Driving risk interaction prompting method, intelligent device and readable storage medium

Abstract

The invention relates to the technical field of driving, in particular to a driving risk interaction prompting method, intelligent equipment and a readable storage medium, and aims to solve the technical problem of how to ensure that a driver can timely, accurately and effectively receive a risk prompt of a driving risk scene. In order to achieve the purpose, a driving risk scene of the intelligent equipment is obtained based on environment perception information of the intelligent equipment; obtaining a visual focus prediction result of the driver based on the driver state of the driver of the intelligent equipment; obtaining a channel availability evaluation result of at least one prompt channel of the intelligent equipment based on the prompt channel state of the intelligent equipment; according to a visual focus prediction result and a channel availability evaluation result, risk prompting is performed on a driving risk scene, so that the situation that the risk prompting is ignored in an unconcerned area of a driver is effectively avoided, and the risk prompting can be adaptively, timely, effectively and accurately perceived and transmitted by the driver; therefore, the driving safety of the intelligent equipment is effectively improved.

Description

Technical Field

[0001] The present application relates to the field of driving technology, and in particular to a driving risk interactive prompting method, an intelligent device, and a readable storage medium. Background Art

[0002] As driving tasks become increasingly complex, the complexity of interactions between vehicles and drivers continues to rise, whether in manual or assisted driving modes. Especially in critical scenarios such as takeover requests, path deviations, and sudden obstacles ahead, it is necessary to ensure that drivers can receive prompt information in a timely and accurate manner.

[0003] However, most traditional prompt systems use fixed area displays or single prompt channels, such as HUD, voice broadcasts, instrument prompts, etc., which may be ignored when the driver's attention is distracted or his vision is diverted. The failure of the prompt will indirectly lead to safety hazards.

[0004] Accordingly, this field requires a new driving risk interactive prompting solution to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problem of how to ensure that drivers can receive risk warnings of driving risk scenarios in a timely, accurate and effective manner.

[0006] In a first aspect, a driving risk interactive prompting method is provided, which is applied to a smart device; the method comprises:

[0007] Obtaining a driving risk scenario of the smart device based on environmental perception information;

[0008] obtaining a prediction result of the driver's visual focus based on the driver's driving state;

[0009] Based on the prompt channel status of the smart device, obtaining a channel availability evaluation result of at least one prompt channel of the smart device;

[0010] A risk warning is provided for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result.

[0011] In one technical solution of the above-mentioned driving risk interactive prompting method, the driving risk scenario includes a risk event and a risk level of the risk event;

[0012] Providing a risk warning for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result includes:

[0013] For each prompt channel, obtaining the channel prompt priority of the risk event in the prompt channel according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event;

[0014] At least one prompt channel for providing risk prompts is selected according to the channel prompt priority, and risk prompts are provided for the driving risk scenario based on the selected prompt channel.

[0015] In one technical solution of the above-mentioned driving risk interactive prompt method, obtaining the channel prompt priority of the risk event in the prompt channel according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event includes:

[0016] acquiring a channel prompt priority of the prompt channel according to the visual focus prediction result, the channel availability assessment result, the risk level, and the channel response preference data of the prompt channel;

[0017] The channel response preference data is data obtained based on the driver's historical behavioral responses to the risk prompts of the prompt channel at historical moments.

[0018] In one technical solution of the above-mentioned driving risk interactive prompt method, it is characterized in that:

[0019] The channel response preference data is obtained based on the historical behavioral responses based on a preset neural network model;

[0020] The method further comprises:

[0021] The driver's current behavioral response to the risk prompt is collected for updating the neural network model.

[0022] In one technical solution of the above-mentioned driving risk interactive prompt method, selecting a prompt channel for providing risk prompts according to the channel prompt priority includes:

[0023] At least one prompt channel is selected for risk prompting according to the channel prompt priority and a preset first priority threshold.

[0024] In one technical solution of the above-mentioned driving risk interactive prompting method, the risk prompting includes:

[0025] Determining the risk prompt intensity and / or prompt duration and / or prompt frequency of the prompt channel according to the channel prompt priority;

[0026] The risk prompt of the prompt channel is performed according to the prompt intensity and / or the prompt duration and / or the displayed prompt frequency.

[0027] In one technical solution of the above-mentioned driving risk interactive prompting method, the risk prompting includes:

[0028] Determining whether the risk event requires an external interaction prompt of the smart device according to the channel prompt priority;

[0029] For risk events that require external interactive prompts, provide external interactive prompts on the smart device;

[0030] The external interaction prompt is a way of providing risk prompts for environmental obstacles outside the smart device.

[0031] In one technical solution of the above-mentioned driving risk interactive prompt method, the method further includes:

[0032] If the driver does not respond to the risk prompt with a behavioral response, the fallback prompt mode of the smart device is activated; the prompt intensity of the fallback prompt mode is greater than the prompt intensity of the risk prompt according to the channel prompt priority.

[0033] In one technical solution of the above-mentioned driving risk interactive prompt method, the method further includes:

[0034] If the driver does not respond to the backup prompt mode, the assisted driving mode of the smart device is activated.

[0035] In one technical solution of the above-mentioned driving risk interactive prompting method, the driver status includes the driver's eye movement analysis results and head posture analysis results;

[0036] The obtaining of the driver's visual focus prediction result based on the driver's state includes:

[0037] Obtaining a heat distribution result of the driver's attention according to the eye movement analysis result and the head posture analysis result;

[0038] The visual focus prediction result is obtained according to the attention heat distribution result.

[0039] In one technical solution of the above-mentioned driving risk interactive prompt method, obtaining a channel availability evaluation result of at least one prompt channel of the smart device based on the prompt channel status of the smart device includes:

[0040] For each prompt channel, obtaining an availability score of the prompt channel according to the prompt channel status;

[0041] An availability evaluation result of the prompt channel is obtained according to the availability score and the preset weight of the prompt channel.

[0042] In one technical solution of the above-mentioned driving risk interactive prompt method, obtaining the driving risk scenario of the smart device based on environmental perception information includes:

[0043] According to the environmental perception information, a risk event of the current environment of the smart device and a risk level of the risk event are obtained as the driving risk scenario.

