Intelligent driving vehicle motion control parameter setting method based on driver style

Abstract

The invention discloses an intelligent driving vehicle motion control parameter setting method based on driver styles, and the method comprises the steps: forming a closed-loop iteration optimization path of a human-vehicle driving strategy through constructing a style modeling system based on multi-dimensional driving behavior analysis; the technical bottlenecks of a traditional intelligent driving system in the aspects of personalized service capability and sustainable evolution capability are broken through, and the change from thousands of people to thousands of people is realized. Specifically, multi-dimensional driving behavior data and intelligent driving scenes are associated, driver styles are classified and correspondingly adjusted into different driving stylization parameters, and a cost value feedback mechanism is formed by learning driver driving behaviors and takeover behaviors and combining driving data before and after takeover with the driving scenes. The driving behavior and the real-time environment are organically linked, and human-in-the-loop parameter optimization is realized. According to the invention, the automatic driving function of the vehicle can effectively meet the real intention of the driver, and the driving experience of the user is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and in particular, to a method for setting motion control parameters of an intelligent driving vehicle based on driver style. Background Art

[0002] With the successive introduction of L3-level intelligent driving regulations, L3-level intelligent driving allows the driver to take hands and feet off, and hand over the longitudinal and lateral motion control of the vehicle to the intelligent driving system. At this time, whether the vehicle travels in line with the driver's intention will be more important. By learning different driver styles and adjusting the motion control parameters of the intelligent driving system, the vehicle can travel more in line with the driver's intention, reduce the takeover caused by inconsistent driving expectations, and also reduce the accidents caused by takeover mistakes, improving the intelligent driving experience.

[0003] There is no reliable technical application in the prior art to adjust the motion control parameters of the intelligent driving system according to the driver style, and the existing prior art related to the driver style has deficiencies: for example, the essence of the solution is only to provide a warning function, and the motion control of the vehicle is still the responsibility of the driver; or only from the perspective of modifying the vehicle operation parameters, but the actual control of the vehicle still depends on the driver's operation; or only define the driver style from the road environment modeling and the driver's preview interval, and cannot accurately define a more realistic driver style.

[0004] After analysis, the existing intelligent driving systems mainly adopt fixed parameter configurations and cannot be adjusted stylistically according to the individual driving habits of drivers, resulting in an operation logic fault when switching between the human and intelligent driving modes. The user needs to adapt to the logic of the intelligent driving system, bringing an unsatisfactory driving experience; at the same time, due to the lack of a learning mechanism for the driver takeover scenario, the algorithm cannot be continuously optimized through real driving feedback, restricting the intelligent evolution of the entire system. Summary of the Invention

[0005] In view of the above, the present invention aims to provide a method for setting motion control parameters of an intelligent driving vehicle based on driver style to solve the problems of insufficient personalized service and lack of dynamic optimization mechanism in the existing intelligent driving systems.

[0006] The technical solution adopted by the present invention is as follows:

[0007] The present invention provides a method for setting motion control parameters of an intelligent driving vehicle based on driver style, which includes:

[0008] After identifying that the current driver is in an unregistered state, obtain the initial motion control parameters and determine whether the automatic driving mode is enabled when the vehicle is running;

[0009] When it is determined that the autonomous driving mode is not enabled, classify the current driver's style based on the collected control data of the current driver during vehicle operation and the current driving environment scene, and adjust the initial motion control parameters based on the classification result to obtain the first driving style motion control parameters;

[0010] When it is determined that the autonomous driving mode is enabled, continuously detect the time point when the current driver takes over and intervenes. After the takeover and intervention occur, record the current driving environment scene corresponding to the time point, the intelligent driving data before the time point, and the takeover driving data after the time point;

[0011] Combined with the current driving environment scene, obtain cost feedback according to the intelligent driving data and the takeover driving data, and use the cost feedback to dynamically adjust the initial motion control parameters or the first driving style motion control parameters to obtain the second driving style motion control parameters;

[0012] Based on the first driving style motion control parameters and / or the second driving style motion control parameters, determine the personalized vehicle motion control parameters corresponding to the current driver, and register and bind the current driver.

