Intelligent cockpit evaluation method and device based on improved genetic algorithm, and storage medium

Abstract

The invention relates to the technical field of smart cabin evaluation, in particular to a smart cabin evaluation method and device based on an improved genetic algorithm and a storage medium. The method comprises the steps that the intelligent cockpit is evaluated to obtain an objective score of a last-stage index in an evaluation system, and the evaluation system comprises multiple levels of indexes; obtaining subjective scores of a user group on each index in the evaluation system for different intelligent cabins; learning the weight of each index in the evaluation system through a genetic algorithm to minimize the difference between the objective score of the superior index and the subjective score of the superior index; wherein the objective score of the superior index is obtained by performing weighted summation on the objective score of the index of the level. According to the method, the comprehensive score calculated based on the objective indexes can effectively approach and fit the real subjective evaluation of the user, so that the self-adaption and continuous optimization of the evaluation system are realized.

Description

Technical Field

[0001] The present application relates to the technical field of smart cockpit evaluation, and specifically to a smart cockpit evaluation method, device, and storage medium based on an improved genetic algorithm. Background Art

[0002] Currently, demand for evaluation systems continues to grow across various industries, and building scientific, efficient, and adaptively optimized evaluation models has become a research focus. For example, for the smart cockpit, evaluation metrics include screen response time, display quality, touch sensitivity, voice recognition accuracy, system startup time, and operational response speed. These tests are designed to ensure that the smart cockpit meets standards in terms of functionality, performance, safety, and user experience, thereby improving product quality, optimizing the user experience, enhancing safety, reducing development costs, meeting regulatory requirements, and enhancing market competitiveness.

[0003] Traditional evaluation systems typically employ fixed evaluation indicators and weighting strategies. However, these approaches have significant limitations in practical application. First, the indicator weights in traditional evaluation systems often rely on manual experience or basic statistical methods, making it difficult to fully reveal the complex inherent relationships between indicators, thereby affecting the accuracy and objectivity of evaluation results. Second, faced with the dynamic evolution of the evaluation object and the external environment, traditional evaluation systems lack effective adaptive optimization mechanisms and are unable to adjust model parameters in a timely manner to respond to new evaluation requirements, thereby weakening the evaluation system's practical application effectiveness and flexibility.

[0004] In view of this, this application is filed. Summary of the Invention

[0005] The purpose of this application is to provide a smart cockpit evaluation method, device and storage medium based on an improved genetic algorithm, so that the comprehensive score calculated based on objective indicators can effectively approximate and fit the user's real subjective evaluation, thereby achieving self-adaptation and continuous optimization of the evaluation system.

[0006] In order to achieve the above objectives, this application adopts the following technical solutions: In a first aspect, the present application provides a smart cockpit evaluation method based on an improved genetic algorithm, comprising: Evaluating the smart cockpit to obtain an objective score for the final level indicator in the evaluation system, wherein the evaluation system includes indicators at multiple levels; Obtaining subjective scores of user groups on various indicators in the evaluation system for different smart cockpits; The weights of the indicators in the evaluation system are learned by a genetic algorithm to minimize the gap between the objective scores of the upper-level indicators and the subjective scores of the upper-level indicators; wherein the objective scores of the upper-level indicators are obtained by weighted summing the objective scores of the indicators at the current level; The genetic algorithm uses the weight of each level indicator as the individual to be optimized, and uses the following fitness function The final weight is obtained after optimizing the individuals; ; Among them, M is the total score, Q is the subjective score of the superior indicator, is the weight of the i-th indicator at this level associated with the upper-level indicator, is the objective score of the i-th indicator at this level, n is the number of indicators at this level, is the weighted objective score of the i-th level indicator of multiple smart cockpits The standard deviation of is the weighted objective score of the i-th level indicator of multiple smart cockpits The mean of It is an adjustable coefficient.