[0044] In one technical solution of the above-mentioned driving risk interactive prompt method, the method further includes obtaining the risk level according to the following steps:

[0045] The risk level is obtained according to at least one parameter of a collision time, a dynamic relative distance, a path overlap, and a behavior intention conflict score between the environmental obstacle corresponding to the risk event and the smart device.

[0046] In one technical solution of the above-mentioned driving risk interactive prompt method, the prompt channel includes at least one of the voice prompt channel of the smart device, the central control screen prompt channel, the head-up display device prompt channel, the ambient light prompt channel, the seat vibration prompt channel, the steering wheel vibration prompt channel, and the seat belt tightening prompt channel.

[0047] In a second aspect, a smart device is provided, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and when the computer program is executed by the at least one processor, the method described in any one of the technical solutions of the above-mentioned driving risk interactive prompt method is implemented.

[0048] In a third aspect, a computer-readable storage medium is provided, which stores a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the method described in any one of the technical solutions of the above-mentioned driving risk interactive prompt method.

[0049] Solution 1. A driving risk interactive prompting method, characterized in that the method is applied to a smart device; the method comprises:

[0050] Obtaining a driving risk scenario of the smart device based on environmental perception information;

[0051] obtaining a prediction result of the driver's visual focus based on the driver's driving state;

[0052] Based on the prompt channel status of the smart device, obtaining a channel availability evaluation result of at least one prompt channel of the smart device;

[0053] A risk warning is provided for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result.

[0054] Solution 2. The driving risk interactive prompt method according to Solution 1, wherein the driving risk scenario includes a risk event and a risk level of the risk event;

[0055] Providing a risk warning for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result includes:

[0056] For each prompt channel, obtaining the channel prompt priority of the risk event in the prompt channel according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event;

[0057] At least one prompt channel for providing risk prompts is selected according to the channel prompt priority, and risk prompts are provided for the driving risk scenario based on the selected prompt channel.

[0058] Solution 3. The driving risk interactive prompt method according to Solution 2 is characterized in that:

[0059] The acquiring, according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event, the channel prompt priority of the risk event in the prompt channel includes:

[0060] acquiring a channel prompt priority of the prompt channel according to the visual focus prediction result, the channel availability assessment result, the risk level, and the channel response preference data of the prompt channel;

[0061] The channel response preference data is data obtained based on the driver's historical behavioral responses to the risk prompts of the prompt channel at historical moments.

[0062] Solution 4. The driving risk interactive prompting method according to Solution 3 is characterized in that:

[0063] The channel response preference data is obtained based on the historical behavioral responses based on a preset neural network model;

[0064] The method further comprises:

[0065] The driver's current behavioral response to the risk prompt is collected for updating the neural network model.

[0066] Solution 5. The driving risk interactive prompting method according to Solution 2 is characterized in that:

[0067] The selecting a prompt channel for providing risk prompts according to the channel prompt priority includes:

[0068] At least one prompt channel is selected for risk prompting according to the channel prompt priority and a preset first priority threshold.

[0069] Solution 6. The driving risk interactive prompting method according to Solution 2 is characterized in that:

[0070] The risk warnings include:

[0071] Determining the risk prompt intensity and / or prompt duration and / or prompt frequency of the prompt channel according to the channel prompt priority;

[0072] The risk prompt of the prompt channel is performed according to the prompt intensity and / or the prompt duration and / or the displayed prompt frequency.

[0073] Solution 7. The driving risk interactive prompting method according to Solution 2 is characterized in that:

[0074] The risk warnings include:

[0075] Determining whether the risk event requires an external interaction prompt of the smart device according to the channel prompt priority;

[0076] For risk events that require external interactive prompts, provide external interactive prompts on the smart device;

[0077] The external interaction prompt is a way of providing risk prompts for environmental obstacles outside the smart device.

[0078] Solution 8. The driving risk interactive prompting method according to Solution 2 is characterized in that:

[0079] The method further comprises:

[0080] If the driver does not respond to the risk prompt with a behavioral response, the fallback prompt mode of the smart device is activated; the prompt intensity of the fallback prompt mode is greater than the prompt intensity of the risk prompt according to the channel prompt priority.

[0081] Solution 9. The driving risk interactive prompting method according to Solution 8 is characterized in that:

[0082] The method further comprises:

[0083] If the driver does not respond to the backup prompt mode, the assisted driving mode of the smart device is activated.

[0084] Solution 10. The driving risk interactive prompting method according to Solution 1, wherein the driver status includes an eye movement analysis result and a head posture analysis result of the driver;

[0085] The obtaining of the driver's visual focus prediction result based on the driver's state includes:

[0086] Obtaining a heat distribution result of the driver's attention according to the eye movement analysis result and the head posture analysis result;

[0087] The visual focus prediction result is obtained according to the attention heat distribution result.

[0088] Solution 11. The driving risk interactive prompting method according to Solution 1 is characterized in that:

[0089] The obtaining, based on the prompt channel status of the smart device, a channel availability evaluation result of at least one prompt channel of the smart device includes:

[0090] For each prompt channel, obtaining an availability score of the prompt channel according to the prompt channel status;

[0091] An availability evaluation result of the prompt channel is obtained according to the availability score and the preset weight of the prompt channel.

[0092] Solution 12. The driving risk interactive prompt method according to Solution 1 is characterized in that:

[0093] The acquiring of the driving risk scenario of the smart device based on the environmental perception information includes:

[0094] According to the environmental perception information, a risk event of the current environment of the smart device and a risk level of the risk event are obtained as the driving risk scenario.

[0095] Solution 13. The driving risk interactive prompting method according to Solution 12 is characterized in that:

[0096] The method further comprises obtaining the risk level according to the following steps:

[0097] The risk level is obtained according to at least one parameter of a collision time, a dynamic relative distance, a path overlap, and a behavior intention conflict score between the environmental obstacle corresponding to the risk event and the smart device.

