KR20200133444A - Autonomous driving apparatus and method - Google Patents

Abstract

The present invention relates to an autonomous driving device and a method thereof. The autonomous driving device includes: a memory that stores an autonomous driving algorithm for autonomous driving control of a user vehicle; and a processor that controls the autonomous driving of the user vehicle according to the autonomous driving algorithm stored in the memory. The processor determines whether the driving operation of a user vehicle driver is involved in a process of controlling the autonomous driving of the user vehicle according to the autonomous driving algorithm, and performs learning of the autonomous driving algorithm based on a comparison result between the control process according to the autonomous driving algorithm and the driving operation of the driver at the time when it is determined that the driver′s driving operation is involved.

Description

Autonomous driving device and method {AUTONOMOUS DRIVING APPARATUS AND METHOD}

The present invention relates to an autonomous driving apparatus and method applied to an autonomous driving vehicle.

Today's automotive industry is moving toward implementing autonomous driving that minimizes driver intervention in vehicle driving. An autonomous vehicle refers to a vehicle that independently determines a driving route by recognizing the surrounding environment through a function of sensing and processing external information during driving, and independently driving by using its own power.

Autonomous vehicles can drive themselves to their destination by preventing collisions with obstacles on the driving path and adjusting the vehicle speed and driving direction according to the shape of the road, even if the driver does not operate the steering wheel, accelerator pedal, or brake. have. For example, acceleration may be performed on a straight road and deceleration may be performed while changing a driving direction in response to a curvature of the road on a curved road.

In order to ensure stable driving of an autonomous vehicle, it is necessary to accurately measure the driving environment through each sensor mounted on the vehicle, and control the driving according to the measured driving environment by continuously monitoring the driving state of the vehicle. To this end, various sensors, such as a lidar sensor, a radar sensor, an ultrasonic sensor, and a camera sensor, are applied to the autonomous vehicle as a sensor for detecting surrounding objects such as surrounding vehicles, pedestrians, and fixed facilities. Data output from these sensors is used to determine information about the driving environment, such as the location, shape, movement direction, and movement speed of surrounding objects.

Furthermore, autonomous vehicles use pre-stored map data to determine and correct the location of the vehicle to optimally determine the driving route and driving lane, control the driving of the vehicle so as not to deviate from the determined path and lane, and abruptly It also provides the function of performing defense and avoidance driving against dangers existing on the vehicle entering or the driving path.

Background art of the present invention is disclosed in Korean Patent Application Publication No. 10-1998-0068399 (published on October 15, 1998).

An object according to an aspect of the present invention is to improve driving stability and driving accuracy of an autonomous vehicle by learning an autonomous driving algorithm applied to autonomous driving control in consideration of a driver's driving operation involved in the autonomous driving control process of the own vehicle. It is to provide a device and method for autonomous driving.

An autonomous driving apparatus according to an aspect of the present invention includes a memory for storing an autonomous driving algorithm for autonomous driving control of the own vehicle, and a processor for controlling the autonomous driving of the own vehicle according to the autonomous driving algorithm stored in the memory. , The processor, in the process of controlling the autonomous driving of the own vehicle according to the autonomous driving algorithm, determines whether a driving operation of the occupant of the host vehicle is involved, and when it is determined that the driving operation of the occupant is involved And learning of the autonomous driving algorithm is performed based on a comparison result between the control process according to the autonomous driving algorithm and the driving operation of the occupant at a time when the driving operation of the occupant is intervened.

In the present invention, the processor is characterized in that when it is determined that the driving operation of the occupant is intervened, the autonomous driving algorithm is learned after stopping the autonomous driving control of the own vehicle.

In the present invention, the processor is characterized in that, when the control process and the occupant's driving operation are different, the learning of the autonomous driving algorithm is performed after verifying the risk of the occupant's driving operation.

The present invention further includes a sensor unit for detecting surrounding objects of the host vehicle, and a driving information detection unit for detecting driving information on a driving state of the host vehicle, wherein the processor comprises: surrounding objects detected by the sensor unit The autonomous driving algorithm is learned based on the information on, driving information of the own vehicle detected by the driving information detection unit, the control process, and a driving operation of the occupant.

In an autonomous driving method according to an aspect of the present invention, a processor controls autonomous driving of the own vehicle according to an autonomous driving algorithm stored in a memory, and the processor controls autonomous driving of the own vehicle according to the autonomous driving algorithm. In the process of determining whether a driving operation of the occupant of the own vehicle is intervened, and when it is determined that the driving operation of the occupant is involved, the processor, at the time when the driving operation of the occupant is intervened. And learning the autonomous driving algorithm based on a comparison result between the control process according to the autonomous driving algorithm and the driving operation of the occupant.

According to an aspect of the present invention, the present invention learns an autonomous driving algorithm applied to autonomous driving control in consideration of the driving operation of a passenger involved in the autonomous driving control process of the own vehicle, and then sleeps according to the learned autonomous driving algorithm. By controlling the autonomous driving of the vehicle, driving stability and driving accuracy of the autonomous driving vehicle can be improved.

1 is an overall block diagram of an autonomous driving control system to which an autonomous driving device according to an embodiment of the present invention can be applied.
2 is a block diagram showing a detailed configuration of an integrated autonomous driving control unit in an autonomous driving device according to an embodiment of the present invention.
3 is an exemplary view showing an example in which an autonomous driving device according to an embodiment of the present invention is applied to a vehicle.
4 is an exemplary view showing an example of an internal structure of a vehicle to which an autonomous driving device according to an embodiment of the present invention is applied.
5 is an exemplary view showing an example of a set distance and a horizontal angle of view at which a lidar sensor, a radar sensor, and a camera sensor can detect surrounding objects in an autonomous driving apparatus according to an embodiment of the present invention.
6 is an exemplary view showing an example in which a sensor unit detects a nearby vehicle in an autonomous driving device according to an embodiment of the present invention.
7 is a flowchart illustrating an autonomous driving method according to an embodiment of the present invention.