[0007] In a second aspect, the present application provides an electronic device, comprising: at least one processor, and a memory communicatively coupled to the at least one processor; In which, the memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute the above-mentioned smart cockpit evaluation method based on the improved genetic algorithm.

[0008] In a third aspect, the present application provides a computer-readable storage medium having computer instructions stored thereon, and the computer instructions are used to enable a computer to execute the above-mentioned smart cockpit evaluation method based on the improved genetic algorithm.

[0009] Compared with the prior art, the present invention has the following advantages: (1) Significantly improve the objectivity and scientific rigor of the evaluation system: The traditional smart cockpit evaluation system is highly dependent on a priori set and fixed indicator weights. Such weight assignments are often based on the experience and judgment of field experts or broad industry reference standards. Their inherent defect is that they may not be able to fully and dynamically reflect the real needs, preferences and evolution trends of end users, which can easily lead to significant deviations between the evaluation results and the actual driving experience of users. This application innovatively introduces genetic algorithms and effectively uses the subjective evaluation of end consumers as the key input of the optimization target to conduct continuous, data-driven dynamic optimization and fine-tuning of the weights of various indicators in the evaluation system. This mechanism enables the evaluation system to adaptively learn and accurately adjust the relative importance of each evaluation indicator in the overall evaluation. Therefore, the method constructed by this application can ensure that the final evaluation results are more realistic and reflect the actual feelings of users, thereby fundamentally improving the objectivity, fairness and scientific rigor of the evaluation system.

[0010] (2) Endowing the evaluation system with endogenous adaptive optimization and continuous evolution capabilities: This application successfully constructs a self-optimization model with endogenous intelligence, which gives the evaluation system the core ability to self-adjust and adaptively optimize based on continuous user feedback data. By using genetic algorithms for iterative learning and calculation, the evaluation system can continuously learn the potential patterns of user group preferences and dynamically adjust the weight parameter configuration of its internal evaluation indicators at all levels. This mechanism ensures that the evaluation system can keep pace with the times and actively adapt to the increasingly complex market environment changes and the high diversity and dynamic evolution characteristics of user needs. This embedded adaptive optimization capability enables the evaluation system proposed in this application to transcend the limitations of traditional static evaluation tools and evolve into an intelligent evaluation system that can respond sensitively and dynamically to user feedback and continuously improve itself.

[0011] (3) Provide accurate data-driven decision support for smart cockpit product iteration and function optimization: The evaluation system optimized by the present application method can not only produce more accurate and reliable comprehensive evaluation results, but more importantly, it can provide clear, data-driven guidance for the subsequent iteration and upgrade of smart cockpit products and the targeted improvement of core functions. By deeply analyzing and exploring the differences and inherent correlations between consumers' subjective evaluation data and the system's objective evaluation results, automakers and suppliers can accurately identify specific performance indicator dimensions or user experience pain points that need to be improved in current product design or function implementation. This continuous improvement strategy based on objective data insights can help guide manufacturers to optimize product design solutions more specifically and improve the user experience of key functions, thereby effectively improving user satisfaction and product core competitiveness in the fierce market competition.

[0012] To sum up, this application uses genetic algorithms to intelligently and adaptively optimize the parameter weights of various core indicators in the smart cockpit evaluation system, significantly improving the accuracy, reliability and user perception fit of the evaluation results, and successfully realizing the dynamic self-optimization and continuous evolution of the evaluation system. Therefore, it has very significant practical application value and broad prospects for industrial promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0014] Figure 1 This is a flow chart of a smart cockpit evaluation method based on an improved genetic algorithm provided in an embodiment of the present application; Figure 2 is a schematic diagram of the evaluation system provided in the embodiment of the present application; Figure 3 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following description of exemplary embodiments of the present application is made in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0016] In recent years, intelligent optimization algorithms, represented by genetic algorithms (GAs), have demonstrated remarkable problem-solving performance in numerous fields. Genetic algorithms, a global optimization algorithm that simulates biological evolutionary processes, have garnered widespread attention in the industry for their inherent efficiency, robustness, and adaptability. By simulating core biological mechanisms such as natural selection, crossover, and mutation, these algorithms can rapidly locate global optimal solutions or high-quality approximate optimal solutions in complex, high-dimensional search spaces. Compared to traditional optimization techniques, genetic algorithms do not rely on precise mathematical models to describe the problem being optimized and can effectively handle nonlinear, multi-objective, and high-dimensional optimization challenges, providing an innovative technical approach for the construction and optimization of evaluation systems.

