Head-up display adjusting method and device, electronic equipment and program product

Abstract

The invention relates to the technical field of vehicle man-machine interaction, and discloses a head-up display adjusting method and device, electronic equipment and a program product. The method comprises the following steps: acquiring multi-source monitoring data; calculating adjustment constraint parameters of head-up display according to the multi-source monitoring data; based on the adjustment constraint parameter, obtaining an adjustment strategy of head-up display; and adjusting the head-up display based on the adjustment strategy. According to the technical scheme, multi-source monitoring data are effectively fused, self-adaptive adjustment of head-up display is achieved, and driving safety and interaction comfort are improved.

Description

Technical Field

[0001] This application relates to the field of vehicle human-machine interaction technology, specifically to a head-up display adjustment method, device, electronic device, and program product. Background Technology

[0002] In-vehicle head-up display systems are not effective in adapting to complex and changing driving environments.

[0003] In related technologies, head-up display systems rely excessively on linear adjustments based on single parameters such as ambient light or vehicle speed, lacking the ability to integrate multi-source data such as navigation data, vehicle status data, driver status monitoring data, and environmental monitoring data. This results in rigid adjustment strategies in dynamic scenarios such as intersections and highway lane changes, making it impossible to respond accurately to real-time road conditions and driver needs. Summary of the Invention

[0004] This application provides a head-up display adjustment method, device, electronic device, and program product that can effectively integrate multi-source monitoring data to achieve adaptive adjustment of the head-up display, thereby improving driving safety and interactive comfort.

[0005] In a first aspect, this application provides a head-up display (HUD) adjustment method, comprising: acquiring multi-source monitoring data; calculating adjustment constraint parameters for the HUD based on the multi-source monitoring data; obtaining an adjustment strategy for the HUD based on the adjustment constraint parameters; and adjusting the HUD based on the adjustment strategy.

[0006] Secondly, this application provides a head-up display adjustment device, the device comprising: an acquisition module for acquiring multi-source monitoring data; a calculation module for calculating adjustment constraint parameters of the head-up display based on the multi-source monitoring data; a strategy module for obtaining an adjustment strategy for the head-up display based on the adjustment constraint parameters; and an adjustment module for adjusting the head-up display based on the adjustment strategy.

[0007] Thirdly, this application provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the head-up display adjustment method of the first aspect or any corresponding embodiment described above.

[0008] Fourthly, this application provides a computer-readable storage medium storing computer instructions for causing a computer to execute the head-up display adjustment method of the first aspect or any corresponding embodiment described above.

[0009] Fifthly, this application provides a computer program product, including computer instructions for causing a computer to execute the head-up display adjustment method of the first aspect or any corresponding embodiment described above.

[0010] As can be seen from the above, the head-up display adjustment method provided in this application achieves adaptive adjustment by acquiring multi-source monitoring data and calculating adjustment constraint parameters. It has the advantages of effectively integrating multi-source monitoring data to achieve adaptive adjustment of the head-up display, thereby improving driving safety and interactive comfort. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a schematic flowchart of a first method for adjusting a head-up display according to an embodiment of this application; Figure 2 This is a second flowchart illustrating the head-up display adjustment method according to an embodiment of this application; Figure 3 This is a structural block diagram of a head-up display adjustment device according to an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0015] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0016] Traditional in-vehicle head-up display (HUD) systems have several limitations in adaptive adjustment. Their adjustment dimensions are singular, relying excessively on parameters such as ambient light or vehicle speed, and lacking the ability to fuse multi-source data for decision-making. The system fails to monitor the driver's physiological state in real time, resulting in inefficient human-machine interaction. Furthermore, it does not adequately consider the spatiotemporal relationship between the vehicle's trajectory and the navigation path, leading to inaccurate timing and location of information presentation. In extreme environments, the system lacks effective fault tolerance and compensation mechanisms, resulting in insufficient information reliability. In addition, the adjustment strategy is fixed, limiting its ability to personalize and adapt.

[0017] In response, this application proposes a method for adjusting the head-up display, such as... Figure 1 As shown, the method includes: Step S101: Obtain multi-source monitoring data.

[0018] Multi-source monitoring data refers to a collection of data acquired from multiple different sources to assess the driving environment, vehicle status, and driver status. This may include, but is not limited to, navigation information, vehicle sensor data, driver physiological data, and external environmental sensor data. By integrating heterogeneous data, a comprehensive information foundation can be provided for the adaptive adjustment of the head-up display.

[0019] In one example, acquiring multi-source monitoring data could include manually inputting some driving scenario information, such as the driver manually selecting the current road type or weather conditions. As one implementation, ambient brightness data could be acquired solely from the vehicle's ambient light sensor as a single data source. Alternatively, a dataset based on typical driving scenarios could be pre-stored and loaded as monitoring data upon system startup.

[0020] Step S102: Calculate the adjustment constraint parameters for the head-up display based on the multi-source monitoring data.

[0021] Among them, the adjustment constraint parameters refer to a series of restrictive or guiding parameters calculated based on multi-source monitoring data to guide the adjustment of the head-up display. These parameters reflect the complexity of the current driving scenario, the driver's cognitive load, and the priority of information presentation, providing a quantitative basis for generating specific adjustment strategies.

[0022] In one example, the adjustment constraint parameters for the head-up display can be calculated based on multi-source monitoring data using a pre-defined lookup table. For instance, when the ambient brightness data reaches a specific threshold, the corresponding brightness adjustment constraint value can be directly retrieved from the lookup table. As another implementation method, a simple linear weighted summation can be performed on the different types of data to obtain a comprehensive constraint parameter, where the weight of each data type is a fixed value. Alternatively, independent threshold rules can be set for each type of monitoring data; when the data exceeds the threshold, the corresponding constraint parameter is triggered.

[0023] Step S103: Based on the adjustment constraint parameters, obtain the adjustment strategy for the head-up display.

[0024] The adjustment strategy refers to a specific and executable head-up display adjustment scheme generated based on adjustment constraint parameters. It can include adjustment instructions for multiple dimensions such as display brightness, display content, display color, display font, and information density, aiming to optimize information presentation and improve driving safety and comfort.

[0025] In one example, the adjustment strategy for the head-up display (HUD) based on adjustment constraint parameters can be achieved through direct mapping. For instance, the calculated brightness constraint parameters can be directly mapped to the specific brightness percentage of the HUD. Alternatively, a set of rules based on logical judgments can be pre-defined. When a specific combination of constraint parameters is met, corresponding adjustment instructions for the displayed content, color, or font are triggered. Furthermore, the adjustment range of the display elements can be determined based on a single constraint parameter and a simple proportional relationship.

