low. May be.

  For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the presence / absence of map information stored in the map database 4. In this case, when the position recognition unit 11 tries to recognize the position of the vehicle M in an area where no map information exists, the position recognition unit 11 may reduce the reliability of the position of the vehicle M compared to the case where the map information exists. Good. For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the freshness (newness) of the map information stored in the map database 4. In this case, the position recognizing unit 11 may lower the reliability of the position of the vehicle M when the time when the map information is updated is older than when the time when the map information is updated.

  For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on a weather map that shows the weather for each place stored in the map database 4. In this case, the position recognition unit 11 determines the position of the vehicle M when the recognized position of the vehicle M is bad weather on the weather map (for example, rain or snow) and when it is not bad weather (for example, sunny). The reliability may be lowered. For example, the position recognition unit 11 may calculate the reliability of the position of the vehicle M based on a disaster map stored in the map database 4 and indicating a place where a disaster occurs. In this case, when the position of the recognized vehicle M is a place where a disaster has occurred on the disaster map, the position recognition unit 11 determines the position of the vehicle M as compared to a place where no disaster has occurred. The reliability may be lowered.

  The position recognition unit 11 may calculate the reliability of the position of the vehicle M based on the plurality of reliability calculated based on the plurality of methods for calculating the reliability of the position of the vehicle M described above.

  The environment recognition unit 12 recognizes the surrounding environment of the vehicle M based on the detection result of the external sensor 2 that detects the external situation of the vehicle M. The environment recognizing unit 12 recognizes the surrounding environment of the vehicle M by a known method based on the captured image of the camera, the obstacle information of the radar, or the obstacle information of the rider. Specifically, the environment recognizing unit 12 detects obstacles around the vehicle M (fixed obstacles, dynamic obstacles) based on camera imaging information, radar obstacle information, or rider obstacle information. recognize. When the environment recognition unit 12 recognizes a vehicle (another vehicle) as an obstacle, the environment recognition unit 12 recognizes the type of the recognized vehicle. Here, as the type of vehicle, for example, it is recognized whether the vehicle is an emergency vehicle (a police car, an ambulance, a fire engine, etc.) or a vehicle other than the emergency vehicle. For example, in the case of using a camera, the type of vehicle may be distinguished by comparing the shape of the recognized vehicle with the shape of a predetermined emergency vehicle. Moreover, the environment recognition part 12 recognizes the state of the recognized vehicle, when a vehicle (another vehicle) is recognized as an obstacle. Here, as the state of the vehicle, for example, it is recognized whether or not the vehicle is lighting a hazard lamp. Further, the environment recognition unit 12 recognizes the position of the white line of the traveling lane in which the vehicle M travels based on the imaging information of the camera or the obstacle information of the rider. The environment recognition unit 12 may further recognize the white line type and the white line curvature.

  In addition, the environment recognition unit 12 calculates the reliability of recognition of the environment around the recognized vehicle M. The reliability of recognition of the surrounding environment of the vehicle M represents the certainty of recognition of the recognized surrounding environment. For example, the environment recognition unit 12 may calculate the reliability of the recognition of the surrounding environment of the vehicle M based on the calculation cycle of the recognition of the surrounding environment of the vehicle M. In this case, when the calculation cycle is longer than the predetermined cycle, the environment recognition unit 12 may lower the reliability of recognition of the surrounding environment of the vehicle M than when the calculation cycle is shorter than the predetermined cycle.

  For example, the environment recognition unit 12 may calculate the reliability of the recognition of the surrounding environment of the vehicle M based on the acquisition period of the detection result of the external sensor 2. In this case, when the acquisition period of the detection result of the external sensor 2 is long, the environment recognition unit 12 may reduce the reliability of recognition of the surrounding environment of the vehicle M compared to the case where the acquisition period is shorter than a predetermined period. . For example, the environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the temporal change in the position or speed of the recognized obstacle. In this case, the environment recognizing unit 12 determines that there is no contradiction when there is a temporal contradiction in the position or speed of the recognized obstacle (when the change is discontinuous or when there is a lot of noise). The reliability of the recognition of the surrounding environment of the vehicle M may be lowered as compared with the case where the noise is low or when there is little noise.

  For example, the environment recognizing unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the number of times the obstacle is divided or combined when the temporal change of the recognized obstacle is seen. Good. In this case, the environment recognizing unit 12 may reduce the reliability of the recognition of the surrounding environment of the vehicle M when the number of times the obstacle is split or coupled is large compared to when the number of times the obstacle is split or coupled is small.

  For example, the environment recognition unit 12 calculates the reliability of recognition of the surrounding environment of the vehicle M based on the ratio of the obstacles that can identify the type of the obstacle among the plurality of obstacles around the vehicle M. Also good. In this case, the environment recognizing unit 12 may reduce the reliability of the recognition of the surrounding environment of the vehicle M when the rate of identifying the type of obstacle is small compared to when the rate of identifying the obstacle is large. For example, the environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the number of pedestrians existing within a predetermined distance (Pm) from the vehicle M among the recognized obstacles. Good. In this case, the environment recognizing unit 12 may reduce the reliability of recognition of the surrounding environment of the vehicle M when the number of pedestrians existing within a predetermined distance is large compared to the case where the number of pedestrians is small. Good. As in the case of pedestrians, the environment recognizing unit 12 determines the surrounding environment of the vehicle M based on the number of obstacles other than pedestrians that exist within a predetermined distance (Pm) from the vehicle M. The reliability of recognition may be calculated. Also in this case, the environment recognition unit 12 is more reliable in recognizing the surrounding environment of the vehicle M when the number of obstacles other than pedestrians present within a predetermined distance is large compared to the case where the number of obstacles other than pedestrians is small. The degree may be lowered.