[0017] In the practice of weighting assessment systems, genetic algorithms can be applied to multiple aspects, such as optimizing the allocation of indicator weights and screening and identifying key indicators, thereby enhancing the scientific rationality and environmental adaptability of the assessment system. For example, the global search characteristics of genetic algorithms can identify the optimal combination of indicator weights, thereby making the assessment results more objective and accurate in reflecting the true condition of the assessed object. By simulating the evolutionary process, the internal structure and core parameters of the assessment model can be dynamically adjusted to adapt to changing assessment scenarios and requirements. In addition, the parallel processing capabilities of genetic algorithms enable them to efficiently process large-scale data sets, providing powerful computational support for the construction of complex assessment systems.

[0018] In light of this, we propose a genetic algorithm-based method for self-optimizing evaluation system weights. This method effectively overcomes the inherent flaws of traditional evaluation systems and enables dynamic optimization and adaptive adjustment of evaluation models. By leveraging the global optimization capabilities of genetic algorithms, we can build a more scientific and efficient evaluation system, thereby improving the accuracy and credibility of evaluation results. Leveraging its adaptive nature, we can empower evaluation models with the ability to dynamically adjust to complex and changing evaluation environments. This method not only opens up new technical paths for the construction of evaluation systems but also provides powerful tools for evaluating practices across various industries, heralding broad application prospects.

[0019] Figure 1 This is a flow chart of a smart cockpit evaluation method based on an improved genetic algorithm provided by an embodiment of the present application. The method can be executed by an electronic device. The method provided in this embodiment is applicable to the case of weight learning of various indicators in the evaluation system of the smart cockpit. Figure 1 , the method provided in this embodiment includes the following operations: S110. Evaluate the smart cockpit to obtain an objective score for the final indicator in the evaluation system.

[0020] The evaluation system includes multiple levels of indicators. For example, first-level indicators include intelligent interaction, intelligent escort, and smart services. Second-level indicators are subordinate indicators of each first-level indicator. For example, intelligent interaction includes touch interaction, digital key, and mobile car connectivity. Third-level indicators are subordinate indicators of each second-level indicator. For example, touch interaction includes response time, sliding smoothness, and ease of operation. This process can be repeated to form a complete evaluation system with multiple levels of indicators, and the last level of indicators is the final level.

[0021] Figure 2 This is a schematic diagram of the evaluation system provided in the embodiment of the present application. The first-level indicators include A, B, and C, the second-level indicators include A1, A2, A3, B1, B2, B3, C1, C2, C3, and C4, and the third-level indicator C4 includes D1, D2, and D3. The relationship between the indicators can be seen in Figure 2 .

[0022] In this embodiment, the smart cockpit is evaluated using relevant evaluation tools and evaluation processes to obtain the physical values of the final indicators, such as the response time of the screen. The physical values of the final indicators are normalized to obtain a score in the range of 0 to 1, i.e., an objective score. Each final indicator has a corresponding weight, i.e., a weight that needs to be optimized using a genetic algorithm in this application. The objective scores of several final indicators are weighted and summed according to the weights to obtain an objective score of a higher-level indicator associated with the several final indicators. For example, Figure 2 In the example above, the objective scores of D1, D2, and D3 are weighted and summed to obtain the objective score of C4. This process is repeated to obtain the objective scores of each indicator in the evaluation system.

[0023] S120. Obtain subjective scores of user groups for each indicator in the evaluation system for different smart cockpits.

