A vehicle-mounted computer interface interaction method based on human interaction

Abstract

The present invention relates to the field of human-computer interaction technology, and particularly to a vehicle-mounted computer interface interaction method based on human-computer interaction. The method includes collecting information such as user voice input and touch operations, and using a vector space model for quantitative representation to evaluate the degree of association between the user and the vehicle-mounted computer functions, so as to accurately identify the user's intention; by combining the recognition result of the user's intention with the vehicle-mounted computer system status information, using a Bayesian network to analyze the interaction scenario, and dynamically adjusting the interaction response mechanism to meet the user's needs in different scenarios; by using the analytic hierarchy process to determine the weights of the vehicle-mounted computer interface function modules and optimize the interface layout to improve the convenience and efficiency of user interaction; by constructing a Markov chain model based on user feedback and adjusting the vehicle-mounted computer feedback strategy, the interaction performance and user experience of the vehicle-mounted computer system are improved. The present invention is used to solve the technical problem of poor user interaction experience in vehicle-mounted information systems.

Description

Technical Field

[0001] The present invention relates to the field of human-computer interaction technology, and particularly to a vehicle-mounted computer interface interaction method based on human-computer interaction. Background Art

[0002] Human-computer interaction technology is a professional field of computer science. It studies the interaction methods between humans and computers, aiming to improve the naturalness, efficiency, and satisfaction of this interaction. This field involves many core technologies such as interaction design, artificial intelligence, speech recognition and synthesis, and natural language processing.

[0003] Currently, the control of in-vehicle information systems in the automotive industry still faces some challenges, including the following aspects: the user interaction experience is poor. The vehicle-mounted computer interface interaction methods of existing solutions may have errors, resulting in the inability to accurately meet user needs, inconvenient user operations, and poor experience; the interaction response mechanism is unreasonable and inflexible. The previous vehicle-mounted computer interface interaction methods have not fully considered the vehicle-mounted computer system status information and user needs in different interaction scenarios, resulting in an insufficiently personalized and flexible interaction response mechanism; the interface layout is not intuitive and has poor usability. The current vehicle-mounted computer interface layout has not been reasonably optimized according to the importance of function modules and user usage frequency, making it difficult for users to quickly find and use the required functions; the adjustment strategy for user feedback lacks pertinence and effectiveness. Some vehicle-mounted computer interface interaction methods may lack dynamic adjustment based on user behavior feedback data in the feedback strategy, resulting in an insufficiently optimized user experience. To address these challenges, corresponding solutions and technical means are needed to improve the interaction efficiency and user experience of the vehicle-mounted computer system. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems in the background art, and a vehicle-mounted computer interface interaction method based on human-computer interaction is proposed.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] A vehicle-mounted computer interface interaction method based on human-computer interaction, comprising:

[0007] Step 1: The vehicle-mounted computer interface collects multi-source interaction information and uses a vector space model for quantitative representation to deeply identify user intentions;

[0008] Among them, the multi-source interaction information includes user voice input and touch operations;

[0009] Step 2: Combine the recognition result of the user intention with the vehicle-mounted computer system status information, analyze the interaction scenario using a Bayesian network, and formulate an interaction response mechanism for the current interaction scenario;

[0010] Among them, the in-vehicle infotainment system status information is sensor data such as the obtained vehicle driving speed, the in-vehicle environment (temperature, wind speed), and the current time. These status information may affect the needs of user interaction and the appropriate response methods. For example, when the vehicle is driving at a high speed, some complex operations (such as long-time text input operations) may be inappropriate, and simpler voice interaction and quickly responsive operations are more preferred.

[0011] Step 3: Based on the results of the interaction scenario analysis, for several function modules included in the in-vehicle infotainment system interface, use the analytic hierarchy process to determine the weights of each function module, and then optimize the layout of the in-vehicle infotainment system interface.

[0012] Among them, several function modules included in the in-vehicle infotainment system interface include navigation functions, multimedia functions, vehicle status display functions, etc. The importance of these function modules may vary in different interaction scenarios. The layout of the in-vehicle infotainment system interface includes display size, position, color contrast, etc.

[0013] Step 4: According to the reactions of users after the in-vehicle infotainment system provides feedback, model the user behavior feedback by constructing a Markov chain, so as to adjust the feedback strategy of the in-vehicle infotainment system.

[0014] It should be noted that an in-vehicle infotainment system interface interaction method proposed by the present invention can be widely applied to the control of in-vehicle information systems in the automotive industry. This method can be used to improve the interaction efficiency and user experience of the in-vehicle infotainment system interface. Specifically, it can comprehensively analyze and evaluate and optimize user interaction information, in-vehicle infotainment system status information, and user behavior feedback data through technologies such as vector space models, Bayesian networks, the analytic hierarchy process, and Markov chains. This process covers multiple links such as user intention recognition, interaction scenario analysis, interface layout optimization, and feedback strategy adjustment. These links jointly ensure the accurate matching of user intentions, the rationality of the interaction response mechanism, the usability of the interface layout, and the effectiveness of the feedback strategy, thereby improving the intelligent level and user satisfaction of the in-vehicle information system.