  The environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the width of the travel lane of the vehicle M. In this case, the environment recognition unit 12 calculates the width based on the position of the white line of the travel lane in which the recognized vehicle M travels. And the environment recognition part 12 makes the reliability of recognition of the surrounding environment of the vehicle M low when the calculated width is narrow compared with the case where the width is wide. The environment recognition unit 12 calculates the width based on the recognized white line, and calculates the reliability of the recognition of the surrounding environment of the vehicle M based on the calculated width, but the environment recognition unit 12 calculates the width. It is not limited to. For example, when the lane width information is included in the map database 4, the environment recognition unit 12 acquires the lane width from the map database 4 based on the position of the vehicle M recognized by the position recognition unit 11. Also good. Then, the environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the width acquired from the map database 4. In addition to the position of the white line of the driving lane, the environment recognition unit 12 calculates the width of the driving lane based on the position of a physical road boundary (curbstone, guardrail, wall, pylon for construction, etc.), for example. May be. In addition to the white line and the physical road boundary, the environment recognizing unit 12 may be regarded as a road width, for example, when an oncoming vehicle comes on a road without a center line, Half of the physical width of the road obtained based on this may be regarded as the width of the driving lane. Alternatively, the environment recognition unit 12 may calculate the virtual width based on a position where a sufficient distance is secured from the pedestrian, for example, when a pedestrian or the like is walking on a road shoulder without a road surface sign. Good. Further, the environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on the distance between the vehicle M and the road boundary. This road boundary may be a lane marking such as a white line, may be the edge of a travelable area defined by curbs, walls, gutters, etc., and should be treated as other road boundaries. May be.

  For example, the environment recognition unit 12 may calculate the reliability of recognition of the surrounding environment of the vehicle M based on whether or not the vehicle around the recognized vehicle M is an emergency vehicle. In this case, the environment recognition unit 12 reduces the reliability of recognition of the surrounding environment of the vehicle M when the ratio or number of emergency vehicles around the vehicle M is large compared to the case where the ratio or number of emergency vehicles is small. May be. For example, the environment recognition unit 12 may calculate the reliability of the recognition of the surrounding environment of the vehicle M based on whether or not the vehicles around the recognized vehicle M are lighting the hazard lamp. In this case, the environment recognizing unit 12 determines that the surrounding environment of the vehicle M is higher when the ratio of the number of the lights around the vehicle M is high or the number of the lights is low than when the number of the lights is low. The reliability of recognition may be lowered.

  The environment recognition unit 12 calculates the reliability of recognition of the surrounding environment of the vehicle M based on the plurality of reliability calculated based on the plurality of calculation methods of the reliability of recognition of the surrounding environment of the vehicle M described above. May be.

  The vehicle state recognition unit 13 recognizes the state of the vehicle M including the vehicle speed and direction of the vehicle M based on the detection results of the internal sensor 3 and the operation detection unit 9. Specifically, the vehicle state recognition unit 13 recognizes the vehicle speed of the vehicle M based on the vehicle speed information of the vehicle speed sensor. Based on the yaw rate information of the vehicle state recognition unit 13 and the yaw rate sensor, the direction of the vehicle M is recognized. The vehicle state recognition unit 13 recognizes the acceleration of the vehicle M based on the acceleration information of the acceleration sensor. The vehicle state recognition unit 13 recognizes whether or not the accelerator pedal is operated by the driver based on the accelerator operation information of the accelerator pedal sensor. The vehicle state recognition unit 13 recognizes whether or not the brake pedal has been operated based on the brake operation information of the brake pedal sensor. The vehicle state recognition unit 13 recognizes the steering angle (steering angle) of the vehicle M based on the steering angle information of the steering angle sensor.

  Further, the vehicle state recognition unit 13 calculates the reliability of recognition of the recognized state of the vehicle M. The reliability of the recognition of the state of the vehicle M represents the certainty of the recognition of the recognized state of the vehicle M. For example, the vehicle state recognition unit 13 may calculate the reliability of the recognition of the state of the vehicle M based on the calculation cycle for the recognition of the state of the vehicle M. In this case, when the calculation cycle is longer than the predetermined cycle, the vehicle state recognition unit 13 may lower the reliability of the recognition of the state of the vehicle M than when the calculation cycle is shorter than the predetermined cycle.

  For example, the vehicle state recognition unit 13 may calculate the reliability of the recognition of the state of the vehicle M based on the acquisition period of the detection result of the internal sensor 3 and the operation detection unit 9 used for the recognition of the state of the vehicle M. Good. In this case, the vehicle state recognition unit 13 reduces the reliability of the recognition of the state of the vehicle M when the acquisition cycle of the detection result of the internal sensor 3 is long compared to when the acquisition cycle is shorter than a predetermined cycle. Also good.

  The vehicle state recognition unit 13 calculates the reliability of recognition of the state of the vehicle M based on the plurality of reliability calculated based on the plurality of calculation methods of the reliability of recognition of the state of the vehicle M described above. Also good.

  The first calculation unit 14 calculates the first travel plan and the reliability of the first travel plan that are used when the automatic operation control of the vehicle M is performed. The first calculation unit 14 includes a first planning unit 14a and a first reliability calculation unit 14b.

  The first planning unit 14 a includes the position of the vehicle M recognized by the position recognition unit 11, the surrounding environment recognized by the environment recognition unit 12, the state of the vehicle M recognized by the vehicle state recognition unit 13, and the map database 4. Based on the map information, a first travel plan for the vehicle M is generated so as to travel along the target route set by the navigation system 5. The first planning unit 14a starts generating the first travel plan when it is determined that automatic driving is possible in the automatic driving system 100 and the driver performs an automatic driving control start operation. The first travel plan is a travel plan from the current position of the vehicle M until the vehicle M reaches a preset destination.