[0024] The user groups here are groups that have been classified based on user portraits to reflect different users' preferences for smart cockpit functions. Optionally, user groups can be divided using a clustering algorithm. The functional requirements of different users for smart cockpits were investigated, and users were clustered according to the same or similar functional requirements, resulting in five typical user groups: rational replacement type, family upgrade type, urban single type, practical experience type, and basic economy type. Different user groups have significantly different functional requirements for smart cockpits (for example, in the range of 0 to 100). For example, family upgrade users have a 25% higher demand for "child safety lock response time" and "rear seat heating speed" than the average, while urban single users have a 30% higher demand for "voice interaction multi-round dialogue capability" than the average.

[0025] In this embodiment, the weights of the evaluation system indicators are optimized for each type of user group respectively, so it is necessary to obtain subjective scores for each type of user group respectively.

[0026] For example, consider 10 car models, each equipped with a different smart cockpit. Each user in a specific user group experiences each smart cockpit and, based on their experience, provides a subjective rating for each indicator in the evaluation system. During subsequent use, the subjective ratings of all users for each indicator in each smart cockpit can be averaged to obtain the user group's subjective rating for each indicator.

[0027] In one specific embodiment, in order to improve the efficiency of collecting subjective scores, the following method is adopted: obtaining evaluation texts of user groups on different types of vehicles equipped with smart cockpits; identifying indicator content and user tendencies in the evaluation texts; and obtaining subjective scores for each indicator based on the identified indicator content and user tendencies.

[0028] Collect user comments on smart cockpits in different car models from web pages or forums, such as "This screen is very smooth." Identify keywords in the comments and compare them with the indicator content one by one. Identify user tendencies in the comments, such as "positive, neutral, and negative." Different tendencies are mapped to different subjective scores, and the user's subjective score for the indicator "sliding smoothness" can be obtained.

[0029] S130. Learning the weight of each indicator in the evaluation system through a genetic algorithm to minimize the gap between the objective score of the upper-level indicator and the subjective score of the upper-level indicator; wherein the objective score of the upper-level indicator is obtained by weighted summing the objective scores of the indicators at this level.

[0030] After the weights of each indicator in the evaluation system are optimized, the objective scores of the indicators at this level should be weighted and summed to obtain the objective scores of the associated superior indicators. The objective scores of the superior indicators should be close to the subjective scores of the superior indicators, indicating that the weights of the indicators at this level are accurate.

[0031] In this embodiment, the indicators in the evaluation system are divided into different groups according to their association relationships. Each group of indicators includes all the indicators at the same level that are associated with the same upper-level indicator. Then, a genetic algorithm is used to optimize the weights of each group of indicators in the evaluation system in order from bottom to top until the weight optimization of all indicators is completed. For example, Figure 2 In this example, the weights of D1, D2, and D3 are optimized together. The objective scores of D1, D2, and D3 are then weighted and summed to obtain the objective score of C4. C1, C2, and C3 are the final indicators, and their objective scores are obtained through evaluation. The weights of C1, C2, C3, and C4 are optimized together, and the objective scores of C1, C2, C3, and C4 are then weighted and summed to obtain the objective score of C. The following details the weight optimization process of the genetic algorithm, using a set of indicators in the evaluation system as an example.

[0032] Step 1: Initialize the weights of a set of indicators to form an initial population. The initial population includes multiple individuals, each of which represents the weight of a set of indicators.

[0033] Taking A1, A2, and A3 as an example, the initial population may include 10 individuals, each of which is a set of initial weights for A1, A2, and A3.

[0034] Optionally, to improve optimization efficiency, the initial weights of all indicators in the evaluation system are derived using the Analytic Hierarchy Process (AHP). Initial weights of indicators at the current level that are associated with the same upper-level indicator are optimized. Specifically, ten industry experts assign a 1-9 scale to each level of indicators, such as intelligent interaction, intelligent escort, and smart services. These indicators are then compared pairwise to construct a judgment matrix. The initial weights of each indicator are calculated using the AHP.