[0015] Furthermore, the process by which the in-vehicle infotainment system interface deeply identifies user intentions by collecting multi-source interaction information and using a vector space model for quantitative representation includes:

[0016] When a user interacts with the in-vehicle infotainment system interface, collect multi-source interaction information: Define the user voice input signal as ; among them, represents a time variable, and is a multi-dimensional vector containing frequency, amplitude, and phase information.

[0017] Mark the user touch operation as ; among them, represents the touch position, Indicates the touch force, Indicates the touch occurrence time;

[0018] The vector space model is used to quantitatively represent the collected multi-source interaction information, specifically including:

[0019] For the user's voice input, it is converted into text form through speech recognition to obtain a text vector , where Indicates the pre-set text feature quantization value, Indicates the pre-set number of features;

[0020] For the user's touch operation, the pre-set functions and corresponding operation instructions of the in-vehicle infotainment system interface are converted into a touch vector , where Indicates the quantization value of the corresponding feature in the pre-set in-vehicle infotainment system function instruction;

[0021] For example, the occurrence frequency of a certain word or phrase in the text after speech recognition conversion (if this word or phrase is in the vocabulary related to the in-vehicle infotainment system functions), or the encoding of the screen area corresponding to the touch operation (according to the pre-divided screen area rules), etc.;

[0022] By calculating the cosine similarity between the text vector and the touch vector, the user's intention is matched to the corresponding in-vehicle infotainment system function to identify the user's intention; among them, the cosine similarity calculation formula is:

[0023] ,

[0024] where Indicates the cosine similarity between the text vector and the touch vector; Is the dot product of the vectors, representing the sum of the products of the two vectors in each dimension; And Are the text vector And the touch vector Of the modulus, that is, the length of the vector; Indicates the feature index;

[0025] It can be understood that the value of the cosine similarity is between -1 and 1. If the value of the cosine similarity is closer to 1, it means that the two vectors are more similar, that is, the user's interaction information is more matched with a certain function instruction of the in-vehicle infotainment system; by calculating the cosine similarity between the two vectors, the correlation degree between the user input (user voice input and touch operation) and the in-vehicle infotainment system function is evaluated, so as to complete the accurate matching of the user's intention to identify the user's intention.

[0026] Furthermore, the process of combining the recognition result of the user's intention with the in-vehicle infotainment system status information and analyzing the interaction scenario using the Bayesian network includes:

[0027] After determining the user's intention, collect the status information of the in-vehicle system through sensors;

[0028] Set the user's intention as , which are different status information of the in-vehicle system; where represents the number of different status information of the in-vehicle system; is a multi-dimensional variable, which can be the vehicle driving speed, or the in-vehicle environment temperature, etc.;

[0029] Obtain the currently observed user input or operation and label it as ; where is a discrete variable;

[0030] Use Bayes' formula to calculate the occurrence probability of the user's intention under different system states; where Bayes' formula is specifically:

[0031] , in the formula, represents the posterior probability of the user's intention given the observed user input or operation and the in-vehicle system state being ; represents the joint probability of the observed user input or operation and the in-vehicle system state being given the user's intention ; represents the prior probability of the occurrence of the user's intention ; represents the joint probability of the observed user input or operation and the in-vehicle system state being .

[0032] Furthermore, the process of formulating the interaction response mechanism in the current interaction scenario includes:

[0033] Step B1: Assign a prior probability to each user intention ;

[0034] Step B2: For each user intention and each possible observed user input or operation as well as the in-vehicle system state , define the conditional probabilities and ;

[0035] Step B3. For a given observed user input or operation and the in - vehicle infotainment system status , use Bayes' formula to calculate the posterior probability of each user intention ; ;

[0036] Step B4. Select the user intention with the highest posterior probability as the intention of the current interaction scenario;

[0037] Step B5. Based on the selected user intention , obtain the corresponding user intention estimate value:

[0038] , where represents the estimated value of the user intention, represents the user intention that maximizes ;

[0039] Step B6. Pre - define a set of responses; among them, this set contains all possible interaction response mechanisms, and each response mechanism corresponds to a specific behavior or operation, such as displaying a certain piece of information, performing a certain action, changing the interface layout, etc.;

[0040] Step B7. According to the user intention estimate value obtained in Step B5, search for the response mechanism that best matches it in the pre - defined set of responses; when the response mechanism that best matches the user intention estimate value is found, use it as the interaction response mechanism in the current interaction scenario;

[0041] Step B8. Execute the selected interaction response mechanism to respond to the user's input or operation.

