Intelligent driving method, control system and device integrating personalized driving style

Abstract

The application provides an intelligent driving method, a control system and a device integrating personalized driving styles, the device comprising an intelligent driving controller, a human-computer interface, a positioning sensor device, a perception sensor device, a vehicle networking control unit and a remote server, the intelligent driving controller being equipped with an intelligent driving control system having a driving style recognition and matching algorithm unit, a perception algorithm unit, a planning algorithm unit and a control algorithm unit; the intelligent driving method having the steps of generating one or more kinds of personalized driving style information according to autonomous driving data collected when a driver autonomously drives, combining pre-stored default driving style information to generate a driving style form for user selection, and generating set driving style information in response to user instructions. The device and method of the application can generate corresponding driving styles according to autonomous driving data, so that the same driving style as the driver can be achieved after the intelligent driving function is activated.

Description

Technical Field

[0001] This invention belongs to the field of intelligent driving design technology, specifically relating to an intelligent driving method, control system, and device that integrates personalized driving styles. Background Technology

[0002] In recent years, intelligent driving has become a major trend in the automotive industry. Various driver assistance and autonomous driving functions have been widely applied in mass-produced vehicles. Similar to human drivers, intelligent driving vehicles also have their own driving styles. Different driving styles perform differently in improving traffic safety, road efficiency, and user comfort.

[0003] Currently, most intelligent driving vehicles are still driven by human drivers, with intelligent driving functions activated only in specific scenarios to enable autonomous vehicle operation. Different drivers have different driving styles; however, specific intelligent driving functions of a particular brand and model share the same driving style. This results in different driving experiences for the same vehicle before and after intelligent driving function activation, negatively impacting the user experience. When driving autonomously, drivers adjust their driving style according to different regions and scenarios to adapt to local conditions. For example, in fast-paced, densely populated cities, most drivers tend to maintain a closer following distance to prevent lane cutting in during congestion; while in slower-paced, less congested cities, most drivers do not drive aggressively even in traffic jams. If an intelligent driving vehicle can only have a single driving style after intelligent driving function activation, or if the driving style cannot switch according to region and scenario, then it may experience frequent lane cutting in during congested conditions in fast-paced cities, or relatively aggressive driving in slower-paced cities, causing additional pressure on other road users.

[0004] To address the above issues, those skilled in the art have made the following improvements:

[0005] Patent 1: Chinese patent application publication number CN 113581188A, publication date November 2, 2021, invention title "A Method for Recognizing Driving Style of Commercial Vehicle Drivers Based on Internet of Vehicles Data". This invention collects commercial vehicle driving data and vehicle status data in natural driving scenarios; performs outlier detection, missing value imputation and removal, feature value extraction and establishment, and data dimensionality reduction on the collected data; clusters the dimensionality-reduced data based on the k-means clustering algorithm, and defines the driver's driving type according to the clustering results; integrates driver characteristics and driver driving type to establish a driving style recognition model based on the random forest algorithm, and trains and tests the driving style recognition model to achieve effective recognition of the driving style of commercial vehicle drivers.

[0006] Patent 2: Chinese patent application publication number CN 113942521A, publication date January 18, 2022, invention title "A Driver Style Recognition Method in an Intelligent Vehicle-Road System". This invention obtains first driving data and image information of the driving vehicle; performs data preprocessing and data fusion on the first driving data to obtain the average vehicle speed, driving time, driver ID, and vehicle latitude and longitude information within a preset time period; performs DB-LSTM calculation on the driving image information to obtain the driver's driving behavior type and obtain a driver style feature dataset; and determines the driver style type based on the driver style feature dataset.

[0007] Patent 3: The Chinese patent authorization announcement number is CN 113895464B, the authorization announcement date is April 8, 2022, and the invention title is "Intelligent Vehicle Driving Map Generation Method and System Integrating Personalized Driving Style". This invention obtains a driving map based on coded vehicle and target information and driving style information; and outputs personalized driving decisions based on the driving map and global path planning generated from vehicle map data.

[0008] Of the solutions in the aforementioned patents, Patents 1 and 2 only identify driving styles. Patent 3 uses a specific driving style as input and output for personalized decision-making, but it doesn't mention how different driving styles are obtained. Overall, existing solutions cannot solve the problems of inconsistency between autonomous driving and intelligent driving styles, nor the problem of adapting driving styles to different regions and scenarios.

[0009] Invention Patent Content

[0010] To address the aforementioned problems, this invention provides an intelligent driving method, control system, and device that integrates personalized driving styles. It generates driving style information adapted to the driver's driving style and / or the style of the local scene by collecting data from the driver's autonomous driving, and provides users with selectable driving style options, thereby enabling the intelligent driving vehicle to have a suitable driving style and improving the user experience. The objective of this invention is achieved through the following technical solutions:

[0011] This invention first provides an intelligent driving method integrating personalized driving styles, which includes:

[0012] Step S100: Generate one or more personalized driving style information based on the autonomous driving data collected by the driver during autonomous driving, generate a driving style form for the user to select from in combination with the pre-stored default driving style information, and generate the set driving style information in response to the user's instructions.

[0013] Step S200: Provide optional driving route information based on driving environment information, wherein the optional driving route information includes drivable areas, target lists, and lane lines;

[0014] Step S300: Generate a driving planning trajectory based on vehicle positioning information, selectable driving route information and set driving style information. The driving planning trajectory includes the vehicle's longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicles, and the lateral safety distance of following vehicles.