[0098] Solution 14. The driving risk interactive prompting method according to any one of Solutions 1 to 13, characterized in that:

[0099] The prompt channel includes at least one of the voice prompt channel of the smart device, the central control screen prompt channel, the head-up display device prompt channel, the ambient light prompt channel, the seat vibration prompt channel, the steering wheel vibration prompt channel, and the seat belt tightening prompt channel.

[0100] Solution 15. A smart device, comprising:

[0101] at least one processor;

[0102] and, a memory communicatively coupled to the at least one processor;

[0103] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the driving risk interactive prompt method described in any one of Schemes 1 to 14 is implemented.

[0104] Solution 16. A computer-readable storage medium storing a plurality of program codes, characterized in that the program codes are suitable for being loaded and run by a processor to execute the driving risk interactive prompt method described in any one of Solutions 1 to 14.

[0105] The above one or more technical solutions of this application have at least one or more of the following beneficial effects:

[0106] In implementing the technical solution of the interactive driving risk prompt method provided by this application, this application obtains the driving risk scenario of the smart device based on environmental perception information; obtains the driver's visual focus prediction result based on the driver's driving status; obtains the channel availability evaluation result of at least one prompt channel of the smart device based on the prompt channel status of the smart device; and provides risk prompts for the driving risk scenario based on the visual focus prediction result and the channel availability evaluation result. Through the above configuration, this application can implement risk prompts for driving risk scenarios by combining the driver's visual focus prediction result and the channel availability evaluation result of the prompt channel, effectively avoiding the situation where the risk prompt is ignored in the area that the driver does not pay attention to, and can also avoid the situation where the risk prompt cannot be conveyed due to failure or obstruction of the prompt channel, etc., and can ensure that the risk prompt can be adaptively perceived and conveyed by the driver in a timely, effective and accurate manner, thereby effectively improving the driving safety of the smart device.

[0107] Furthermore, the present application can combine the driver's visual focus prediction results, the channel availability assessment results of the prompt channel, the risk level of the risk event and the channel response preference data of the prompt channel to determine the channel prompt priority of the prompt channel, and determine at least one prompt channel for risk prompts based on the channel prompt priority. It can realize risk prompts based on multiple prompt channels, further improve the effectiveness of the prompt channel, and effectively avoid safety hazards caused by the failure of prompts from a single prompt channel.

[0108] Furthermore, the present application can determine the prompt intensity and / or prompt duration of the risk prompt of the prompt channel according to the channel prompt priority, and can realize adaptive adjustment of the prompt intensity and / or prompt duration according to the channel prompt priority.

[0109] Furthermore, the present application can determine whether a risk event requires an external interaction prompt of the smart device based on the channel prompt priority. For risk events with a higher channel prompt priority, the external interaction prompt of the smart device is triggered to express the current status of the smart device and the next intention to the pedestrians, non-motor vehicles, motor vehicles, etc. in the environment where the smart device is located through the external interaction prompt, so as to effectively improve the friendliness and explainability of the external interaction prompt of the smart device. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] The disclosure of this application will become more easily understood with reference to the accompanying drawings. Those skilled in the art will readily appreciate that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Among them:

[0111] Figure 1 This is a flowchart of the main steps of the driving risk interactive prompt method according to an embodiment of the present application;

[0112] Figure 2 This is a schematic diagram of the main components of a method for implementing interactive driving risk prompts according to an embodiment of the present application;

[0113] Figure 3 is a schematic diagram of an application scenario according to an embodiment of the present application;

[0114] Figure 4 is a schematic diagram of another application scenario according to an embodiment of the present application;

[0115] Figure 5 This is a scenario diagram of the third application scenario according to an embodiment of the present application. DETAILED DESCRIPTION

[0116] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the scope of protection of the present application.

[0117] In the description of this application, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories, and may also include software components, such as program code, or a combination of software and hardware. The term "A and / or B" represents all possible combinations of A and B, such as just A, just B, or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B" and may include just A, just B, or A and B. The singular terms "a" and "the" may also include plural forms.

[0118] The relevant user personal information that may be involved in the various embodiments of this application is strictly in accordance with the requirements of laws and regulations, following the principles of legality, legitimacy and necessity, and based on the reasonable purposes of business scenarios, to process the personal information that users actively provide during the use of products / services or generated due to the use of products / services, as well as the personal information obtained with the user's authorization.

[0119] The user personal information processed by this application will vary depending on the specific product / service scenario and must be based on the specific scenario in which the user uses the product / service. This may involve the user's account information, device information, driving information, vehicle information, or other related information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0120] This application attaches great importance to the security of user personal information and has taken reasonable and feasible security protection measures that comply with industry standards to protect user information and prevent personal information from being accessed, disclosed, used, modified, damaged or lost without authorization.

[0121] See attached Figure 1 , Figure 1 This is a flow chart of the main steps of the driving risk interactive prompt method according to an embodiment of the present application. Figure 1 As shown, the driving risk interactive prompt method in the embodiment of the present application is applied to a smart device, and the driving risk interactive prompt method in the embodiment of the present application mainly includes the following steps S101 to S104.

[0122] Step S101: Acquire driving risk scenarios of smart devices based on environmental perception information.

[0123] In this embodiment, the environment in which the smart device is located can be sensed by an environment sensor provided on the smart device to obtain environment perception information of the smart device. Based on the environment perception information, the driving risk scenario currently existing in the smart device can be determined.

[0124] In one embodiment, environmental perception sensors may include cameras, millimeter-wave radars, lidars, GPS (Global Positioning System), inertial measurement units (IMUs), and V2X (Vehicle to Everything). Environmental perception can be performed based on one or more of these sensors to build a dynamic environmental model, enabling real-time object (environmental obstacle) recognition, trajectory prediction, and driving scenario recognition in the smart device's environment.

[0125] In one implementation, environmental perception information can be used to obtain risk events and their risk levels in the smart device's current environment, thereby determining the driving risk scenario for the smart device. Specifically, the smart device's environmental perception information and road structure information can be combined to predict the device's own behavior and that of surrounding obstacles. This allows for the creation of an interactive risk map, which quantitatively assesses the device's current potential driving risk scenarios, as well as the risk events and risk levels within those scenarios.