[0042] Furthermore, based on the results of the interaction scenario analysis, for several functional modules included in the in - vehicle infotainment interface, the process of determining the weights of each functional module using the analytic hierarchy process and then optimizing the in - vehicle infotainment interface layout includes:

[0043] Assume that there are functional modules in the in - vehicle infotainment interface, and construct a judgment matrix ; among them, represents the importance of the in - vehicle infotainment interface functional module with respect to the functional module , are all functional module indices, represents the number of functional modules; for example, if , it means that functional module 1 is three times more important than functional module 2 in the current interaction scenario;

[0044] It is understandable that the judgment matrix is a matrix used in the analytic hierarchy process to represent the results of pairwise comparisons of the importance of factors at the same level with respect to a certain criterion in the upper level;

[0045] Use the characteristic equation , calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector , and solve it by an iterative method;

[0046] It is understandable that in this context, the features refer to the features related to the in-vehicle infotainment system interface function modules, and these features reflect the relative importance of the function modules. For example, for the navigation function and the multimedia function in the in-vehicle infotainment system interface, their relative importance in different interaction scenarios (such as urgently needing to navigate to a destination during driving or being more inclined to the multimedia function when parking for leisure) is a kind of feature; the in the judgment matrix quantifies the importance feature of the in-vehicle infotainment system interface function module relative to the function module

[0047] This importance may be based on factors such as the frequency and urgency of function usage by users in specific scenarios; When normalizing the eigenvector , the normalized vector of the eigenvector is , where is the Euclidean norm of the eigenvector , and in the formula,

[0048] represents the number of eigenvectors;

[0049] Furthermore, the process of using the characteristic equation to calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector and solving it by an iterative method includes:

[0050] Step C1: Rewrite the equation as , that is , where represents the identity matrix; by solving the characteristic polynomial , find the eigenvalue , where represents the eigenvalue quantity;

[0051] Step C2: Select an initial vector ;

[0052] Step C3: For calculate , where represents the number of iterations;

[0053] Step C4: Normalize , that is ;

[0054] Step C5: Calculate ;

[0055] Step C6: Repeat steps C3 - C5 until converges to a preset value, thereby obtaining the maximum eigenvalue ;

[0056] Among them, the selection of the initial vector will affect the convergence speed of the iterative process and the finally obtained eigenvector. Selecting a unit vector as the initial vector can ensure that the modulus of the vector remains within a reasonable range during the iterative process; the more iterations, the more likely the algorithm is to converge to the exact eigenvalue and eigenvector. In practical applications, the iteration can be stopped by setting an upper limit on the number of iterations or according to the change in the eigenvalue; during the iterative process, the eigenvector will gradually converge to the eigenvector corresponding to the maximum eigenvalue.

[0057] Furthermore, according to the user's reaction after the in - vehicle unit provides feedback, the process of adjusting the feedback strategy of the in - vehicle unit by constructing a Markov chain to model the user behavior feedback includes:

[0058] Let the feedback state space be , representing the set of states of possible user behavior feedback (for example represents satisfaction, represents dissatisfaction, represents the need for further operations, etc.), where represents the number of possible user behavior feedback states;

[0059] Establish a transition probability matrix , and complete the initialization: The transition probability matrix is a matrix, representing the probability of transitioning from the user behavior feedback state to the state ; The element of the transition probability matrix Defined as: , that is, the probability that the user will transfer to state in the next step under the current behavior feedback state . In the formula, are all indices of the user behavior feedback state, represents the probability value; in the initial stage, the transition probability matrix is initialized. For example, it is assumed that initially the probability of the user transferring from the "satisfied" feedback state to the "satisfied" state is relatively high, and the probability of transferring from the "dissatisfied" feedback state to the "need further operation" state is also relatively high;

[0060] Based on the interaction between the user and the in-vehicle system, continuously collect the user's behavior feedback data, and update the transition probability matrix according to the collected user behavior feedback data;

[0061] After obtaining the current transition probability matrix and the user's current behavior feedback state , predict the user's possible behavior feedback state in the next step, that is ;

[0062] For the predicted possible behavior feedback of the user in the next step, adjust the feedback strategy of the in-vehicle system; for example, if the probability of the user being in the "dissatisfied" state is relatively high, the feedback method can be changed, such as adjusting the intonation of the voice prompt or re-optimizing the interface display;

[0063] It can be understood that by defining the feedback state space and the transition probability matrix, a Markov chain model is constructed; by initializing and updating the transition probability matrix, the model can reflect the changes in actual data; by predicting the user's next behavior and adjusting the in-vehicle feedback strategy, the user experience is optimized.