Hereinafter, embodiments of an autonomous driving apparatus and method according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of the lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described later are terms defined in consideration of functions in the present invention and may vary according to the intention or custom of users or operators. Therefore, definitions of these terms should be made based on the contents throughout the present specification.

1 is an overall block diagram of an autonomous driving control system to which an autonomous driving device according to an embodiment of the present invention can be applied, and FIG. 2 is a detailed diagram of an autonomous driving integrated control unit in an autonomous driving device according to an embodiment of the present invention. A block diagram showing the configuration, FIG. 3 is an exemplary view showing an example in which an autonomous driving device according to an embodiment of the present invention is applied to a vehicle, and FIG. 4 is an autonomous driving device according to an embodiment of the present invention applied. It is an exemplary view showing an example of the internal structure of the vehicle, and FIG. 5 is a set distance and a horizontal angle of view at which a lidar sensor, a radar sensor, and a camera sensor can detect surrounding objects in an autonomous driving device according to an embodiment of the present invention. 6 is an exemplary diagram illustrating an example in which a sensor unit detects a nearby vehicle in an autonomous driving apparatus according to an embodiment of the present invention.

First, the structure and function of the autonomous driving control system to which the autonomous driving apparatus according to the present embodiment can be applied will be described with reference to FIGS. 1 and 3. As shown in FIG. 1, the autonomous driving control system is used to control autonomous driving of a vehicle through a driving information input interface 101, a driving information input interface 201, a passenger output interface 301, and a vehicle control output interface 401. It may be implemented around the autonomous driving integrated control unit 600 that transmits and receives necessary data.

The integrated autonomous driving control unit 600 may obtain driving information according to a user's manipulation of the user input unit 100 in the autonomous driving mode or the manual driving mode of the vehicle through the driving information input interface 101. The user input unit 100 includes a driving mode switch 110 and a user terminal 120 (e.g., a navigation terminal mounted on a vehicle, a smartphone or a tablet PC, etc.) possessed by a passenger, as shown as an example in FIG. 1 Accordingly, the driving information may include driving mode information and navigation information of the vehicle. For example, the driving mode of the vehicle determined according to the operation of the occupant on the driving mode switch 110 (i.e., autonomous driving mode/manual driving mode, or sports mode/eco mode/safety A mode (Safe Mode)/Normal Mode) may be transmitted to the autonomous driving integrated control unit 600 through the driving information input interface 101 as the driving information described above. In addition, navigation information such as the passenger's destination input through the user terminal 120 by the occupant and the route to the destination (the shortest route selected by the occupant among candidate routes to the destination, or a preferred route, etc.) is driving information as described above. It may be transmitted to the integrated autonomous driving controller 600 through the input interface 101. Meanwhile, the user terminal 120 may be implemented as a control panel (eg, a touch screen panel) that provides a user interface (UI) for the driver to input or modify information for autonomous driving control of the vehicle. In this case, the driving mode switch 110 described above may be implemented as a touch button on the user terminal 120.

In addition, the autonomous driving integrated control unit 600 may obtain driving information indicating the driving state of the vehicle through the driving information input interface 201. The driving information includes the steering angle formed as the passenger manipulates the steering wheel, the accelerator pedal stroke or the brake pedal stroke formed by depressing the accelerator pedal or the brake pedal, and the vehicle speed, acceleration, yaw, and pitch as the behavior formed in the vehicle. And various information indicating the driving state and behavior of the vehicle such as a roll, and each of the driving information includes a steering angle sensor 210, an APS (Accel Position Sensor)/PTS (Pedal Travel Sensor) as shown in FIG. 220, the vehicle speed sensor 230, the acceleration sensor 240, it may be detected by the driving information detection unit 200 including the yaw / pitch / roll sensor 250. Furthermore, the driving information of the vehicle may include location information of the vehicle, and the location information of the vehicle may be obtained through the GPS (Global Positioning Sysetm) receiver 260 applied to the vehicle. The driving information may be transmitted to the integrated autonomous driving controller 600 through the driving information input interface 201 and used to control the driving of the vehicle in the autonomous driving mode or the manual driving mode.

In addition, the integrated autonomous driving controller 600 may transmit driving state information provided to the occupant in the autonomous driving mode or the manual driving mode of the vehicle to the output unit 300 through the occupant output interface 301. That is, the integrated autonomous driving control unit 600 transmits the driving state information of the vehicle to the output unit 300, so that the occupant can drive the vehicle autonomously or manually based on the driving state information output through the output unit 300. The state may be checked, and the driving state information may include various information indicating the driving state of the vehicle, such as a current driving mode, a shift range, and a vehicle speed. In addition, the autonomous driving integrated control unit 600 outputs warning information through the occupant output interface 301 when it is determined that the driver needs a warning in the autonomous driving mode or the manual driving mode of the vehicle together with the driving status information. By passing it to 300, the output unit 300 may output a warning to the driver. In order to audibly and visually output such driving state information and warning information, the output unit 300 may include a speaker 310 and a display device 320 as shown in FIG. 1. In this case, the display device 320 may be implemented as the same device as the user terminal 120 described above, or may be implemented as a separate and independent device.

In addition, the autonomous driving integrated control unit 600 may transmit control information for driving control of the vehicle in the autonomous driving mode or the manual driving mode of the vehicle to the lower control system 400 applied to the vehicle through the vehicle control output interface 401. have. The lower control system 400 for driving control of a vehicle may include an engine control system 410, a braking control system 420, and a steering control system 430 as shown in FIG. 1, and an integrated autonomous driving control unit 600 may transmit engine control information, braking control information, and steering control information as the control information to each of the sub-control systems 410, 420, and 430 through the vehicle control output interface 401. Accordingly, the engine control system 410 may increase or decrease the fuel supplied to the engine to control the vehicle speed and acceleration, and the braking control system 420 may control the braking of the vehicle by adjusting the braking force of the vehicle. In addition, the steering control system 430 may control the steering of the vehicle through a steering device applied to the vehicle (eg, a Motor Driven Power Steering (MDPS) system).