[0126] In a specific example, risk events can be divided into five risk levels (R1-R5) according to the interaction urgency of risk events in driving risk scenarios, specifically:

[0127] R5: Emergency physical conflict (e.g., the risk event is the vehicle in front suddenly stops or a pedestrian crosses the road)

[0128] R4: High potential interference (e.g., risk events such as vehicles approaching from a blind spot or complex intersections)

[0129] R3: Medium-risk operation (e.g., risk events include a neighboring vehicle changing lanes or an oncoming vehicle deviating)

[0130] R2: Attention risk (e.g., risk events include driver distraction, closing eyes, or vision deviation)

[0131] R1: Low risk warning (e.g., risk event is slight speeding, front vehicle brakes remotely).

[0132] In one embodiment, the risk level of a risk event can be obtained based on at least one parameter of the collision time (TTC, Time-to-Collision), dynamic relative distance (DTC, Distance To Collision), path overlap, and behavioral intention conflict score between the environmental obstacle and the smart device corresponding to the risk event. The collision time refers to the time required for the smart device to collide with the obstacle in front. The dynamic relative distance indicates the minimum distance between the predicted collision point of the target vehicle relative to the main vehicle and the rear of the main vehicle in the current motion state. The path overlap refers to the degree of path overlap between the smart device and the environmental obstacle. The behavioral intention conflict score is an evaluation score of the behavioral conflict between the smart device and the environmental obstacle, which is determined based on the behavioral prediction results of the smart device and the behavioral prediction results of the environmental obstacle. After comprehensively evaluating the risk event based on the above parameters, the risk level of the risk event and the confidence level of the risk level can be obtained.

[0133] For a specific example, see Table 1 below, which shows an example table of risk event judgment criteria and risk levels. As shown in Table 1, the risk level of a risk event can be determined based on parameters such as collision time, dynamic relative distance, path overlap, and behavioral intention conflict score.

[0134] Table 1 Example of risk event judgment conditions and risk levels Judgment conditions Risk level range TTC≤1.0s, and the object is a pedestrian 5 There is an obstruction ahead, there may be pedestrians crossing, and the vehicle (i.e., smart device) is accelerating. 4 Smash the accelerating merging vehicles in the surrounding lanes 3 Risks are not immediately relevant (e.g., green light not yet taken) 1-2

[0135] It should be noted that the above-mentioned risk level judgment conditions and risk level settings are exemplary, and those skilled in the art can set the judgment conditions and risk levels according to the needs of actual applications.

[0136] In one embodiment, the smart device may be a driving device, a smart vehicle, a robot, or the like.

[0137] Step S102: Based on the driver's driving state, obtain the driver's visual focus prediction result.

[0138] In this embodiment, the driver's visual focus prediction result can be determined according to the driver's state of the smart device.

[0139] In one embodiment, the driver status may include the driver's eye movement analysis results and head posture analysis results. The driver's attention distribution results may be determined based on the driver's eye movement analysis results and head posture analysis results, and the driver's visual focus prediction result may be obtained based on the driver's attention distribution results.

[0140] Specifically, the driver's eye movements can be analyzed to obtain eye movement analysis results, and the driver's head posture can be analyzed to obtain head posture analysis results. The eye movement and head posture analysis results are combined to obtain the driver's attention heat distribution results. Based on the driver's attention heat distribution results, the driver's current attention area and attention blind spot are determined, thereby obtaining a prediction result of the driver's visual focus.

[0141] In one embodiment, the driver's attention heat distribution result can be obtained by combining the driver's eye movement analysis results, head posture analysis results, and the driver's historical behavior.

[0142] In a specific example, the driver's attention areas in the smart device can be divided, such as the HUD, instrument panel, central control screen, road area, interior and exterior rearview mirrors, A-pillars, etc. The driver's attention intensity can be determined based on the following rules:

[0143] If the driver stares at an area for more than 1.5 seconds, then this area can be regarded as the main attention area, that is, the attention intensity is high.

[0144] The attention blind spot is defined as: the area outside the driver's average line of sight angle of ±60° can be considered as the driver's attention blind spot; the area where the angle between the driver's average line of sight angle and the direction of the dangerous event is greater than 90° can be considered as the driver's attention blind spot.

[0145] Among them, those skilled in the art can set the parameters in the above rules to other values ​​according to the needs of actual applications.

[0146] Step S103: Based on the prompt channel status of the smart device, a channel availability evaluation result of at least one prompt channel of the smart device is obtained.

[0147] In this embodiment, the prompt channel status of each prompt channel of the smart device may be detected in real time, and the channel availability evaluation result of the prompt channel of the smart device may be obtained according to the prompt channel status.

[0148] In one embodiment, the prompt channel of the smart device may include at least one of a voice prompt channel, a central control screen prompt channel, a head-up display (HUD) prompt channel, an ambient light prompt channel, a seat vibration prompt channel, a steering wheel vibration prompt channel, and a seat belt tightening prompt channel.

[0149] For example, for the voice prompt channel, the channel availability evaluation result of the voice prompt channel can be obtained based on the volume status of the voice device corresponding to the voice prompt channel, call conflict detection, and background noise assessment (noise input analysis based on the microphone of the smart device).

[0150] For the HUD prompt channel, the channel availability assessment result of the HUD prompt channel can be obtained based on whether the HUD is blocked (e.g., due to sunlight interference) and the display status feedback code. The display status feedback code indicates the status identification code of the current operating status of the HUD device.

[0151] For the ambient light prompt channel, the channel availability assessment result can be obtained based on the ambient light's color signal strength. Color signal strength refers to the brightness value or percentage intensity of the ambient light's current color channel (e.g., red, green, and blue), which is used to control or indicate the actual color saturation and brightness of the light.

[0152] In a specific example, the channel availability evaluation result for each prompt channel can be a value between 0 and 1. For example, if the voice prompt channel has moderate interference, the channel availability evaluation result is 0.6. If the HUD prompt channel is in good condition, the channel availability evaluation result is 0.9.

