Driving strategy selection method and device based on human-computer interaction, equipment and medium

Abstract

The invention discloses a driving strategy selection method and device based on human-computer interaction, equipment and a medium, and the method comprises the steps: generating an outflow quantity of a strategy scheme according to a comprehensive priority value of the strategy scheme relative to a reference scheme and an inflow quantity generation model, generating an inflow amount of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow amount generation model; according to the outflow and inflow, net flow corresponding to the strategy schemes is determined, and the strategy scheme corresponding to the maximum net flow is selected as the optimal driving strategy. According to the invention, by constructing the behavior model of the person, the decision-making characteristics and behavior modes of the person are collected and learned in various interactions of the intelligent agent represented by the intelligent automobile, and the foreground theory is introduced into the traditional algorithm, so that the accuracy of evaluating and selecting the driving decision is improved; the driving strategy which better conforms to the preference tendency of the decision-making person can be provided in the intelligent driving process, and then the user experience of the whole man-machine interaction process is improved.

Description

Technical Field

[0001] The present application relates to the field of human-computer interaction technology, and in particular to a driving strategy selection method, device, equipment and medium based on human-computer interaction. Background Art

[0002] In recent years, the fields of artificial intelligence and human-computer interaction have risen rapidly, and digitalization and intelligence have become the core themes of the future development of various industries. As cars owned by thousands of households become increasingly intelligent, the era of a vehicle being a complete intelligent entity has arrived, and the interaction between drivers and vehicles will become one of the most common human-computer interaction scenarios in the future.

[0003] However, the existing driving strategy selection process often ignores the driver's personal habits and behavior patterns. Each driver will develop a unique driving style and habits in the long-term driving process. Different drivers may make different decisions in similar traffic situations. This individual difference is often not taken into account in traditional decision-making methods. Therefore, the existing driving strategy selection process is not conducive to improving the efficiency of driving strategy selection, and is not conducive to improving the reliability and accuracy of the selected driving strategy.

[0004] Therefore, there is an urgent need for a driving strategy that meets the decision maker's preferences in an intelligent driving mode that is adapted to human-computer interaction, so as to improve the user experience of the entire human-computer interaction process and thereby enhance the intelligence and personalization of driving decisions. Summary of the invention

[0005] The embodiments of the present application provide a driving strategy selection method, device, equipment and medium based on human-computer interaction to solve the technical problems that the above-mentioned existing driving strategy selection process is not conducive to improving the selection efficiency of the driving strategy and is not conducive to improving the reliability and accuracy of the selected driving strategy.

[0006] In a first aspect, an embodiment of the present application provides a driving strategy selection method, which is applied to an electronic device on a vehicle, wherein the electronic device stores an automatic driving system, and the driving strategy selection method includes: Obtaining a strategy scheme of the autonomous driving system and obtaining attributes of the strategy scheme; Generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; Obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; Determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate a comprehensive deviation between the strategy solution and the reference solution with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generation model; Generate a comprehensive priority value of the strategy scheme relative to the reference scheme according to the comprehensive deviation, the comprehensive weight parameter and a preset comprehensive priority value generation model; Generate the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset inflow generation model, and generate the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset outflow generation model; The net flow corresponding to the strategy scheme is determined according to the outflow and the inflow, the net flow corresponding to multiple strategy schemes are sorted, and the strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy.

[0007] In a possible implementation manner of the first aspect, before obtaining the strategy scheme of the autonomous driving system and obtaining the attributes of the strategy scheme, the driving strategy selection method includes: Acquire multiple alternative plans for the autonomous driving system, and acquire two of the multiple alternative plans, use one of the two alternative plans as a strategy plan, and use the other of the two alternative plans as a reference plan.

[0008] In a possible implementation manner of the first aspect, the obtaining a strategy scheme of the autonomous driving system and obtaining an attribute of the strategy scheme include: Obtaining sensor data, vehicle logs, and vehicle monitoring data corresponding to the strategy solution; The sensor data, the vehicle log and the vehicle monitoring data are integrated into evaluation data, and the attribute value of the strategy solution is obtained in the evaluation data.

[0009] In a possible implementation manner of the first aspect, generating a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute, and a preset weight model includes: Acquire language data, and determine a subjective weight corresponding to the attribute value based on the language data and a preset probabilistic language term set; A comprehensive weight parameter corresponding to the attribute is generated according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model.

[0010] In a possible implementation manner of the first aspect, after determining a net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: The driving strategy is transmitted to the preset automatic driving system, and the automatic driving system is controlled to execute the driving strategy.

[0011] In a possible implementation manner of the first aspect, after determining a net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: Read the preset upload time and determine whether the current time is the upload time; If the current time is the upload time, the preset server is connected and the driving strategy is uploaded to the server.

[0012] In a possible implementation manner of the first aspect, after determining a net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: A preset storage area is obtained, and the driving strategy is stored in the storage area.