As described above, the autonomous driving integrated control unit 600 according to the present embodiment provides driving information according to the driver's manipulation and driving information indicating the driving state of the vehicle through the driving information input interface 101 and the driving information input interface 201, respectively. The driving state information and warning information generated according to the autonomous driving algorithm acquired and processed by the internal processor 610 may be transmitted to the output unit 300 through the occupant output interface 301, and the internal processor ( Control information generated according to the autonomous driving algorithm processed by 610 may be transmitted to the lower control system 400 through the vehicle control output interface 401 to perform driving control of the vehicle.

Meanwhile, in order to ensure stable autonomous driving of the vehicle, it is necessary to continuously monitor the driving state by accurately measuring the driving environment of the vehicle and control the driving according to the measured driving environment. To this end, the autonomous driving apparatus of this embodiment As shown in FIG. 1, it may include a sensor unit 500 for detecting objects around the vehicle, such as surrounding vehicles, pedestrians, roads, or fixed facilities (eg, traffic lights, milestones, traffic signs, construction fences, etc.). The sensor unit 500 may include one or more of a lidar sensor 510, a radar sensor 520, and a camera sensor 530 to detect surrounding objects outside the vehicle as shown in FIG. 1.

The lidar sensor 510 can detect surrounding objects outside the vehicle by transmitting a laser signal around the vehicle and receiving a signal that is reflected and returned to the object. It is possible to detect surrounding objects located within a vertical field of view and a set horizontal field of view. The lidar sensor 510 may include a front lidar sensor 511, an upper lidar sensor 512, and a rear lidar sensor 513 respectively installed at the front, upper and rear of the vehicle, but the installation position thereof And the number of installations is not limited to a specific embodiment. The threshold value for determining the validity of the laser signal reflected and returned from the object may be stored in advance in the memory 620 of the integrated autonomous driving controller 600, and the processor 610 of the integrated autonomous driving controller 600 Is to determine the position (including the distance to the object), speed, and movement direction of the object through a method of measuring the time when the laser signal transmitted through the lidar sensor 510 is reflected on the object and returns. I can.

The radar sensor 520 can detect surrounding objects outside the vehicle by emitting electromagnetic waves around the vehicle and receiving a signal reflected back from the object, and a preset distance, a set vertical angle of view, and Nearby objects located within the set horizontal angle of view can be detected. The radar sensor 520 includes a front radar sensor 521, a left radar sensor 521, a right radar sensor 522, and a rear radar sensor 523 respectively installed on the front, left, right and rear of the vehicle. However, the installation location and the number of installations are not limited to specific embodiments. The processor 610 of the autonomous driving integrated control unit 600 analyzes the power of the electromagnetic waves transmitted and received through the radar sensor 520 to determine the location of the object (including the distance to the object) and speed. And the moving direction can be determined.

The camera sensor 530 may detect surrounding objects outside the vehicle by photographing the surroundings of the vehicle, and may detect surrounding objects located within a preset distance, a set vertical angle of view, and a set horizontal angle of view according to the specifications. . The camera sensor 530 includes a front camera sensor 531, a left camera sensor 532, a right camera sensor 533, and a rear camera sensor 534 respectively installed on the front, left, right and rear of the vehicle. However, the installation location and the number of installations are not limited to specific embodiments. The processor 610 of the autonomous driving integrated control unit 600 applies a predefined image processing processing to the image captured through the camera sensor 530, so that the location of the object (including the distance to the object) and speed And the moving direction can be determined. In addition, an internal camera sensor 535 for capturing the inside of the vehicle may be mounted at a predetermined position (eg, a rear view mirror) inside the vehicle, and the processor 610 of the autonomous driving integrated control unit 600 is an internal camera A guide or warning may be output to the occupant through the above-described output unit 300 by monitoring the occupant's behavior and condition based on the image acquired through the sensor 535.

In addition to the lidar sensor 510, the radar sensor 520, and the camera sensor 530, the sensor unit 500 may further include an ultrasonic sensor 540 as shown in FIG. Various types of sensors for detecting surrounding objects may be further employed in the sensor unit 500. 3 is a front lidar sensor 511 or a front radar sensor 521 installed on the front of the vehicle, and a rear lidar sensor 513 or a rear radar sensor 524 is installed at the rear of the vehicle to aid understanding of this embodiment. The front camera sensor 531, the left camera sensor 532, the right camera sensor 533, and the rear camera sensor 534 are respectively installed on the front, left, right and rear sides of the vehicle. , As described above, the installation position and number of each sensor are not limited to a specific embodiment. FIG. 5 shows an example of a set distance and a horizontal angle of view at which the lidar sensor 510, the radar sensor 520, and the camera sensor 530 can detect surrounding objects in front, and FIG. 6 is An example of detecting an object is shown. 6 is only an example of detecting surrounding objects, and a method of detecting surrounding objects is determined depending on the installation location and number of sensors. According to the configuration of the sensor unit 500 described above, surrounding vehicles and surrounding objects in an omnidirectional region of the host vehicle may be detected.

Furthermore, the sensor unit 500 detects voice and bio signals of the occupant (eg, heart rate, electrocardiogram, respiration, blood pressure, body temperature, brain waves, blood flow (pulse wave), blood sugar, etc.) to determine the state of the occupant in the vehicle. It may further include a microphone and a biometric sensor, and the biometric sensor includes a heart rate sensor, an electrocardiogram sensor, a breathing sensor, a blood pressure sensor, a body temperature sensor, an electroencephalogram sensor, a photoplethysmography sensor, and a blood sugar sensor. This can be.

4 shows an example of the internal structure of the vehicle, and the state of the vehicle is controlled by the operation of the occupant, such as the driver or passenger of the vehicle, to facilitate the driver's driving or convenience (e.g., rest, entertainment activities, etc.) Internal devices may be installed to support. Such an internal device may include a vehicle seat (S) on which the occupant is seated, a lighting device such as an interior light and mood lamp (L), the aforementioned user terminal 120 and a display device 320, an interior table, and the like. The state of the internal device may be controlled by the processor 610.