[0153] In one embodiment, preset weights may be set for different prompt channels. An availability score of each prompt channel may be obtained based on its channel status, and a channel availability evaluation result of the prompt channel may be obtained based on the availability score of the prompt channel and the preset weight.

[0154] Specifically, the product of the availability score of the prompt channel and the preset weight can be used as the channel availability evaluation result of the prompt channel. That is, the channel availability evaluation result = availability score × preset weight.

[0155] Please refer to Table 2 below, which is an example table of channel availability evaluation results. Table 2 exemplifies the channel status, availability score, and preset weight of each prompt channel.

[0156] Table 2 Example of channel availability evaluation results Prompt Channel Usability score Preset weights Exemplary interference conditions that suggest channel status HUD 0-1 0.4 Insufficient HUD brightness in direct sunlight during the day voice 0-1 0.3 Background noise in the cockpit>75dB Steering wheel vibration 0-1 0.2 The driver does not have his hands on the steering wheel Ambient lighting 0-1 0.1 Ambient light brightness offsets atmosphere light

[0157] It should be noted that the above-mentioned criteria for judging the channel availability evaluation results are merely exemplary, and those skilled in the art may adjust the criteria, scoring, and weighting of the channel availability evaluation results according to the needs of actual applications.

[0158] Step S104: Providing risk warnings for driving risk scenarios based on the visual focus prediction results and channel availability assessment results.

[0159] In this embodiment, the driver's visual focus prediction result and the channel availability evaluation result of the prompt channel can be combined to determine the prompt channel for risk prompting, and risk prompts for driving risk scenarios can be performed based on the prompt channel.

[0160] In one embodiment, step S104 may further include the following steps S1041 and S1042:

[0161] Step S1041: For each prompt channel, the channel prompt priority of the risk event in the prompt channel is obtained according to the visual focus prediction result, the channel availability assessment result and the risk level of the risk event.

[0162] In this embodiment, the channel prompt priority of the prompt channel may be calculated by combining the driver's visual focus prediction result, the channel availability evaluation result of the prompt channel, and the risk level of the risk event.

[0163] In one embodiment, channel response preference data of the prompt channel may also be considered when calculating the channel prompt priority of the prompt channel, that is, data is obtained based on the driver's historical behavioral responses to risk prompts of the prompt channel at historical moments.

[0164] In one embodiment, the channel prompt priority can be calculated according to the following formula (1):

[0165] Channel prompt priority = risk level × visual blind spot weight × channel availability assessment result × historical response weighting factor (1)

[0166] The visual blind spot weight is determined based on the driver's visual focus prediction results. The closer the location of the prompt channel is to the driver's attention blind spot in the driver's visual focus prediction results, the larger the visual blind spot weight is. The historical response weighting factor is determined based on the driver's channel response preference data. If the driver has historically paid attention to prompts for the current prompt channel, the historical response weighting factor has a lower value; if the driver has historically ignored prompts for the current prompt channel, the historical response weighting factor has a higher value.

[0167] Specifically, please refer to Table 3 below, which is an example table of judgment basis and weight comparison of visual blind spot weights.

[0168] Table 3 Judgment basis and weight comparison example of visual blind spot weight

[0169] Please refer to Appendix 4, which is an example table of historical response weighting factors and judgment criteria.

[0170] Table 4 Example of historical response weighting factors and judgment criteria

[0171] In a specific example, if the risk level of a risk event is 5, the risk event is located in the driver's visual blind spot, the visual blind spot weight is 1.3, the channel availability evaluation result of the HUD prompt channel is 0.8, the driver's historical response to the HUD is poor, and the historical response weighting factor is 1.1, combined with formula (1), it can be obtained that the HUD channel prompt priority for this risk event is:

[0172] Channel prompt priority: 5×1.3×0.8×1.1=5.72.

[0173] It should be noted that the above-mentioned judgment criteria and numerical settings of the visual blind spot weight and historical response weighting factor are merely exemplary. Those skilled in the art may adjust the judgment criteria and numerical settings of the visual blind spot weight and historical response weighting factor according to the needs of actual applications.

[0174] Step S1042: Select at least one prompt channel for risk prompting according to the channel prompt priority, and provide risk prompting for the driving risk scenario based on the selected prompt channel.

[0175] In this embodiment, the prompt channel for risk prompting can be selected according to the channel prompt priority.

[0176] In one embodiment, the channel prompt priority may be compared with a preset first priority threshold, and a prompt channel for risk prompting may be selected based on the comparison result.

[0177] In a specific example, the first priority threshold may be 4.0. Taking the HUD prompt channel's channel prompt priority of 5.72 calculated above as an example, since the HUD prompt channel's channel prompt priority of 5.72 is greater than the first priority threshold of 4.0, the HUD prompt channel may be selected for risk warning. Those skilled in the art may set the first priority threshold based on actual application needs.

[0178] In one embodiment, if the channel prompt priority is lower than the first priority threshold, the driving risk scenario can be controlled to enter the buffer queue and the risk prompt for the driving risk scenario is temporarily not performed.

[0179] In one embodiment, the prompt intensity, prompt duration and prompt frequency of the risk prompt of the prompt channel can be determined based on the channel prompt priority, and the risk prompt of the prompt channel can be performed according to the determined prompt intensity, prompt duration and prompt frequency. That is, for risk events with higher channel prompt priority, a more intense and more perceptible prompt channel combination can be assigned to make it easier for the driver to perceive the risk event. For example, a prompt channel combination can be a seat / steering wheel vibration + voice + HUD prompt channel combination. At the same time, for risk events with higher channel prompt priority, the prompt duration of the risk prompt can be longer, and the prompt frequency of the risk prompt can be higher, so that the driver can more easily perceive the risk event.