[0015] Step S400: Generate driving control commands based on vehicle positioning information, driving trajectory planning, and set driving style information.

[0016] Furthermore, the personalized driving style information includes the driver's driving style information and the driving style information for specific geographical locations.

[0017] Furthermore, the method for generating the vehicle driver's driving style information includes:

[0018] Step S111: Read the autonomous driving data of the driver once. The autonomous driving data includes autonomous driving driving data, autonomous driving environment data, and autonomous driving positioning data. The autonomous driving driving data includes the accelerator pedal opening and its changes, the brake pedal opening and its changes, and the steering wheel angle and its changes. The autonomous driving environment data includes information such as the target list and lane lines. The autonomous driving positioning data includes information such as the vehicle's speed, position, acceleration, azimuth angle, and yaw rate.

[0019] Step S112: Determine whether a driving style boundary has been formed based on the autonomous driving data collected in step S111. If the determination result is yes, proceed to step S150; if the determination result is no, proceed to step S113.

[0020] Step S113: Determine whether the current number of reads exceeds the preset number M. If it exceeds the preset number M, proceed to step S114; if it does not exceed the preset number M, proceed to step S111.

[0021] Step S114: Plot the M autonomous driving data read onto the same coordinate system, solve the boundary Limit, so that a certain preset proportion of the data in the M data are all within the boundary Limit, and generate the vehicle driver's driving style information based on the boundary Limit.

[0022] Step S115: Read the current real-time driving data, determine whether the real-time driving data falls within the boundary Limit, and count the data that exceeds the boundary Limit;

[0023] Step S116: Determine whether the number of data exceeding the current boundary Limit exceeds the preset threshold N. If the determination result is that it exceeds the preset threshold N, then proceed to step S114; if the determination result is that it does not exceed the threshold N, then proceed to step S111.

[0024] Furthermore, the method for generating the regional driving style information includes:

[0025] Step S121: Determine whether there is a regional scene driving style boundary. If the result is no, proceed to step S122; if the result is yes, proceed to step S126.

[0026] Step S122: Read the regional scene data, which includes geographical range, time period and vehicle quantity information;

[0027] Step S123: Determine whether the regional scene data exceeds the preset size ZS. If it exceeds the preset size ZS, proceed to step S124. If it does not exceed the preset size, proceed to step S122.

[0028] Step S124: Divide the regional scene into partitions based on the regional scene data obtained in step S122;

[0029] Step S125: Based on the regional driving style partitions defined in Step S124, solve the driving style boundary LimitZ for different regional scenarios to define the regional driving style.

[0030] Step S126: Determine whether the preset update time interval has been exceeded. If it has, proceed to step S122; otherwise, proceed to the end.

[0031] Furthermore, the driving style boundary includes at least two of the following: longitudinal acceleration boundary, speed limit deviation boundary, longitudinal safety distance boundary, and lateral safety distance boundary.

[0032] A second aspect of the present invention is to provide an intelligent driving control system that integrates personalized driving styles, comprising:

[0033] The driving style recognition and matching algorithm unit generates personalized driving style information based on the manual driving information transmitted by the vehicle actuators and the positioning information transmitted by the vehicle positioning sensor when the driver is driving autonomously. It also generates a driving style form for the user to select by combining the pre-stored default driving style information.

[0034] The perception algorithm unit provides optional driving route information based on the driving environment information transmitted by the vehicle perception sensing device. The optional driving route information includes drivable areas, target lists, and lane lines.

[0035] The planning algorithm unit generates a driving planning trajectory based on the vehicle positioning information transmitted by the vehicle positioning sensor, the optional driving path information transmitted by the perception algorithm unit, and the selected driving style information transmitted by the driving style recognition and matching algorithm unit. The driving planning trajectory includes the vehicle's longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicles, and the lateral safety distance of following vehicles.

[0036] The control algorithm unit generates driving control commands based on the vehicle positioning information transmitted by the vehicle positioning sensor and the driving planning trajectory transmitted by the planning algorithm unit.

[0037] Furthermore, the driving style recognition and matching algorithm unit includes:

[0038] The vehicle-mounted driver data acquisition module collects autonomous driving data information during autonomous driving. The autonomous driving data information includes the action information of the vehicle actuators, the positioning information of the positioning sensor, and the driving environment information transmitted by the sensing sensor.

[0039] The in-vehicle driver style recognition module generates in-vehicle driver driving style information based on information from the data acquisition module.

[0040] The regional scene information collection module collects time information, positioning information from positioning sensors, speed information, and action information from vehicle actuators during driving.

[0041] The driving style adaptation module provides the human-machine interface with a form containing information on the driver's driving style, regional scene driving style, and default driving style for the user to select. It generates the set driving style in response to user commands and transmits the set driving style parameters to the control algorithm unit.

[0042] Based on this, a third aspect of the present invention provides an intelligent driving device integrating personalized driving styles, comprising an intelligent driving controller, a human-machine interface, a positioning sensor, a perception sensor, a vehicle networking control unit, and a remote server: the intelligent driving controller is a domain controller equipped with the aforementioned intelligent driving control system; the vehicle networking control unit is used to connect vehicles to the network, send information, and receive network data; the positioning sensor includes a global positioning system and an inertial measurement unit; the perception sensor includes at least two of a camera, millimeter-wave radar, and lidar; the intelligent driving controller is a domain controller; and the remote server communicates with the vehicle networking control unit via a wireless network device.

