Vehicle early warning control method and system for pedestrians
By detecting pedestrian movement data in real time and assessing collision risks, and dynamically adjusting vehicle light projection, the problem of information lag and insufficient scene adaptability in vehicle-pedestrian interaction is solved, achieving accurate early warning of pedestrian risks and improving driving and pedestrian traffic safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-23
- Publication Date
- 2026-04-07
AI Technical Summary
Existing vehicle lighting systems cannot anticipate risks when interacting with pedestrians, information transmission is delayed and one-way, scene adaptability is insufficient, and the coordination accuracy between ADAS systems and vehicle lighting control terminals is lacking, resulting in low pedestrian safety.
By detecting target pedestrians in real time and collecting their movement data, the system uses a preset algorithm to determine the collision risk and configures the vehicle headlight projection parameters based on the risk index and environmental characteristics to provide forward warnings. This includes a multi-dimensional assessment of distance, time, behavioral intent, and environmental correction index, and dynamically adjusts the projection content in conjunction with the characteristics of the vehicle's surrounding environment and weather conditions.
It enables real-time and accurate identification of pedestrian collision risks, reduces the probability of collisions in human-vehicle interaction scenarios, and improves the safety of driving and pedestrian traffic.
Smart Images
Figure CN121799285A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle technology, and in particular to a vehicle warning and control method and system for pedestrians. Background Technology
[0002] Existing passenger vehicle lighting technology has evolved from simple illumination to intelligent interaction, forming two main forms: pixelated lighting and graphic projection. It can display basic information such as turn arrows, reversing warnings, and navigation guidance. Some high-end products already support full-color streaming media projection and preliminary linkage with ADAS systems.
[0003] However, existing technologies have significant pain points in interaction scenarios involving vulnerable road users (pedestrians, cyclists, etc.):
[0004] 1. Delayed and one-way information transmission: Existing vehicle lights can only respond to actions that have already been performed by the vehicle (such as projecting arrows when turning), and cannot predict risks in advance or convey the vehicle's subsequent intentions, resulting in a lack of reaction time for vulnerable participants.
[0005] 2. Insufficient scene adaptability: The system does not have dedicated interaction logic designed for high-risk scenarios such as low visibility (rain, fog, night) and complex intersections (no traffic lights or zebra crossings), and the information conveyed by relying on a single symbol is easily overlooked.
[0006] 3. Lack of coordination accuracy: There are cross-domain coordination barriers between the ADAS system and the vehicle lighting control terminal. There is a lack of special detection algorithms for weak participants, which makes it impossible to accurately identify their movement trajectory and dynamically adjust the interaction content. Summary of the Invention
[0007] The purpose of this invention is to provide a vehicle warning and control method and system for pedestrians, solving the technical problems of low collision probability and low traffic safety for both vehicles and pedestrians in existing human-vehicle interaction scenarios. The specific solution is as follows:
[0008] A vehicle warning and control method for pedestrians, the method comprising the following steps:
[0009] S1: Real-time detection of whether there are target pedestrians within a preset range of the target vehicle;
[0010] S2: If a target pedestrian is present, collect the pedestrian's movement data; the movement data includes at least: pedestrian position, movement speed, and movement direction;
[0011] S3: Based on the acquired movement data, the first preset algorithm is used to determine whether there is a collision risk to the pedestrian;
[0012] S4: If there is a risk of collision, the second preset algorithm is used to obtain the vehicle's projection parameters in order to warn the target pedestrian in advance.
[0013] Optionally, step S3, based on the acquired movement data, uses a first preset algorithm to determine whether there is a collision risk with the pedestrian, specifically including:
[0014] Based on the acquired mobile data, a risk index is obtained according to preset rules; wherein, the risk index includes at least: distance index, time index, behavioral intention index, and environmental correction index pre-configured based on weather characteristics;
[0015] Based on the obtained risk index, the first formula is used to obtain the target risk index of the target pedestrian.