[0180] In one embodiment, it can be determined whether a risk event requires an external interactive prompt of the smart device based on the channel prompt priority. For risk events that require external interactive prompts, external interactive prompts of the smart device are performed. That is, for risk events with a higher channel prompt priority, the external interactive prompt function of the smart device can be triggered. Among them, the external interactive prompt can be a way of providing risk prompts to environmental obstacles such as pedestrians and non-motor vehicles outside the smart device through external voice broadcasts (with adjustable tone), light language, and external graphic icon projections of the smart device (including in multiple locations such as the front, side, rear, and ground of the smart device).

[0181] In a specific example, taking the smart device as a vehicle, the external interaction prompts may include the following combinations:

[0182] If the vehicle is slowing down to avoid the vehicle, the "Please Pass" pattern + blue light bar + external voice will be displayed.

[0183] If the vehicle is making an emergency stop or is not yielding → "Please wait" + red border light + external voice notification will be displayed

[0184] If the assisted driving is taking over → the "system in control" animated icon + external voice will be displayed

[0185] Among them, the prompt channel inside the smart device and the prompt strategy of the external interactive prompt need to be synchronized to avoid conflicts between internal and external risk prompts.

[0186] In one embodiment, the channel prompt priority can be compared with a preset second priority threshold. If the channel prompt priority is greater than the second priority threshold, the external interaction prompt of the smart device can be enabled. Those skilled in the art can set the second priority threshold according to actual application needs.

[0187] In another embodiment, whether to enable external interaction prompts on smart devices can be determined based on the risk level of the risk event. Alternatively, whether to enable external interaction prompts on smart devices can be determined based on the type of environmental obstacle in the risk event. Alternatively, multiple parameters (e.g., channel prompt priority, risk level, or environmental obstacle type) can be combined to determine whether to enable external interaction prompts on smart devices.

[0188] In a specific example, the intensity of the risk prompt and whether to enable the external interactive prompt of the smart device can be determined according to the channel prompt priority. Specifically,

[0189] Prompt intensity - low: Prompt channel 1 + Prompt channel 2 + external interactive prompt (optional)

[0190] Prompt intensity - Medium: Prompt channel 1 + Prompt channel 2 + Prompt channel 3 + External interactive prompt (optional)

[0191] Prompt intensity - high: Prompt channel 1 + Prompt channel 2 + Prompt channel 3 + Emergency prompt channel + External interactive prompt (optional)

[0192] In one embodiment, if the driver does not respond to the risk warning, the smart device activates a fallback warning mode. The fallback warning intensity is greater than the risk warning intensity based on the channel prompt priority. In other words, if the driver does not respond to the risk warning, the risk warning intensity can be increased.

[0193] In one embodiment, if the driver does not respond behaviorally within 3 seconds after the risk prompt, the prompt intensity of the risk prompt can be increased.

[0194] In a specific example, the backup prompt mode can be: starting the seat vibration prompt channel + the ambient light prompt channel light flashing + the alarm sound + the instrument panel prompt channel red frame prompt.

[0195] In one embodiment, if the driver does not respond to the fallback prompt mode, the assisted driving mode of the smart device is activated. That is, if the driver does not respond to the fallback prompt mode after a preset period of time, the assisted driving mode of the smart device can be activated to reduce the risk of risk events. For example, the assisted driving mode can be activated to achieve the deceleration of the smart device, activate the LKA (Lane Keeping Assist) function, and record abnormal events (i.e., prompt channel failure) to facilitate the subsequent processing and optimization of driving risk interaction prompts.

[0196] In one embodiment, the maximum value of the preset duration does not exceed 10 seconds.

[0197] In one embodiment, the channel response preference data is obtained based on historical behavioral responses according to a preset neural network model. The driver's current behavioral responses to risk prompts can be collected for updating the neural network model, and the updated neural network model can be used when subsequently acquiring channel response preference data.

[0198] The neural network model can be deployed on a smart device. The neural network model can also be set up on a server. The smart device communicates with the server. When obtaining channel response preference data, the smart device can communicate with the server to call the neural network model to obtain the channel response preference data.

[0199] In one embodiment, a knowledge graph node can be established for the driver's behavioral response to risk prompts. The node may include driving risk scenario type, driving intention, risk level, prompt channel, driver's driving style (cautious / aggressive), etc. A neural network model is established based on the above knowledge graph nodes to predict the driver's response preference data for risk prompts of different prompt channels. For example, if the driver's behavioral response to the voice prompt channel is faster, then a lower historical response weighting factor can be set for the voice prompt channel. A data closed loop can be formed through the driver's risk prompt data, and the neural network model can be updated based on the driver's behavioral response data each time to update the weight distribution in the neural network model.

[0200] Based on the method described in steps S101 to S104 above, the embodiment of the present application obtains the driving risk scenario of the smart device based on the environmental perception information of the smart device; obtains the visual focus prediction result of the driver based on the driver status of the driver of the smart device; obtains the channel availability evaluation result of at least one prompt channel of the smart device based on the prompt channel status of the smart device; and provides risk prompts for driving risk scenarios based on the visual focus prediction result and the channel availability evaluation result. Through the above configuration, the embodiment of the present application can implement risk prompts for driving risk scenarios by combining the visual focus prediction result of the driver and the channel availability evaluation result of the prompt channel, effectively avoiding the situation where the risk prompt is ignored in the area that the driver does not pay attention to, and can also avoid the situation where the risk prompt cannot be conveyed due to failure or obstruction of the prompt channel, etc., and can ensure that the risk prompt can be adaptively perceived and conveyed by the driver in a timely, effective and accurate manner, thereby effectively improving the driving safety of the smart device.

[0201] Furthermore, the embodiments of the present application can combine the driver's visual focus prediction results, the channel availability assessment results of the prompt channel, the risk level of the risk event, and the channel response preference data of the prompt channel to determine the channel prompt priority of the prompt channel, and determine at least one prompt channel for risk prompts based on the channel prompt priority. It can realize risk prompts based on multiple prompt channels, further improve the effectiveness of the prompt channel, and effectively avoid safety hazards caused by the failure of a single prompt channel.

[0202] Furthermore, the embodiment of the present application can determine the prompt intensity and / or prompt duration of the risk prompt of the prompt channel according to the channel prompt priority, and can realize adaptive adjustment of the prompt intensity and / or prompt duration according to the channel prompt priority.