[0035] Optionally, the initial weights can be modified based on user group preferences. Because different user groups prioritize different smart cockpit features, the weights of the indicators corresponding to these features should be increased to reflect their preferences. For example, for the family upgrade group, the weight of the smart escort indicator is increased by 30% compared to the initial weight, prompting the algorithm to optimize these indicators, reduce errors in these indicators, and improve their performance. For the urban single group, the weight of the smart interaction indicator is increased by 25% compared to the initial weight.

[0036] Optionally, the initial weights can be adjusted based on weather impacts and emerging technologies. Real-time monitoring of seasonal changes (such as low temperatures in winter) automatically increases the initial weights of indicators such as "seat heating speed" and "air conditioning heating efficiency" by 10% to 15%. If a cockpit incorporates emerging technologies, the weight of these indicators is dynamically adjusted from an initial 5% to 12%.

[0037] Step 2: Use the fitness function to select individuals in the population.

[0038] The genetic algorithm takes the weight of each level indicator as the individual to be optimized, and uses the following fitness function The final weight is obtained after optimizing the individuals; ;Formula (1) Where M is the total score. In this embodiment, the objective score and subjective score of the indicator have the same total score (or full score, such as 10 points). The score difference is normalized by dividing it by the total score. Q is the subjective score of the parent indicator (A). is the weight of the i-th indicator at this level (e.g. A1) associated with the upper-level indicator, is the objective score of the i-th indicator at this level (e.g. A1), n is the number of indicators at this level (e.g. 3), Is an adjustable coefficient, such as 0.2, used to adjust and proportion.

[0039] is the weighted objective score of the i-th level indicator of multiple smart cockpits The standard deviation of is the weighted objective score of the i-th level indicator of multiple smart cockpits Assuming there are 10 different smart cockpits corresponding to 10 car models, and the objective score of the i-th indicator at this level can be obtained for each smart cockpit, then for each indicator at this level, we can get the number of objective scores of smart cockpits, for example, 10 objective scores for indicator A1, 10 objective scores for indicator A2, and 10 objective scores for indicator A3.

[0040] Calculate the standard deviation and mean of all weighted objective scores for each current-level metric. The standard deviation of a data series divided by its mean is commonly referred to as the coefficient of variation (CV). It is a dimensionless ratio that measures the degree of dispersion of the data relative to its mean. Specifically, it represents the relative difference between a data point and its mean and is suitable for comparing the dispersion of data sets with different means or dimensions. Therefore, in this embodiment, the CV is calculated for each current-level metric. If the CV is large, it indicates that the weighted objective scores for that current-level metric are relatively dispersed. This metric's weighted objective scores for different smart cockpits are significantly different, clearly reflecting the differences between them. Therefore, the weight for that current-level metric should be retained, indicating a high fitness level. If the CV is small, it indicates that the weighted objective scores for that current-level metric are relatively concentrated. This metric's weighted objective scores for different smart cockpits are similar, failing to reflect the differences between them. Therefore, the weight for that current-level metric should not be retained, indicating a low fitness level.

[0041] Compared with the traditional method of constructing a fitness function based on the "difference between individual values and target values", this embodiment also considers that the weighted objective score should be able to reflect the differences between different smart cockpits, making the weighted objective score more meaningful for evaluation.

[0042] The selection strategy uses elite retention combined with dynamic roulette: the top 150 high-fitness individuals (accounting for 10%) are retained in each generation, while low-fitness individuals are retained with a probability of 5% to maintain diversity.