  The first planning unit 14a generates the first travel plan so that the vehicle M travels appropriately on the target route in accordance with standards such as safety, legal compliance, and travel efficiency. The first travel plan includes, for example, driving behavior such as obstacle avoidance and lane change in addition to following the preceding vehicle and steering along the shape of the travel lane. The first planning unit 14a generates, as the first travel plan, for example, a travel plan including a plurality of target positions along the target route and the speed at each target position. That is, the first travel plan includes a route plan and a speed plan.

  The first reliability calculation unit 14b calculates the reliability of the first travel plan generated by the first planning unit 14a. The reliability of the first travel plan represents the degree of whether or not the vehicle M can travel appropriately in accordance with standards such as safety, legal compliance, and travel efficiency. For example, the higher reliability of the first travel plan indicates that the vehicle can travel while maintaining a distance from the surrounding obstacles at a predetermined distance or more, or that the travel efficiency is good. Good traveling efficiency here means, for example, that the distance traveled per unit time is long. Alternatively, the fact that the traveling efficiency is good may include that the fuel efficiency is good.

  Specifically, the first reliability calculation unit 14b includes the reliability of the position of the vehicle M recognized by the position recognition unit 11, the reliability of the recognition of the surrounding environment of the vehicle M recognized by the environment recognition unit 12, the vehicle The reliability of the first travel plan is calculated based on at least one of the reliability of recognition of the state of the vehicle M recognized by the state recognition unit 13 and the first travel plan generated by the first planning unit 14a. .

  For example, the first reliability calculation unit 14b may calculate the reliability of the first travel plan based on the generation cycle of the first travel plan in the first planning unit 14a. In this case, the first reliability calculation unit 14b lowers the reliability of the first travel plan when the generation cycle of the first travel plan is longer than the predetermined cycle, compared to when the generation cycle is shorter than the predetermined cycle. May be.

  For example, the first reliability calculation unit 14b calculates the reliability of the first travel plan based on various reliability calculated by the position recognition unit 11, the environment recognition unit 12, and the vehicle state recognition unit 13. Also good. In this case, when the reliability of the position of the vehicle M calculated by the position recognition unit 11 is low, the first reliability calculation unit 14b compares the first travel plan with that of the first travel plan. The reliability may be lowered. When the reliability of the recognition of the surrounding environment of the vehicle M calculated by the environment recognition unit 12 is low, the first reliability calculation unit 14b is compared with the case where the reliability of the recognition of the surrounding environment of the vehicle M is high. The reliability of the travel plan may be lowered. When the reliability of the recognition of the state of the vehicle M calculated by the vehicle state recognition unit 13 is low, the first reliability calculation unit 14b increases the reliability of the first travel plan as compared with the case where the recognition reliability is high. It may be lowered.

  For example, the first reliability calculation unit 14b may calculate the reliability of the first travel plan based on the number of obstacles present around the route in the first travel plan. In this case, the first reliability calculation unit 14b has a large number of obstacles around the route in the first travel plan based on the generated first travel plan and the surrounding environment recognized by the environment recognition unit 12. The reliability of the first travel plan may be lowered as compared with the case where the number of obstacles is small. For example, the first reliability calculation unit 14b may calculate the reliability of the first travel plan based on the distance between an obstacle present around the route in the first travel plan and the vehicle M. In this case, the first reliability calculation unit 14b determines that the obstacle M and the vehicle M present around the route in the first travel plan are based on the generated first travel plan and the surrounding environment recognized by the environment recognition unit 12. When the distance between the vehicle and the vehicle M is less than the predetermined distance, the reliability of the first travel plan may be lowered as compared with the case where the distance between the obstacle and the vehicle M is greater than the predetermined distance.

  For example, the first reliability calculation unit 14b is based on the route in the first travel plan, the obstacle recognized by the environment recognition unit 12, and the physical blind spot of the external sensor 2 caused by the obstacle. The reliability of the travel plan may be calculated. Here, the physical blind spot of the external sensor 2 refers to an area that cannot be detected by the external sensor 2 due to the presence of an obstacle even though it is a detection area of the external sensor 2. The detection area of the external sensor 2 is determined in advance based on the direction in which the external sensor 2 is installed on the vehicle M or the like. A specific example of the physical blind spot of the external sensor 2 will be described. For example, it is assumed that the road ahead of the vehicle M is a curved road and a building exists near the road ahead. The building is present on the front side (vehicle M side) with respect to the tip of the curve, and is hidden behind the building so that the portion of the curve beyond the vehicle M cannot be seen. In this case, the tip of the curve cannot be detected by the external sensor 2 because the building near the road is blocked. An area obstructed by the building (an area that cannot be detected by the external sensor 2) is a physical blind spot of the external sensor 2. In this case, the first reliability calculation unit 14b may reduce the reliability of the first travel plan when the blind angle of the external sensor 2 is large compared to when the blind angle is small.

  For example, the first reliability calculation unit 14b may calculate the reliability of the first travel plan based on the route in the first travel plan, the map information in the map database 4, and the sensor blind spot of the external sensor 2. Good. The sensor blind spot of the external sensor 2 here refers to an area that is not a detection area of the external sensor 2 in the first place. Further, the sensor blind spot of the external sensor 2 refers to an area that is not an area to be detected by the external sensor 2 in spite of being an area that should be detected by the external sensor 2 when traveling according to the first travel plan. For example, when there is a lane that merges from the side on the route traveled by the vehicle M (route in the first travel plan), the lane that merges from the side is an area to be detected by the external sensor 2. And when the detection area of the external sensor 2 provided in the vehicle M is only the front, the lanes joining from the side cannot be detected by the external sensor 2 because the detection directions are different. For this reason, the lane that merges from the side becomes the sensor blind spot of the external sensor 2. In this case, the first reliability calculation unit 14b may lower the reliability of the first travel plan when the sensor blind spot is large compared to when the sensor blind spot is low.