[0203] Furthermore, the embodiment of the present application can determine whether a risk event requires an external interaction prompt of the smart device based on the channel prompt priority. For risk events with a higher channel prompt priority, the external interaction prompt of the smart device is triggered to express the current status of the smart device and the next intention to the pedestrians, non-motor vehicles, motor vehicles, etc. in the environment where the smart device is located through the external interaction prompt, so as to effectively improve the friendliness and explainability of the external interaction prompt of the smart device.

[0204] In one embodiment, you can participate in the Figure 2 ,like Figure 2 As shown in FIG, the system architecture for implementing the driving risk interactive prompt method may include a multi-source environmental perception system, a visual focus prediction engine, a channel health management subsystem, a driving scenario risk modeling module, a prompt decision and scheduling module, a prompt measurement knowledge graph engine, a prompt space migration mechanism, a smart device multimodal prompt interface, and a smart device external interactive behavior expression module.

[0205] In this embodiment, the multi-source environmental perception system can integrate information from sensors such as cameras, millimeter-wave radars, lidars, GPS, IMUs, and vehicle-to-everything (V2X) to build a dynamic environmental model, obtain environmental perception information, and support real-time object recognition, trajectory prediction, and driving scenario recognition.

[0206] The driving scenario risk modeling module can establish an interaction risk map between smart devices and obstacles in the environment based on environmental perception information, road structure and behavior prediction results, and quantitatively evaluate the potential risk scenarios of smart devices (i.e., driving risk scenarios), such as pedestrian-vehicle intersection, blind spot lane change, and intersections.

[0207] The visual focus prediction engine can predict the driver's current attention area and attention blind spot through eye movement analysis and head posture analysis, using multimodal input and historical behavior modeling, as the driver's visual focus prediction result.

[0208] The channel health management subsystem can dynamically detect the operating status and historical stability of each prompt channel (voice, central control screen, HUD, ambient light, seat, steering wheel vibration, seat belt tightening, etc.), and generate channel availability assessment results for the prompt channel to assist in risk prompt scheduling.

[0209] The prompt strategy knowledge graph engine constructs a mapping diagram of driving risk scenarios, risk events, and prompt channels. It integrates parameters such as the driver's driving style (conservative / aggressive) and interaction feedback to generate channel response preference data. This engine supports adaptive strategy generation and enables remote OTA (Over-the-Air) updates.

[0210] The prompt strategy and scheduling module is responsible for comprehensively analyzing the current situation based on the received risk events, the driver's visual focus prediction results, the channel availability assessment results of each prompt channel, and other information, and deciding whether to issue prompts for risk events, how to issue prompts, and in which area (prompt channel) to issue prompts, and controlling the prompt intensity and frequency.

[0211] The prompt space migration mechanism can migrate risk prompts to the HUD, central control screen, and instrument panel based on the driver's attention, and give priority to prompt channels corresponding to positions around the driver's current attention projection area for risk prompts, thereby increasing the probability of perceiving risk prompts.

[0212] The multimodal prompt interface of smart devices can control the multimodal prompt effects of multiple prompt channels such as voice broadcast (adjustable tone), graphic icons (dynamic adjustment of visual brightness), seat vibration (adjustable frequency level), ambient light (color / flashing), steering wheel vibration (adjustable frequency level), and seat belt tightening through a unified prompt middle platform API (Application Programming Interface).

[0213] The external interactive behavior expression module of the smart device can combine the status of the smart device to present visual symbols such as "courtesy", "passable", and "attention" on the external display screen or ground projection of the smart device to assist external communication, and can also play corresponding voice prompts to the outside.

[0214] The following combination Figures 3 to 5 , taking the smart device as a smart car (ego) as an example, three different practical application scenarios of the embodiments of the present application are described.

[0215] The first practical application scenario, such as Figure 3As shown in the figure, the driving risk scenario is that at an intersection, the vehicle turns right, the vehicle in the straight lane (RU) blocks the view, and a pedestrian crosses the zebra crossing from the left side of the vehicle. The pedestrian is in the left front blind spot of the driver of the vehicle. The driver is observing the pedestrian and vehicle on the right and does not pay attention to the crosswalk on the left front.

[0216] By integrating the vehicle's camera to detect pedestrian trajectories and modeling the occlusion position, the risk level of the risk event in the driving risk scenario is determined to be 5;

[0217] Identifying the driver's driving state and determining the driver's visual focus prediction result as follows: the driver's attention area is the right front direction, and the left front direction is the attention blind spot;

[0218] Based on the risk level, visual focus prediction results and other parameters, the risk warning intensity is judged to be high;

[0219] The determined risk warning strategy is: Prompt channel 1: a flashing red frame prompt in the right area of ​​the cockpit (central control screen / co-pilot screen) + Prompt channel 2: a highlighted voice "there is a pedestrian in front of the left" + Prompt channel 3: right side ambient light + Emergency prompt channel: slight steering wheel vibration + outside vehicle channel (external interactive prompt): "Be careful to avoid" is projected on the ground, and a prompt voice outside the vehicle is played to enhance pedestrian perception.

[0220] The second practical application scenario, such as Figure 4 As shown in the figure, the driving risk scenario is that the vehicle is driving in the middle lane of a three-lane highway with sparse traffic around it. The driver looks straight ahead for a long time without moving his head, and the vehicle begins to deviate to the right and approaches the right lane line.

[0221] Based on the detection that the vehicle is drifting without turning on the lights and the speed of drifting in the center of the lane is fast, the risk level of the risk event in the driving risk scenario is determined to be 4.

[0222] Eye movement analysis detected that the driver lowered his head and his eye movement frequency decreased, indicating that the driver's fatigue index increased.

[0223] The HUD prompt channel availability is low (interference caused by strong light during the day), so choose the prompt channel combination of seat vibration + voice prompt.

[0224] Based on parameters such as the driver's visual prediction results and risk level, the prompt intensity is determined to be medium.