[0064] Compared with the existing technology, the advantages of the in-vehicle interface interaction method based on human-computer interaction provided by the present invention are as follows:

[0065] By using the vector space model to quantitatively represent the user's voice input and touch operations, and using the calculation of cosine similarity to match the user's intention with the in-vehicle functions, the present invention can more accurately identify the user's intention, reduce the misrecognition rate, thereby improving the user experience. At the same time, considering both voice and touch interaction methods, the in-vehicle interface can interact with the user more naturally, conform to the user's daily operation habits, and improve the convenience and comfort of the interaction;

[0066] By combining the user's intention and the status information of the in-vehicle system, and analyzing the interaction scenario using a Bayesian network, the present invention can formulate an interaction response mechanism that better conforms to the current scenario and user needs, achieve personalized interaction, and effectively improve user satisfaction; by using the analytic hierarchy process to determine the weights of each functional module and adjusting the interface layout according to the weights, the important functional modules are made more prominent, facilitating the user to quickly find and use them, and improving the interaction efficiency;

[0067] By constructing a Markov chain model based on the user behavior feedback data and continuously updating the transition probability matrix, the present invention can predict the possible next behavior feedback of the user, thereby dynamically adjusting the feedback strategy of the in-vehicle system, improving the pertinence and effectiveness of the feedback. By dynamically adjusting the feedback strategy, it can better meet the user's needs, reduce user dissatisfaction and complaints, and enhance user satisfaction and loyalty.

[0068] In summary, through the above steps being interrelated and mutually promoting, the present invention realizes the accurate recognition of the user's intention, the dynamic analysis of the interaction scenario, the reasonable optimization of the interface layout, and the continuous adjustment of the feedback strategy, significantly improving the interaction performance and user experience of the in-vehicle system, and ensuring the normal implementation of a subsequent in-vehicle interface interaction method based on human-computer interaction. Brief Description of the Drawings

[0069] Figure 1 It is a flowchart of an in-vehicle interface interaction method based on human-computer interaction proposed by the present invention. Detailed Embodiment

[0070] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0071] Refer to Figure 1 , an in-vehicle interface interaction method based on human-computer interaction, includes:

[0072] Step 1: The in-vehicle interface collects multi-source interaction information and uses a vector space model for quantitative representation to deeply identify the user's intention;

[0073] Among them, the multi-source interaction information includes user voice input and touch operations;

[0074] Step 2: Combine the recognition result of the user's intention with the status information of the in-vehicle system, analyze the interaction scenario using a Bayesian network, and formulate an interaction response mechanism for the current interaction scenario;

[0075] Among them, the in-vehicle system status information is sensor data such as the obtained vehicle driving speed, the in-vehicle environment (temperature, wind speed), and the current time. These status information may affect the user interaction requirements and appropriate response methods. For example, when the vehicle is driving at a high speed, some complex operations (such as long-term text input operations) may be inappropriate, and more concise voice interaction and quickly responsive operations are preferred;

[0076] Step 3: Based on the results of the interaction scenario analysis, for several function modules included in the in-vehicle interface, use the analytic hierarchy process to determine the weights of each function module, and then optimize the in-vehicle interface layout;

[0077] Among them, several function modules included in the in-vehicle interface include navigation functions, multimedia functions, vehicle status display functions, etc. The importance of these function modules may vary in different interaction scenarios; the in-vehicle interface layout includes display size, position, color contrast, etc.;

[0078] Step 4: According to the user's reaction after the in-vehicle system provides feedback, model the user behavior feedback by constructing a Markov chain, so as to adjust the feedback strategy of the in-vehicle system.

[0079] It should be noted that an in-vehicle interface interaction method based on artificial interaction proposed by the present invention can be widely applied to the control of in-vehicle information systems in the automotive industry. This method can be used to improve the interaction efficiency and user experience of the in-vehicle interface. Specifically, it can comprehensively analyze and evaluate and optimize user interaction information, in-vehicle system status information, and user behavior feedback data through technologies such as vector space models, Bayesian networks, the analytic hierarchy process, and Markov chains. This process covers multiple links such as user intention recognition, interaction scenario analysis, interface layout optimization, and feedback strategy adjustment; these links jointly ensure the accurate matching of user intentions, the rationality of the interaction response mechanism, the usability of the interface layout, and the effectiveness of the feedback strategy, thereby improving the intelligence level and user satisfaction of the in-vehicle information system.

[0080] Please refer to Figure 1 , the present invention provides an in-vehicle interface interaction method based on artificial interaction. In step 1, the steps for the in-vehicle interface to deeply identify user intentions by collecting multi-source interaction information and using a vector space model for quantitative representation include:

[0081] Step 101: When the user interacts with the in-vehicle interface, collect multi-source interaction information:

[0082] Define the user voice input signal as ; among them, represents a time variable, and is a multi-dimensional vector containing frequency, amplitude, and phase information;

[0083] Mark the user's touch operation as ; where represents the touch position, represents the touch force, represents the touch occurrence time;

[0084] Step 102: Use the vector space model to quantitatively represent the collected multi-source interaction information, specifically including:

[0085] For the user's voice input, convert it to text form through speech recognition to obtain a text vector , where represents the pre-set text feature quantization value, represents the pre-set number of features;

[0086] For the user's touch operation, convert the pre-set functions and corresponding operation instructions of the in-vehicle infotainment system into a touch vector , where represents the quantization value of the corresponding feature in the pre-set in-vehicle infotainment system function instructions;