[0043] To ensure the reasonable convergence of the genetic algorithm optimization process and the stability and practical usability of the final optimization results, the embodiment of the present application introduces and sets the following key constraints on the basis of the above-mentioned fitness function: First, the weight of each indicator is strictly limited to the closed interval [0,1] to ensure its clear physical meaning and probabilistic interpretation; Second, to prevent excessive and unrealistic fluctuations in the weight during the optimization iteration process, it is stipulated that the absolute difference between the weight of each indicator after optimization and its initial weight before optimization shall not exceed a preset threshold (for example, 0.15); Third, the sum of the weights of all current-level indicators associated with the same upper-level indicator must be strictly equal to 1, that is, the weight normalization condition is met. The comprehensive setting of the above constraints fully considers the intrinsic validity of the indicators and the stability requirements of the entire evaluation system, ensuring that the fitness function can truly and accurately reflect the actual contribution of each indicator in the smart cockpit evaluation system to the final comprehensive score during the iterative process of global optimization search, thereby effectively supporting and realizing the self-optimization process of the evaluation system.

[0044] Step 3: Perform crossover and mutation on the selected individuals to obtain a new generation of individuals.

[0045] The selected individuals are crossed, including: dividing all indicators of the same level in a group of indicators into different functional domains; setting the crossover probability of indicator weights of different functional domains to be smaller than the crossover probability of indicator weights of the same functional domain; and crossing the weights of all indicators of the same level in a group of indicators according to the set crossover probability.

[0046] At least one metric in the same functional domain is related to a function of the smart cockpit, while metrics in different functional domains are related to different functions of the smart cockpit. For example, metrics at this level include: wakeup success rate, wakeup time, interaction success rate, and interaction response time. The wakeup success rate and wakeup time are related to the wakeup function and belong to the same functional domain; the interaction success rate and interaction response time are related to the interaction function and belong to the same functional domain.

[0047] For example, the crossover probability of the indicators "wake-up success rate and interaction success rate" in different functional domains is a smaller value, such as 0.6, and the crossover probability of the indicators in the same functional domain is a larger value, such as 0.8, which strengthens the collaborative optimization between indicators with the same function.

[0048] Optionally, the selected individuals are mutated, including: dividing all indicators at this level in the set of indicators into safety and non-safety categories; setting the variation range of the weights of safety indicators to be smaller than the variation range of the weights of non-safety indicators; setting the mutation probability to decay linearly as the iteration proceeds; and mutating the weights of all indicators at this level in a set of indicators according to the set variation range and mutation probability.

[0049] In this embodiment, the mutation strategy is dynamically adjusted based on indicator attributes: safety-related indicators (such as surround view obstacle detection) are limited to a ±3% variation range, while non-safety indicators, such as entertainment indicators (such as ambient lighting color switching speed), are allowed to fluctuate by ±12%. The mutation probability decays linearly from 0.1 to 0.01 with each iteration, balancing global exploration and local refinement. For example, the mutation probability of the population is 0.1 in the first iteration and 0.09 in the second iteration, with a linear decay in steps of 0.01.

[0050] Step 4: Termination condition. If the termination condition is not met, return to the operation of selecting individuals in the population using the fitness function until the termination condition is met and the optimal individual is obtained.

[0051] Optionally, the termination condition includes a fixed number of iterations, such as 200 generations. If the fitness function value fluctuation is less than 2% for 30 consecutive generations, convergence is considered early. When the number of user reviews of emerging features (such as head-up display technology) exceeds a set threshold, such as 400, a weight reset is automatically triggered and emergency optimization is initiated, ensuring the evaluation system's rapid response to market trends.

[0052] The optimized weighting system was deployed in a phased, real-vehicle validation phase: Subjective user ratings (e.g., voice interaction satisfaction and surround view clarity) were collected in real time from the in-vehicle systems of 200 test vehicles. The new weightings were deployed on 100 vehicles, while the old weightings were retained on 100. On-board sensors collected real-time dynamic parameters such as acceleration, braking, and steering. Combined with user subjective ratings (e.g., voice interaction satisfaction and surround view clarity), the difference in pre- and post-optimization scores was calculated. Data showed that the new weighting reduced the error between the Intelligent Interaction module's scores and user subjective ratings by 12%, and the correlation coefficient between the Intelligent Escort module's scores and user subjective ratings increased from 0.71 to 0.88. Following validation, the Smart Cockpit Evaluation application was compressed to 896 bytes and pushed to the in-vehicle controllers of production vehicles via over-the-air download, achieving lightweight deployment.