[0026] Step S104: Adjust the head-up display based on the adjustment strategy.

[0027] A head-up display (HUD) is a device that projects vehicle information, navigation information, or other relevant data into the driver's field of vision. It allows the driver to obtain necessary information without looking down at the instrument panel, thereby reducing eye strain and improving driving safety.

[0028] In one example, adjusting the head-up display (HUD) based on an adjustment strategy can include sending the generated adjustment commands to the HUD hardware all at once to set brightness, content, or font. As one implementation, a fixed time interval can be set to periodically execute the adjustment strategy and update the HUD's display state. Alternatively, adjustments can be made only to a single display dimension specified in the adjustment strategy (e.g., brightness only), while other dimensions remain unchanged.

[0029] It is understood that the technical solution in this embodiment acquires multi-source monitoring data, calculates and adjusts constraint parameters based on the data, and then generates and executes an adjustment strategy, thereby achieving intelligent and adaptive adjustment of the head-up display. This effectively solves the problem of existing head-up display systems having a single adjustment dimension and lacking real-time perception of driver status and vehicle dynamics. The technical solution in this embodiment can provide comprehensive and accurate information presentation, improving driving safety and comfort.

[0030] In some embodiments, multi-source monitoring data includes navigation data and vehicle status data.

[0031] Navigation data refers to information related to the vehicle's driving path provided by the vehicle navigation system. This includes real-time route planning information, road type, curve curvature, number of lanes, and speed limits from in-vehicle navigation systems (such as GPS / BeiDou positioning modules combined with map data), or more refined road topology, traffic event information, and suggested lanes from cloud-based navigation services or V2X (vehicle-to-everything) systems. Navigation data provides macro-level guidance and structured information for the vehicle's future driving path. Vehicle status data refers to data collected in real-time by the vehicle's own sensors reflecting its current movement and operating status. This includes vehicle speed from speed sensors, acceleration from accelerometers, angular velocity from gyroscopes, steering angle from steering wheel angle sensors, wheel speed from wheel speed sensors, and braking status from brake pedal sensors. It may also include engine speed, throttle opening, gear information, and vehicle yaw rate obtained from vehicle bus systems (such as CAN bus). Vehicle status data describes the vehicle's current dynamic behavior and movement trends.

[0032] Based on multi-source monitoring data, the adjustment constraint parameters for the head-up display are calculated, including: Step a1: Based on the preset prediction model, predict the vehicle trajectory according to navigation data and vehicle status data to obtain the vehicle trajectory prediction result.

[0033] The pre-defined prediction model refers to a mathematical model or algorithm used to predict the future trajectory of a vehicle based on input data. For example, it can be based on a state estimation and prediction model using a Kalman filter (KF) or an extended Kalman filter (EKF), predicting the future trajectory by iteratively updating the vehicle's state; or it can be based on more complex nonlinear state estimation methods such as an unscented Kalman filter (UKF) or a particle filter to handle the nonlinear characteristics of vehicle motion; or it can be based on machine learning or deep learning methods, such as recurrent neural networks (RNNs) or long short-term memory networks (LSTMs), predicting the vehicle's trajectory by learning from historical data. The pre-defined prediction model establishes a mapping relationship between input data (navigation data, vehicle state data) and the vehicle's future trajectory, providing a quantitative prediction mechanism.

[0034] The preset prediction model includes: X_{k|k-1}=f(X_{k-1},u_k)+w_k, where k is the time index, X is the state vector, X=[x,y,v,a,θ,ω]^T, x and y are the vehicle positions, v is the vehicle speed, a is the acceleration, ω is the heading angle, ω is the steering angular velocity, X_{k-1} is the state estimate of the previous time step, u_k is the control input vector, X_{k|k-1} is the state prediction of the current time step, and w_k is the process noise vector.

[0035] The preset prediction model is a state-space model expression used to describe the evolution of the system state over time. Specifically, it represents the predicted state X_{k|k-1} of the system at time k under the influence of the state X_{k-1} at time k-1 and the control input u_k. The function f represents the kinematic or dynamic model of the vehicle. For example, it can be a simple linear motion model (such as a uniform linear motion model or a uniform turning model), a complex nonlinear motion model (such as a bicycle model or an Ackermann steering model), or a data-driven learning model. w_k represents unpredictable random disturbances (process noise) in the model, usually modeled as Gaussian white noise, and its covariance matrix reflects the uncertainty of the model. Here, k is the time index, representing the current time in the discrete time step or the position in the sequence, used to distinguish data and states at different time points, enabling the model to make predictions and updates on the time series. X is the state vector, X=[x, y, v, a, θ, ω]^T, which contains the key physical quantities describing the motion of the vehicle in the two-dimensional plane. x and y represent the vehicle's absolute position, v is the vehicle's velocity, a is the acceleration, θ is the heading angle (the angle between the vehicle's direction of travel and a fixed axis), and ω is the steering angular velocity (the rate of change of the heading angle). These variables together constitute a complete description of the vehicle's motion state at a given moment, providing a comprehensive foundation for trajectory prediction. X_{k-1} is the state estimate from the previous moment, referring to the estimated value of the vehicle's state at a time step k-1 before the current moment k. As input to the prediction model, it provides the starting point information for the vehicle's motion and is the basis for the next prediction step. u_k is the control input vector, referring to the external control quantity acting on the vehicle at time k. This can include direct control inputs such as the driver's accelerator pedal opening, brake pedal opening, and steering wheel angle, or control commands generated by the vehicle's autonomous driving system or driver assistance system, such as target acceleration and target steering angle. The control input vector affects the vehicle's motion state. X_{k|k-1} represents the current state prediction, calculated by the prediction model based on the state estimate from the previous moment and the control input at the current moment. It reflects the model's inference about the vehicle's future motion trend. w_k is the process noise vector, representing the state prediction error caused by factors such as model inaccuracies, unmodeled disturbances, and sensor noise during vehicle movement. Introducing w_k improves the model's robustness, enabling it to better cope with various uncertainties in the real-world environment and making the prediction results more consistent with reality. The vehicle trajectory prediction result refers to the vehicle's driving path and state sequence over a future period, calculated using a pre-set prediction model based on navigation data and vehicle state data. It provides detailed information about the vehicle's future motion, including future position, speed, and heading angle, serving as a crucial basis for subsequently adjusting the timing and location of the head-up display information.