In the case of the vehicle seat S, the angle can be adjusted by the processor 610 (or by manual operation of the occupant), and the vehicle seat S is the front row seat S1 and the rear row seat S2 If it is composed of, only the angle of the front row seat (S1) can be adjusted. If the rear row seat (S2) is not provided and the front row seat (S1) is divided into a seat structure and a footrest structure, the seat structure of the front row seat (S1) is physically separated from the footrest structure. It can be implemented so that the angle is adjusted. In addition, an actuator (eg, a motor) for adjusting the angle of the vehicle seat S may be provided. In the case of the lighting device L, the on/off may be controlled by the processor 610 (or by manual operation of the occupant), and the lighting device L may operate a plurality of lighting units such as internal lights and mood lights. When included, each lighting unit can be independently controlled on and off. The user terminal 120 or the display device 320 may adjust the angle according to the occupant's viewing angle by the processor 610 (or by the occupant's manual operation), for example, the screen in the direction of the occupant's gaze. The angle can be adjusted so that there is. In this case, an actuator (eg, a motor) for adjusting the angles of the user terminal 120 and the display device 320 may be provided.

The integrated autonomous driving controller 600 may communicate with the server 700 through a network as shown in FIG. 1. As a network method between the autonomous driving integrated control unit 600 and the server 700, various communication methods such as a wide area network (WAN), a local area network (LAN), or a personal area network (PAN) may be employed. In addition, in order to secure wide network coverage, LPWAN (Low Power Wide Area Network, a network with very wide coverage among Internet of Things, including commercialized technologies such as LoRa, Sigfox, Ingenu, LTE-M, NB-IOT, etc.) communication The scheme can be employed. For example, LoRa (low-power communication is possible, but has a wide coverage of up to 20Km), or Sigfox (which has a coverage of 10Km (city) to 30Km (outskirts outside the city) depending on the environment) is adopted. In addition, 3GPP (3rd Generation) such as LTE-MTC (Machine-type Communications) (or LTE-M), NB (Narrowband) LTE-M, and NB IoT having a power saving mode (PSM) Partnership Project) Release 12, 13-based LTE network technology may be employed. The server 700 may provide map information that is kept up to date (various map information such as two-dimensional navigation map data, three-dimensional remote map data, or three-dimensional high-precision electronic map data may correspond), and further It can also provide various information such as accident information, road control information, traffic volume information and weather information. The autonomous driving integrated control unit 600 may receive the latest map information from the server 700 and update the map information stored in the memory 620, and receive accident information, road control information, traffic volume information, and weather information to It can also be used for autonomous driving control.

Next, the structure and function of the integrated autonomous driving controller 600 according to the present embodiment will be described with reference to FIG. 2. As shown in FIG. 2, the integrated autonomous driving controller 600 may include a processor 610 and a memory 620.

The memory 620 may store basic information necessary for the autonomous driving control of the vehicle, or may store information generated in the process of controlling the autonomous driving of the vehicle by the processor 610, and the processor 610 may store the memory 620 ), you can control the autonomous driving of the vehicle by accessing (read, access) the information stored in it. The memory 620 may be implemented as a computer-readable recording medium and may operate to allow the processor 610 to access it. Specifically, the memory 620 is a hard drive, magnetic tape, memory card, read-only memory (ROM), random-access memory (RAM), digital video disc (DVD), or optical disk. It may be implemented as an optical data storage device such as.

Map information required for autonomous driving control by the processor 610 may be stored in the memory 620. The map information stored in the memory 620 may be a navigation map (a numerical topographic map) that provides road-level information, but a precision road map that provides road information in units of lanes to improve the precision of autonomous driving control, That is, it may be desirable to be implemented as 3D high-precision electronic map data. Accordingly, the map information stored in the memory 620 is dynamic and necessary for autonomous driving control of the vehicle, such as lanes, lane center lines, regulatory lines, road boundaries, road center lines, traffic signs, road signs, road shapes and heights, and lane widths. Static information can be provided.

In addition, the memory 620 may store an autonomous driving algorithm for autonomous driving control of a vehicle. The autonomous driving algorithm is an algorithm (recognition, determination and control algorithm) that recognizes the surroundings of an autonomous vehicle, determines its state, and controls the driving of the vehicle according to the determination result, and the processor 610 is an autonomous vehicle stored in the memory 620 By executing the driving algorithm, it is possible to perform active autonomous driving control in the surrounding environment of the vehicle.

The processor 610 includes driving information and driving information respectively input from the driving information input interface 101 and the driving information input interface 201 described above, information on surrounding objects detected through the sensor unit 500, and memory. The autonomous driving of the vehicle may be controlled based on the map information stored in 620 and the autonomous driving algorithm. The processor 610 may be implemented as a dedicated semiconductor circuit such as an embedded processor such as a complex instruction set computer (CISC) or a reduced instruction set computer (RISC), or an application specific integrated circuit (ASIC). .

In this embodiment, the processor 610 may control autonomous driving of the own vehicle by analyzing each driving trajectory of the own vehicle and the surrounding vehicle. To this end, as shown in FIG. 2, the processor 610 includes a sensor processing module ( 611), a driving trajectory generation module 612, a driving trajectory analysis module 613, a driving control module 614, a trajectory learning module 615, and a passenger status determination module 616. Although FIG. 2 shows each module as an independent block according to a function, each module may be integrated into one module to be implemented in a configuration in which each function is integrated.

The sensor processing module 611 includes driving information of the surrounding vehicle (that is, the location of the surrounding vehicle, and includes the location of the surrounding vehicle and the speed of the surrounding vehicle along with the location) based on the result of detecting the surrounding vehicle of the host vehicle through the sensor unit 500. It may further include a moving direction). That is, the position of the surrounding vehicle is determined based on the signal received through the lidar sensor 510, the position of the surrounding vehicle is determined based on the signal received through the radar sensor 520, or the camera sensor 530 The location of the surrounding vehicle may be determined based on the image captured through the device, or the location of the surrounding vehicle may be determined based on a signal received through the ultrasonic sensor 540. To this end, the sensor processing module 611 as shown in FIG. 1 may include a lidar signal processing module 611a, a radar signal processing module 611b, and a camera signal processing module 611c (ultrasonic signal processing A module may be further added to the sensor processing module 611). A method of determining a location of a nearby vehicle using the lidar sensor 510, the radar sensor 520, and the camera sensor 530 is a specific embodiment, and the implementation method is not limited thereto. In addition, the sensor processing module 611 may determine attribute information such as the size and type of the surrounding vehicle as well as the location, speed, and movement direction of the surrounding vehicle, and the position, speed, movement direction, and size of the surrounding vehicle as described above. And an algorithm for determining information such as type may be predefined.