[0043] Preferably, the positioning sensing device adopts a fusion positioning scheme of Global Positioning System and Inertial Measurement Unit to provide more accurate and reliable positioning information; the perception sensing device adopts a fusion perception scheme of multiple cameras + millimeter-wave radar + lidar to provide more comprehensive and reliable driving environment information data for the control system decision-making.

[0044] Furthermore, the remote server contains a regional scene driving style recognition module; the information sent by the vehicle network control unit includes information collected by the regional scene information acquisition module; the received network data includes the results recognized by the regional scene driving style recognition module.

[0045] Preferably, the specific process of implementing step S100 using the above-mentioned intelligent driving device is as follows:

[0046] Step S110: The intelligent driving domain controller takes the vehicle information, perception information and positioning information obtained by the vehicle driver data acquisition module after reading the data M times in a preset number as input, and generates the current driver's driving style information through the vehicle driver style recognition algorithm in the vehicle driver style recognition module.

[0047] Step S120: The regional scene information acquisition module transmits the acquired information to the remote server. The remote server takes the regional scene information of all vehicles in the preset scale ZS as input and calculates the current regional scene driving style information through the algorithm of the regional scene driving style recognition module. Preferably, the preset scale ZS includes a preset geographical range P1, a preset time period T3, and a preset number of vehicles K.

[0048] Step S130: The human-machine interface receives the current driving style Limit of the onboard driver sent by the onboard driver style recognition module and the current regional scene driving style LimitZ sent by the regional scene driving style recognition module, and prompts the user with the currently available driving styles through the display screen of the human-machine interface 4. Typically, the driving styles are divided into three types: default driving style, onboard driver driving style, and regional scene driving style. The driver operates on the human-machine interface 4 to select a specific driving style, and the human-machine interface 4 sends the selected driving style to the intelligent driving domain controller 3 via the CAN bus.

[0049] Step S140: The intelligent driving controller 3 receives the driving style selected by the user and adjusts the parameter values ​​of the planning algorithm and control algorithm to the values ​​corresponding to that style based on the corresponding driving style information, so as to reasonably control the vehicle to achieve style switching in intelligent driving mode. The adjusted parameter values ​​include: Limit for driver driving style; LimitZ for regional scenario driving style.

[0050] The beneficial effects of this invention are as follows:

[0051] 1. This invention solves the problem of inconsistent driving styles between autonomous driving and intelligent driving; it allows users to customize the driving style so that intelligent driving, once activated, has the same driving style as when the driver is autonomously driving the vehicle.

[0052] 2. This invention solves the problem of adapting driving styles to different regions and scenarios; it allows users to customize the driving style so that the intelligent driving function, once activated, has the same driving style as other vehicles in the current vehicle's area. Attached Figure Description

[0053] Figure 1 The diagram shown is a schematic representation of the intelligent driving device in Embodiment 1 of the present invention.

[0054] Figure 2 The diagram shown is a schematic representation of the overall process of the intelligent driving method in Embodiment 2 of the present invention.

[0055] Figure 3 The diagram shown is a flowchart of the driving style recognition and matching algorithm in Embodiment 2 of the present invention;

[0056] Figure 4 The diagram shown is a flowchart illustrating the process of generating vehicle driver driving style information in Embodiment 2 of the present invention.

[0057] Figure 5 The diagram shown is a flowchart illustrating the process of generating regional driving style information in Embodiment 2 of the present invention.

[0058] Figure 6The image shows an example of generating the vehicle driver style boundary in Embodiment 2 of the present invention;

[0059] Figure 7 The image shows an example of generating regional scene driving style boundaries in Embodiment 2 of the present invention. Detailed Implementation

[0060] The preferred embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0061] Example 1:

[0062] This embodiment provides an intelligent driving device that integrates personalized driving styles, and its configuration is as follows: Figure 1 As shown, it includes an intelligent driving controller 3, a human-machine interface 4, a positioning sensor 1, a perception sensor 2, a vehicle networking control unit 5, and a remote server 6.

[0063] In a preferred embodiment, the positioning sensor device 1 employs a combined positioning scheme commonly used in the field of intelligent driving. A specific embodiment uses a combined positioning scheme of GPS (Global Positioning System) + IMU (Inertial Measurement Unit). The output port of the positioning sensor device 1 is a CAN (Controller Area Network) interface, and the output signal is the current positioning information of the vehicle.

[0064] The sensing device 2 includes at least two of the following: a camera, millimeter-wave radar, and lidar. In a preferred embodiment, the sensing device 2 employs a multi-sensor scheme; a specific embodiment uses a fusion sensing scheme combining multiple cameras, millimeter-wave radar, and lidar. The output ports of the sensing sensor module 2 vary depending on the sensors used, and are common interfaces in the field of intelligent driving. For example, when the fusion sensing scheme is a multi-camera + millimeter-wave radar + lidar fusion sensing scheme, its output interfaces include an LVDS (Low Voltage Differential Signaling) interface, a CAN interface, and an Ethernet interface. The output signal of the sensing sensor module 2 is the raw sensor signal.

[0065] In a preferred embodiment, the intelligent driving controller 3 is a domain controller. The intelligent driving controller 3 has interfaces commonly found in the field of intelligent driving, including CAN interface, LVDS interface, and Ethernet interface. This domain controller carries an integrated intelligent driving algorithm for personalized driving styles. To realize the personalized driving style setting function, in the illustrated preferred embodiment, the intelligent driving control system mounted on the intelligent driving controller 3 includes a driving style recognition and matching algorithm unit 31, a perception algorithm unit 32, a planning algorithm unit 33, and a control algorithm unit 34, such as... Figure 2 As shown, the functions of each unit are as follows:

[0066] The driving style recognition and matching algorithm unit 31 generates personalized driving style information based on the manual driving information transmitted by the vehicle actuators and the positioning information transmitted by the vehicle positioning sensor when the driver is driving autonomously, and generates a driving style form for the user to choose from by combining it with the pre-stored default driving style information.