[0036] In some embodiments, the multi-source monitoring data further includes driver status monitoring data. Based on the multi-source monitoring data, the adjustment constraint parameters for the head-up display are calculated, including: Step b1: Obtain the driver's visual load index based on the driver's condition monitoring data.

[0037] Driver state monitoring data refers to real-time data reflecting the driver's physiological and psychological state, revealing information such as attention level, fatigue level, emotional state, and visual focus. Driver state monitoring data can be acquired in various ways. For example, it can be obtained through in-vehicle camera systems that capture visual information such as facial expressions, eye movements, and head posture, and analyze this information using image processing and pattern recognition technologies. Alternatively, driver state monitoring data can be acquired through wearable devices or seat sensors, which can monitor physiological signals such as heart rate, brain waves, and skin conductance, thus indirectly reflecting the driver's physiological state. The visual load index is an indicator that quantifies the workload or cognitive burden on the driver's visual system in a specific driving situation. A high visual load index usually means that the driver needs to invest more visual resources to process information, which may lead to distraction or fatigue. The visual load index can be obtained by analyzing the driver's eye movement data (such as fixation point, saccade path, and pupil size change frequency). For example, frequent shifts in the driver's fixation point, narrowing of the saccade range, or continuous pupil dilation may indicate a high visual load. Alternatively, the visual load index can also be evaluated by combining information such as the driver's head posture, blinking frequency, and facial micro-expressions through machine learning models. For example, frequent blinking or head shaking may be associated with visual fatigue or increased cognitive load.

[0038] It is understood that the technical solution in this embodiment incorporates driver status monitoring data into a multi-source monitoring data system and calculates a visual load index based on this, effectively solving the problem of disconnect between human-computer interaction and data acquisition. Specifically, by acquiring real-time physiological and psychological state data of the driver, the system can dynamically perceive the driver's visual load level. When the driver's visual load is high, the system can identify that the driver may be in a state of distraction or fatigue, thus providing key constraint parameters for subsequent head-up display adjustments. This real-time perception of the driver's state overcomes the limitations of relying solely on vehicle and navigation data for prediction, making head-up display adjustments no longer unidirectional and passive, but proactively adaptable to the driver's needs, improving the efficiency and safety of human-computer interaction. For example, when the driver's visual load is high, adjusting the head-up display's brightness, content density, or font size can reduce the driver's visual burden, ensure the effective transmission of key information, and thus avoid safety hazards caused by information overload or improper display.

[0039] In some embodiments, the multi-source monitoring data also includes environmental monitoring data. Environmental monitoring data refers to data describing the external driving environment conditions of the vehicle. Its function is to provide real-time external environmental context information for head-up display adjustments to cope with complex and changing driving scenarios. Environmental monitoring data can be acquired in various ways. For example, it can utilize onboard sensors, such as visible light cameras, infrared cameras, lidar, millimeter-wave radar, ultrasonic sensors, rain sensors, light sensors, and temperature sensors, to collect real-time information on road conditions, weather conditions, light intensity, and obstacles; or it can be acquired from external information sources, such as receiving environmental information from roadside units or other vehicles via vehicle-to-everything (V2X) communication, or obtaining real-time weather forecasts and traffic congestion information through network interfaces.

[0040] Based on multi-source monitoring data, the adjustment constraint parameters for the head-up display are calculated, including: Step c1: Based on the preset spatiotemporal weight prediction model, the spatiotemporal weight parameters are obtained according to the vehicle trajectory prediction results, visual load index and environmental monitoring data.

[0041] The pre-defined spatiotemporal weight prediction model is a model used to dynamically assess and allocate the importance of different information sources (such as vehicle dynamics, driver status, and environmental conditions) in a specific spatiotemporal context. The spatiotemporal weight parameters are numerical values ​​or indicators output by this model, used to quantify the relative impact or priority on head-up display (HUD) adjustments. Methods for obtaining spatiotemporal weight parameters can include, but are not limited to: training a machine learning model, such as a deep neural network or support vector machine, using vehicle trajectory prediction results, visual load index, and environmental monitoring data as input, and outputting comprehensive spatiotemporal weight parameters; or establishing a fuzzy logic system based on expert experience or predefined rules to dynamically generate corresponding spatiotemporal weight parameters according to different combinations and thresholds of input data.

[0042] The pre-defined spatiotemporal weighted prediction model includes: P = α·E + β·V + η·T + δ·S + ε·P_f, where E is the environmental adaptability coefficient, used to quantify the degree of challenge or adaptability required for the presentation of head-up display information in the current driving environment. For example, the E value will increase in environments such as strong light, rain, fog, or tunnels, indicating a need for stronger environmental adaptability adjustment. V is the vehicle trajectory prediction result, which represents the vehicle's movement trend and path over a future period of time, such as predicting that the vehicle is about to make a sharp turn or change lanes. T is the time urgency coefficient, used to measure the time sensitivity of specific information or operations. For example, the T value will increase when approaching a navigation command point or when emergency avoidance is required. S is the spatial importance coefficient, used to assess the criticality of information in its current or future spatial location. For example, the S value will increase when approaching complex intersections or areas with potential collision risks. P_f is the driver's visual load index, which reflects the driver's current cognitive and visual burden level. α, β, η, δ, and ε are all weighting coefficients, used to adjust the relative contribution and influence of each factor (E, V, T, S, P_f) on the final spatiotemporal weight parameter P in the model. The weighting coefficients can be pre-set fixed values, or they can be dynamically adjusted or optimized based on driver preferences, historical data, or real-time driving scenarios.

[0043] In some embodiments, calculating the adjustment constraint parameters for the head-up display based on multi-source monitoring data further includes: Step d1: Based on the modified compensation model, the adjustment constraint parameters are modified and compensated according to environmental monitoring data and historical monitoring data to obtain the modified constraint parameters.