[0087] For example, the occurrence frequency of a certain word or phrase in the text after speech recognition conversion (if this word or phrase is in the vocabulary related to in-vehicle infotainment system functions), or the encoding of the screen area corresponding to the touch operation (according to the pre-divided screen area rules), etc.;

[0088] Step 103: By calculating the cosine similarity between the text vector and the touch vector, match the user's intention to the corresponding in-vehicle infotainment system function to identify the user's intention; where the cosine similarity calculation formula is:

[0089] ,

[0090] where represents the cosine similarity between the text vector and the touch vector; is the dot product of the vectors, representing the sum of the products of the two vectors in each dimension; and are the norms of the text vector and the touch vector respectively, that is, the length of the vector; represents the feature index;

[0091] In step 103, the value of the cosine similarity is between -1 and 1. The closer the value of the cosine similarity is to 1, the more similar the two vectors are, that is, the more the user's interaction information matches a certain function instruction of the in-vehicle computer. By calculating the cosine similarity between two vectors, the correlation degree between the user input (user voice input and touch operation) and the in-vehicle computer function is evaluated, so as to complete the accurate matching of the user intention and identify the user intention.

[0092] Please refer to Figure 1 , the present invention provides a method for in-vehicle computer interface interaction based on human interaction. In the second step, the step of combining the recognition result of the user intention with the in-vehicle computer system state information, analyzing the interaction scenario by using a Bayesian network, and formulating an interaction response mechanism in the current interaction scenario includes:

[0093] Step 201, after determining the user intention, collect the in-vehicle computer system state information through sensors;

[0094] Step 202, set the user intention as , is different state information of the in-vehicle computer system; wherein, represents the number of different state information of the in-vehicle computer system; is a multi-dimensional variable, can be the vehicle driving speed, can be the in-vehicle environment temperature, etc.;

[0095] Step 203, obtain the currently observed user input or operation, and mark it as ; wherein, is a discrete variable;

[0096] Step 204, use the Bayesian formula to calculate the occurrence probability of the user intention under different system states; wherein, the Bayesian formula is specifically:

[0097] , in the formula, represents the posterior probability of the user intention under the condition of the given observed user input or operation and the in-vehicle computer system state being ; represents the joint probability of the observed user input or operation and the in-vehicle computer system state being under the condition of the given user intention ; represents the prior probability of the occurrence of the user intention ; represents the observed user input or operation and the in-vehicle computer system state being Joint probability;

[0098] Step 205. The steps for formulating the interaction response mechanism in the current interaction scenario are as follows:

[0099] Step B1. Assign a prior probability to each user intention ; ;

[0100] Step B2. For each user intention and each possible observed user input or operation as well as the in-vehicle system state , define the conditional probabilities and ;

[0101] Step B3. For the given observed user input or operation and the in-vehicle system state , use Bayes' formula to calculate the posterior probability of each user intention ;

[0102] Step B4. Select the user intention with the highest posterior probability as the intention of the current interaction scenario;

[0103] Step B5. Based on the selected user intention , obtain the corresponding user intention estimate:

[0104] , where represents the estimate of the user intention, represents the user intention that maximizes ;

[0105] Step B6. Pre-define a set of responses;

[0106] In Step B6, the set contains all possible interaction response mechanisms, and each response mechanism corresponds to a specific behavior or operation, such as displaying a certain piece of information, performing a certain action, changing the interface layout, etc.;

[0107] Step B7. According to the user intention estimate obtained in Step B5, search for the response mechanism that best matches it in the pre-defined set of responses; when the response mechanism that best matches the user intention estimate is found, it is used as the interaction response mechanism in the current interaction scenario;

[0108] Step B8. Execute the selected interaction response mechanism to respond to the user's input or operation.

[0109] Please refer to Figure 1, the present invention provides a vehicle-mounted computer interface interaction method based on human interaction. In step 3, according to the results of interaction scenario analysis, for several function modules included in the vehicle-mounted computer interface, the analytic hierarchy process is used to determine the weights of each function module, and the steps of optimizing the layout of the vehicle-mounted computer interface include:

[0110] Step 301, assume that there are function modules in the vehicle-mounted computer interface, and construct a judgment matrix ; where represents the importance of the function module of the vehicle-mounted computer interface function module for the function module are all function module indexes, represents the number of function modules; for example, if , it means that function module 1 is three times more important than function module 2 in the current interaction scenario;

[0111] In step 301, the judgment matrix is a matrix in the analytic hierarchy process used to represent the results of pairwise comparisons of the importance of each factor at the same level with respect to a certain criterion at the upper level;

[0112] Step 302, use the characteristic equation to calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector , and solve it by an iterative method. The specific steps are as follows:

[0113] Step C1, rewrite the equation as , that is , where represents the identity matrix; by solving the characteristic polynomial , find the eigenvalue , where represents the eigenvalue quantity;

[0114] Step C2, select an initial vector ;

[0115] Step C3, for calculate , where represents the number of iterations;

[0116] Step C4, normalize , that is ;