[0067] The perception algorithm unit 32 provides optional driving route information based on the driving environment information transmitted by the vehicle perception sensing device. The optional driving route information includes drivable areas, target lists, and lane lines.

[0068] The planning algorithm unit 33 generates a driving planning trajectory based on the vehicle positioning information transmitted by the vehicle positioning sensor, the optional driving path information transmitted by the perception algorithm unit, and the selected driving style information transmitted by the driving style recognition and matching algorithm unit. The algorithm performance is adjusted by specific parameters, including the vehicle longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicle, and the lateral safety distance of following vehicle.

[0069] The control algorithm unit 34 generates driving control commands based on the vehicle positioning information transmitted by the vehicle positioning sensor and the driving planning trajectory transmitted by the planning algorithm unit.

[0070] The human-machine interface 4 is a common device in the automotive field; a specific embodiment uses an in-vehicle central control screen for displaying information and reading driver commands. The human-machine interface 4 includes an intelligent driving style selection module. This module first obtains driver style recognition results from the in-vehicle driver style recognition module of the intelligent driving domain controller 3 and regional scene driving style recognition results from the remote server 6. These results are then displayed to the driver on the screen. The driver selects the desired driving style using physical buttons on the human-machine interface 4 or virtual buttons on the touchscreen.

[0071] The vehicle networking control unit 5 is used to connect vehicles to the network, send information, and receive network data. The information sent by the vehicle networking control unit 5 includes: information collected by the regional scene driving style recognition information acquisition module; the information received includes: the recognition results from the regional scene driving style recognition module.

[0072] The remote server 6 communicates with the vehicle network control unit 5 via a wireless network device. The remote server 6 establishes a connection with the vehicle network control unit 5 via the wireless network, sending and receiving data. The remote server 6 carries a regional scene driving style recognition algorithm; the information it receives includes information collected by the regional scene information acquisition module; the information it sends includes the results of regional scene driving style recognition.

[0073] In a preferred embodiment, the connections between the components of the intelligent driving device are as follows: the CAN interface of the positioning sensor 1 is connected to the CAN interface of the intelligent driving domain controller 3 via a twisted-pair cable; the Ethernet interface of the sensing sensor 2 is connected to the Ethernet interface of the intelligent driving domain controller 3 via a network cable; the LVDS interface of the sensing sensor 2 is connected to the LVDS interface of the intelligent driving domain controller 3 via a wire; the CAN interface of the sensing sensor 2 is connected to the CAN interface of the intelligent driving domain controller 3 via a twisted-pair cable; the CAN interface of the intelligent driving domain controller 3 and the CAN interface of the human-machine interface 4 are connected via a twisted-pair cable; the CAN interface of the intelligent driving domain controller 3 and the CAN interface of the vehicle networking control unit 5 are connected via a twisted-pair cable; and the vehicle networking control unit 5 and the remote server 6 are connected via a wireless network.

[0074] In a preferred embodiment, the driving style recognition and matching algorithm unit includes an onboard driver data acquisition module 31a, an onboard driver driving style recognition module 31b, a regional scene information acquisition module 31c, and a driving style adaptation module 31d, etc., and the functions of each module are as follows:

[0075] The onboard driver data acquisition module 31a is used to collect information required for onboard driver driving style recognition. Typically, it needs to collect autonomous driving data, including the vehicle's actuator motion information, positioning information from the positioning sensor device, and driving environment information transmitted by the perception sensor device. In one specific embodiment, the onboard driver data acquisition module 31a obtains vehicle information from the vehicle via a CAN bus; perception information from the perception sensor device 2 via LVDS, CAN, and Ethernet interfaces; and positioning information from the positioning sensor device 1 via a CAN bus. The vehicle information includes: accelerator pedal opening, accelerator pedal opening rate of change, brake pedal opening, brake pedal opening rate of change, steering wheel angle, and steering wheel angle rate of change. The perception information includes a target list and lane lines. The positioning information includes the vehicle's speed, position, acceleration, azimuth angle, and yaw rate.

[0076] The vehicle driver style recognition module 31b generates vehicle driver autonomous driving style information based on the information from the data acquisition module, which is used to identify the driver's driving style when driving autonomously.

[0077] The regional scene information acquisition module 31c is used to collect information required for regional scene driving style recognition, specifically including driving time information, positioning information from the positioning sensor, speed information, and vehicle actuator action information. In a preferred embodiment, the information collected by the regional scene information acquisition module 31c includes time information obtained from the human-machine interface 4; position information and speed information obtained from the positioning sensor 1; and accelerator pedal opening, accelerator pedal opening rate of change, brake pedal opening, brake pedal opening rate of change, steering wheel angle, and steering wheel angle rate of change information obtained from the vehicle CAN line.

[0078] The driving style adaptation module 31d provides the human-machine interface 4 with a form containing information on the driver's driving style, regional scene driving style, and default driving style for the user to select. It generates the set driving style in response to user commands and transmits the set driving style parameters to the control algorithm unit. Its purpose is to adjust the corresponding parameter configuration based on the driving style selected in the intelligent driving style selection module of the human-machine interface 4, thereby adjusting the performance of the planning algorithm and control algorithm, so that the vehicle has the selected style in intelligent driving mode.