[0044] The correction and compensation process addresses the issue of inaccurate environmental monitoring data due to interference in extreme scenarios such as tunnel entry / exit and rainy / foggy weather, thereby improving the robustness and reliability of adjusting constraint parameters. Specifically, correction and compensation can be implemented in the following ways: One approach is for the system to monitor output data from environmental sensors (e.g., light sensors, rain sensors, fog sensors) in real time and compare and calibrate it with pre-stored historical environmental data under similar environmental conditions or geographical locations. When there is a significant deviation between real-time sensor data and historical data, or when the confidence level reported by the sensor itself is low, the system will activate the correction and compensation mechanism to adjust the current environmental monitoring data. Another approach is to fuse environmental data from different types of sensors (e.g., visible light sensors, infrared sensors, millimeter-wave radar) and use data fusion algorithms (e.g., Kalman filtering, Bayesian estimation) to comprehensively evaluate environmental parameters. When data from a single sensor is abnormal or unreliable, the system can rely on information from other sensors for cross-validation and compensation to obtain more accurate environmental parameters.

[0045] The corrected compensation model includes: L = Σ(C_i·L_i) / ΣC_i, where L is the estimated light intensity, L_i is the light intensity value from the i-th information source, and C_i is the confidence factor of the i-th information source. This model obtains a more accurate and reliable light intensity estimate L by weighting the light intensity values ​​L_i from different information sources and using the confidence factor C_i of each source as the weight. The weighted averaging mechanism effectively avoids the bias or error that may exist with a single data source, enhancing the accuracy of the light intensity estimate. The confidence factor C_i can be determined in several ways: one method is to dynamically evaluate and set it based on the type of information source, the sensor's calibration status, current environmental conditions (e.g., whether the sensor is blocked, or whether it is in a strong or weak light interference environment), and historical data performance. For example, in sunny and well-lit environments, the confidence level of the in-vehicle light sensor might be set high; while in rainy or foggy weather, or when the sensor may be obstructed, the confidence level of the external light sensor might decrease accordingly. In such cases, the confidence level of light information from auxiliary sources such as navigation maps or weather forecasts might be relatively higher. Another approach is to use multiple sources for L_i, such as: in-vehicle ambient light sensors, external ambient light sensors, light information obtained through image analysis from the vehicle's forward-facing camera, light information provided by the navigation system based on location and time, and ambient light data from surrounding vehicles or infrastructure obtained through vehicle-to-everything (V2X) communication. By integrating multi-source information and weighting it according to its respective confidence level, a more comprehensive and accurate estimate of light intensity can be obtained.

[0046] It is understood that the technical solution in this embodiment introduces a correction and compensation mechanism when calculating the adjustment constraint parameters for the head-up display. Given that environmental monitoring data is susceptible to interference in extreme scenarios, leading to inaccurate adjustment constraint parameters, this application improves the robustness and accuracy of the adjustment constraint parameters by using a correction and compensation model based on environmental monitoring data and historical monitoring data. Specifically, the correction and compensation model L=Σ(C_i·L_i) / ΣC_i, by weighting the light intensity values ​​L_i from different information sources using a confidence factor C_i, comprehensively utilizes multi-source information and dynamically adjusts their weights according to the reliability of each information source, thereby obtaining a more accurate light intensity estimate L. This avoids the potential bias of a single data source and ensures that the obtained correction constraint parameters accurately reflect the actual situation even in complex, variable, or extreme environments. Based on the revised adjustment constraint parameters, the subsequent head-up display adjustment strategy will be more accurate and adaptable, thereby improving the adaptive adjustment capability and user experience of the head-up display system. Especially in scenarios with drastic changes in lighting or low visibility, it can provide a more stable and clearer display effect, ensuring driving safety.

[0047] In some embodiments, the adjustment strategy for the head-up display is obtained based on the adjustment constraint parameters, including: Step d1: Based on the Q-learning model, adjust the brightness, content, color, and font of the head-up display according to the correction constraint parameters. The Q-learning model includes: Q(s, a) ← Q(s, a) + λ[R + γ·max_a'Q(s', a') - Q(s, a)], where s is the state space, S = {scene type, driver state, environmental conditions}, a is the action space, a = {font adjustment, brightness adjustment, contrast adjustment, information density adjustment}, R is the reward function, R = ·U + ω_2·C - ω_3·D, where U is the negative correlation function of user manual intervention frequency, C is the information recognition accuracy, D is the duration of attention distraction, ω_1, ω_2, and ω_3 are weight parameters, γ is the discount factor, and λ is the learning rate.

[0048] Among them, the Q-based learning model refers to using a model-free reinforcement learning algorithm to guide the head-up display system in making policy decisions. This model learns an action-value function Q(s, a) to evaluate the expected cumulative reward obtained by performing action a in a given state s, thereby finding the optimal head-up display adjustment strategy under the goal of maximizing long-term rewards. Its implementation can include: one approach is offline training, where a Q-table or Q-network is pre-trained in a simulated environment and then deployed to the actual in-vehicle system; the other approach is online learning, where the system continuously updates the Q-value based on real-time driver feedback and environmental changes during actual operation, achieving dynamic optimization and adaptation of the strategy.

[0049] The corrected constraint parameters are adjusted constraint parameters after being modified using environmental monitoring data and historical monitoring data. They provide the Q-learning model with a reliable basis for adjusting the head-up display in the current driving scenario. The parameters integrate various information such as vehicle trajectory prediction results, driver visual load index, and environmental monitoring data, and are optimized through a correction and compensation model to improve its accuracy and robustness. For example, the corrected constraint parameters may include corrected light intensity estimates and corrected driver fatigue levels, providing more accurate state input to the Q-learning model and ensuring the accuracy of policy generation.

[0050] Adjusting the brightness, content, color, and font of the head-up display (HUD) refers to the system's multi-dimensional and refined control of the HUD's visual presentation based on the actions output by the Q-learning model. Brightness adjustment ensures information is clearly visible under different lighting conditions, such as increasing brightness in strong light and decreasing it in low light. Content adjustment dynamically adjusts the priority and presentation of information based on driving scenarios, such as highlighting navigation information at complex intersections and simplifying the display on highways. Color adjustment uses different colors to distinguish information types or importance, such as using red for hazard warnings and white for general information. Font adjustment adjusts font size and thickness according to the driver's visual load or preferences to improve readability.

[0051] The update rule Q(s, a)←Q(s, a)+λ[R+γmax_a'Q(s', a')-Q(s, a)] in the Q-learning model is the core of the Q-learning algorithm, used to iteratively update the Q-value of state-action pairs. Here, Q(s, a) represents the expected cumulative reward obtained by performing action a in state s. R is the immediate reward, γ is the discount factor, max_a'Q(s', a') represents the maximum Q-value among all possible actions a' in the next state s', and λ is the learning rate. This formula gradually converges to the optimal Q-value by weighting the current Q-value with an estimate based on the immediate reward and the maximum expected future reward. Its implementation can include: one approach is based on a Q-table, storing the Q-values ​​of all state-action pairs in a table for direct lookup and update; another approach is based on a Deep Q-Network (DQN), which uses a neural network to approximate the Q-function when the state space or action space is too large, updating the Q-value by training the neural network.