[0013] In at least one possible implementation manner, the recording of the current driving environment scene corresponding to the time point, the intelligent driving data before the time point, and the takeover driving data after the time point includes: recording the surrounding environment at the time of takeover and intervention and labeling it as the corresponding scene type, and recording the intelligent driving planned path in the first period before the takeover and intervention and the driver's operation driving trajectory in the second period after the takeover and intervention.

[0014] In at least one possible implementation manner, the obtaining of the cost feedback includes: comparing the driver's operation driving trajectory with the intelligent driving planned path to generate a cost value.

[0015] In at least one possible implementation manner, the parameter setting method further includes: fine-tuning several parameters in the personalized vehicle motion control parameters according to the manual driving style in the non-autonomous driving mode within a preset period or the trend of the driver taking over from the autonomous driving mode to obtain temporary stylized parameters;

[0016] During the subsequent intelligent driving process using the temporary stylized parameters, determine whether to solidify the temporary stylized parameters into the target vehicle motion control parameters corresponding to the current driver.

[0017] In at least one possible implementation, the control objects of the initial motion control parameters include at least one of the following: left-right spacing control, following distance control, acceleration / deceleration control, average speed control, steering rate control, and kinetic energy recovery intensity control.

[0018] In at least one possible implementation, the control data includes at least one of the following: following distance, steering rate, accelerator pedal opening, brake pedal opening, pedal opening change rate, average speed, lane change operation, and overtaking frequency.

[0019] In at least one possible implementation, the classification of the current driver's style at least includes: a radical driving style, a neutral driving style, and a steady driving style defined in ascending order of the established scores.

[0020] Compared with the prior art, the main design concept of the present invention is to form a closed-loop iterative optimization path for the vehicle-driver driving strategy by constructing a style modeling system based on multi-dimensional driving behavior analysis, break through the technical bottlenecks of traditional intelligent driving systems in terms of personalized service capabilities and continuous evolution capabilities, and achieve the transformation from one-size-fits-all to personalized for each individual. Specifically, the present invention correlates multi-dimensional driving behavior data and intelligent driving scenarios, classifies the driver's style, and correspondingly adjusts it to different driving style parameters. By learning the driver's driving behavior and takeover behavior, and using the driving data before and after takeover in combination with the driving scenario, a cost value feedback mechanism is formed to establish an organic connection between the driving behavior and the real-time environment (such as weather and road conditions), and realize the parameter optimization with the driver in the loop. The present invention can effectively make the vehicle's autonomous driving function more in line with the driver's true intention and significantly improve the user's driving experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described below with reference to the drawings, where:

[0022] Figure 1 is a schematic diagram of a method for setting motion control parameters of an intelligent driving vehicle based on driver style provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0024] The present invention proposes an embodiment of a method for setting motion control parameters of an intelligent driving vehicle based on driver style. Specifically, as Figure 1As shown in the figure, which includes:

[0025] Step S1: After identifying that the current driver is in an unregistered state, obtain the initial motion control parameters and determine whether the autonomous driving mode is enabled when the vehicle is running;

[0026] In actual operation, if the driver is not registered in the intelligent driving system of this vehicle, it can be considered that the current driver has not been bound to the terminal of the intelligent driving system of this vehicle and belongs to a driver with an unknown style. Therefore, it is proposed to first call the default vehicle motion control parameters related to the driving style.

[0027] After that, for the currently identified driver, it is necessary to construct a personalized driving style motion control mode for him, that is, to determine the vehicle autonomous driving motion control parameters that conform to the current driver's personal driving habits. Due to this requirement, the present invention proposes to first distinguish whether the vehicle is in manual driving or autonomous driving during operation. In other words, the learning of the aforementioned motion control parameters in the present invention is comprehensively learned through different driving modes, rather than relying solely on a single driving mode.