  For example, the first reliability calculation unit 14b is based on the number of lane changes of the vehicle M obtained based on the route in the first travel plan and the map information in the map database 4, or the number of merging of driving lanes. The reliability of the first travel plan may be calculated. In this case, the first reliability calculation unit 14b increases the reliability of the first travel plan when the number of lane changes of the vehicle M or the number of merging of the driving lanes is large, compared to the case where the number of times is small. It may be lowered.

  For example, the first reliability calculation unit 14b determines the reliability of the first travel plan based on the degree of coincidence between the generated first travel plan and a predetermined travel plan pattern that allows stable travel. It may be calculated. In this case, the first reliability calculation unit 14b has a lower coincidence between the generated first travel plan and a predetermined travel plan pattern that allows stable travel than when the coincidence is high. Thus, the reliability of the first travel plan may be lowered. For example, when the first travel plan includes a merger with the priority road, the first reliability calculation unit 14b may reduce the reliability of the first travel plan.

  For example, the first reliability calculation unit 14b includes the recognition result of the position recognition unit 11 used when generating the first travel plan, the recognition result of the environment recognition unit 12, and the recognition result of the vehicle state recognition unit 13. The reliability of the first travel plan may be calculated based on the degree of coincidence with a predetermined condition that can stably generate the travel plan. In this case, if the degree of coincidence between the recognition result of the position recognizing unit 11 used when generating the first travel plan and a predetermined condition is low, the first reliability calculation unit 14b has a degree of coincidence. Compared to a high case, the reliability of the first travel plan may be lowered.

  For example, the first reliability calculation unit 14b is based on the degree of coincidence between the movement of the vehicle around the vehicle M recognized by the environment recognition unit 12 and the movement of the vehicle M in the generated first travel plan. The reliability of one travel plan may be calculated. In this case, the first reliability calculation unit 14b may reduce the reliability of the first travel plan when the degree of coincidence between the motion of the surrounding vehicles and the motion of the vehicle M in the first travel plan is low. For example, when the movement of the vehicle M in the first travel plan continues to stop despite the start of surrounding vehicles at an intersection where a traffic signal exists, the first reliability calculation unit 14b Reduce reliability.

  The first reliability calculation unit 14b may calculate the reliability of the first travel plan based on the plurality of reliability calculated based on the plurality of methods for calculating the reliability of the first travel plan described above. .

  The second computing unit 15 computes the second travel plan and the reliability of the second travel plan used when performing the automatic operation control of the vehicle M. The second calculation unit 15 generates a second travel plan by a method different from that of the first calculation unit 14. The 2nd calculating part 15 is provided with the 2nd plan part 15a and the 2nd reliability calculation part 15b.

  The second planning unit 15a generates a travel plan as a second travel plan for the purpose of easily taking over the driving operation to the driver from the state in which the automatic operation control is performed based on the second travel plan. To do. Specifically, the travel plan is such that the state of the vehicle M is not significantly changed per unit time. Unlike the first travel plan, for example, the second travel plan does not include driving behavior such as obstacle avoidance and lane change. If the movement of the vehicle M is slow as in the case where the vehicle M travels based on the second travel plan, the state of the vehicle M when the driving operation is taken over when the driving operation is taken over by the driver. The driver can easily adapt. That is, when the automatic driving control of the vehicle M is performed based on the second travel plan, the driver can easily take over the driving operation from the state where the automatic driving is performed.

  The second plan unit 15 a generates a second travel plan for the vehicle M based on the state of the vehicle M recognized by the vehicle state recognition unit 13. Specifically, the second planning unit 15a determines the amount of change in the acceleration / deceleration of the vehicle M per unit time from the current state (vehicle speed, acceleration, steering angle, etc.) of the vehicle M recognized by the vehicle state recognition unit 13. The second travel plan is generated so that at least one of the steering change amount is smaller than a predetermined value. That is, the second travel plan is a travel plan in which the movement of the vehicle M does not change significantly from the current state of the vehicle M. Thus, the 2nd plan part 15a should just produce | generate a travel plan looser than the 1st travel plan which the 1st plan part 14a produces | generates as a 2nd travel plan. For example, the second planning unit 15a generates a second travel plan for decelerating the vehicle M at a deceleration smaller than a predetermined value from the current vehicle speed of the vehicle M. Thus, the 2nd plan part 15a generates the 2nd run plan using only the recognition result in vehicle state recognition part 13, without using the recognition result in position recognition part 11 and environment recognition part 12.

  As the second travel plan, the second plan unit 15a generates, for example, a travel plan including a plurality of target positions where the vehicle M travels and the speed at each target position, as in the first travel plan. That is, the second travel plan includes a route plan and a speed plan. Moreover, the 2nd plan part 15a starts the production | generation of a 2nd travel plan, when a driver performs start operation of automatic driving | operation control.

  The second reliability calculation unit 15b calculates the reliability of the second travel plan generated by the second planning unit 15a. The reliability of the second travel plan represents the degree of whether or not the vehicle M can travel suitably in accordance with standards such as safety, legal compliance, and travel efficiency. Unlike the first travel plan, the second travel plan does not consider the surrounding environment. For this reason, generally, the reliability of the second travel plan is lower than the reliability of the first travel plan.