[0016] Based on the comparison results between the target risk index and the index threshold, it is determined whether there is a collision risk to the target pedestrian;
[0017] The distance index is configured to be obtained based on a comparison between a first distance between the target pedestrian and the target vehicle and a preset interval;
[0018] The time index is configured to be obtained by comparing the time difference between the target pedestrian and the target vehicle at the intersection of the trajectory with the corresponding time interval.
[0019] The behavioral intent index is configured to be obtained based on the collected behavioral characteristics of the target pedestrian; wherein, the behavioral characteristics are key features of the pedestrian's intention to cross towards a vehicle or road;
[0020] The behavioral characteristics include at least: the pedestrian's body orientation, the pedestrian's pausing behavior, and the pedestrian's walking behavior.
[0021] Optionally, the step of obtaining the target risk index of the target pedestrian using a first formula based on the acquired risk index specifically includes:
[0022] The first formula is as follows:
[0023] Target Risk Index = Distance Index + Time Index + Behavioral Intent Index + Environmental Correction Index;
[0024] Correspondingly, if the target risk index is less than or equal to the index threshold, then it is determined that there is no risk of collision with the target pedestrian.
[0025] If the target risk index is greater than the index threshold, then the target pedestrian is determined to be at risk of collision.
[0026] Optionally, step S4 specifically includes:
[0027] S401: Obtain environmental features within a preset range; the environmental features include at least: intersection type, road type, and road width;
[0028] S402: Based on the acquired environmental features, input the pre-trained scene classification model and output the target scene of the target pedestrian;
[0029] S403: Based on the target scene of the target pedestrian, the acquired weather features, and the target risk index, obtain the target projection parameters corresponding to the target scene by looking up the scene and projection mapping table; wherein, the weather features include at least: visibility and light intensity;
[0030] S404: Based on the acquired target projection parameters, control the vehicle headlight projection module to project onto a preset part of the target vehicle to warn the target pedestrian in advance; the target projection parameters include at least: flashing frequency, color intensity, marking pattern, marking pattern size and visual complexity of the marking pattern; wherein, the marking pattern includes at least: warning signs, distance signs and guidance signs.
[0031] Optionally, step S4 further includes:
[0032] The facial image of the target pedestrian is acquired and input into the facial recognition model to obtain key facial features; the key facial features include at least: facial contour features, hair features, eye features, and wrinkle features;
[0033] Based on the acquired key facial features, the data is input into a pre-trained age classification model, which outputs the age category of the target pedestrian. The age category includes at least: children, the elderly, and middle-aged people. The age classification model is pre-trained based on key visual features of facial age.
[0034] Based on the age level of the target pedestrian, a corresponding projection strategy is adopted to control the projection of the vehicle headlights projection module.
[0035] Optionally, the step of controlling the vehicle headlight projection module to project according to the age level of the target pedestrian using a corresponding projection strategy specifically includes:
[0036] If the target pedestrian's age category is child, the corresponding projection auxiliary parameters are retrieved and transmitted to the vehicle headlight projection module to perform auxiliary projection; the projection auxiliary parameters include at least: warning cartoon pattern, number of cartoon patterns, display rhythm and display speed of the cartoon pattern;
[0037] If the target pedestrian's age category is elderly, then the corresponding projection auxiliary parameters are retrieved, and the visual complexity of the target projection parameter's sign pattern is adjusted to the lowest level. Based on the projection auxiliary parameters, the color vibrancy is adjusted to a high level, and the color saturation is adjusted to a low level. The projection auxiliary parameters include at least: the vibrancy and color saturation of the sign pattern.
[0038] If the target pedestrian's age category is middle-aged, then maintain the target projection parameters and control the vehicle headlight projection module to perform projection.
[0039] A vehicle warning and control system for pedestrians, the system comprising:
[0040] The data acquisition module is configured to detect in real time whether there are target pedestrians within a preset range of the target vehicle;
[0041] The first judgment module is configured to collect the movement data of the target pedestrian if the target pedestrian exists; the movement data includes at least: pedestrian position, movement speed and movement direction;
[0042] The second judgment module is configured to determine whether there is a collision risk to the pedestrian based on the acquired movement data and using the first preset algorithm.