[0028] Step S2: When it is determined that the autonomous driving mode is not enabled, classify the current driver based on the collected control data of the current driver when operating the vehicle and the current driving environment scene, and adjust the initial motion control parameters based on the classification result to obtain the first driving style motion control parameters;

[0029] During the manual driving stage, by recording information such as the following-distance time, steering rate, accelerator pedal opening, brake pedal opening, pedal opening change rate, average speed, lane-changing operation, and overtaking frequency when the driver drives the vehicle, combined with the current driving scene identified by the intelligent driving system, classify the driver's style (for example, classify from aggressive, neutral to steady, on a scale of 0-100); then, use the classification result to adjust the default stylized control parameters of the intelligent driving system (including but not limited to left and right spacing control, following-distance time control, acceleration and deceleration control, average speed control, steering rate control, kinetic energy recovery intensity, etc.) to make it as consistent as possible with the current driver's style.

[0030] Step S3: When it is determined that the autonomous driving mode is enabled, continuously detect the time point when the current driver takes over and intervenes, and after the takeover and intervention occur, record the current driving environment scene corresponding to the time point, the intelligent driving data before the time point, and the takeover driving data after the time point;

[0031] The present invention believes that it is still insufficient to learn the driving style only through the aforementioned non-autonomous driving mode. Therefore, the present invention proposes that during the autonomous driving stage, when the driver takes over or intervenes in the intelligent driving state, the system records the surrounding environment at the time of takeover or intervention, identifies the scene type and labels it as the corresponding scene, and records the planned path of the intelligent driving system 10 seconds before the takeover or intervention and the driving trajectory, steering, and pedal operations of the vehicle driven by the driver 5 seconds after that.

[0032] Regarding what can be supplemented in the above steps S2 and S3, the cloud can train a stylized parameter model through driver data and send it to the vehicle terminal. When the vehicle terminal feeds back the takeover scene data in real time, it can also trigger the cloud model to be iteratively updated in reverse. Specifically, in some preferred embodiments of the present invention, in combination with the data closed-loop, after collecting the driver's driving style data, a stylized parameter configuration is obtained by training a model in the cloud and deployed to the production vehicle through OTA, so that the learning ability of the driving style covers automotive products, thereby greatly improving the intelligence of the vehicle. Specifically, in combination with the end-to-end intelligent driving solution, the driver's driving style data collected can be used to train a model in the cloud and fine-tune the model, so as to directly output an end-to-end intelligent driving model optimized for the individual and deploy it to the production vehicle.

[0033] Step S4: Combine the current driving environment scene, obtain a cost feedback according to the intelligent driving data and the takeover driving data, and dynamically adjust the initial motion control parameter or the first driving style motion control parameter by using the cost feedback to obtain a second driving style motion control parameter;

[0034] Specifically, in combination with the foregoing embodiments, on the basis of the current scene at the time of takeover, the driving path of the manual takeover driving can be compared with the planned path of the intelligent driving system to generate a cost value as a parameter affecting the stylization, so as to dynamically adjust the stylization parameter of the intelligent driving system.

[0035] Step S5: Based on the first driving style motion control parameter and / or the second driving style motion control parameter, determine the intelligent driving system personal stylization parameter corresponding to the current driver, and register and bind the current driver.

[0036] In actual operation, when the DMS identifies that the driver is a registered and bound driver, the stylized vehicle motion control parameters of the driver that have been stabilized through the above two adjustment environments are automatically matched, so as to achieve the intelligent driving effect that fits the real driving intention.

[0037] Finally, it can also be added that after forming the above-mentioned stable personalized parameters of the intelligent driving system, partial parameters can be further fine-tuned on the basis of the above-mentioned stable personalized parameters according to the manual driving style of the preset period (such as the day of this driving) or the trend of taking over from the intelligent driving system, so as to generate better temporary personalized parameters; during the subsequent intelligent driving process using the temporary personalized parameters, if the temporary personalized parameters are more in line with the driver's true style, the temporary personalized parameters can be solidified into the target personalized vehicle motion control parameters; this concept can make the above-mentioned personalized parameters updated at any time to fit the driving expectations.