  Specifically, the second reliability calculation unit 15b is based on at least one of the reliability of the recognition result used when the second planning unit 15a generates the second travel plan and the second travel plan. 2. Calculate the reliability of the travel plan. Here, the second reliability calculation unit 15b is at least one of the reliability of the recognition of the state of the vehicle M recognized by the vehicle state recognition unit 13 and the second travel plan generated by the second planning unit 15a. Based on this, the reliability of the second travel plan is calculated.

  For example, the second reliability calculation unit 15b may calculate the reliability of the second travel plan based on the generation cycle of the second travel plan in the second planning unit 15a. In this case, when the generation period of the second travel plan is longer than the predetermined period, the second reliability calculation unit 15b lowers the reliability of the second travel plan compared to the case where the generation period is shorter than the predetermined period. May be. For example, the second reliability calculation unit 15b may calculate the reliability of the second travel plan based on the reliability of the recognition of the state of the vehicle M calculated by the vehicle state recognition unit 13. In this case, when the reliability of the recognition of the state of the vehicle M calculated by the vehicle state recognition unit 13 is low, the second reliability calculation unit 15b is compared with the case where the reliability of the recognition of the state of the vehicle M is high. The reliability of the second travel plan may be lowered.

  The second reliability calculation unit 15b may calculate the reliability of the second travel plan based on the plurality of reliability calculated based on the plurality of methods of calculating the reliability of the second travel plan described above. .

  The selection unit 16 compares the reliability of the first travel plan calculated by the first reliability calculation unit 14b with the reliability of the second travel plan calculated by the second reliability calculation unit 15b. A travel plan having a high reliability of the travel plan is selected from the travel plan and the second travel plan. After the selection unit 16 selects the second travel plan, the selection unit 16 further accepts an input operation for selecting the first travel plan by the input unit 8a when the presentation unit 8b presents the first travel plan. Then, the first travel plan is selected instead of the second travel plan.

  The travel control unit 17 performs automatic driving control of the vehicle M based on the travel plan selected by the selection unit 16. Specifically, the travel control unit 17 outputs a control signal corresponding to the travel plan to the actuator 7. Thereby, the traveling control unit 17 controls the traveling of the vehicle M so that the vehicle M automatically travels along the traveling plan.

  The presentation control part 18 produces | generates the display image displayed on the presentation part 8b, and displays the produced | generated display image on the presentation part 8b. When the second travel plan is selected by the selection unit 16, the presentation control unit 18 generates a display image so that the first travel plan is presented to the driver of the vehicle and causes the presentation unit 8 b to display the display image.

  Here, an example of a display image presented to the driver through the presentation unit 8b will be described. FIG. 2 shows an example of an image displayed on the display of the presentation unit 8b. As shown in FIG. 2, a vehicle X serving as an obstacle is stopped on the shoulder in front of the vehicle M. In this case, the first planner 14a generates a first travel plan having a route L1 that avoids the vehicle X, as indicated by a broken line. Moreover, the 2nd plan part 15a produces | generates the 2nd driving plan which has the path | route L2 for stopping before the vehicle X, as shown with a continuous line.

  In the example shown in FIG. 2, a physical blind spot area Z of the external sensor 2 that cannot be detected from the vehicle M because it is blocked by the vehicle X exists on the right side of the vehicle X. Since the route L1 of the first travel plan passes through the blind spot area Z, the reliability of the first travel plan is reduced. On the other hand, since the route L2 of the second travel plan does not pass through the blind spot area Z, there is no decrease in the reliability of the second travel plan due to the presence of the blind spot area Z. Therefore, for example, in the example illustrated in FIG. 2, the reliability of the second travel plan is higher than the reliability of the first travel plan, and the selection unit 16 selects the second travel plan. In this case, the presentation control unit 18 generates a display image in which the route L2 of the second travel plan selected by the selection unit 16 is displayed with a solid line, and causes the presentation unit 8b to display the display image. Further, since the second travel plan is selected by the selection unit 16, the presentation control unit 18 generates a display image for presenting the route L1 of the first travel plan to the driver, and the route L1 is also presented by the presentation unit 8b. To display. Here, the presentation control unit 18 generates a display image indicating the route L1 of the first travel plan with a broken line as a travel plan candidate that can be selected by the driver.

  Thus, even when the second travel plan is selected by the selection unit 16 and the automatic operation control of the vehicle M is performed based on the second travel plan, the driver recognizes the first travel plan through the presentation unit 8b. Can do. Then, based on the display image displayed on the presentation unit 8b, the driver can determine whether or not the first travel plan is acceptable. When the driver determines that the first travel plan is acceptable and performs an input operation for selecting the first travel plan on the input unit 8a, the selection unit 16 replaces the second travel plan with the first travel plan. Select. Thereby, instead of the second travel plan, automatic driving control of the vehicle M is performed based on the first travel plan.

  Next, the flow of processing executed by the ECU 6 will be described using the flowchart of FIG. This process is executed by the ECU 6 when the driver of the vehicle M instructs execution of automatic driving control. When the process of the flowchart reaches the end, the ECU 6 repeats the process from the start again. The ECU 6 ends the process shown in FIG. 3 when the execution of the automatic driving control is canceled by the driver of the vehicle M.

  As shown in FIG. 3, the position recognition unit 11 recognizes the position of the vehicle M (S101). The environment recognition unit 12 recognizes the surrounding environment of the vehicle M (S102). The vehicle state recognition unit 13 recognizes the state of the vehicle M (S103). The processing order of S101 to S103 is not limited to being executed in this arrangement order, and the processes may be executed in parallel. The first planning unit 14a generates a first travel plan based on the recognition result of the position recognition unit 11 and the like (S104). The first reliability calculation unit 14b calculates the reliability of the generated first travel plan (S105).