[0053] This embodiment uses a closed-loop feedback mechanism to optimize the dynamic weighting of the evaluation system, triggering continuous evolution through multi-source data. When the cumulative number of valid user reviews exceeds 1,200, a new round of genetic algorithm iterations is automatically initiated. The version management system stores the past eight generations of weighted versions, annotating user classification tags such as "mainstream users," "family users," "young groups," "urban beauties," "affordable," and "basic transportation." It supports one-click rollback and records key parameter changes for each optimization (e.g., the intelligent interaction weight increased from 58% to 63%) and user feedback trends (e.g., voice interaction satisfaction increased from 82% to 89%).

[0054] Through genetic algorithm iteration, the weight update cycle has been shortened from the traditional monthly approach to weekly, with extreme scenarios enabling daily updates. The optimized evaluation scores are 32% more consistent with consumer survey data, and the latency in weight adjustments responding to user needs has been reduced to within 48 hours. This mechanism transcends the limitations of traditional static evaluation systems, achieving a fully closed-loop evolutionary process of "user demand collection - algorithm weight optimization - actual vehicle performance verification - further optimization." This provides a data-driven, scientific path for the continuous iteration of the smart cockpit, in line with the core goal of "dynamic weight adjustment and user experience adaptation."

[0055] like Figure 3 As shown, this embodiment provides an electronic device, including: at least one processor; and a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to perform the above method. The at least one processor in the electronic device is capable of performing the above method, thereby having at least the same advantages as the above method.

[0056] Optionally, the electronic device also includes interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of a GUI (Graphical User Interface) on an external input / output device (such as a display device coupled to the interface). In other embodiments, if necessary, multiple processors can be used with multiple memories, and / or multiple buses can be used with multiple memories. Similarly, multiple electronic devices can be connected (for example, as a server array, a group of blade servers, or a multi-processor system), with each device providing part of the necessary operations. Figure 3 A processor 301 is taken as an example.

[0057] Memory 302, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the improved genetic algorithm-based smart cockpit evaluation method in the embodiments of this application. Processor 301 executes the software programs, instructions, and modules stored in memory 302 to perform various functional applications and data processing of the device, thereby implementing the aforementioned improved genetic algorithm-based smart cockpit evaluation method.

[0058] The memory 302 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function; the data storage area may store data created based on the use of the terminal, etc. Furthermore, the memory 302 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 302 may further include a memory remotely located relative to the processor 301, and these remote memories may be connected to the device via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0059] The electronic device may further include: an input device 303 and an output device 304. The processor 301, the memory 302, the input device 303 and the output device 304 may be connected via a bus or other means. Figure 3 The bus connection is taken as an example.

[0060] The input device 303 can receive input digital or character information, and the output device 304 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light-emitting diode (LED) display, and a plasma display. In some embodiments, the display device may be a touch screen.

[0061] This embodiment provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to cause a computer to execute the above method. The computer instructions on the computer-readable storage medium are used to cause a computer to execute the above method, thereby having at least the same advantages as the above method.

[0062] The medium in this application may be any combination of one or more computer-readable media. The medium may be a computer-readable signal medium or a computer-readable storage medium. The medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0063] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0064] The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the foregoing.

[0065] Computer program code for performing the operations of the present application can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0066] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired connection, such as a coaxial cable, optical fiber, digital subscriber line (DSL), or wireless connection, such as infrared, wireless, or microwave. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device, such as a server or data center, that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, or magnetic tape), an optical medium, or a semiconductor medium. It is worth noting that the computer-readable storage medium mentioned in the embodiments of the present application may be a non-volatile storage medium, in other words, a non-transitory storage medium.

[0067] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.