[0117] Step C5, calculate ;

[0118] Step C6, repeat steps C3-C5 until Converge to a preset value to obtain the maximum eigenvalue ;

[0119] In step 302, the selection of the initial vector affects the convergence speed of the iterative process and the finally obtained eigenvector. Selecting a unit vector as the initial vector ensures that the norm of the vector remains within a reasonable range during the iterative process; the more iterations, the more likely the algorithm is to converge to the exact eigenvalue and eigenvector. In practical applications, the iteration can be stopped by setting an upper limit on the number of iterations or by judging according to the change amount of the eigenvalue; during the iterative process, the eigenvector will gradually converge to the eigenvector corresponding to the maximum eigenvalue;

[0120] In steps 301-302, in this context, features refer to the features related to the in-vehicle infotainment system interface function modules, and these features reflect the relative importance features between function modules. For example, for the navigation function and multimedia function in the in-vehicle infotainment system interface, their relative importance in different interaction scenarios (such as urgently needing to navigate to a destination during driving or being more inclined to the multimedia function when parking and relaxing) is a kind of feature; in the judgment matrix quantifies the importance features of the in-vehicle infotainment system interface function modules relative to the function module This importance may be based on factors such as the frequency and urgency of function usage by users in specific scenarios;

[0121] Step 303. When normalizing the eigenvector , the normalized vector of the eigenvector is , where is the Euclidean norm of the eigenvector , and in the formula, represents the number of eigenvectors;

[0122] Step 304. According to the normalized weight vector, determine the relative importance of each function module in the interface layout and adjust the function modules in the interface layout;

[0123] In step 304, adjusting the function modules in the interface layout specifically includes but is not limited to the display size, position, etc. of the function modules; for example, if the weight of a certain function module is high, its display area can be increased or its position can be swapped. Specifically, the function module with a larger weight can be designed with a larger display size or a more prominent position to attract the user's attention, while the function module with a smaller weight can be designed with a smaller display size or a less conspicuous position.

[0124] Please refer to Figure 1, the present invention provides a vehicle-mounted interface interaction method based on human interaction. In step 4, according to the user's reaction after the vehicle-mounted system provides feedback, the steps of modeling the user's behavior feedback by constructing a Markov chain and then adjusting the feedback strategy of the vehicle-mounted system include:

[0125] Step 401, set the feedback state space as , which represents the set of states of possible user behavior feedback (for example, represents satisfaction, represents dissatisfaction, represents the need for further operation, etc.). Among them, represents the number of possible user behavior feedback states;

[0126] Step 402, establish a transition probability matrix , and complete the initialization: The transition probability matrix is a matrix, which represents the probability of transitioning from the user behavior feedback state to the state ; The element of the transition probability matrix is defined as: , that is, the probability that the user will transition to the state in the next step under the current behavior feedback state . In the formula, are all indexes of the user behavior feedback state, represents the probability value; In the initial stage, the transition probability matrix is initialized. For example, it is assumed that initially the probability of the user transitioning from the "satisfied" feedback state to the "satisfied" state is relatively high, and the probability of transitioning from the "dissatisfied" feedback state to the "need for further operation" is also relatively high;

[0127] Step 403, based on the interaction between the user and the vehicle-mounted system, continuously collect the user's behavior feedback data, and update the transition probability matrix according to the collected user behavior feedback data;

[0128] Step 404, after obtaining the current transition probability matrix and the user's current behavior feedback state , predict the possible next behavior feedback state of the user, that is, ;

[0129] Step 405, adjust the feedback strategy of the vehicle-mounted system for the predicted possible next behavior feedback of the user; For example, if the probability of the user being in the "dissatisfied" state is relatively high, the feedback method can be changed, such as adjusting the intonation of the voice prompt or re-optimizing the interface display;

[0130] For example, assume a feedback state space , where represents satisfaction, represents dissatisfaction; the initial transition probability matrix is defined as: , which means the probability that the user transfers from the "satisfied" state to the "satisfied" state is 0.8, and the probability of transferring to the "dissatisfied" state is 0.2; the probability of transferring from the "dissatisfied" state to the "satisfied" state is 0.3, and the probability of transferring to the "dissatisfied" state is 0.7; when the current behavior feedback state of the user is (dissatisfied), then the predicted possible behavior feedback state of the user in the next step is: , ; According to this prediction result, the in-vehicle system can take corresponding measures to adjust the feedback strategy to improve user satisfaction;

[0131] In steps 401 - 405, a Markov chain model is constructed by defining the feedback state space and the transition probability matrix; by initializing and updating the transition probability matrix, the model can reflect the changes in actual data; by predicting the user's next behavior and adjusting the in-vehicle feedback strategy, the user experience is optimized.