  The second planning unit 15a generates a second travel plan for the vehicle M based on the state of the vehicle M (S106). The second reliability calculation unit 15b calculates the reliability of the generated second travel plan (S107). In addition, the process of S104-S105 and the process of S106-S107 are not limited to performing the process of S106-S107 after the process of S104-S105. For example, the processes of S104 to S105 and the processes of S106 to S107 may be executed in parallel.

  The selection unit 16 compares the reliability of the first travel plan with the reliability of the second travel plan, and selects a travel plan with a high reliability of the travel plan (S108). The travel control unit 17 performs automatic driving control of the vehicle M based on the travel plan selected by the selection unit 16 (S109). The presentation control unit 18 determines whether or not the second travel plan is selected by the selection unit 16 (S110). When the second travel plan is not selected (S110: NO), that is, when the first travel plan is selected, the ECU 6 performs the process from S101 again. When the second travel plan is selected (S110: YES), the presentation control unit 18 determines that the first travel plan is a driver of the vehicle as shown by a route L1 (route of the first travel plan) indicated by a broken line in FIG. The presenting unit 8b is controlled so as to be presented (S111).

  The selection unit 16 determines whether or not the driver has performed an input operation for selecting the first travel plan using the input unit 8a when the first travel plan is presented in the presentation unit 8b (S112). If the driver has not performed an input operation for selecting the first travel plan (S112: NO), the ECU 6 performs the process from S101 again. When the driver performs an input operation for selecting the first travel plan (S112: YES), that is, when the driver performs an input operation for permitting the first travel plan, the selection unit 16 selects the second travel plan. Instead, the first travel plan is selected (S113). The travel control unit 17 performs automatic driving control of the vehicle M based on the first travel plan selected by the selection unit 16 (S114).

  The present embodiment is configured as described above. In the automatic driving system 100, the travel control unit 17 automatically controls the vehicle M based on the travel plan selected by the selection unit 16 from the first travel plan and the second travel plan. Perform operation control. Here, the 1st plan part 14a produces | generates a 1st driving | running plan using all the recognition results in the position recognition part 11, the environment recognition part 12, and the vehicle state recognition part 13. FIG. For this reason, for example, when the reliability of any of the recognition results of the position recognition unit 11, the environment recognition unit 12, and the vehicle state recognition unit 13 decreases, the reliability of the first travel plan generated using all of these results The degree decreases.

  On the other hand, the 2nd plan part 15a generates the 2nd run plan based on the kind of recognition result fewer than the 1st plan part 14a. Specifically, the second planning unit 15a generates the second travel plan based on the recognition result of the vehicle state recognition unit 13 without using the recognition results of the position recognition unit 11 and the environment recognition unit 12. Since the second travel plan is generated based on fewer types of recognition results than the first travel plan, when the second travel plan is generated without using the recognition result having reduced reliability, the recognition result The second travel plan is generated without being affected by the decrease in reliability. When the second travel plan is generated in this way, since the recognition result with reduced reliability is not used, the reliability of the second travel plan is reduced due to the decrease in the reliability of the recognition result. Absent. Specifically, even if the reliability of the recognition result of the position recognition unit 11 and the recognition result of the environment recognition unit 12 is reduced, the second travel plan is generated without using these recognition results. The generation of the travel plan is not affected by the decrease in reliability.

  For this reason, the reliability of the first travel plan decreases with a decrease in the reliability of the recognition result of any one of the position recognition unit 11, the environment recognition unit 12, and the vehicle state recognition unit 13, and the automatic operation based on the first travel plan is performed. When the driving control is difficult and a situation occurs in which the driver takes over the driving operation, the second travel plan is selected by the selection unit 16. Thus, even if the situation where the automatic driving control based on the first travel plan becomes difficult and the driver takes over the driving operation occurs, the vehicle M is based on the second driving plan until the driver takes over the driving operation. The automatic operation control is continued. The driver can take over the driving operation while the vehicle is traveling based on the second travel plan.

  Here, even if the reliability of the first travel plan is reduced and the second travel plan is selected, the driver may determine that travel based on the first travel plan is acceptable. Therefore, the presentation unit 8b presents the first travel plan when the second travel plan is selected by the selection unit 16. Thereby, the driver can determine whether or not the first travel plan is acceptable. The input unit 8a receives an input operation for selecting the first travel plan performed by the driver. When the input operation for selecting the first travel plan is received, the selection unit 16 selects the first travel plan instead of the second travel plan. Thereby, the traveling control unit 17 performs automatic driving control of the vehicle M based on the first traveling plan instead of the second traveling plan. As described above, since the first travel plan is acceptable, when the driver performs an input operation for selecting the first travel plan, automatic operation control based on the first travel plan is performed. As described above, since the automatic operation control based on the first travel plan is resumed, it is possible to suppress an increase in the frequency with which the driver performs the manual operation as the automatic operation control based on the first travel plan ends.

(First modification)
Next, as a first modified example, a modified example of the second travel plan generation method in the second calculation unit 15 and a modified example of the reliability calculation of the second travel plan in the second reliability calculation unit 15b will be described. . In the first modification, the second planning unit 15a includes the position of the vehicle M recognized by the position recognition unit 11, the surrounding environment recognized by the environment recognition unit 12, and the state of the vehicle M recognized by the vehicle state recognition unit 13. Based on the map information in the map database 4, a second travel plan for the vehicle M is generated so as to travel along the target route set by the navigation system 5. The second planning unit 15a generates a second travel plan so that at least one of the change amount of acceleration / deceleration of the vehicle M and the change amount of steering per unit time is smaller than that of the first travel plan. A second travel plan is generated. That is, the second travel plan is a travel plan in which the movement of the vehicle M does not change significantly compared to the first travel plan.