[0013] In a second aspect, an embodiment of the present application provides a driving strategy selection device, which is applied to an electronic device on a vehicle, wherein the electronic device stores an automatic driving system, including: An acquisition module, used to acquire a strategy scheme of the autonomous driving system and acquire attributes of the strategy scheme; A first generating module, used to generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; A setting module, used for obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; A second generating module is used to determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate a comprehensive deviation between the strategy scheme and the reference scheme with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generating model; A third generation module, configured to generate a model according to the comprehensive deviation, the comprehensive weight parameter and a preset comprehensive priority value, and to generate a comprehensive priority value of the strategy scheme relative to the reference scheme; A fourth generation module, used to generate the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model, and to generate the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model; A selection module is used to determine the net flow corresponding to the strategy scheme according to the outflow and the inflow, sort the net flows corresponding to multiple strategy schemes, and select the strategy scheme corresponding to the largest net flow as the optimal driving strategy.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the driving strategy selection method of the first aspect when executing the computer program.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the driving strategy selection method in the above-mentioned first aspect is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the driving strategy selection method in the first aspect above.

[0017] The beneficial effects of the embodiments of the present application lie in two aspects. On the one hand, the outflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model. The inflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model. The net flow corresponding to the strategy scheme is determined according to the outflow and the inflow, and the net flows corresponding to multiple strategy schemes are sorted. The strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy. Since no manual selection is required, a large amount of human resources and time resources are saved, which is conducive to improving the efficiency of selecting driving strategies. On the other hand, the comprehensive priority value generation model can be used to eliminate the influence of human factors, which is conducive to improving the reliability and accuracy of the selected driving strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 An application scenario diagram of the driving strategy selection method provided in an embodiment of the present application; Figure 2 is a flowchart of a driving strategy selection method provided in an embodiment of the present application; Figure 3 A flowchart for obtaining evaluation data provided in an embodiment of the present application; Figure 4 A schematic block diagram of a driving strategy selection device provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0021] The technical solutions between the various embodiments can be combined with each other, but they must be based on the fact that they can be implemented by ordinary technicians in the field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0022] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0023] The driving strategy selection method provided in the embodiment of the present application can be applied to electronic devices such as mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptop computers, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), etc. The embodiment of the present application does not impose any restrictions on the specific type of electronic devices.

[0024] See also Figure 1 , Figure 1 The application scenario diagram of the driving strategy selection method provided in the embodiment of the present application is described in detail as follows: The electronic device obtains a strategy plan of the autonomous driving system and obtains attributes of the strategy plan.

[0025] Among them, the multiple alternative solutions can come from one file of the autonomous driving system or from multiple files of the autonomous driving system.

[0026] For the sake of explanation, we take the multiple alternatives from two files as an example, as follows: The two files include a first file and a second file. The electronic device is connected to the first file and the second file respectively, and obtains a plurality of alternative solutions from the first file and the second file.

[0027] In an embodiment of the present application, the electronic device can connect to different files at the same time, obtain the strategy plan of the autonomous driving system and the attributes of the strategy plan from the different files.

[0028] See also Figure 2 , Figure 2 It is a flow chart of a driving strategy selection method provided in an embodiment of the present application. The method can be applied to electronic devices on a vehicle, where the electronic devices store an automatic driving system.

[0029] like Figure 2 As shown, the driving strategy selection method provided in the embodiment of the present application includes the following steps, which are described in detail as follows: S201, obtaining a strategy scheme of the autonomous driving system, and obtaining attributes of the strategy scheme; Wherein, before obtaining the strategy scheme of the automatic driving system and obtaining the attributes of the strategy scheme, the driving strategy selection method includes: Acquire multiple alternative plans for the autonomous driving system, and acquire two of the multiple alternative plans, use one of the two alternative plans as a strategy plan, and use the other of the two alternative plans as a reference plan.

[0030] The step of obtaining the strategy scheme of the autonomous driving system and obtaining the attributes of the strategy scheme includes: Obtaining sensor data, vehicle logs, and vehicle monitoring data corresponding to the strategy solution; The sensor data, the vehicle log and the vehicle monitoring data are integrated into evaluation data, and the attributes of the strategy solution are obtained in the evaluation data.

[0031] The attributes include one or a combination of safety data, stability data, congestion rate, and energy consumption.

[0032] For the sake of illustration, an example is given below: The autonomous driving system has n alternatives , has 4 attributes The four attributes are safety data, stability data, congestion rate, and energy consumption. The solution is obtained based on sensor data, vehicle logs, and real-time monitoring. About Attributes The attribute value of , the attribute values ​​of all schemes A decision matrix is ​​formed.

[0033] In order to scientifically determine the superiority of the solution, it is necessary to process the attribute values ​​in the above decision matrix to obtain a standardized decision matrix. and driving stability The attribute value of the benefit-type indicator is standardized as follows: ; in, For the plan About Attributes In the decision matrix The actual attribute value corresponding to This is the first result after standardization.