[0079] Example 2:

[0080] This embodiment provides an intelligent driving method integrating personalized driving styles based on the device of Embodiment 1, such as... Figure 2 As shown, the method includes the following steps:

[0081] Step S100 - Driving Style Recognition and Matching Algorithm: Generate one or more personalized driving style information based on autonomous driving data collected during autonomous driving, generate a driving style form for user selection by combining pre-stored default driving style information, and generate set driving style information in response to user commands; preferably, the personalized driving style information includes in-vehicle driver driving style information and regional scene driving style information.

[0082] Step S200 - Perception Algorithm: Provide optional driving route information based on driving environment information, wherein the optional driving route information includes drivable areas, target lists, and lane lines;

[0083] Step S300 - Planning Algorithm: Generate a driving planning trajectory based on vehicle positioning information, selectable driving route information, and set driving style information. The driving planning trajectory includes the vehicle's longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicles, and the lateral safety distance of following vehicles.

[0084] Step S400 - Control Algorithm: Generate driving control commands based on vehicle positioning information, driving trajectory planning, and set driving style information.

[0085] In a preferred embodiment, the implementation process of the driving style recognition and matching algorithm is as follows: Figure 3 As shown, step S100 can be specifically broken down into the following processes:

[0086] Step S110: The intelligent driving domain controller 3 takes the vehicle information, perception information and positioning information obtained by the vehicle driver data acquisition module after reading the data M times in a preset number as input, and generates the current driver's driving style information through the vehicle driver style recognition algorithm in the vehicle driver style recognition module.

[0087] Step S120: The regional scene information acquisition module 31c transmits the acquired information to the remote server 6. The remote server 6 takes the regional scene information of all vehicles in the preset scale ZS as input and calculates the driving style information of the current regional scene through the regional scene driving style recognition algorithm. In a preferred embodiment, the preset scale ZS includes a preset geographical range P1, a preset time period T3, and a preset number of vehicles K. One embodiment of the preset geographical range P1 is: Changchun City; one embodiment of the preset time period T3 is: winter; one embodiment of the preset number of vehicles K is: 100,000 vehicles.

[0088] Step S130: The human-machine interface 4 receives the current driving style Limit of the onboard driver sent by the onboard driver style recognition module 31b and the current regional scene driving style LimitZ sent by the regional scene driving style recognition module, and prompts the user with the currently available driving styles through the display screen of the human-machine interface 4. Typically, the driving styles are divided into three types: default driving style, driver driving style, and regional scene driving style; wherein, the default driving style is the style determined by parameters set by the algorithm developers in the intelligent driving algorithm; the driver driving style is the driving style identified in the onboard driver driving style recognition; and the regional scene driving style is the driving style identified by the regional scene driving style algorithm in the server 6; the driver operates on the human-machine interface 4 to select a specific driving style, and the human-machine interface 4 sends the selected driving style to the intelligent driving domain controller 3 via the CAN bus.

[0089] Step S140: The intelligent driving controller 3 receives the driving style selected by the user and adjusts the parameter values ​​of the planning algorithm and control algorithm to the values ​​corresponding to that style according to the set driving style information, so as to reasonably control the vehicle to achieve style switching in intelligent driving mode. The adjusted parameter values ​​include: Limit for driver driving style; LimitZ for regional scenario driving style.

[0090] In a preferred embodiment, the method steps for generating the vehicle driver's driving style information in step S110 are as follows: Figure 4 As shown, the specific process includes the following:

[0091] Step S111: Read the autonomous driving data of the driver once. The autonomous driving data includes autonomous driving driving data, autonomous driving environment data, and autonomous driving positioning data. The autonomous driving driving data includes the accelerator pedal opening and its changes, the brake pedal opening and its changes, and the steering wheel angle and its changes. The autonomous driving environment data includes a target list and lane line information. The autonomous driving positioning data includes the vehicle's speed, position, acceleration, azimuth angle, and yaw rate information.

[0092] Step S112: Determine whether a driving style boundary has been formed based on the autonomous driving data collected in step S111. If the determination result is yes, proceed to step S150; if the determination result is no, proceed to step S113.

[0093] Step S113: Determine whether the current number of reads exceeds the preset number M. If it exceeds the preset number M, proceed to step S114; if it does not exceed the preset number M, proceed to step S111.

[0094] Step S114: Plot the read M autonomous driving data onto the same coordinate system, solve for the boundary Limit, so that a certain preset proportion of the data in the M data are all within the boundary Limit, and generate the vehicle driver's driving style information based on the boundary Limit; accordingly, the intelligent driving domain controller 3 can send the calculated current driver's driving style to the human-machine interface 4 via the CAN bus, where the driver style is the aforementioned boundary Limit;

[0095] Step S115: Read the current real-time driving data, determine whether the real-time driving data falls within the boundary Limit, and count the data that exceeds the boundary Limit;

[0096] Step S116: Determine whether the number of data exceeding the current boundary Limit exceeds the preset threshold N. If the determination result is that it exceeds the preset threshold N, then proceed to step S114; if the determination result is that it does not exceed the threshold N, then proceed to step S111.

[0097] In a preferred embodiment, the method steps for generating the regional scene driving style information in step S120 are as follows: Figure 5 As shown, it specifically includes the following process steps:

[0098] Step S121: Determine whether there is a regional scene driving style boundary. If the result is no, proceed to step S122; if the result is yes, proceed to step S126.