[0038] In summary, the main design concept of the present invention is to build a style modeling system based on multi-dimensional driving behavior analysis, form a closed-loop iterative optimization path for the driving strategies of people and vehicles, break through the technical bottlenecks of traditional intelligent driving systems in terms of personalized service capabilities and continuous evolution capabilities, and realize the transformation from one-size-fits-all to personalized for each individual. Specifically, the present invention correlates multi-dimensional driving behavior data and intelligent driving scenarios, classifies the driver's style, and correspondingly adjusts it to different driving personalized parameters. By learning the driver's driving behavior and takeover behavior, and using the driving data before and after takeover in combination with the driving scenario, a cost value feedback mechanism is formed to establish an organic connection between the driving behavior and the real-time environment, and realize the parameter optimization with the driver in the loop. The present invention can effectively make the vehicle's autonomous driving function more in line with the driver's true intention and significantly improve the user's driving experience.

[0039] In the embodiments of the present invention, if there are any expressions of directions, they are relative concepts based on the embodiments. In addition, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent the situation where A exists alone, A and B exist simultaneously, and B exists alone. Where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the front and rear associated objects. "At least one of the following" and its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0040] The structure, features, and effects of the present invention have been described in detail based on the embodiments shown in the drawings. However, the above are only the preferred embodiments of the present invention. It should be noted that for the technical features involved in the above embodiments and their preferred modes, those skilled in the art can reasonably combine and match them into various equivalent solutions without departing from or changing the design concept and technical effects of the present invention. Therefore, the scope of implementation of the present invention is not limited by the drawings. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to equivalent changes, should still be within the protection scope of the present invention as long as they do not exceed the spirit covered by the specification and the drawings.

Claims

1. A method for setting motion control parameters of an intelligent driving vehicle based on driver style, characterized in that: include: After recognizing that the current driver is in an unregistered state, obtaining initial motion control parameters and determining whether the automatic driving mode is turned on when the vehicle is running; When it is determined that the automatic driving mode is not turned on, the current driver is classified according to the collected control data of the current driver when operating the vehicle and the current driving environment scene, and the initial motion control parameters are adjusted based on the classification result to obtain the first driving style motion control parameters; When it is determined that the automatic driving mode is turned on, the time point at which the current driver takes over the intervention is continuously detected, and after the takeover intervention occurs, the current driving environment scene corresponding to the time point is recorded, and the intelligent driving data before the time point and the takeover driving data after the time point are recorded; In combination with the current driving environment scene, cost feedback is obtained according to the intelligent driving data and the takeover driving data, and the initial motion control parameter or the first driving style motion control parameter is dynamically adjusted by using the cost feedback to obtain a second driving style motion control parameter; Based on the first driving style motion control parameter and / or the second driving style motion control parameter, a personal stylized vehicle motion control parameter corresponding to the current driver is determined, and the current driver is registered and bound.

2. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to claim 1, characterized in that: The recording of the current driving environment scene corresponding to the time point and the recording of the intelligent driving data before the time point and the takeover driving data after the time point includes: recording the surrounding environment at the time of takeover intervention and marking it as the corresponding scene type, and recording the intelligent driving planning path in the first time period before the takeover intervention and the driver's operation driving trajectory in the second time period after the takeover intervention.

3. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to claim 2, characterized in that: The obtaining of cost feedback includes: comparing the driver's operating trajectory with the intelligent driving planning path to generate a cost value.

4. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to claim 1, characterized in that: The parameter setting method further includes: fine-tuning several parameters of the personal stylized vehicle motion control parameters according to the manual driving style in the non-automatic driving mode or the tendency of the driver to take over from the automatic driving mode within a preset period to obtain temporary stylized parameters; In a subsequent intelligent driving process using the temporary stylized parameters, it is determined whether to solidify the temporary stylized parameters as target vehicle motion control parameters corresponding to the current driver.

5. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to claim 1, characterized in that: The control object of the initial motion control parameter includes at least one of the following: left and right spacing control, following vehicle distance control, acceleration and deceleration control, average speed control, steering rate control, and kinetic energy recovery intensity control.

6. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to claim 5, characterized in that: The control data includes at least one of the following: following distance, turning rate, accelerator pedal start, brake pedal opening, pedal opening change rate, average speed, lane change operation, and overtaking frequency.

7. The method for setting motion control parameters of an intelligent driving vehicle based on driver style according to any one of claims 1 to 6, characterized in that: The style classification of the current driver at least includes: defining an aggressive driving style, a neutral driving style, and a steady driving style in ascending order of predetermined scores.

Images