[0052] Here, s represents the state space, s = {scene type, driver state, environmental conditions}, defining all relevant information about the current environment perceived and understood by the Q-learning model. Scene type can include urban roads, highways, tunnels, intersections, etc., identified through onboard sensors (such as GPS, map data, and cameras); driver state can include fatigue level, attention level, and mood, acquired through driver monitoring systems (such as eye tracking and facial recognition); environmental conditions can include light intensity, weather conditions (rain, fog, snow), and traffic density, acquired through environmental sensors (such as light sensors, radar, and cameras). This state information collectively constitutes the basis for the model's decision-making.

[0053] Here, 'a' represents the action space, a = {font adjustment, brightness adjustment, contrast adjustment, information density adjustment}, defining all discrete or continuous adjustment operations that the Q-learning model can perform. Font adjustment can include increasing / decreasing font size or changing font style; brightness adjustment can include increasing / decreasing the display brightness percentage; contrast adjustment can include increasing / decreasing the display contrast; and information density adjustment can include increasing / decreasing the amount or level of detail of displayed information. Actions are the specific means by which the system adaptively adjusts the head-up display. By selecting different combinations of actions, the display effect is optimized.

[0054] Here, R is the reward function, R = ·U + ω_2·C - ω_3·D, which quantifies the degree of good or bad the agent receives after performing a certain action. This reward function comprehensively considers user experience, information recognition efficiency, and driving safety. Its implementation can include: one method is real-time calculation, where the system calculates the reward value in real time based on user feedback, driver behavior data, and vehicle status data after each adjustment; another method is offline evaluation, where the effects of different adjustment strategies are evaluated through user surveys, simulation tests, etc., and used as a reference for the reward function. U is a negative correlation function of the frequency of user manual intervention, designed to measure user satisfaction with the system's automatic adjustment strategy. When users frequently manually adjust the head-up display, it indicates that the system's current automatic adjustment strategy does not meet user expectations, resulting in a lower U value (or a larger negative value), thus giving the model a negative reward and prompting it to learn a better strategy. Conversely, if users rarely intervene manually, it indicates that the system adjustment is appropriate, resulting in a higher U value and giving the model a positive reward. The implementation methods can include: one method is to record the number of manual operations by the user, statistically analyzing the frequency of manual adjustments to brightness, content, etc., within a certain time window; another method is to record the magnitude of manual operations, not only counting the number of operations but also considering the magnitude of each adjustment, with a larger magnitude indicating a stronger negative correlation. C represents the information recognition accuracy rate, used to evaluate the effectiveness of the information presented by the head-up display. High information recognition accuracy means that the driver can quickly and accurately understand the displayed content, thereby improving driving efficiency and safety. The implementation methods can include: one method is based on eye tracking, using driver eye tracking devices to monitor the driver's gaze time and gaze point distribution for key information to determine whether the information is effectively recognized; another method is based on task completion rate, evaluating whether the driver can correctly perform driving tasks (such as turning on time or avoiding obstacles) based on the head-up display information in specific driving scenarios. D represents the duration of attention distraction, used to measure whether the head-up display causes driver distraction. A longer duration of attention distraction indicates that the head-up display may have design flaws or be improperly adjusted, thus posing a potential risk to driving safety. The implementation methods can include: one method is based on driver facial recognition, using a camera to monitor the driver's head posture and eye movements to determine if they have been deviating from the road for an extended period, or if they are looking at a head-up display or other non-driving-related areas; another method is based on physiological signals, combining heart rate, EEG, and other physiological signals to assist in judging the driver's attention state. ω_1, ω_2, and ω_3 are all weight parameters used to balance the importance of different components in the reward function. By adjusting the weights, the Q-learning model can be optimized to focus more on user satisfaction, information recognition accuracy, or driving safety. For example, in scenarios emphasizing safety, the weight of ω_3 can be appropriately increased to penalize inattention; in scenarios emphasizing user experience, the weight of ω_1 can be increased.The weight parameters can be set by expert experience, or they can be learned through offline optimization or online adaptive adjustment algorithms.

[0055] Here, γ is the discount factor, a value between 0 and 1, used to measure the importance of future rewards. A γ value closer to 1 indicates that the model prioritizes long-term rewards; a γ value closer to 0 indicates that the model prioritizes immediate rewards. λ is the learning rate, a value between 0 and 1, used to control the step size of Q-value updates. A larger λ value leads to faster Q-value updates but may be unstable; a smaller λ value leads to slower Q-value updates but is more stable. An appropriate learning rate ensures that the Q-learning model effectively converges to the optimal policy within a finite number of interactions.

[0056] It is understandable that the technical solution in this embodiment, by introducing a Q-learning model, achieves adaptive optimization of the head-up display adjustment strategy, solving the problems of personalized adaptation limitations and low human-computer interaction efficiency caused by traditional fixed strategies. Specifically, the Q-learning model utilizes the trial-and-error mechanism of reinforcement learning to continuously learn and optimize the strategy based on real-time feedback, thereby adapting to changes in the driver's personal habits and the needs of different driving scenarios, avoiding the rigidity of traditional fixed strategies. Using modified constraint parameters as input to the Q-learning model ensures that strategy generation is based on reliable data corrected by environmental monitoring and historical data, significantly improving the accuracy and robustness of adjustments, especially in extreme environments or complex traffic conditions. Multi-dimensional adjustments are made to the head-up display's brightness, content, color, and font, avoiding the shortcomings of single-dimensional adjustments, enhancing the adaptability and readability of information presentation, and enabling drivers to clearly and comfortably obtain information under different conditions. The state space S integrates multi-source data such as scene type, driver state, and environmental conditions, achieving deep perception of the driving environment and driver state, which helps to coordinate information guidance in time and space, such as automatically reducing information density when the driver is fatigued, or highlighting key navigation information at complex intersections. The action space 'a' defines specific executable actions such as font adjustment, brightness adjustment, contrast adjustment, and information density adjustment, enabling the system to flexibly respond to different driving needs. The reward function R, by combining the negative correlation function U (frequency of user manual intervention), information recognition accuracy C, and attention distraction duration D, supplemented by weight parameters ω_1, ω_2, and ω_3, constructs an optimization objective that comprehensively considers user satisfaction, information effectiveness, and driving safety. This allows the system to autonomously learn driver preferences, dynamically optimize and adjust strategies, and improve driving safety, human-computer interaction efficiency, and user experience. The introduction of the discount factor γ and learning rate λ further ensures that the model can stably and efficiently learn the optimal strategy, achieving continuous evolution and personalized adaptation of the head-up display system.