[0099] Step S122: Read the regional scene data, which includes geographical range, time period and vehicle quantity information;

[0100] Step S123: Determine whether the regional scene data exceeds the preset scale ZS. If it exceeds the preset scale ZS, proceed to step S124. If it does not exceed the preset scale, proceed to step S122. The preset scale ZS includes the preset geographical range P1, the preset time period T3, and the preset number of vehicles K.

[0101] Step S124: Divide the regional driving style into zones based on the regional scene data obtained in step S122; some typical scenarios of the regional driving style zones include zones divided according to location, time period, vehicle size, and road segment, such as: Changchun winter congested expressway, Changchun winter unobstructed expressway, Changchun highway, etc.; the corresponding data obtained are Changchun winter congested expressway data, Changchun winter unobstructed expressway data, Changchun highway data, etc.

[0102] Step S125: Based on the regional driving style partitions defined in step S124, calculate the driving style boundary LimitZ for different regional scenarios to define the regional scenario driving style. Typically, the remote server 6 sends the calculated current regional scenario driving style to the vehicle network control unit 5 via wireless network; the vehicle network control unit 5 then sends this information to the human-machine interface 4 via the CAN bus. The aforementioned regional scenario driving style is the boundary LimitZ.

[0103] Step S126: Determine whether the preset update time interval has been exceeded. If it has, proceed to step S122; otherwise, proceed to the end.

[0104] In a preferred embodiment, when generating the vehicle driver's driving style information, the lateral and longitudinal acceleration ranges, the following road speed limit difference, the following longitudinal safety distance, and the following lateral safety distance in the planning and control algorithm unit are adjusted according to the lateral and longitudinal acceleration boundaries, speed limit deviation boundaries, longitudinal safety distance boundaries, and lateral safety distance boundaries, respectively. The solution process for the applied boundary parameters, including the lateral acceleration boundary, speed limit deviation boundary, longitudinal safety distance boundary, and lateral safety distance boundary, is as follows (see...). Figure 6 ):

[0105] (1) Transverse and longitudinal acceleration boundaries

[0106] See Figure 6 As shown in Figure a, the intelligent driving domain controller 3 reads the maximum longitudinal acceleration and the corresponding lateral acceleration, minimum longitudinal acceleration and the corresponding lateral acceleration, minimum lateral acceleration and the corresponding longitudinal acceleration, and maximum lateral acceleration and the corresponding longitudinal acceleration during a single driving action of the intelligent driving vehicle in the driver's autonomous driving mode, from vehicle start-up to vehicle shutdown. These four points are then plotted as a single data point in a coordinate system with the horizontal axis representing lateral acceleration alat and the vertical axis representing longitudinal acceleration alon.

[0107] Read M data points and plot them in the same coordinate system. Using the y-axis as the center of symmetry, solve for the elliptical combination aLimit as the boundary. This elliptical combination can cover all points plotted from the M data points with a preset ratio Mpercent. A specific embodiment of the preset ratio is 90%.

[0108] (2) Speed ​​limit deviation boundary

[0109] See Figure 6As shown in Figure b, the intelligent driving domain controller 3 reads the road speed limit Vlimit and the vehicle's maximum speed Vmax at that speed limit during a single driving action of the intelligent driving vehicle in autonomous driving mode, from vehicle start-up to vehicle shutdown. The data points for a single driving action are plotted on a coordinate system with the road speed limit Vlimit as the horizontal axis and the difference between the road speed limit and the maximum speed at that speed, Vlimit-Vmax, as the vertical axis.

[0110] Read M data points and plot them in the same coordinate system. Solve for a straight line bLimit as the boundary. This line can divide all points plotted from the M data points onto one side of the boundary at a preset ratio Mpercent. A specific embodiment of this preset ratio is 90%.

[0111] (3) Longitudinal safety distance boundary

[0112] See Figure 6 As shown in Figure c, the intelligent driving domain controller 3 reads the longitudinal distance between the vehicle and the vehicle in front at different speeds during a single driving action of the intelligent driving vehicle in autonomous driving mode, from vehicle start-up to vehicle shutdown. The data points from this single driving action are plotted on a coordinate system with vehicle speed v as the horizontal axis and the longitudinal distance between the vehicle and the vehicle in front (distLon) as the vertical axis.

[0113] Read M data points and plot them in the same coordinate system. Solve for a straight line cLimit as the boundary. This line can divide all points plotted from the M data points onto one side of the boundary at a preset ratio Mpercent. A specific embodiment of this preset ratio is 90%.

[0114] (4) Lateral safety distance boundary

[0115] See Figure 6 As shown in diagram d, the intelligent driving domain controller 3 reads the lateral distance between the vehicle and the adjacent vehicle at different speeds during a single driving action of the intelligent driving vehicle in autonomous driving mode, from vehicle start-up to vehicle shutdown. The data points from this single driving action are plotted on a coordinate system with vehicle speed v as the abscissa and the lateral distance distLat between the vehicle and the adjacent vehicle as the ordinate.

[0116] Read M data points and plot them in the same coordinate system. Solve for a straight line dLimit as the boundary. This line can divide all points plotted from the M data points onto one side of the boundary at a preset ratio Mpercent. A specific embodiment of the preset ratio is 90%.