[0132] In the embodiments of the present invention, by collecting multi-source interaction information such as user voice input and touch operations, the user's intention can be understood more comprehensively, improving the accuracy and flexibility of interaction. By using a vector space model to quantitatively represent the interaction information, the correlation degree between the user input and the in-vehicle system functions can be accurately evaluated, achieving precise matching of the user's intention. By calculating the cosine similarity between the text vector and the touch vector, the user's intention can be accurately recognized, reducing the possibility of misoperations and enhancing the user experience. By combining the user intention recognition result with the in-vehicle system status information, the interaction scenario can be analyzed more accurately, and a suitable response mechanism can be formulated. By using a Bayesian network to analyze the interaction scenario, the occurrence probability of the user's intention in different system states can be calculated, improving the formulation accuracy of the response mechanism. Dynamically adjusting the interaction response mechanism according to the analysis results can meet the user's needs in different scenarios, improving the convenience and efficiency of interaction. By using the analytic hierarchy process to determine the weights of each functional module, the importance of different functional modules in different interaction scenarios can be reflected, providing a basis for optimizing the interface layout. Adjusting the interface layout according to the weights of the functional modules can highlight the functional modules with higher weights, improving the convenience and efficiency of user operations. By optimizing the interface layout, the interface becomes more intuitive and easy to use, thus enhancing the user experience and satisfaction. By constructing a Markov chain to model the user behavior feedback, the possible next behavior feedback state of the user can be predicted, providing a basis for adjusting the in-vehicle feedback strategy. Adjusting the in-vehicle feedback strategy according to the prediction results can actively adapt to the user's needs and feedback, improving the intelligence and personalization level of interaction. By continuously collecting user behavior feedback data and updating the transition probability matrix, the interaction experience can be continuously optimized, enhancing the user's usage effect. In summary, the embodiments of the present invention involve data processing, comprehensive analysis, and intelligent adjustment decisions, solving the technical problem of poor user interaction experience in in-vehicle information systems. In actual situations, more data and context information may be required to make specific decisions and optimization plans.

[0133] In addition, the formulas involved above are all calculated by removing the dimension and taking their numerical values, which is a formula obtained by collecting a large amount of data for software simulation to be closest to the actual situation. The proportionality coefficients in the formulas and each preset threshold in the analysis process are set by those skilled in the art according to the actual situation or obtained by simulating a large amount of data; the size of the proportionality coefficient is a specific value obtained by quantifying each parameter for subsequent comparison. Regarding the size of the proportionality coefficient, it depends on the amount of sample data and the processing coefficients initially set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0134] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device embodiments, since they are basically based on the method embodiments, they are described relatively simply, and the relevant parts can refer to the descriptions in the method embodiments.

[0135] For the sake of convenience in description, when describing the above device, it is divided into various units according to functions for separate description. Of course, when implementing the present application, the functions of each unit can be realized in the same or multiple software and / or hardware.

[0136] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0137] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0138] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the functions in the process Figure 1One or more processes and / or boxes Figure 1 Steps of the functions specified in one or more boxes.

[0140] Secondly: In the accompanying drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, reference may be made to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0141] Finally: The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent replacements or changes, and should be covered by the protection scope of the present invention.

Claims

1. A vehicle-mounted device interface interaction method based on human interaction, characterized in that: Step 1: The vehicle-mounted device interface collects multi-source interaction information and uses a vector space model for quantitative representation to deeply identify the user's intention; among them, the multi-source interaction information includes the user's voice input and touch operations; Step 2: Combine the recognition result of the user's intention with the vehicle-mounted device system status information, analyze the interaction scenario using a Bayesian network, and formulate an interaction response mechanism for the current interaction scenario; Step 3: According to the result of the interaction scenario analysis, for several function modules included in the vehicle-mounted device interface, use the analytic hierarchy process to determine the weight of each function module, and then optimize the layout of the vehicle-mounted device interface; Step 4: According to the user's reaction after the vehicle-mounted device provides feedback, model the user's behavior feedback by constructing a Markov chain, so as to adjust the feedback strategy of the vehicle-mounted device; In the said Step 1, the process that the vehicle-mounted device interface collects multi-source interaction information and uses a vector space model for quantitative representation to deeply identify the user's intention includes: When the user interacts with the vehicle-mounted device interface, collect multi-source interaction information: Define the user voice input signal as ; where represents the time variable, and is a multi-dimensional vector containing frequency, amplitude, and phase information; Mark the user's touch operation as ; where represents the touch position, represents the touch force, represents the touch occurrence time; Use a vector space model to quantitatively represent the collected multi-source interaction information, specifically including: For the user's voice input, it is converted into text form through speech recognition to obtain a text vector , where represents a preset text feature quantization value, represents the preset number of features; For a user's touch operation, convert the functions preset in the in-vehicle infotainment system interface and the corresponding operation instructions into touch vectors , where represents the quantization value of the corresponding feature in the preset in-vehicle infotainment system function instructions; By calculating the cosine similarity between the text vector and the touch vector, match the user's intention to the corresponding vehicle-mounted device function to identify the user's intention; among them, the cosine similarity calculation formula is: ， In the formula, represents the cosine similarity between the text vector and the touch vector; is the dot product of vectors, representing the sum of the products of two vectors in each dimension; and are the moduli of the text vector and the touch vector respectively, that is, the lengths of the vectors; represents the feature index.