[0057] In some embodiments, this application provides a method for adjusting a head-up display, such as... Figure 2 As shown, the method includes: Step S201: Obtain multi-source monitoring data. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0058] Step S202: Calculate the adjustment constraint parameters for the head-up display based on the multi-source monitoring data. For details, please refer to [link to relevant documentation]. Figure 1 Step S102 of the illustrated embodiment will not be described again here.

[0059] Step S203: Based on the adjustment constraint parameters, obtain the adjustment strategy for the head-up display. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0060] Step S204: Adjust the head-up display based on the adjustment strategy. For details, please refer to [link / reference]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0061] Step S205: Determine a reminder strategy based on the duration the driver gazes at the head-up display.

[0062] Specifically, the duration of a driver's gaze at the head-up display (HUD) refers to the length of time the driver's gaze remains on the HUD area. This duration is a key indicator for assessing the driver's attention allocation, used to determine whether the driver is overly focused on the HUD and neglecting road conditions ahead. A warning strategy is then determined, which involves pre-setting a series of tiered intervention measures based on this gaze duration. This can be achieved through an in-vehicle eye-tracking system, such as using an infrared camera with image recognition algorithms, to capture the driver's eye movement trajectory in real time and calculate the time their gaze lingers within the HUD area. Alternatively, the duration of the driver's gaze at the HUD can be indirectly estimated by analyzing the driver's head posture and facial orientation, combined with a pre-defined gaze model.

[0063] Based on the duration of the driver's gaze at the head-up display, a reminder strategy is determined, which specifically includes: Step e1: When the driver's gaze duration on the head-up display is greater than or equal to the first duration and less than the second duration, control the head-up display to provide a visual reminder.

[0064] Specifically, when the driver's gaze at the head-up display (HUD) is greater than or equal to the first duration but less than the second duration, it indicates that the driver's attention is beginning to wander, at which point the system will trigger the first level of alert. Controlling the HUD for visual alerts aims to attract the driver's attention in a gentle, non-intrusive way through changes in the HUD itself. For example, the HUD content can flicker slightly, the color can change from the normal display color to a soft warning color, or a small warning icon that does not obscure the main information can appear at the edge of the display area. The system can also briefly adjust the brightness or contrast of the HUD to visually alert the driver without causing discomfort.

[0065] Step e2: When the driver's gaze duration on the head-up display is greater than or equal to the second duration and less than the third duration, control the head-up display to provide a visual reminder and control the vehicle to provide an auditory reminder.

[0066] Specifically, when the driver's gaze duration on the head-up display further increases, exceeding the second duration but falling short of the third, it indicates a deepening of driver inattention and necessitates stronger intervention. At this point, while continuing the aforementioned visual alerts, the system will activate an auditory alert. Controlling the vehicle's auditory alerts involves playing preset warning sounds through the vehicle's audio system, such as a short beep, or a brief voice prompt, such as "Please be aware of the road ahead." The volume and frequency of the auditory alerts can adaptively adjust based on the ambient noise level inside the vehicle or the driver's hearing habits to ensure the effectiveness of the alert.

[0067] Step e3: When the driver gazes at the head-up display for a duration greater than or equal to the third duration, control the head-up display to provide a visual reminder, and simultaneously control the vehicle to provide an auditory and tactile reminder.

[0068] Specifically, when the driver's gaze at the head-up display reaches or exceeds the third time, it indicates that the driver's attention is severely distracted, posing a high safety risk. At this point, the system will activate the highest level of multi-sensory collaborative alert, adding a tactile alert to the visual and auditory alerts. Controlling the vehicle to provide tactile alerts aims to forcibly awaken the driver through physical sensation. For example, the vehicle seat may vibrate slightly, or the steering wheel may provide brief haptic feedback. Additionally, the seatbelt pretensioning system can also provide a slight pretension for a short period, alerting the driver through the feeling of restraint. The intensity, frequency, or pretensioning force of the tactile alert can be dynamically adjusted based on factors such as vehicle speed and the driver's physiological state to achieve the best warning effect.

[0069] It is understood that the technical solution in this embodiment addresses the problem that drivers may become distracted by excessive focus on the displayed content after adjusting the head-up display (HUD), thereby increasing safety risks. Specifically, by monitoring the duration of the driver's gaze at the HUD in real time, the system can dynamically assess the driver's attention state and trigger visual, auditory, and tactile alerts in a tiered manner based on different gaze duration intervals. When the driver's attention begins to wander, a gentle visual alert is first provided through the HUD to guide the driver with minimal interference. As the degree of distraction deepens, the system gradually increases the auditory alerts to enhance the warning intensity. When the driver's attention is severely distracted, a multi-sensory coordinated visual, auditory, and tactile alert is activated to form a strong comprehensive stimulus, forcibly waking the driver and ensuring that they refocus their attention on the driving task. This tiered, multi-modal alert mechanism not only optimizes the closed-loop feedback of human-computer interaction, enabling the system to intervene precisely according to the driver's actual state, but also significantly improves driving safety, effectively avoiding potential dangers caused by excessive focus on the HUD, thus ensuring driver safety while providing rich information.

[0070] In one example, a specific case will be used to illustrate the adjustment method of the head-up display in more detail: In a real-world application scenario, a vehicle is equipped with a head-up display system. When user A is driving on a city road and is about to enter a complex intersection (location A),

[0071] First, the system continuously acquires multi-source monitoring data, including: navigation data such as the vehicle's current location, preset route, type of intersection ahead, turning direction, and speed limit information.

[0072] Vehicle status data: real-time speed, acceleration, steering angular velocity, braking status, etc.

[0073] Driver status monitoring data: Through in-vehicle cameras and sensors, the system monitors user A's eye gaze direction, blink frequency, head posture, etc. in real time to assess their visual load index. For example, when user A's eyes frequently scan the interior of the vehicle, or their blink frequency is abnormal, the system determines that their visual load is high.