  The second reliability calculation unit 15 b includes the reliability of the position of the vehicle M recognized by the position recognition unit 11, the reliability of the recognition of the surrounding environment of the vehicle M recognized by the environment recognition unit 12, and the vehicle state recognition unit 13. The reliability of the second travel plan is calculated based on at least one of the recognized reliability of the state of the vehicle M and the second travel plan generated by the second planning unit 15a. The calculation of the reliability of the second travel plan performed by the second reliability calculation unit 15b can be calculated in the same manner as the first reliability calculation unit 14b in the embodiment. Further, the second reliability calculation unit 15b corrects the calculated reliability of the second travel plan, and sets the corrected reliability of the second travel plan as the reliability in the second travel plan. The second reliability calculation unit 15b corrects the reliability of the second travel plan so that the reliability of the second travel plan becomes small when correcting the reliability of the second travel plan. For example, the second reliability calculation unit 15b may calculate the reliability of the second travel plan by correcting the reliability of the calculated second travel plan using a predetermined coefficient.

  In the case of the first modification, the second planning unit 15a uses the same recognition result as the first planning unit 14a, but the amount of change in the acceleration / deceleration of the vehicle M per unit time and the steering change are higher than those in the first travel plan. A second travel plan in which at least one of the changes is small is generated. In other words, the second plan unit 15a generates a second travel plan that causes the vehicle M to travel without significantly changing the state of the vehicle M, as compared to the first travel plan.

  When the state of the vehicle M is not significantly changed, for example, a travel plan can be generated even when the recognition accuracy of the surrounding environment of the vehicle M is lower than when the state of the vehicle M is largely changed. For this reason, compared with the case where the 1st plan part 14a produces | generates a 1st travel plan, the 2nd plan part 15a has the reliability of the recognition result in the position recognition part 11, the environment recognition part 12, and the vehicle state recognition part 13. Even if it is low, the second travel plan can be generated. Further, since the second travel plan can be generated even if the reliability of the recognition result is low, the reliability of the second travel plan is higher than that of the first travel plan even if the reliability of the recognition result is reduced. The effect is small. In this way, it is possible to generate the second travel plan in the second planning unit 15a and calculate the reliability of the second travel plan in the second reliability calculation unit 15b.

(Second modification)
Next, as a second modification, a modification of the second travel plan generation method in the second calculation unit 15 and a modification of the second travel plan reliability calculation in the second reliability calculation unit 15b will be described. . In the second modification, the second planning unit 15a generates a second travel plan based on the past first travel plan generated by the first planning unit 14a.

  Here, a first example in which the second planning unit 15a generates the second travel plan based on the past first travel plan will be described. The second planning unit 15a acquires the first travel plan and the reliability of the first travel plan from the first calculation unit 14 every time they are generated and calculated. And the 2nd plan part 15a uses the 1st travel plan just before the reliability of the acquired 1st travel plan becomes below a predetermined value as a 2nd travel plan. The second reliability calculation unit 15b corrects the reliability of the first travel plan used as the second travel plan by the second planning unit 15a, and sets the corrected reliability of the first travel plan to the second travel plan. Reliability.

  Since the second travel plan is generated using the first travel plan generated in the past, the reliability of the second travel plan is the same as that of the first travel plan used when generating the second travel plan. It becomes lower than the reliability. For this reason, the second reliability calculation unit 15b reduces the reliability by multiplying the reliability of the first travel plan used for generating the second travel plan by a predetermined correction coefficient. The reliability of the first travel plan after correction is set as the reliability of the second travel plan. In addition, when correcting the reliability of the first travel plan to be low, the reliability may be corrected to be lower as time passes from the present (older).

  Next, a second example in which the second planning unit 15a generates the second travel plan based on the past first travel plan will be described. The second planning unit 15a acquires the reliability of the first travel plan and the first travel plan from the first calculation unit 14 each time they are generated and calculated. The 2nd plan part 15a uses the 1st travel plan with the highest reliability as the 2nd travel plan among the reliability of the 1st travel plan acquired from the present to predetermined time ago. The second planning unit 15a corrects the reliability of the first travel plan so that the value decreases with time, and uses the first travel plan with the highest reliability after the correction as the second travel plan. May be.

  The second reliability calculation unit 15b corrects the reliability of the first travel plan used as the second travel plan to be low, and sets the corrected reliability of the first travel plan as the reliability of the second travel plan. can do. In addition, when correcting the reliability of the first travel plan to be low, the reliability may be corrected to be lower as time passes from the present (older).

  In addition, although the 1st example and the 2nd example were shown as an example which produces | generates a 2nd travel plan based on the past 1st travel plan, if based on the past 1st travel plan, other than these The second travel plan may be generated by a method.

  In the case of the second modification, the second planning unit 15a generates a second travel plan based on the first travel plan generated in the past by the first planning unit 14a. Further, since the second reliability calculation unit 15b calculates the reliability of the second travel plan based on the reliability of the first travel plan generated in the past, the reliability of the recognition result or the like in the position recognition unit 11 is high. Even if it decreases, the reliability of the second travel plan does not immediately decrease. For this reason, when the reliability of the first travel plan is reduced as the reliability of the recognition result in the position recognition unit 11 is reduced, the second travel plan is selected by the selection unit 16. Using the reliability of the second travel plan calculated in this way and the reliability of the first travel plan, the selection unit 16 can select a travel plan used for automatic operation control.

  In the second modification, the second planning unit 15a does not use the previous first travel plan as it is as the second travel plan, but the amount of change in acceleration / deceleration of the vehicle M and the amount of change in steering per unit time. The first travel plan may be corrected so that at least one of the two becomes smaller, and the corrected first travel plan may be set as the second travel plan.