[0034] Similarly, for the attribute congestion rate , fuel consumption The attribute value of the cost-type indicator is standardized as follows: ; in, For the plan About Attributes In the decision matrix The actual attribute value corresponding to The first result and the second result are combined to obtain the standardized decision matrix ; S202, generating a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute, and a preset weight model; According to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model, a comprehensive weight parameter corresponding to the attribute is generated, including: Acquiring language data, and determining a subjective weight corresponding to the attribute based on the language data and a preset probabilistic language term set; A comprehensive weight parameter corresponding to the attribute is generated according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model.

[0035] Since different attributes generally play different roles in the decision-making process, in order to overcome the limitations of the traditional method of obtaining weights based solely on the entropy weight method, we construct a probabilistic language terminology set and build a human behavior model based on it to further collect and learn user decision preferences and behavior patterns to obtain more reliable subjective weights. The subjective weights are combined with the objective weights obtained by the entropy weight method to determine the comprehensive weight parameters of the four main attributes in the four autonomous driving scenarios: safety data, stable data, congestion rate, and energy consumption.

[0036] The process of obtaining the objective weight corresponding to the attribute is described in detail as follows: For the above standardized decision matrix , entropy can determine the relative importance of each attribute by quantifying the information uncertainty of each attribute, and then measure the dispersion or variability of each attribute in the decision matrix. Entropy is defined as: ;

[0037] in, ,when hour, .

[0038] Entropy is used to quantify the degree of uncertainty of an attribute. If the entropy value of an attribute is large, it means that the information distribution of the attribute is more dispersed and the uncertainty is high. The smaller the entropy value, the more concentrated the information of the attribute and the lower the uncertainty, which means that this attribute should be given a higher weight in decision-making because it provides more deterministic information.

[0039] Therefore, for The objective weights of the attributes are: ; for The objective weight of the attribute, For the The entropy value of the attribute. Reflects the The distribution uncertainty of the attribute in all options. If the entropy value of an attribute is large, it means that the information distribution of the attribute is more dispersed and the uncertainty is high; the smaller the entropy value, the more concentrated the information of the attribute and the lower the uncertainty, which means that this attribute should be given a higher weight in decision making because it provides more certain information.

[0040] The process of obtaining the subjective weight corresponding to the attribute is described in detail as follows: Step 1: Data collection.

[0041] Collecting data is a key step in building a human behavior model. We invite multiple smart car users to independently evaluate and judge the relative importance of the two indicators in each subsystem and their driving habits.

[0042] Step 2: Data preprocessing.

[0043] The user's evaluation and judgment data in the previous step are usually expressed in language, so we need to preprocess the language data and express it intuitively. Specifically, we need to count the language data and obtain the probability of each type of evaluation. , then convert the language evaluation into language labels and further into numerical values, and use the converted numerical values ​​to establish a judgment matrix Then, the judgment matrix The probability corresponding to each element in the matrix The combination of these two results results in a probabilistic language term set, which is an intuitive reflection of the preferences implied in the user reviews collected in the previous step. The detailed process is as follows: Assume the following language term set S= { : Very unimportant, : Very unimportant, : Not very important, : Slightly less important, : Equally important, : Slightly more important, : More importantly, : Very important, : Extremely important}; Regulation With numerical value There is a mapping relationship, and the larger the value, the greater the preference contained in the language data.

[0044] For a specific evaluation of a user, such as "safety is more important than driving stability", "safety is extremely important than fuel efficiency", "driving stability is more important than fuel consumption", the user's language data is mapped to the language term set using mapping rules, and the corresponding value is further mapped. .

[0045] The probabilistic language term set quantifies the user's language data into the probability of language terms, which in turn reflects the user's evaluation of the importance of different attributes, making the subjective weight calculation in subsequent steps more in line with the actual situation.

[0046] Step 3: Calculate users’ subjective preferences for different attributes.

[0047] In order to balance the user's subjective preferences for different autonomous driving attributes, so that the behavior of the autonomous driving system is more in line with the user's expectations and needs, we establish a human behavior model based on the language data collected above and the established probabilistic language terminology set. They are safety, driving stability, congestion rate, and fuel consumption. These weights reflect the user's subjective preferences for different attributes in the autonomous driving system.

[0048] The model seeks the optimal subjective weights of each attribute by minimizing the degree of dissatisfaction, as follows: ; ; In the objective function of this model, and The user attributes Relative to the attribute The positive and negative preference differences; The goal is to find a balance point that minimizes the user's overall dissatisfaction with all attribute preferences.

[0049] is the number of language terms contained in the probabilistic language term set; is the Kth language term; is the probability value corresponding to the Kth language term; is a constant used to adjust the ratio between weights, and the default setting is 4;

[0050] Is an attribute The subjective weight of Is an attribute The subjective weight of is the expectation function, which reflects the user's Relative to the attribute The expected value of the preference intensity is given by To calculate the expectation; It is used to ensure that the deviation will not have logically unreasonable negative values; constraint It is used to ensure that the sum of all subjective weights is 1 and each weight is non-negative, so as to ensure the rationality and interpretability of the weights.