[0074] Environmental monitoring data: external light intensity, weather conditions (e.g., the system detects that the vehicle is about to enter a tunnel, or that the external light suddenly increases / decreases), information on obstacles ahead, etc.

[0075] Historical monitoring data: Stored user A's past driving habits, head-up display adjustment preferences, and system adjustment records in similar scenarios.

[0076] Next, the system calculates the adjustment constraint parameters for the head-up display based on the multi-source monitoring data. The specific process is as follows: Vehicle trajectory prediction: Based on a preset prediction model, combined with navigation data and vehicle status data, the system predicts the vehicle's trajectory over the next few seconds, obtaining the vehicle trajectory prediction result. For example, it predicts that the vehicle will enter the left-turn lane in 3 seconds at its current speed and steering angle. The prediction model is X_{k|k-1}=f(X_{k-1},u_k)+w_k.

[0077] Visual load index acquisition: Based on driver status monitoring data, the system obtains the current visual load index of user A.

[0078] Spatiotemporal weight parameter calculation: The system comprehensively considers vehicle trajectory prediction results, visual load index, and environmental monitoring data to calculate spatiotemporal weight parameters. For example, in a scenario where a user is about to turn left and has a high visual load, the system will assign a higher spatial importance coefficient and a higher time urgency coefficient to the left-turn prompt information. The preset spatiotemporal weight prediction model is P=α·E+β·V+η·T+δ·S+ε·P_f.

[0079] Correction and Compensation: Based on environmental monitoring data and historical monitoring data, the system corrects and compensates for the aforementioned adjustment constraint parameters to obtain corrected constraint parameters. For example, if the ambient light intensity drops sharply, the system will quickly and accurately correct the estimated light intensity value based on historical and current environmental data to avoid lag in display brightness adjustment due to sensor delays or errors. The correction and compensation model is L=Σ(C_i·L_i) / ΣC_i.

[0080] Subsequently, based on the corrected constraint parameters, the system derives an adjustment strategy for the head-up display. Utilizing a Q-learning model, the system adjusts the brightness, content, color, and font of the head-up display according to the corrected constraint parameters. For example, in the scenario described above, where a user is about to turn left and experiences high visual load, the system might decide: Display brightness: Appropriately reduce the overall brightness to avoid excessive brightness and stimulation, while ensuring that key information is clearly visible.

[0081] Display content: Highlight the left turn arrow and distance information, and temporarily hide non-critical entertainment information.

[0082] Display color: Adjust the color of the left turn arrow to a high-contrast green to enhance visibility.

[0083] Display font: Increase the font size of the left turn prompt to make it easier to recognize quickly. The Q-learning model is Q(s,a)←Q(s,a)+λ[R+γ·max_a'Q(s',a')-Q(s,a)].

[0084] Finally, the system adjusted the head-up display based on this strategy. User A observed that the left-turn prompts on the head-up display became more prominent, with larger fonts, brighter colors, and moderate brightness. This enabled User A to quickly obtain key navigation information at complex intersections, reducing the time spent searching for information and improving driving safety.

[0085] It is understood that the technical solution in this application, by integrating multi-source information such as navigation data, vehicle status data, driver status monitoring data, environmental monitoring data, and historical data, and combining predictive models, spatiotemporal weight models, correction and compensation models, and Q-learning models, achieves multi-dimensional, adaptive, and personalized adjustment of the head-up display's brightness, content, color, and font. This solves the problems of existing systems, such as limited adjustment dimensions, fragmented human-computer interaction, lack of spatiotemporal coordination, insufficient adaptability to extreme scenarios, and limitations in personalized adaptation.

[0086] After the head-up display is adjusted, the system will continue to monitor the duration of user A's gaze on the head-up display and intervene according to preset reminder strategies. For example: When user A gazes at the head-up display for a duration greater than or equal to the first duration (e.g., 3 seconds) and less than the second duration (e.g., 5 seconds), the system will control the head-up display to provide visual reminders, such as by flashing or changing color to attract user A's attention.

[0087] When user A gazes at the head-up display for a duration greater than or equal to the second duration but less than the third duration (e.g., 8 seconds), the system will not only control the head-up display to provide visual alerts but also control the vehicle to provide auditory alerts, such as emitting a soft prompting sound to remind user A to pay attention to the road conditions ahead.

[0088] When user A gazes at the head-up display for a duration equal to or greater than the third duration, the system will control the head-up display to provide visual alerts, while simultaneously controlling the vehicle to provide auditory and tactile alerts, such as slight seat vibrations, to more strongly alert user A in a multimodal way and ensure that their attention returns to the driving task.

[0089] The above technical solutions further improve the safety and efficiency of human-computer interaction, and prevent drivers from being distracted from the road ahead due to excessive attention to the head-up display.

[0090] In one example, as the vehicle approaches the tunnel exit, the duration of the visual adaptation lag (e.g., 1-3 seconds) is dynamically predicted based on the difference in illumination inside and outside the tunnel, the duration of tunnel exposure, and the driver's individual adaptation speed. Simultaneously, the potential impact on the driver's visual perception during this lag is assessed.

[0091] Specifically, when the vehicle has just exited the tunnel and entered the initial stage of the visual adaptation lag period, the HUD will temporarily hide all non-critical information (such as vehicle speed, fuel consumption, etc.) and only display the core and urgent single navigation command in the current scenario, such as a bright, flashing "turn right" arrow with a halo effect, precisely anchored on the target lane.

[0092] The command text will be slightly enlarged and feature a high-contrast yellow-black color scheme. Simultaneously, the system will trigger a very short, slight, high-frequency (e.g., 5-10Hz) visual flicker, below the frequency perceived by the driver, but sufficient to stimulate retinal cells, accompanied by a slight tactile pulse in the corresponding direction of the steering wheel. This multi-sensory, coordinated instantaneous stimulation ensures that critical information is efficiently captured by the driver even during moments of impaired visual perception.

[0093] As the visual adaptation lag progresses, the system, based on instructions from the prediction module, will gradually and smoothly increase the brightness and contrast of the HUD display over the next 1-2 seconds, rather than abruptly changing it, to avoid secondary stimulation to the driver's eyes. Simultaneously, the system will gradually introduce secondary key information (such as distance to the exit and lane markings) and gradually reduce the flickering effect of core instructions, restoring them to normal high-brightness display.