The driving trajectory generation module 612 can generate the actual driving trajectory and the expected driving trajectory of the surrounding vehicle, and the actual driving trajectory of the own vehicle, and for this purpose, as shown in FIG. 2, the surrounding vehicle driving trajectory generation module 612a And a driving trajectory generation module 612b of the own vehicle.

First, the surrounding vehicle driving trajectory generation module 612a may generate an actual driving trajectory of the surrounding vehicle.

Specifically, the surrounding vehicle driving trajectory generation module 612a is based on the driving information of the surrounding vehicle detected by the sensor unit 500 (that is, the position of the surrounding vehicle determined by the sensor processing module 611). You can create an actual driving trajectory. In this case, in order to generate the actual driving trajectory of the surrounding vehicle, the surrounding vehicle driving trajectory generation module 612a may refer to the map information stored in the memory 620, and the position of the surrounding vehicle detected by the sensor unit 500 And a random location on the map information stored in the memory 620 may be cross-referenced to generate an actual driving trajectory of the surrounding vehicle. For example, when a nearby vehicle is detected at a specific point by the sensor unit 500, the surrounding vehicle driving trajectory generation module 612a is a location of the detected surrounding vehicle and an arbitrary location on the map information stored in the memory 620. By cross-referencing, it is possible to specify the location of the currently detected surrounding vehicle on the map information, and by continuously monitoring the location of the surrounding vehicle as described above, the actual driving trajectory of the surrounding vehicle can be generated. That is, the surrounding vehicle driving trajectory generation module 612a maps and accumulates the location of the surrounding vehicle detected by the sensor unit 500 to the location on the map information stored in the memory 620 based on the cross reference. The actual driving trajectory of the vehicle can be created.

Meanwhile, the actual driving trajectory of the surrounding vehicle may be compared with the predicted driving trajectory of the surrounding vehicle to be described later, and used to determine whether the map information stored in the memory 620 is incorrect. In this case, when comparing the actual driving trajectory of a specific surrounding vehicle with the predicted driving trajectory, a problem of erroneously determining that the map information is accurate but inaccurate may arise. For example, when the actual driving trajectory of a plurality of surrounding vehicles and the predicted driving trajectory coincide with each other, and the actual driving trajectory and the predicted driving trajectory of a specific surrounding vehicle are different, only the actual driving trajectory of the specific surrounding vehicle is matched with the expected driving trajectory. In comparison, even though the map information is accurate, it can be misjudged as inaccurate. Therefore, there is a need to determine whether the tendency of the actual driving trajectory of the plurality of surrounding vehicles deviates from the expected driving trajectory, and for this purpose, the driving trajectory generation module 612a of the surrounding vehicles generates the actual driving trajectories of the plurality of surrounding vehicles, respectively. You may. Furthermore, considering that the driver of the surrounding vehicle tends to move the steering wheel to the left and right during the driving process for a straight path driving, the actual driving trajectory of the surrounding vehicle may be generated in a curved shape rather than a straight line. In order to calculate the error between the predicted driving trajectories, the driving trajectory generation module 612a of the surrounding vehicle may generate the actual driving trajectory in the form of a straight line by applying a predetermined smoothing technique to the original actual driving trajectory that is generated in a curved form. Various techniques such as interpolation for each position of the surrounding vehicle can be employed as the smoothing technique.

Also, the surrounding vehicle driving trajectory generation module 612a may generate a predicted driving trajectory of the surrounding vehicle based on map information stored in the memory 620.

As described above, the map information stored in the memory 620 may be 3D high-precision electronic map data. Therefore, the map information includes lanes, lane center lines, regulatory lines, road boundaries, road center lines, traffic signs, road signs, and road shapes. And it is possible to provide dynamic and static information necessary for autonomous driving control of the vehicle, such as height and lane width. In general, considering that the vehicle runs in the center of the lane, it may be expected that the surrounding vehicles running in the vicinity of the own vehicle also travel in the center of the lane. Therefore, the surrounding vehicle driving trajectory generation module 612a The vehicle's predicted driving trajectory can be generated as a lane center line reflected in the map information.

The own vehicle driving trajectory generation module 612b may generate an actual driving trajectory that the own vehicle has traveled to date based on the driving information of the own vehicle acquired through the driving information input interface 201 described above.

Specifically, the host vehicle driving trajectory generation module 612b includes a location of the host vehicle obtained through the driving information input interface 201 (that is, location information of the host vehicle obtained through the GPS receiver 260) and a memory 620 ), it is possible to create an actual driving trajectory of the own vehicle by cross-referencing an arbitrary location on the map information stored in ). For example, by cross-referencing the location of the own vehicle acquired through the driving information input interface 201 and an arbitrary location on the map information stored in the memory 620, the current location of the own vehicle can be specified on the map information, As described above, by continuously monitoring the position of the own vehicle, an actual driving trajectory of the own vehicle can be generated. That is, the host vehicle driving trajectory generation module 612b maps and accumulates the location of the host vehicle acquired through the driving information input interface 201 to the location on the map information stored in the memory 620 based on the cross reference. By doing so, it is possible to generate an actual traveling trajectory of the own vehicle.

In addition, the own vehicle driving trajectory generation module 612b may generate an expected driving trajectory in which the own vehicle should travel to the destination based on map information stored in the memory.