[0051] In the constraints, the deviation under the subjective weight is calculated by the main constraint formula and , constantly updated and weight Iterate until the objective function converges to the optimal solution. That is the optimal value of the final output of the model, and the optimal value obtained That is the attribute The required subjective weight.

[0052] According to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model, a comprehensive weight parameter corresponding to the attribute is generated, including: generating the first comprehensive weight parameter corresponding to the security data according to the subjective weight corresponding to the security data, the objective weight corresponding to the security data, and a preset weight model; generating the second comprehensive weight parameter corresponding to the stationary data according to the subjective weight corresponding to the stationary data, the objective weight corresponding to the stationary data, and the weight model; generating the third comprehensive weight parameter corresponding to the congestion rate according to the subjective weight corresponding to the congestion rate, the objective weight corresponding to the congestion rate, and the weight model; The fourth comprehensive weight parameter corresponding to the energy consumption is generated according to the subjective weight corresponding to the energy consumption, the objective weight corresponding to the energy consumption, and the weight model.

[0053] Exemplarily, the weight model is: ; Among them, the parameters Indicates the proportional coefficient corresponding to the subjective weight; Among them, the parameters Indicates the proportional coefficient corresponding to the objective weight; in, , which is used to indicate the final comprehensive importance of different attributes, namely the importance of the four attributes of safety data, stability data, congestion rate and energy consumption.

[0054] For the sake of illustration, an example is given below: in, Indicates the serial number; ; in, Indicates Comprehensive weight parameters; ; Said represents the first comprehensive weight parameter; Said represents the second comprehensive weight parameter; Said represents the third comprehensive weight parameter; Said represents the fourth comprehensive weight parameter; in, represents the jth subjective weight; ; Said is the subjective weight corresponding to the safety data; Said is the subjective weight corresponding to the stationary data; Said is the subjective weight corresponding to the congestion rate; Said is the subjective weight corresponding to the energy consumption; in, represents the jth objective weight; ; is the objective weight corresponding to the safety data; is the objective weight corresponding to the stationary data; is the objective weight corresponding to the congestion rate; is the objective weight corresponding to the energy consumption.

[0055] Among them, the parameters , the parameters It can be set by the user or by the system default, and there is no restriction here.

[0056] S203, obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; For the sake of illustration, an example is given below: In order to more accurately reflect the relationship between the pros and cons of different decision options and take into account the limited rationality of users, the following two functions are set as corresponding priority functions for selection in different situations. Indicates selection function as the priority function.

[0057] For nonlinear features, it is more appropriate to set the Gaussian criterion function: Among them, nonlinear features refer to characteristic variables that do not conform to linear relationships during data analysis and model building.

[0058] When there are different levels of importance among the evaluation criteria, the hierarchical priority function is more applicable.

[0059] S204, determining the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determining the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generating a comprehensive deviation between the strategy solution and the reference solution with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generation model; The process of setting the preference weight of the priority function is described in detail as follows: Prospect theory is used to describe people's actual behavior when faced with risky choices. It takes into account people's irrational preferences when evaluating probabilities and outcomes. In the autonomous driving strategy decision scenario, we combine prospect theory with the evaluation of driving habits by smart car users in the previous steps to obtain a more realistic and accurate preference evaluation of the user on the priority function.

[0060] The process of setting the preference weight of the priority function is described in detail as follows: Prospect theory is used to describe people's actual behavior when faced with risky choices. It takes into account people's irrational preferences when evaluating probabilities and outcomes. In the autonomous driving strategy decision scenario, we combine prospect theory with the evaluation of driving habits by smart car users in the previous steps to obtain a more realistic and accurate preference evaluation of the user on the priority function.

[0061] Here are the steps: Step 1: Set is the weight vector of each priority function determined by the user under the influence of prospect theory. The T in the upper right corner of the matrix represents the transpose of the matrix. Based on the previous collection of evaluations and judgments of smart car users on driving habits, we have obtained a large number of existing evaluations of priority functions. , which indicates the user's The priority function is Step 2: To quantify the value perception of users on various outcomes and characterize the different ways people perceive gains and losses, the following value function is used To indicate the user's In using When there are priority functions, evaluate Perceived value after conversion: ; Among them, when the value When the perceived value is in the benefit part, the default setting parameters , representing the decreasing marginal satisfaction brought by the benefits; when the evaluation value When the perceived value is in the loss part, the default setting parameters Represents the decreasing marginal dissatisfaction caused by loss, and the loss aversion coefficient is set by default ,Reflecting that in real autonomous driving scenarios, users are more averse and sensitive to risks and losses due to their own safety considerations.

[0062] Step 3: In the scenario of autonomous driving, users often have inaccurate perceptions of low-probability events and high-probability events. The impact of low-probability events is overestimated, while the impact of high-probability events is underestimated. To truly reflect this psychological bias, a probability weighting function is used To adjust the perceived value to make it more consistent with the psychological preferences in actual decision-making: ; in, The value perception after converting the user's specific evaluation in step 2. For specific evaluation. Parameters Used here to describe the user's distorted perception of value in an autonomous driving scenario, the default setting , indicating the general situation of overestimating low-probability events and underestimating high-probability events.