  As mentioned above, although embodiment and various modifications of the present invention were explained, the present invention is not limited to the above-mentioned embodiment and various modifications. For example, in the automatic driving system 100A illustrated in FIG. 4, the selection unit 16A may select the first travel plan based on the detection result of the operation detection unit 9 instead of the input unit 8a. In the automatic driving system 100A shown in FIG. 4, the same components as those in the automatic driving system 100 shown in FIG. When the selection unit 16A selects the second travel plan by comparing the reliability of the first travel plan with the reliability of the second travel plan, the presenting unit 8b presents the first travel plan. Based on the detection result of the operation detector 9, it is determined whether the driver has performed a driving operation (accelerator operation, brake operation, steering) in accordance with the first travel plan. For example, in the example illustrated in FIG. 2, when the steering angle sensor of the operation detection unit 9 detects that the driver has steered in the right direction, the selection unit 16A causes the driver to drive according to the first travel plan. It is determined that the operation has been performed. When the driver performs a driving operation according to the first travel plan displayed on the presentation unit 8b, it can be said that the driver has determined that the first travel plan is acceptable. For this reason, the selection unit 16 </ b> A is used when the operation detection unit 9 detects that the driver has performed a driving operation along the first travel plan when the presentation unit 8 b is presenting the first travel plan. The first travel plan is selected instead of the second travel plan. Thereby, the traveling control unit 17 performs automatic driving control of the vehicle M based on the first traveling plan.

  In the embodiment, the second planning unit 15a generates the second travel plan for the vehicle M based on the state of the vehicle M recognized by the vehicle state recognition unit 13, but the recognition result of the vehicle state recognition unit 13 It is not limited to producing | generating a 2nd driving plan using only. For example, the second planning unit 15a includes the position of the vehicle M recognized by the position recognition unit 11, the surrounding environment of the vehicle M recognized by the environment recognition unit 12, and the state of the vehicle M recognized by the vehicle state recognition unit 13. Of these, the second travel plan may be generated based on one or two of them. Then, the second reliability calculation unit 15 b includes the position of the vehicle M recognized by the position recognition unit 11, the surrounding environment of the vehicle M recognized by the environment recognition unit 12, and the vehicle M recognized by the vehicle state recognition unit 13. Among the states, the reliability of the second travel plan is calculated based on at least one of the reliability of the recognition result used when the second planning unit 15a generates the second travel plan and the second travel plan. May be.

  In addition, when the second travel plan is selected by the selection units 16 and 16A, the display image displayed by the presentation control unit 18 through the presentation unit 8b is a specific example of the first travel plan such as the shape of the route of the first travel plan. It may be an image that displays typical information, and the one that causes the second travel plan to be selected (vehicle X in the example of FIG. 2) may be highlighted. The presentation control unit 18 also presents the second travel plan only when the reliability of the first travel plan is equal to or greater than a predetermined threshold when presenting the first travel plan when the second travel plan is selected. In addition, the first travel plan may be presented.

  Further, in the embodiment and various modifications, the first planning unit 14a does not use the position of the vehicle M recognized by the position recognition unit 11, but based on the recognition result other than the position of the vehicle M. May be generated.

  DESCRIPTION OF SYMBOLS 1 ... GPS receiving part (position measurement part), 2 ... External sensor, 3 ... Internal sensor, 8a ... Input part, 8b ... Presentation part, 11 ... Position recognition part, 12 ... Environment recognition part, 13 ... Vehicle state recognition part, DESCRIPTION OF SYMBOLS 14a ... 1st plan part, 14b ... 1st reliability calculation part, 15a ... 2nd plan part, 15b ... 2nd reliability calculation part, 16 ... Selection part, 17 ... Travel control part, 100,100A ... Automatic driving system , M ... Vehicle.

Claims (1)

An automatic driving system that performs automatic driving control of a vehicle,
A position recognition unit for recognizing the position of the vehicle based on a measurement result of the position measurement unit of the vehicle;
An environment recognition unit for recognizing the surrounding environment of the vehicle based on a detection result of an external sensor that detects an external situation of the vehicle;
A vehicle state recognition unit that recognizes the state of the vehicle based on a detection result of an internal sensor that detects the state of the vehicle;
A first planning unit that generates a first travel plan of the vehicle based on the recognized position of the vehicle, the recognized surrounding environment of the vehicle, and the recognized state of the vehicle;
At least one of the recognized reliability of the position of the vehicle, the recognized reliability of the surrounding environment of the vehicle, the recognized reliability of the state of the vehicle, and the generated first travel plan A first reliability calculation unit for calculating the reliability of the first travel plan,
Based on one or two of the recognized position of the vehicle, the recognized surrounding environment of the vehicle, and the recognized state of the vehicle, the state of the vehicle is more than the first travel plan. A second planning unit that generates a second travel plan that does not change significantly per time ;
Of the recognized position of the vehicle, the recognized surrounding environment of the vehicle, and the recognized state of the vehicle, the reliability of the recognition result used when the second planning unit generates the second travel plan. A second reliability calculation unit that calculates the reliability of the second travel plan based on at least one of the second travel plan and the generated second travel plan;
A selection unit for selecting one of the first travel plan and the second travel plan;
A presentation unit that presents the first travel plan to the driver of the vehicle when the second travel plan is selected by the selection unit;
An input unit for accepting an input operation for selecting the first travel plan by the driver;
A travel control unit that performs automatic driving control of the vehicle based on the travel plan selected by the selection unit;
The selection unit compares the reliability of the first travel plan with the reliability of the second travel plan, and the reliability of the travel plan is high among the first travel plan and the second travel plan. When the travel plan is selected and the second travel plan is selected based on the reliability of the travel plan, the first travel plan is displayed by the input unit when the first travel plan is presented by the presenting unit. When the selection input operation is accepted, the automatic driving system selects the first travel plan instead of the second travel plan.

Images