[0225] The confirmed risk warning strategy is: Prompt channel 1: flashing red frame prompt in the area in front of the driver's seat (instrument screen) + Prompt channel 2: highlighted voice "The vehicle has deviated from the lane, please pay attention to safety" + Prompt channel 3: continuous vibration on the right side of the seat.

[0226] The third practical application scenario, such as Figure 5As shown, the driving risk scenario is that assisted driving (L2) is enabled on a congested road. Through environmental perception, it is detected that there is a construction area ahead and needs to be taken over. The driver lowers his head to check his mobile phone and there is no sign of taking over.

[0227] Based on environmental perception information, the risk level of the risk event in the driving risk scenario is determined to be 2.

[0228] The eye movement analysis results showed that the driver was looking down and not paying attention, and was judged to be distracted driving.

[0229] If the HUD prompt channel availability is low (interference from strong light during the day) and the driver is distracted, select the prompt channel combination of seat vibration and voice prompt.

[0230] The confirmed risk warning strategy is: Prompt channel 1: flashing red frame prompt in the area in front of the driver's seat (instrument screen) + Prompt channel 2: highlighted voice "Please take over the vehicle in time."

[0231] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of the present application, different steps do not have to be performed in such an order. They can be performed simultaneously (in parallel) or in other orders. These adjusted solutions are equivalent to the technical solutions described in this application, and therefore will also fall within the scope of protection of this application.

[0232] It will be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above-mentioned various method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0233] Another aspect of the present application provides a computer-readable storage medium.

[0234] In an embodiment of a computer-readable storage medium according to the present application, the computer-readable storage medium can be configured to store a program for executing the driving risk interactive prompt method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned driving risk interactive prompt method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method section of the embodiment of the present application. The computer-readable storage medium can be a storage device formed by various electronic devices, such as a magnetic disk, hard disk, optical disk, flash memory, read-only memory, random access memory, etc. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-transitory computer-readable storage medium.

[0235] Another aspect of the present application provides a smart device.

[0236] In an embodiment of an intelligent device according to the present application, the intelligent device may include at least one processor; and a memory in communication with the at least one processor; wherein the memory stores a computer program that, when executed by the at least one processor, implements the method described in any of the above embodiments. The intelligent device described in the present application may include a driving device, a smart car, a robot, and other devices.

[0237] In some embodiments of the present application, the smart device may further include at least one sensor for sensing information. The sensor is communicatively connected to any type of processor mentioned in this application. Optionally, the smart device may further include an assisted driving system for guiding the smart device to complete assisted driving. The processor communicates with the sensor and / or assisted driving system to complete the method described in any of the above embodiments. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware, or a combination of the two.

[0238] Thus far, the technical solution of the present application has been described in conjunction with an embodiment shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A driving risk interactive prompting method, characterized in that: The method is applied to a smart device; the method comprises: Obtaining a driving risk scenario of the smart device based on environmental perception information; obtaining a prediction result of the driver's visual focus based on the driver's driving state; Based on the prompt channel status of the smart device, obtaining a channel availability evaluation result of at least one prompt channel of the smart device; A risk warning is provided for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result.

2. The driving risk interactive prompting method according to claim 1, characterized in that: The driving risk scenario includes a risk event and a risk level of the risk event; Providing a risk warning for the driving risk scenario based on the visual focus prediction result and the channel availability assessment result includes: For each prompt channel, obtaining the channel prompt priority of the risk event in the prompt channel according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event; At least one prompt channel for providing risk prompts is selected according to the channel prompt priority, and risk prompts are provided for the driving risk scenario based on the selected prompt channel.

3. The driving risk interactive prompting method according to claim 2, characterized in that: The acquiring, according to the visual focus prediction result, the channel availability assessment result, and the risk level of the risk event, the channel prompt priority of the risk event in the prompt channel includes: acquiring a channel prompt priority of the prompt channel according to the visual focus prediction result, the channel availability assessment result, the risk level, and the channel response preference data of the prompt channel; The channel response preference data is data obtained based on the driver's historical behavioral responses to the risk prompts of the prompt channel at historical moments.

4. The driving risk interactive prompting method according to claim 3, characterized in that: The channel response preference data is obtained based on the historical behavioral responses based on a preset neural network model; The method further comprises: The driver's current behavioral response to the risk prompt is collected for updating the neural network model.

5. The driving risk interactive prompting method according to claim 2, characterized in that: The selecting a prompt channel for providing risk prompts according to the channel prompt priority includes: At least one prompt channel is selected for risk prompting according to the channel prompt priority and a preset first priority threshold.

6. The driving risk interactive prompting method according to claim 2, characterized in that: The risk warnings include: Determining the risk prompt intensity and / or prompt duration and / or prompt frequency of the prompt channel according to the channel prompt priority; The risk prompt of the prompt channel is performed according to the prompt intensity and / or the prompt duration and / or the displayed prompt frequency.

7. The driving risk interactive prompting method according to claim 2, characterized in that: The risk warnings include: Determining whether the risk event requires an external interaction prompt of the smart device according to the channel prompt priority; For risk events that require external interactive prompts, provide external interactive prompts on the smart device; The external interaction prompt is a way of providing risk prompts for environmental obstacles outside the smart device.

8. The driving risk interactive prompting method according to claim 2, characterized in that: The method further comprises: If the driver does not respond to the risk prompt with a behavioral response, the fallback prompt mode of the smart device is activated; the prompt intensity of the fallback prompt mode is greater than the prompt intensity of the risk prompt according to the channel prompt priority.

9. The driving risk interactive prompting method according to claim 8, characterized in that: The method further comprises: If the driver does not respond to the backup prompt mode, the assisted driving mode of the smart device is activated.

10. The driving risk interactive prompting method according to claim 1, characterized in that: The driver status includes an eye movement analysis result and a head posture analysis result of the driver; The obtaining of the driver's visual focus prediction result based on the driver's state includes: Obtaining a heat distribution result of the driver's attention according to the eye movement analysis result and the head posture analysis result; The visual focus prediction result is obtained according to the attention heat distribution result.

Images