[0063] Step 4: Calculate the user's preference weight for the priority function based on the above results: ; The weight vector of each priority function determined by the user under the influence of prospect theory can be obtained .

[0064] represents the evaluation value of the Gaussian criterion function on the i-th attribute, Represents the evaluation value of the hierarchical priority function on the i-th attribute.

[0065] represents the preference weight of the Gaussian criterion function, represents the preference weight of the hierarchical priority function. Wherein, the deviation generation model is defined as: ; in, Indicates the selection priority function Time plan With the plan About attributes The deviation For the A priority function, , is the Gaussian criterion function is a hierarchical priority function, , Represents alternatives In Properties The normalized evaluation value.

[0066] The greater the deviation, the greater the degree of difference; the smaller the deviation, the smaller the degree of difference.

[0067] Combined with the user's preference weights for the priority function, the comprehensive deviation generation model is: ; in, Representation scheme About the Solution In Properties The combined deviation on For the A priority function, , is the Gaussian criterion function, is a hierarchical priority function, represents the preference weight corresponding to the tth priority function, represents the preference weight of the Gaussian criterion function, represents the preference weight of the hierarchical priority function, Representation strategy plan In Properties The evaluation value on Representation strategy plan In Properties The evaluation value on Representation strategy plan In Properties The standardized evaluation value is Indicates the reference plan In Properties The normalized evaluation value.

[0068] S205, generating a comprehensive priority value of the strategy scheme relative to the reference scheme according to the comprehensive deviation, the comprehensive weight parameter and a preset comprehensive priority value generation model; For the sake of illustration, an example is given below: For any two alternatives in the autonomous driving scenario , the comprehensive priority value generation model is: ; in, Is an attribute The comprehensive weight parameter, For the strategy scheme, For the reference scheme, is the comprehensive priority value of the strategy scheme relative to the reference scheme, Representation scheme About the Solution In Properties The combined deviation on Representation strategy plan In Properties The evaluation value on Representation strategy plan In Properties The evaluation value on Representation strategy plan In Properties The standardized evaluation value is Indicates the reference plan In Properties The normalized evaluation value.

[0069] S206, generating the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model, and generating the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model; For the sake of illustration, an example is given below: For each strategy plan of the user in the autonomous driving scenario , define the inflow generation model and the outflow generation model.

[0070] Among them, the inflow generation model is: ; Among them, the outflow generation model is: ; in, Indicates the reference plan Strategic Plan The overall priority value.

[0071] Where n is the number of solutions; Among them, the inflow Indicates that all reference solutions take precedence over strategic solutions of traffic.

[0072] in, Representation strategy plan For reference scheme The overall priority value.

[0073] Among them, the outflow Representation scheme Prioritizes traffic over all reference solutions; S207, determining a net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy.

[0074] For the sake of illustration, an example is given below: Based on the strategy The inflow and outflow of the traffic can be used to calculate the strategy in the autonomous driving scenario. Priority : ; in, is the net flow, To reflect the strategy Priority in autonomous driving scenarios, is the outflow, Inflow

[0075] in, The larger the value, the better the strategy. The higher the priority level relative to other reference solutions, the higher the priority level is. The net flows corresponding to the plurality of strategy solutions are sorted, and the strategy solution corresponding to the largest net flow is selected as the optimal driving strategy in the autonomous driving scenario.

[0076] For the sake of illustration, an example is given below: For example, the net flows corresponding to the plurality of said strategy schemes include PA, PB, and PC; Wherein, PA represents the net flow of the strategy scheme A; Wherein, PB represents the net flow of the strategy B; Wherein, PC represents the net flow of the strategy scheme C; When PA is 10, PB is 11, and PC is 12, PA, PB, and PC are sorted, PC is the largest net flow, and strategy C corresponding to PC is selected as the optimal driving strategy.

[0077] Wherein, after determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: Step A: transmitting the driving strategy to the preset automatic driving system, and controlling the automatic driving system to execute the driving strategy.

[0078] The driving strategy and preset control instructions are transmitted to the preset automatic driving system, and the automatic driving system is controlled to execute the driving strategy through the control instructions.

[0079] Wherein, after determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: Step B, reading the preset upload time, and determining whether the current time is the upload time; If the current time is the upload time, the preset server is connected and the driving strategy is uploaded to the server.

[0080] Wherein, after determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: Step C, obtaining a preset storage area, and storing the driving strategy in the storage area.

[0081] Among them, step A can be executed before or after step B and step C, step B can be executed before or after step A and step C, step C can be executed before or after step A and step B, and step A, step B, and step C can be executed simultaneously. The specific execution order is not limited here.