That is, the own vehicle driving trajectory generation module 612b is stored in the current position of the own vehicle obtained through the driving information input interface 201 (that is, the current position information of the own vehicle obtained through the GPS receiver 260) and the memory. An estimated driving trajectory to the destination can be generated using the stored map information, and the predicted driving trajectory of the own vehicle will be generated as a center line of the lane reflected in the map information stored in the memory 620 like the predicted driving trajectory of nearby vehicles. I can.

The driving trajectory generated by the surrounding vehicle driving trajectory generating module 612a and the own vehicle driving trajectory generating module 612b may be stored in the memory 620, and the autonomous driving of the own vehicle is controlled by the processor 610. It can be used for various purposes in the process.

The driving trajectory analysis module 613 is generated by the driving trajectory generation module 612 and stored in the memory 620 (that is, the actual driving trajectory and the expected driving trajectory of the surrounding vehicle, the actual driving trajectory of the own vehicle). By analyzing, it is possible to diagnose the reliability of autonomous driving control for the current own vehicle. The reliability diagnosis of autonomous driving control may be performed by analyzing a trajectory error between an actual driving trajectory and an expected driving trajectory of nearby vehicles.

The driving control module 614 may perform a function of controlling autonomous driving of the own vehicle, and specifically, driving information and driving information input from the driving information input interface 101 and the driving information input interface 201 described above, respectively. Wow, the autonomous driving algorithm is processed by comprehensively using information on surrounding objects detected through the sensor unit 500 and map information stored in the memory 620, and control information is transmitted through the vehicle control output interface 401. It is possible to allow the lower control system 400 to control the autonomous driving of the own vehicle by transmitting the vehicle, and also transmit the driving state information and warning information of the own vehicle to the output unit 300 through the occupant output interface 301 to the driver Can be made aware of. In addition, when the driving control module 614 integrally controls the autonomous driving as described above, the own vehicle analyzed by the sensor processing module 611, the driving trajectory generation module 612, and the driving trajectory analysis module 613 And by controlling the autonomous driving in consideration of the driving trajectory of the surrounding vehicle, the precision of the autonomous driving control may be improved and the stability of the autonomous driving control may be improved.

The trajectory learning module 615 may learn or correct the actual driving trajectory of the own vehicle generated by the own vehicle driving trajectory generation module 612b. For example, if the trajectory error between the actual driving trajectory of the surrounding vehicle and the expected driving trajectory is greater than or equal to a preset threshold, it is determined that the map information stored in the memory 620 is inaccurate, and it is determined that the actual driving trajectory of the own vehicle needs to be corrected. Accordingly, a lateral shift value for correcting the actual traveling trajectory of the own vehicle may be determined to correct the traveling trajectory of the own vehicle.

The occupant state determination module 616 may determine the occupant's state and behavior based on the occupant's state and bio-signals detected by the aforementioned internal camera sensor 535 and the biometric sensor. The state of the occupant determined by the occupant state determination module 616 may be used in the process of controlling the autonomous driving of the own vehicle or outputting a warning to the occupant.

Based on the above, an embodiment of learning an autonomous driving algorithm applied to autonomous driving control will be described below. The'autonomous driving algorithm' indicated below refers to an algorithm applied to the autonomous driving control of the own vehicle, and the'autonomous driving algorithm for nearby vehicles' refers to an algorithm applied to the autonomous driving control of nearby vehicles. I decided to classify. In the present embodiment below, the process of updating the autonomous driving algorithm applied to the own vehicle by comparing the accuracy of the autonomous driving algorithm applied to the own vehicle and the autonomous driving algorithm of surrounding vehicles applied to the surrounding vehicle, and the autonomous driving algorithm applied to the own vehicle. It is implemented as a process of performing learning, and will be described in detail below.

First, the processor 610 can control the autonomous driving of the own vehicle according to the map information stored in the memory 620 and the autonomous driving algorithm, and in the autonomous driving control process of the own vehicle, Vehicle autonomous driving algorithm can be delivered. In this case, the processor 610 may compare the autonomous driving algorithm stored in the memory 620 with the autonomous driving algorithm of the surrounding vehicle received from the surrounding vehicle to determine whether an update of the autonomous driving algorithm stored in the memory 620 is necessary. .

In the present embodiment, a first accuracy index, which indicates the accuracy of autonomous driving control for the own vehicle, is mapped to the autonomous driving algorithm, and the second accuracy index, which indicates the accuracy of the autonomous driving control for surrounding vehicles, is mapped to the autonomous driving algorithm. May be mapped. The accuracy index is a quantitative index calculated based on the history of autonomous driving control of the vehicle based on the autonomous driving algorithm.For example, the frequency of accidents when autonomous driving control is performed based on the autonomous driving algorithm, to the destination It may mean an index calculated to indicate the control accuracy of the autonomous driving algorithm by comprehensively considering the time required to reach, the driving distance and fuel efficiency, and the frequency of the occupant's driving operation intervention. An algorithm for calculating an accuracy index through the above analysis of the accumulated history of autonomous driving control performed based on the autonomous driving algorithm may also be stored in the memory 620, and the calculated accuracy index is mapped to the autonomous driving algorithm. Then, it may be stored in the memory 620.

Accordingly, when the second accuracy index mapped to the autonomous driving algorithm of the surrounding vehicle is greater than the first accuracy index mapped to the autonomous driving algorithm, the processor 610 stores the autonomous driving algorithm of the surrounding vehicle in the memory 620 to drive autonomously. Algorithm can be updated. That is, when the second accuracy index is greater than the first accuracy index, it can be seen that the autonomous driving algorithm of the surrounding vehicle has higher accuracy and reliability than the autonomous driving algorithm, so that the processor 610 stores the autonomous driving algorithm of the surrounding vehicle in the memory 620. Saved in to update the autonomous driving algorithm. The update of the autonomous driving algorithm may be performed in real time or periodically while autonomous driving of the own vehicle is controlled.