[0068] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A smart cockpit evaluation method based on an improved genetic algorithm, characterized in that: include: Evaluating the smart cockpit to obtain an objective score for the final level indicator in the evaluation system, wherein the evaluation system includes indicators at multiple levels; Obtaining subjective scores of user groups on various indicators in the evaluation system for different smart cockpits; The weights of the indicators in the evaluation system are learned by a genetic algorithm to minimize the gap between the objective scores of the upper-level indicators and the subjective scores of the upper-level indicators; wherein the objective scores of the upper-level indicators are obtained by weighted summing the objective scores of the indicators at the current level; The genetic algorithm uses the weight of each level indicator as the individual to be optimized, and uses the following fitness function The final weight is obtained after optimizing the individuals; ； Among them, M is the total score, Q is the subjective score of the superior indicator, is the weight of the i-th indicator at this level associated with the upper-level indicator, is the objective score of the i-th indicator at this level, n is the number of indicators at this level, is the weighted objective score of the i-th level indicator of multiple smart cockpits The standard deviation of is the weighted objective score of the i-th level indicator of multiple smart cockpits The mean of It is an adjustable coefficient.

2. The smart cockpit evaluation method based on improved genetic algorithm according to claim 1 is characterized in that: Obtain subjective ratings of various indicators in the evaluation system from user groups for different smart cockpits, including: Obtain user evaluation texts on different models equipped with smart cockpits; Performing indicator content recognition and user tendency recognition on the evaluation text; Based on the identified indicator content and user tendencies, a subjective score for each indicator is obtained.

3. The smart cockpit evaluation method based on improved genetic algorithm according to claim 1 is characterized in that: The weights of the various indicators in the evaluation system are learned through a genetic algorithm, including: Using a genetic algorithm, each group of indicators in the evaluation system is weighted and optimized in order from bottom to top until the weight optimization of all indicators is completed; Among them, each group of indicators includes all indicators at this level associated with the same upper-level indicators.

4. The smart cockpit evaluation method based on improved genetic algorithm according to claim 2 is characterized in that: A genetic algorithm is used to optimize the weights of a set of indicators in the evaluation system, including: Initialize the weights of a set of indicators to form an initial population, which includes multiple individuals, each of which represents the weight of a set of indicators; Use fitness function to select individuals in the population; Perform crossover and mutation on the selected individuals to obtain a new generation of individuals; Returns the operation of selecting individuals from the population using the fitness function until the termination condition is met and the optimal individual is obtained.

5. The smart cockpit evaluation method based on improved genetic algorithm according to claim 4 is characterized in that: Perform crossover on selected individuals, including: Divide all indicators at this level in the set of indicators into different functional domains; The crossover probability of indicator weights of different functional domains is set to be smaller than the crossover probability of indicator weights of the same functional domain; According to the set crossover probability, the weights of all indicators at this level in a group of indicators are crossed.

6. The smart cockpit evaluation method based on improved genetic algorithm according to claim 5 is characterized in that: Perform mutations on selected individuals, including: Classify all indicators at this level in the set of indicators into safety and non-safety categories; Set the variation range of the weight of safety indicators to be smaller than that of the weight of non-safety indicators; As the iteration proceeds, the mutation probability is set to decay linearly; According to the set variation range and variation probability, the weights of all indicators at this level in a group of indicators are mutated.

7. The smart cockpit evaluation method based on improved genetic algorithm according to claim 6 is characterized in that: After initializing the weights of the set of indicators to form an initial population, the method further includes: The initialized weights are modified based on the preferences of the user group, the impact of weather on the smart cockpit, and the new technologies of the smart cockpit.

8. An electronic device, characterized in that: include: at least one processor, and a memory communicatively coupled to the at least one processor; Wherein, the memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors so that at least one of the processors can execute the smart cockpit evaluation method based on the improved genetic algorithm described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that The medium stores computer instructions, which are used to enable a computer to execute the smart cockpit evaluation method based on the improved genetic algorithm according to any one of claims 1 to 7.

Images