[0082] The beneficial effects of the embodiments of the present application lie in two aspects. On the one hand, the outflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model. The inflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model. The net flow corresponding to the strategy scheme is determined according to the outflow and the inflow, and the net flows corresponding to multiple strategy schemes are sorted. The strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy. Since no manual selection is required, a large amount of human resources and time resources are saved, which is conducive to improving the efficiency of selecting driving strategies. On the other hand, the comprehensive priority value generation model can be used to eliminate the influence of human factors, which is conducive to improving the reliability and accuracy of the selected driving strategy.

[0083] See also Figure 3 , Figure 3 The flowchart for obtaining evaluation data provided in the embodiment of the present application is described in detail as follows: S301, obtaining sensor data, vehicle logs and vehicle monitoring data corresponding to the strategy scheme; Among them, sensor data is obtained from sensors of the autonomous driving system.

[0084] For ease of explanation, sensor data is taken as an example. Sensor data includes the following: Point cloud data: Provides an accurate 3D image of the vehicle’s surroundings for obstacle detection, path planning, and terrain recognition.

[0085] Camera image data: captures visual information from the front, rear, and sides of the vehicle for use in traffic sign recognition, pedestrian detection, lane line detection, etc.

[0086] Inertial measurement unit data: Contains acceleration and angular velocity information, which is used to determine the vehicle's attitude, speed, and direction changes, and is critical to vehicle dynamic control and positioning.

[0087] Other sensor data: such as radar data and ultrasonic sensor data.

[0088] Among them, vehicle monitoring data is obtained through the vehicle-mounted system.

[0089] For the sake of explanation, taking vehicle monitoring data as an example, the vehicle monitoring data includes the following contents: Vehicle speed: record the vehicle's speed in real time.

[0090] Engine status: including engine speed, temperature, fuel consumption, fault codes, etc., used to monitor engine performance and predict potential failures.

[0091] Braking system performance: monitor brake pad wear, brake fluid status, brake pressure, etc. to ensure driving safety.

[0092] External environmental parameters: such as weather conditions, visibility, temperature, humidity, etc., affect vehicle performance and driving decisions.

[0093] Battery and energy management system: battery level, charging status, current and voltage, etc., manage vehicle energy usage.

[0094] Among them, vehicle log data is obtained through the vehicle-mounted system.

[0095] For ease of explanation, take vehicle log data as an example. The vehicle log data includes the following contents: System startup and shutdown records: record the time, location and status of each vehicle startup and shutdown.

[0096] Fault log: records faults in various vehicle systems and the corresponding fault codes to help quickly locate problems.

[0097] Maintenance records: including maintenance cycles, time and details of parts replacement, reflecting the vehicle maintenance status.

[0098] Driver behavior log: records the driver's operating habits, such as sudden acceleration, sudden braking, speeding, etc., for evaluating driving safety and optimizing driving assistance systems.

[0099] Mileage and route: Record the total mileage of the vehicle, the specific route and time of each trip, which is helpful for route optimization and energy consumption analysis.

[0100] S302, integrating the sensor data, the vehicle log and the vehicle monitoring data into the evaluation data; S302: Obtain the attribute value of the strategy solution from the evaluation data.

[0101] In an embodiment of the present application, in the evaluation data, the attribute value of the strategy scheme is obtained, and the attribute value includes safety data, stability data, congestion rate, and energy consumption. The safety data, stability data, congestion rate, and energy consumption all have important application value, and these data can help electronic devices select driving strategies.

[0102] For the driving strategy selection method described in the above embodiment, please refer to Figure 4 , Figure 4 A schematic block diagram of a driving strategy selection device provided in an embodiment of the present application, Figure 4 The driving strategy selection device 400 shown can be applied to Figure 1The electronic device in the application scenario diagram shown below takes the electronic device as an example. Figure 4 The driving strategy selection device 400 shown is described in detail. The driving strategy selection device 400 may include an acquisition module 401, a first generation module 402, a setting module 403, a second generation module 404, a third generation module 405, a fourth generation module 406, and a selection module 407.

[0103] An acquisition module 401 is used to acquire a strategy scheme of the autonomous driving system and acquire attributes of the strategy scheme; A first generating module 402, configured to generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; A setting module 403 is used to obtain a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; The second generating module 404 is used to determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate the comprehensive deviation between the strategy solution and the reference solution with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and the comprehensive deviation generating model; The third generation module 405 is used to generate a model according to the comprehensive deviation, the comprehensive weight parameter and the preset comprehensive priority value, and generate a comprehensive priority value of the strategy scheme relative to the reference scheme; A fourth generating module 406, configured to generate the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generating model, and to generate the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generating model; The selection module 407 is used to determine the net flow corresponding to the strategy scheme according to the outflow and the inflow, sort the net flows corresponding to multiple strategy schemes, and select the strategy scheme corresponding to the largest net flow as the optimal driving strategy.

[0104] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0105] The beneficial effects of the embodiments of the present application lie in two aspects. On the one hand, the outflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model. The inflow of the strategy scheme is generated according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model. The net flow corresponding to the strategy scheme is determined according to the outflow and the inflow, and the net flows corresponding to multiple strategy schemes are sorted. The strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy. Since no manual selection is required, a large amount of human resources and time resources are saved, which is conducive to improving the efficiency of selecting driving strategies. On the other hand, the comprehensive priority value generation model can be used to eliminate the influence of human factors, which is conducive to improving the reliability and accuracy of the selected driving strategy.