Autonomy of the own vehicle according to the autonomous driving algorithm currently stored in the memory 620 (i.e., the autonomous driving algorithm previously stored in the memory 620 or the autonomous driving algorithm of the surrounding vehicle received from the surrounding vehicle) through the update as described above. In the process of controlling the driving, the processor 610 may determine whether a driving operation of the occupant of the own vehicle is involved. In this case, the processor 610 determines whether manual driving operations such as steering operation, accelerator pedal operation, or brake pedal operation of the occupant are intervened through the steering angle sensor 210 of the driving information detection unit 200 or the APS/PTS 220. You can judge whether or not. When it is determined that the occupant's driving operation is involved, learning of an autonomous driving algorithm described later may be performed, and as a prerequisite for learning of the autonomous driving algorithm, the processor 610 stops the autonomous driving control of the own vehicle. It can (that is, the autonomous driving mode of the own vehicle can be turned off).

After the autonomous driving control is stopped due to the occupant's driving operation being intervened, the processor 610 may allow the learning of the autonomous driving algorithm currently stored in the memory 620 to be performed in consideration of the driving operation of the occupant. Learning of the autonomous driving algorithm may be performed based on a comparison result between the control process according to the autonomous driving algorithm at the time when the driving operation is intervened and the driving operation of the occupant. As an example of the above control process and the driving operation of the occupant, i) when the control process is a lane change process through right steering and deceleration, and the driving operation of the occupant is right steering against the steering wheel and depression of the brake pedal, ii) control When the process is the lane change process through right steering and deceleration and the occupant's driving operation is the depression of the left steering and brake pedal to the steering wheel, or iii) the control process is the lane change process through right steering and deceleration and the occupant's driving There may be cases where the operation is left-hand steering against the steering wheel and depression of the accelerator pedal.

If the control process and the driving organization of the occupant are the same as in the case of i) in the above example, the processor 610 returns to the autonomous driving mode again, and autonomous driving of the own vehicle based on the autonomous driving algorithm currently stored in the memory 620 Control can be performed. If, as in the case of ii) and iii) of the above example, the control process and the occupant's driving operation are different, the processor 610 may verify the risk of the occupant's driving operation and then allow the learning of the autonomous driving algorithm to be performed. have. The verification of the risk of the occupant's driving operation may be performed through a process of determining whether an accident has been caused by the occupant's driving operation. Ii) In this case, if an accident is not caused by the driver's left steering and the depression of the brake pedal, it is determined that the risk of the occupant's driving operation has been verified, and the processor 610 autonomously drives based on the occupant's driving operation. The learning of the algorithm can be performed, and in the case of iii) of the above example, if an accident is caused through the depression of the left steering and accelerator pedal, it is determined that the risk of the occupant's driving operation has not been verified, and the processor 610 By returning to the autonomous driving mode again, autonomous driving control of the own vehicle may be performed based on the autonomous driving algorithm currently stored in the memory 620.

When the control process and the occupant's driving operation are different and the risk of the occupant's driving operation is verified, the processor 610 includes information on surrounding objects detected by the sensor unit 500 and the driving information detection unit 200. Learning of an autonomous driving algorithm may be performed based on the detected driving information of the own vehicle, a control process, and a driving operation of an occupant. In other words, it is possible to learn the autonomous driving algorithm through the control process that has been reserved in advance according to the autonomous driving algorithm and the driving operation of the passenger whose risk has been verified, and furthermore, it is possible to actively learn about the surrounding environment and driving state of the own vehicle. To enable autonomous driving control, learning of the autonomous driving algorithm is performed by considering information on surrounding objects detected by the sensor unit 500 and driving information of the own vehicle detected by the driving information detection unit 200 together. I can.

The processor 610 may itself perform the learning of the autonomous driving algorithm, but when considering the computational load of the learning, the server 700 provides information on the surrounding objects so that the learning of the autonomous driving algorithm can be performed. The driving information of the vehicle, the control process, and the driving operation of the occupant are transmitted to the server 700, and the autonomous driving algorithm, which has been learned by the server 700, is transmitted from the server 700 to control the autonomous driving of the own vehicle. May be. Furthermore, in order to share the learned autonomous driving algorithm transmitted from the server 700 with surrounding vehicles, the learned autonomous driving algorithm may be propagated to surrounding vehicles.

7 is a flowchart illustrating an autonomous driving method according to an embodiment of the present invention.

Referring to FIG. 7 to describe the autonomous driving method according to an embodiment of the present invention, first, the processor 610 controls the autonomous driving of the own vehicle according to the autonomous driving algorithm stored in the memory 620 (S100).

Subsequently, the processor 610 compares the autonomous driving algorithm stored in the memory 620 with the autonomous driving algorithm of the surrounding vehicle transmitted from the surrounding vehicle to determine whether to update the autonomous driving algorithm stored in the memory 620 (S200). As described above, in the autonomous driving algorithm, a first accuracy index indicating the accuracy of autonomous driving control for the own vehicle is mapped, and in the autonomous driving algorithm for surrounding vehicles, a second accuracy index indicating the accuracy of autonomous driving control for surrounding vehicles is mapped. Accordingly, in step S200, the processor 610 determines that the autonomous driving algorithm needs to be updated when the second accuracy index mapped to the autonomous driving algorithm of the surrounding vehicle is larger than the first accuracy index mapped to the autonomous driving algorithm. .

If it is determined in step S200 that the autonomous driving algorithm needs to be updated, the processor 610 stores the autonomous driving algorithm of nearby vehicles in the memory 620 to update the autonomous driving algorithm (S300).

Subsequently, the processor 610 determines whether an occupant's driving operation is involved in the process of controlling the autonomous driving of the own vehicle according to the autonomous driving algorithm currently stored in the memory 620 through the update (S400).

If it is determined that the occupant's driving operation is involved in step S400, the processor 610 stops autonomous driving control of the own vehicle (S500).

Thereafter, the processor 610 allows learning of the autonomous driving algorithm currently stored in the memory 620 in consideration of the driving operation of the occupant, and specifically, a control process according to the autonomous driving algorithm at the time when the driving operation of the occupant is intervened. The learning of the autonomous driving algorithm is performed based on the comparison result between the driving operation of the occupant and (S600).