[0117] In a preferred embodiment, for regional scene driving style recognition, the following road speed limit difference and following longitudinal safety distance in the planning and control module are adjusted according to the longitudinal safety distance boundary and the speed limit deviation boundary, respectively. In a specific embodiment, the example of defining the regional scene driving style in step S125 is divided into the following two aspects (see...). Figure 7 ):

[0118] (1) See Figure 7 As shown in Figure a, taking the solution of driving style boundary on a congested expressway in Changchun during winter as an example, in the data of the congested expressway in Changchun during winter, the vehicle speed v is used as the horizontal axis and the longitudinal distance distLon between the vehicle and the vehicle in front is used as the vertical axis. The N data points are plotted on the same coordinate system, and a straight line eLimitZ is solved as the boundary. This straight line can divide all the points plotted from the N data points to one side of the boundary at a preset ratio Npercent. A specific embodiment of the preset ratio is 85%.

[0119] (2) See Figure 7 As shown in Figure b, taking the solution of driving style boundaries on a congested expressway in Changchun during winter as an example, in the data of the congested expressway in Changchun during winter, the road speed limit Vlimit is used as the horizontal axis, and the difference between the road speed limit and the maximum speed under that speed limit, Vlimit-Vmax, is used as the vertical axis. The N data points are plotted on the same coordinate system, and a straight line fLimit is calculated as the boundary. This straight line can divide all the points plotted from the N data points to one side of the boundary at a preset ratio Npercent. A specific embodiment of the preset ratio is 85%.

[0120] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of the present invention; the dimensions described in the drawings and embodiments are not related to the specific physical object and are not used to limit the protection scope of the present invention. The physical dimensions can be selected and changed according to actual needs.

Claims

1. An intelligent driving method integrating personalized driving styles, comprising: Step S100: Generate one or more personalized driving style information based on the autonomous driving data collected during autonomous driving, generate a driving style form for the user to select from by combining it with pre-stored default driving style information, and generate the set driving style information in response to user commands. The personalized driving style information includes in-vehicle driver driving style information and regional scene driving style information. The method for generating the in-vehicle driver driving style information includes: Step S111: Read the autonomous driving data of the driver once. The autonomous driving data includes autonomous driving driving data, autonomous driving environment data, and autonomous driving positioning data. The autonomous driving driving data includes the accelerator pedal opening and its changes, the brake pedal opening and its changes, and the steering wheel angle and its changes. The autonomous driving environment data includes a target list and lane line information. The autonomous driving positioning data includes the vehicle's speed, position, acceleration, azimuth angle, and yaw rate information. Step S112: Determine whether a driving style boundary has been formed based on the autonomous driving data collected in step S111. If the determination result is yes, proceed to step S115; if the determination result is no, proceed to step S113. Step S113: Determine whether the current number of reads exceeds the preset number M. If it exceeds the preset number M, proceed to step S114; if it does not exceed the preset number M, proceed to step S111. Step S114: Plot the M autonomous driving data read onto the same coordinate system, solve the boundary, so that a certain preset proportion of the data in the M data are all within the boundary, and generate the vehicle driver's driving style information based on the boundary. Step S115: Read the current real-time driving data, determine whether the real-time driving data falls within the boundary, and count the data that exceeds the boundary; Step S116: Determine whether the number of data exceeding the boundary exceeds the preset threshold N. If the determination result is that it exceeds the preset threshold N, then proceed to step S114; if the determination result is that it does not exceed the threshold N, then proceed to step S111. Step S200: Provide optional driving route information based on driving environment information, wherein the optional driving route information includes drivable areas, target lists, and lane lines; Step S300: Generate a driving planning trajectory based on vehicle positioning information, selectable driving route information and set driving style information. The driving planning trajectory includes the vehicle's longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicles, and the lateral safety distance of following vehicles. Step S400: Generate driving control commands based on vehicle positioning information, driving trajectory planning, and set driving style information. 2.The intelligent driving method of claim 1, wherein, The method for generating the aforementioned regional driving style information includes: Step S121: Determine whether there is a regional scene driving style boundary. If the result is no, proceed to step S122; if the result is yes, proceed to step S126. Step S122: Read the regional scene data, which includes geographical range, time period and vehicle quantity information; Step S123: Determine whether the regional scene data exceeds the preset size ZS. If it exceeds the preset size ZS, proceed to step S124. If it does not exceed the preset size, proceed to step S122. Step S124: Divide the regional scene into partitions based on the regional scene data obtained in step S122; Step S125: Based on the regional driving style partitions defined in step S124, solve the driving style boundaries for different regional scenarios to define the regional scenario driving style. Step S126: Determine whether the preset update time interval has been exceeded. If it has, proceed to step S122; otherwise, proceed to the end. 3.The intelligent driving method of claim 1 or 2, wherein The driving style boundary includes at least two of the following: longitudinal acceleration boundary, speed limit deviation boundary, longitudinal safety distance boundary, and lateral safety distance boundary.