[0094] Once the predicted visual adaptation lag period ends, the HUD will revert to non-tunnel display mode, providing rich information adapted to the current complex road conditions.

[0095] It is understood that the technical solution in this embodiment avoids the driver's eyes from being subjected to a secondary impact due to sudden changes in the brightness or amount of information of the HUD display, allowing the human eye to complete visual adaptation in a gentler and more gradual process, thereby improving the comfort and safety of the entire transition period.

[0096] This embodiment also provides a head-up display adjustment device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0097] This embodiment provides a head-up display adjustment device, such as... Figure 3 As shown, it includes: Module 301 is used to acquire multi-source monitoring data; Calculation module 302 is used to calculate the adjustment constraint parameters of the head-up display based on multi-source monitoring data; Strategy module 303 is used to obtain the adjustment strategy for the head-up display based on the adjustment constraint parameters; The adjustment module 304 is used to adjust the head-up display based on the adjustment strategy.

[0098] The head-up display adjustment device provided in this application embodiment can execute the head-up display adjustment method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0099] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0100] The following is a detailed reference. Figure 4 This diagram illustrates a suitable structural schematic for implementing the electronic device described in the embodiments of this application. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 401, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 402 or a program loaded from memory 408 into random access memory (RAM) 403. RAM 403 also stores various programs and data required for the operation of the electronic device. The processor 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0101] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0102] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a memory 408, or installed from a ROM 402. When the computer program is executed by the processor 401, it performs the functions defined in the head-up display adjustment method of embodiments of this application.

[0103] Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0104] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the head-up display adjustment method shown in the above embodiments is implemented.

[0105] A portion of this application can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0106] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method of adjusting a head-up display, characterized by, The method comprises: acquiring multi-source monitoring data; calculating adjustment constraint parameters of the head-up display according to the multi-source monitoring data; obtaining an adjustment strategy of the head-up display based on the adjustment constraint parameters; adjusting the head-up display based on the adjustment strategy.

2. The method of claim 1, wherein, The multi-source monitoring data comprises navigation data and vehicle state data, The calculation of the adjustment constraint parameters of the head-up display according to the multi-source monitoring data comprises: predicting a vehicle trajectory based on a preset prediction model according to the navigation data and the vehicle state data to obtain a vehicle trajectory prediction result, wherein the preset prediction model comprises: X_{k|k-1}=f(X_{k-1},u_k)+w_k, wherein k is a time index, X is a state vector, X=[x,y,v,a,θ,ω]^T, x and y are vehicle positions, v is a vehicle speed, a is an acceleration, ω is a heading angle, ω is a steering angle speed, X_{k-1} is a last time state estimation, u_k is a control input vector, X_{k|k-1} is a current time state prediction, and w_k is a process noise vector.

3. The method of claim 2, wherein, The multi-source monitoring data further comprises driver state monitoring data, The calculation of the adjustment constraint parameters of the head-up display according to the multi-source monitoring data further comprises: obtaining a visual load index of the driver according to the driver state monitoring data.

4. The method of claim 3, wherein, The multi-source monitoring data further comprises environmental monitoring data, The calculation of the adjustment constraint parameters of the head-up display according to the multi-source monitoring data further comprises: obtaining a spatio-temporal weight parameter based on a preset spatio-temporal weight prediction model according to the vehicle trajectory prediction result, the visual load index, and the environmental monitoring data, wherein the preset spatio-temporal weight prediction model comprises: P=α·E+β·V+η·T+δ·S+ε·P_f, wherein E is an environmental adaptability coefficient, V is the vehicle trajectory prediction result, T is a time urgency coefficient, S is a spatial importance coefficient, P_f is the visual load index of the driver, and α, β, η, δ, and ε are weight coefficients.

5. The method according to any of claims 2-4, characterized by, The calculation of the adjustment constraint parameters of the head-up display according to the multi-source monitoring data further comprises: correcting and compensating the adjustment constraint parameters based on a correction and compensation model according to the environmental monitoring data and historical monitoring data to obtain a corrected constraint parameter, wherein the correction and compensation model comprises: L=Σ(C_i·L_i) / ΣC_i, wherein L is an estimated value of light intensity, L_i is a light intensity value of the i-th information source, and C_i is a confidence factor of the i-th information source.

6. The method of claim 5, wherein, The obtaining of the adjustment strategy of the head-up display based on the adjustment constraint parameters comprises: adjusting display brightness, display content, display color, and display font of the head-up display based on a Q learning model according to the corrected constraint parameter, wherein the Q learning model comprises: Q(s, a)←Q(s, a)+λ[R+γ·max_a'Q(s', a')-Q(s, a)], s is a state space, s={scene type, driver state, environmental condition}, a is an action space, a={font adjustment, brightness adjustment, contrast adjustment, information density adjustment}, R is a reward function, R=·U+ω_2·C-ω_3·D, U is a negative correlation function of user manual intervention frequency, C is information recognition accuracy, D is attention distraction duration, ω_1, ω_2 and ω_3 are weight parameters, γ is a discount factor, and λ is a learning rate.

7. The method of claim 5, wherein, After adjusting the head-up display based on the adjustment strategy, the method further comprises: determining a reminder strategy based on the length of time that the driver gazes at the head-up display; controlling the head-up display to visually remind when the length of time that the driver gazes at the head-up display is greater than or equal to a first length of time and less than a second length of time; controlling the head-up display to visually remind and controlling the vehicle to audibly remind when the length of time that the driver gazes at the head-up display is greater than or equal to the second length of time and less than a third length of time; controlling the head-up display to visually remind, and controlling the vehicle to audibly remind and haptically remind when the length of time that the driver gazes at the head-up display is greater than or equal to the third length of time.

8. An adjustment device for head-up display, characterized by The device comprises: an acquisition module configured to acquire multi-source monitoring data; a calculation module configured to calculate adjustment constraint parameters of the head-up display based on the multi-source monitoring data; a strategy module configured to obtain an adjustment strategy of the head-up display based on the adjustment constraint parameters; and an adjustment module configured to adjust the head-up display based on the adjustment strategy.

9. An electronic device, comprising: comprise: a memory and a processor, which are communicatively connected, and the memory stores computer instructions, and the processor executes the computer instructions to perform the head-up display adjustment method of any one of claims 1 to 7.

10. A computer program product, characterised in that, comprise computer instructions for causing a computer to perform the head-up display adjustment method of any one of claims 1 to 7.

Images