In step S600, the processor 610 compares the driving operation of the occupant with the control process according to the autonomous driving algorithm at the time when the driving operation of the occupant is intervened (S610), and when the control process and the driving operation of the occupant are different, The risk of driving operation is verified (S620), and when the risk of driving operation of the occupant is verified, learning of the autonomous driving algorithm is performed based on the control process and the driving operation of the occupant (S630). In step S630, the processor 610 may allow the learning of the autonomous driving algorithm to be performed by further considering information on the surrounding objects of the own vehicle and the driving information of the own vehicle, and information on the surrounding objects, driving information of the own vehicle , By transmitting the control process, and the driving operation of the occupant to the server 700, the learning of the autonomous driving algorithm may be performed by the server 700.

Thereafter, the processor 610 receives the autonomous driving algorithm learned by the server 700 from the server 700 and controls the autonomous driving of the own vehicle (S700), and the learned autonomous driving algorithm transmitted from the server 700 In order to share with the surrounding vehicles, the learned autonomous driving algorithm is propagated to the surrounding vehicles (S800).

As described above, in this embodiment, after learning an autonomous driving algorithm applied to autonomous driving control in consideration of the driving operation of the occupant involved in the autonomous driving control process of the own vehicle, autonomous driving of the own vehicle is performed according to the learned autonomous driving algorithm. By controlling, driving stability and driving accuracy of an autonomous vehicle can be improved.

The present invention has been described with reference to the embodiments shown in the drawings, but these are only exemplary, and those of ordinary skill in the art to which the present technology pertains, various modifications and other equivalent embodiments are possible. I will understand. Therefore, the true technical protection scope of the present invention should be determined by the following claims.

100: user input unit 101: driving information input interface
110: drive mode switch 120: user terminal
200: driving information detection unit 201: driving information input interface
210: steering angle sensor 220: APS/PTS
230: vehicle speed sensor 240: acceleration sensor
250: yaw/pitch/roll sensor 260: GPS receiver
300: output unit 301: occupant output interface
310: speaker 320: display device
400: sub-control system 401: vehicle control output interface
410: engine control system 420: braking control system
430: steering control system 500: sensor unit
510: lidar sensor 511: front lidar sensor
512: upper lid sensor 513: rear lid sensor
520: radar sensor 521: front radar sensor
522: left radar sensor 523: right radar sensor
524: rear radar sensor 530: camera sensor
531: front camera sensor 532: left camera sensor
533: right camera sensor 534: rear camera sensor
535: internal camera sensor 540: ultrasonic sensor
600: autonomous driving integrated control unit 610: processor
611: sensor processing module 611a: lidar signal processing module
611b: radar signal processing module 611c: camera signal processing module
612: driving trajectory generation module 612a: driving trajectory generation module of surrounding vehicles
612b: host vehicle driving trajectory generation module 613: driving trajectory analysis module
614: driving control module 615: trajectory learning module
616: occupant status determination module 620: memory
700: server

Claims (9)

A memory for storing an autonomous driving algorithm for autonomous driving control of the own vehicle; And
A processor for controlling the autonomous driving of the own vehicle according to the autonomous driving algorithm stored in the memory; and
The processor,
In the process of controlling the autonomous driving of the host vehicle according to the autonomous driving algorithm, it is determined whether a driving operation of the occupant of the host vehicle is involved,
When it is determined that the occupant's driving operation is involved, learning of the autonomous driving algorithm based on a comparison result between the control process according to the autonomous driving algorithm and the driving operation of the occupant at the time when the driving operation of the occupant is intervened Autonomous driving device, characterized in that to be performed.
The method of claim 1,
The processor,
When it is determined that the driving operation of the occupant is involved, the autonomous driving algorithm is learned after stopping the autonomous driving control of the own vehicle.
The method of claim 1,
The processor,
When the control process and the driving operation of the occupant are different, the learning of the autonomous driving algorithm is performed after verifying the risk of the driving operation of the occupant.
The method of claim 3,
A sensor unit detecting objects around the host vehicle; And
Further comprising a; driving information detection unit for detecting driving information on the driving state of the host vehicle,
The processor,
Learning of the autonomous driving algorithm is performed based on information on surrounding objects detected by the sensor unit, driving information of the own vehicle detected by the driving information detection unit, the control process, and a driving operation of the occupant. An autonomous driving device characterized by.
Controlling, by the processor, autonomous driving of the own vehicle according to the autonomous driving algorithm stored in the memory;
Determining, by the processor, whether a driving operation of an occupant of the host vehicle is involved in the process of controlling the autonomous driving of the host vehicle according to the autonomous driving algorithm; And
When it is determined that the occupant's driving operation is intervened, the processor may perform the autonomy based on a comparison result between the control process according to the autonomous driving algorithm and the occupant's driving operation at a time when the occupant's driving operation is intervened. Allowing learning of a driving algorithm to be performed;
Autonomous driving method comprising a.
The method of claim 5,
If it is determined that the driving operation of the occupant is involved, the processor, stopping autonomous driving control of the own vehicle; further comprising,
The step of allowing the host vehicle to be performed is performed after the autonomous driving control of the host vehicle is stopped.
The method of claim 5,
The step of causing the above to be performed,
Comparing, by the processor, a control process according to the autonomous driving algorithm and the driving operation of the occupant at a time when the driving operation of the occupant is intervened;
If the control process and the occupant's driving operation are different, verifying, by the processor, a risk of the occupant's driving operation; And
And when the risk of the occupant's driving operation is verified, causing the processor to learn the autonomous driving algorithm based on the control process and the occupant's driving operation.
The method of claim 7,
In the step to be performed, the processor,
The autonomous driving method, characterized in that the learning of the autonomous driving algorithm is performed by further considering information on the surrounding objects of the own vehicle and the driving information of the own vehicle.
The method of claim 8,
In the step to be performed, the processor,
Transmitting information on the surrounding object, driving information of the own vehicle, the control process, and a driving operation of the occupant to the server so that the learning of the autonomous driving algorithm can be performed by the server,
And controlling the autonomous driving of the own vehicle by receiving the autonomous driving algorithm learned by the server from the server.

Images