[0106] See also Figure 5 , Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0107] like Figure 5 As shown, Figure 5 The electronic device 2 includes: at least one processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the at least one processor 20, and when the processor 20 executes the computer program 22, the steps in any of the above-mentioned method embodiments are implemented.

[0108] The electronic device 2 may include, but is not limited to, a processor 20 and a memory 21. Those skilled in the art will appreciate that Figure 5 It is only an example of the electronic device 2 and does not constitute a limitation on the electronic device 2. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0109] The processor 20 is used to run the computer program 22 stored in the memory 21, and implement the following steps when executing the computer program 22: Obtaining a strategy scheme of the autonomous driving system and obtaining attributes of the strategy scheme; Generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; Obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; Determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate a comprehensive deviation between the strategy solution and the reference solution with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generation model; According to the comprehensive deviation, the comprehensive weight parameter and the preset comprehensive priority value generation model, the comprehensive priority value of the strategy scheme relative to the reference scheme is generated; according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset inflow generation model, the outflow of the strategy scheme is generated; according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset outflow generation model, the inflow of the strategy scheme is generated; according to the outflow and the inflow, the net flow corresponding to the strategy scheme is determined, the net flows corresponding to the multiple strategy schemes are sorted, and the strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy.

[0110] In some embodiments, the processor 20 is configured to implement: Obtaining sensor data, vehicle logs, and vehicle monitoring data corresponding to the strategy solution; integrating the sensor data, the vehicle log and the vehicle monitoring data into the evaluation data; In the evaluation data, the attribute value of the strategy solution is obtained.

[0111] In some embodiments, the processor 20 is configured to implement: generating the first comprehensive weight parameter corresponding to the security data according to the subjective weight corresponding to the security data, the objective weight corresponding to the security data, and a preset weight model; generating the second comprehensive weight parameter corresponding to the stationary data according to the subjective weight corresponding to the stationary data, the objective weight corresponding to the stationary data, and the weight model; generating the third comprehensive weight parameter corresponding to the congestion rate according to the subjective weight corresponding to the congestion rate, the objective weight corresponding to the congestion rate, and the weight model; The fourth comprehensive weight parameter corresponding to the energy consumption is generated according to the subjective weight corresponding to the energy consumption, the objective weight corresponding to the energy consumption, and the weight model.

[0112] In some embodiments, the processor 20 is configured to implement: Obtaining a first score of the policy scheme on the security data, obtaining a second score of the reference scheme on the security data, subtracting the second score from the first score, and generating a first priority value of the policy scheme relative to the reference scheme on the security data; Obtaining a third score of the strategy scheme on the stable data, obtaining a fourth score of the reference scheme on the stable data, subtracting the fourth score from the third score, and generating the second priority value of the strategy scheme relative to the reference scheme on the stable data; Obtaining a fifth score of the strategic solution on the congestion rate, obtaining a sixth score of the reference solution on the congestion rate, subtracting the sixth score from the fifth score, and generating the third priority value of the strategic solution relative to the reference solution on the congestion rate; Obtain a seventh score of the strategy scheme on the energy consumption, obtain an eighth score of the reference scheme on the energy consumption, subtract the eighth score from the seventh score, and generate the fourth priority value of the strategy scheme relative to the reference scheme on the energy consumption.

[0113] In some embodiments, the processor 20 is configured to implement: The driving strategy is transmitted to the preset automatic driving system, and the automatic driving system is controlled to execute the driving strategy.

[0114] In some embodiments, the processor 20 is configured to implement: Read the preset upload time and determine whether the current time is the upload time; If the current time is the upload time, the preset server is connected and the driving strategy is uploaded to the server.

[0115] In some embodiments, the processor 20 is configured to implement: A preset storage area is obtained, and the driving strategy is stored in the storage area.

[0116] The processor 20 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0117] In some embodiments, the memory 21 may be an internal storage unit of the electronic device 2, such as a hard disk or memory of the electronic device 2. In other embodiments, the memory 21 may also be an external storage device of the electronic device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 2. Further, the memory 21 may also include both an internal storage unit and an external storage device of the electronic device 2. The memory 21 is used to store an operating system, an application program, a boot loader, data, and other programs, such as the program code of the computer program. The memory 21 may also be used to temporarily store data that has been output or is to be output.

[0118] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0119] An embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0120] The computer-readable storage medium stores program codes, which can be called by a processor to execute the driving strategy selection method described in the above method embodiment.

[0121] The computer-readable storage medium has a storage space for program codes.

[0122] The program code includes the code of any step in the driving strategy selection method described in the above method embodiment.

[0123] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0124] The computer-readable storage medium may also be an external storage device of the driving strategy selection device or the electronic device, for example, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), a non-transitory computer-readable storage medium (non-transitory computer-readable storage medium), etc., equipped on the driving strategy selection device or the electronic device.