2. The vehicle-mounted interface interaction method based on human-computer interaction according to claim 1, wherein: In the said Step 2, the process of combining the recognition result of the user's intention with the vehicle-mounted device system status information and analyzing the interaction scenario using a Bayesian network includes: After determining the user's intention, collect the vehicle-mounted device system status information through a sensor; Set the user intention to , which are different status information of the in-vehicle system; among them, represents the number of different status information of the in-vehicle system; Obtain the currently observed user input or operation and mark it as ; Use Bayes' formula to calculate the occurrence probability of the user's intention in different system states; among them, Bayes' formula is specifically: , where represents the posterior probability of the user intention given the observed user input or operation and the in-vehicle system state being ; represents the joint probability of the observed user input or operation and the in-vehicle system state being given the user intention ; represents the prior probability of the user intention occurring; represents the joint probability of the observed user input or operation and the in-vehicle system state being .

3. The vehicle-mounted interface interaction method based on human-computer interaction according to claim 2, characterized in that: In the said Step 2, the process of formulating an interaction response mechanism for the current interaction scenario includes: Step B1: Assign a prior probability to each user intention ; ; Step B2. For each user intention and each possible user input or operation that can be observed as well as the in-vehicle system state , define the conditional probability and ; Step B3. For a given observed user input or operation and the in-vehicle system state , use Bayes' formula to calculate the posterior probability of each user intention ; ; Step B4: Select the user intention with the highest posterior probability as the intention for the current interaction scenario; Step B5, based on the selected user intention , obtain the corresponding user intention estimation value: , where represents the estimated value of the user intention, represents making the user intention that maximizes the value; Step B6: Pre-define a response set; Step B7: According to the user intention estimation value obtained in Step B5, find the most matching response mechanism in the pre-defined response set; when the most matching response mechanism with the user intention estimation value is found, use it as the interaction response mechanism for the current interaction scenario; Step B8: Execute the selected interaction response mechanism to respond to the user's input or operation.

4. A method for in-vehicle infotainment (IVI) interface interaction based on human-computer interaction according to claim 1, characterized in that: In the said Step 3, the process of, according to the result of the interaction scenario analysis, using the analytic hierarchy process to determine the weight of each function module for several function modules included in the vehicle-mounted device interface and then optimizing the layout of the vehicle-mounted device interface includes: Assume that there are functional modules in the in-vehicle infotainment (IVI) system interface, and construct a judgment matrix ; among them, represents the functional module of the IVI system interface Regarding the importance of the functional module , are all functional module indices, represents the number of functional modules; Use the characteristic equation to calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector , and solve it by the iterative method; When normalizing the feature vector , the normalized vector of the feature vector is , where is the Euclidean norm of the feature vector . In the formula, represents the number of feature vectors; According to the normalized weight vector, determine the relative importance of each function module in the interface layout and adjust the function modules in the interface layout.

5. The vehicle-mounted interface interaction method based on human-computer interaction according to claim 4, wherein: The process of using the characteristic equation to calculate the maximum eigenvalue of the judgment matrix and its corresponding eigenvector and solving it through an iterative method includes: Step C1: Rewrite the equation as , that is , where represents the identity matrix; find the eigenvalues by solving the characteristic polynomial , where represents the number of eigenvalues ; Step C2: Select an initial vector ; Step C3. For calculate , where represents the number of iterations; Step C4, Normalization , namely ; Step C5, calculate ; Step C6. Repeat steps C3 - C5 until converging to a preset value, thereby obtaining the maximum eigenvalue .

6. The vehicle-mounted interface interaction method based on human-computer interaction according to claim 1, wherein: In the said Step 4, the process of, according to the user's reaction after the vehicle-mounted device provides feedback, modeling the user's behavior feedback by constructing a Markov chain and then adjusting the feedback strategy of the vehicle-mounted device includes: Let the feedback state space be , which represents the set of states of possible behavioral feedbacks of the user, where represents the number of possible behavioral feedback states of the user; Establish a transition probability matrix , and complete the initialization, specifically: the transition probability matrix is a matrix, representing the probability of transitioning from the user behavior feedback state to the state ; the element of the transition probability matrix is defined as: , that is, the probability that the user will transition to the state in the next step under the current behavior feedback state . In the formula, are all indices of the user behavior feedback state, represents the probability value; in the initial stage, initialize the transition probability matrix . Continuously collect the user's behavioral feedback data based on the interaction between the user and the in-vehicle system, and update the transition probability matrix according to the collected user behavioral feedback data ; After obtaining the current transition probability matrix and the current behavior feedback status of the user predict the possible behavior feedback status of the user in the next step, that is ; For the predicted possible behavior feedback of the user next step, adjust the feedback strategy of the vehicle-mounted device.

Images