4. An intelligent driving control system integrating personalized driving styles, which has the following features: The driving style recognition and matching algorithm unit generates personalized driving style information based on the manual driving information transmitted by the vehicle actuators and the positioning information transmitted by the vehicle positioning sensor during autonomous driving. It then combines this information with pre-stored default driving style information to generate a driving style form for the user to choose from. The personalized driving style information includes the onboard driver's driving style information and the driving style information for specific geographical locations. The driving style recognition and matching algorithm unit is configured to execute steps S111-S116 to generate the onboard driver's driving style information. Step S111: Read the autonomous driving data of the driver once. The autonomous driving data includes autonomous driving driving data, autonomous driving environment data, and autonomous driving positioning data. The autonomous driving driving data includes the accelerator pedal opening and its changes, the brake pedal opening and its changes, and the steering wheel angle and its changes. The autonomous driving environment data includes a target list and lane line information. The autonomous driving positioning data includes the vehicle's speed, position, acceleration, azimuth angle, and yaw rate information. Step S112: Determine whether a driving style boundary has been formed based on the autonomous driving data collected in step S111. If the determination result is yes, proceed to step S115; if the determination result is no, proceed to step S113. Step S113: Determine whether the current number of reads exceeds the preset number M. If it exceeds the preset number M, proceed to step S114. If the preset number of times M is not exceeded, proceed to step S111; Step S114: Plot the M autonomous driving data read onto the same coordinate system, solve the boundary, so that a certain preset proportion of the data in the M data are all within the boundary, and generate the vehicle driver's driving style information based on the boundary. Step S115: Read the current real-time driving data, determine whether the real-time driving data falls within the boundary, and count the data that exceeds the boundary; Step S116: Determine whether the number of data exceeding the boundary exceeds the preset threshold N. If the determination result is that it exceeds the preset threshold N, then proceed to step S114; if the determination result is that it does not exceed the threshold N, then proceed to step S111. The perception algorithm unit provides optional driving route information based on the driving environment information transmitted by the vehicle perception sensing device. The optional driving route information includes drivable areas, target lists, and lane lines. The planning algorithm unit generates a driving planning trajectory based on the vehicle positioning information transmitted by the vehicle positioning sensor, the optional driving path information transmitted by the perception algorithm unit, and the selected driving style information transmitted by the driving style recognition and matching algorithm unit. The driving planning trajectory includes the vehicle's longitudinal acceleration range, the speed difference of the following road speed limit, the longitudinal safety distance of following vehicles, and the lateral safety distance of following vehicles. The control algorithm unit generates driving control commands based on the vehicle positioning information transmitted by the vehicle positioning sensor and the driving planning trajectory transmitted by the planning algorithm unit.

5. The intelligent driving control system integrating personalized driving style according to claim 4, characterized in that, The driving style recognition and matching algorithm unit includes: The vehicle-mounted driver data acquisition module collects autonomous driving data information during autonomous driving. The autonomous driving data information includes the action information of the vehicle actuators, the positioning information of the positioning sensor, and the driving environment information transmitted by the sensing sensor. The in-vehicle driver style recognition module generates in-vehicle driver driving style information based on information from the data acquisition module. The regional scene information collection module collects time information, positioning information from positioning sensors, speed information, and action information from vehicle actuators during driving. The driving style adaptation module provides the human-machine interface with a form containing information on the driver's driving style, regional scene driving style, and default driving style for the user to select. It generates the set driving style in response to user commands and transmits the set driving style parameters to the control algorithm unit.

6. An intelligent driving device integrating personalized driving styles, comprising an intelligent driving controller, a human-machine interface, a positioning sensor, a perception sensor, a vehicle networking control unit, and a remote server: The intelligent driving controller is equipped with an intelligent driving control system, which includes a driving style recognition and matching algorithm unit, a perception algorithm unit, a planning algorithm unit, and a control algorithm unit. The driving style recognition and matching algorithm unit generates personalized driving style information based on the manual driving information transmitted by the vehicle's actuators and the positioning information transmitted by the vehicle's positioning sensors during autonomous driving. It also generates a driving style form for the user to select from, combining this with pre-stored default driving style information. The personalized driving style information includes the onboard driver's driving style information and the driving style information for specific geographical scenarios. The driving style recognition and matching algorithm unit is configured to execute steps S111-S116 to generate the onboard driver's driving style information. Step S111: Read the autonomous driving data of the driver once. The autonomous driving data includes autonomous driving driving data, autonomous driving environment data, and autonomous driving positioning data. The autonomous driving driving data includes the accelerator pedal opening and its changes, the brake pedal opening and its changes, and the steering wheel angle and its changes. The autonomous driving environment data includes a target list and lane line information. The autonomous driving positioning data includes the vehicle's speed, position, acceleration, azimuth angle, and yaw rate information. Step S112: Determine whether a driving style boundary has been formed based on the autonomous driving data collected in step S111. If the determination result is yes, proceed to step S115; if the determination result is no, proceed to step S113. Step S113: Determine whether the current number of reads exceeds the preset number M. If it exceeds the preset number M, proceed to step S114. If the preset number of times M is not exceeded, proceed to step S111; Step S114: Plot the M autonomous driving data read onto the same coordinate system, solve the boundary, so that a certain preset proportion of the data in the M data are all within the boundary, and generate the vehicle driver's driving style information based on the boundary. Step S115: Read the current real-time driving data, determine whether the real-time driving data falls within the boundary, and count the data that exceeds the boundary; Step S116: Determine whether the number of data exceeding the boundary exceeds the preset threshold N. If the determination result is that it exceeds the preset threshold N, then proceed to step S114; if the determination result is that it does not exceed the threshold N, then proceed to step S111. The vehicle networking control unit is used to connect vehicles to the network, send information, and receive network data. The positioning sensing device includes a global positioning system and an inertial measurement unit; The sensing device includes at least two of the following: a camera, a millimeter-wave radar, and a lidar. The intelligent driving control system employs a domain controller; The remote server communicates with the vehicle network control unit via a wireless network device.

7. The intelligent driving device of claim 6, wherein the driving style is integrated into the driving of the vehicle. The remote server contains a regional driving style recognition module; the information sent by the vehicle network control unit includes information collected by the regional scene information acquisition module; the received network data includes the results of the regional driving style recognition module.

Images