[0125] Since the computer program stored in the computer-readable storage medium can execute any one of the driving strategy selection methods provided in the embodiments of the present application, the computer-readable storage medium can achieve the beneficial effects that can be achieved by any one of the driving strategy selection methods provided in the embodiments of the present application. Please refer to the previous embodiments for details and will not be repeated here.

[0126] An embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the above-mentioned driving strategy selection method.

[0127] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0128] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0129] Based on such understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying the computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0130] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A driving strategy selection method, characterized in that: An electronic device applied to a vehicle, wherein the electronic device stores an automatic driving system, and the driving strategy selection method comprises: Obtaining a strategy scheme of the autonomous driving system and obtaining attributes of the strategy scheme; Generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; Obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; Determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate a comprehensive deviation between the strategy solution and the reference solution with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generation model; Generate a comprehensive priority value of the strategy scheme relative to the reference scheme according to the comprehensive deviation, the comprehensive weight parameter and a preset comprehensive priority value generation model; Generate the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset inflow generation model, and generate the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and the preset outflow generation model; The net flow corresponding to the strategy scheme is determined according to the outflow and the inflow, the net flow corresponding to multiple strategy schemes are sorted, and the strategy scheme corresponding to the largest net flow is selected as the optimal driving strategy.

2. The driving strategy selection method according to claim 1, characterized in that: Before obtaining the strategy scheme of the automatic driving system and obtaining the attributes of the strategy scheme, the driving strategy selection method includes: Acquire multiple alternative plans for the autonomous driving system, and acquire two of the multiple alternative plans, use one of the two alternative plans as a strategy plan, and use the other of the two alternative plans as a reference plan.

3. The driving strategy selection method according to claim 1, characterized in that: The obtaining of the strategy scheme of the autonomous driving system and the obtaining of the attributes of the strategy scheme include: Obtaining sensor data, vehicle logs, and vehicle monitoring data corresponding to the strategy solution; The sensor data, the vehicle log and the vehicle monitoring data are integrated into evaluation data, and the attribute value of the strategy solution is obtained in the evaluation data.

4. The driving strategy selection method according to claim 1, characterized in that: According to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model, a comprehensive weight parameter corresponding to the attribute is generated, including: Acquire language data, and determine a subjective weight corresponding to the attribute value based on the language data and a preset probabilistic language term set; A comprehensive weight parameter corresponding to the attribute is generated according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model.

5. The driving strategy selection method according to claim 1, characterized in that: After determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: The driving strategy is transmitted to the preset automatic driving system, and the automatic driving system is controlled to execute the driving strategy.

6. The driving strategy selection method according to claim 1, characterized in that: After determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: Read the preset upload time and determine whether the current time is the upload time; If the current time is the upload time, the preset server is connected and the driving strategy is uploaded to the server.

7. The driving strategy selection method according to claim 1, characterized in that: After determining the net flow corresponding to the strategy scheme according to the outflow and the inflow, sorting the net flows corresponding to a plurality of the strategy schemes, and selecting the strategy scheme corresponding to the largest net flow as the optimal driving strategy, the driving strategy selection method includes: A preset storage area is obtained, and the driving strategy is stored in the storage area.

8. A driving strategy selection device, characterized in that: An electronic device applied to a vehicle, wherein the electronic device stores an automatic driving system, comprising: An acquisition module, used to acquire a strategy scheme of the autonomous driving system and acquire attributes of the strategy scheme; A first generating module, used to generate a comprehensive weight parameter corresponding to the attribute according to the subjective weight corresponding to the attribute, the objective weight corresponding to the attribute and a preset weight model; A setting module, used for obtaining a priority function, wherein the priority function includes a Gaussian criterion function and a hierarchical priority function; A second generating module is used to determine the preference weight of the Gaussian criterion function according to the evaluation value of the Gaussian criterion function on the attribute, determine the preference weight of the hierarchical priority function according to the evaluation value of the hierarchical priority function on the attribute, and generate a comprehensive deviation between the strategy scheme and the reference scheme with respect to the attribute according to the preference weight of the Gaussian criterion function, the preference weight of the hierarchical priority function and a comprehensive deviation generating model; A third generation module, configured to generate a model according to the comprehensive deviation, the comprehensive weight parameter and a preset comprehensive priority value, and to generate a comprehensive priority value of the strategy scheme relative to the reference scheme; A fourth generation module, used to generate the outflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset inflow generation model, and to generate the inflow of the strategy scheme according to the comprehensive priority value of the strategy scheme relative to the reference scheme and a preset outflow generation model; A selection module is used to determine the net flow corresponding to the strategy scheme according to the outflow and the inflow, sort the net flows corresponding to multiple strategy schemes, and select the strategy scheme corresponding to the largest net flow as the optimal driving strategy.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the driving strategy selection method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the driving strategy selection method according to any one of claims 1 to 7 is implemented.

Images