[0043] The processing module is configured to use a second preset algorithm to obtain the vehicle's projection parameters if there is a collision risk, so as to warn the target pedestrian in advance.
[0044] An electronic device includes: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method.
[0045] A computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method described herein.
[0046] A simulation platform, comprising:
[0047] An electronic device for implementing the steps of the method described herein;
[0048] A processor that runs a program, and when the program runs, it executes the steps of the method from data output by the electronic device.
[0049] A storage medium for storing a program that, when run, executes the steps of the method on data output from an electronic device.
[0050] The above solution achieves the following beneficial technical effects:
[0051] This application provides a vehicle warning and control method and system for pedestrians. By detecting target pedestrians within a preset range in real time, and collecting movement data such as pedestrian position, speed, and direction, a first preset algorithm is applied to determine collision risk. When a collision risk is detected, a second preset algorithm is used to obtain projection parameters and execute a forward warning. This achieves real-time and accurate identification of pedestrian collision risks. Furthermore, a risk-triggered projection warning design is applied to avoid ineffective projection interference. The advantage of this design is that, based on pedestrian movement data, when a collision risk is detected, the second preset algorithm retrieves pre-configured projection parameters, making the warning action more consistent with the actual movement state of the pedestrian. This effectively reduces the probability of collisions in pedestrian-vehicle interaction scenarios and improves the safety of both vehicles and pedestrians. Attached Figure Description
[0052] Figure 1 This is a flowchart of a vehicle warning and control method for pedestrians. Detailed Implementation
[0053] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1 This application will be described in further detail. It is obvious that the described embodiments are merely some, not all, of the embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments described herein without inventive effort are within the scope of protection of this application.
[0054] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.
[0055] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.
[0056] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.
[0057] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”
[0058] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.
[0059] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.
[0060] The optional embodiments of this application are described in detail below with reference to the accompanying drawings.
[0061] Figure 1 The method shown is a vehicle warning and control method for pedestrians, the method comprising the following steps:
[0062] S1: Real-time detection of whether there are target pedestrians within a preset range of the target vehicle;
[0063] S2: If a target pedestrian is present, collect the pedestrian's movement data; the movement data includes at least: pedestrian position, movement speed, and movement direction;
[0064] S3: Based on the acquired movement data, the first preset algorithm is used to determine whether there is a collision risk to the pedestrian;
[0065] S4: If there is a risk of collision, the second preset algorithm is used to obtain the vehicle's projection parameters in order to warn the target pedestrian in advance.
[0066] Specifically, this application detects pedestrians within a preset range in real time, collects pedestrian position, speed, and direction data, and applies a first preset algorithm to assess collision risk. When a collision risk is detected, a second preset algorithm retrieves projection parameters to execute a forward warning. This achieves real-time and accurate identification of pedestrian collision risks. Furthermore, the application of a risk-triggered projection warning design avoids ineffective projection interference. The advantage of this design is that, based on pedestrian movement data, when a collision risk is detected, the second preset algorithm retrieves pre-configured projection parameters, making the warning action more closely aligned with the actual pedestrian movement. This effectively reduces the probability of collisions in pedestrian-vehicle interaction scenarios, improving traffic safety for both vehicles and pedestrians.
[0067] This application constructs a 360° all-around perception network using lidar, high-definition cameras, and millimeter-wave radar to accurately collect data on the location, speed, and direction of movement of vulnerable road users within a 10-50 meter radius around a vehicle.
[0068] In one specific embodiment, step S3, based on the acquired movement data and using a first preset algorithm, determines whether there is a collision risk with the pedestrian, specifically including:
[0069] Based on the acquired mobile data, a risk index is obtained according to preset rules; wherein, the risk index includes at least: distance index, time index, behavioral intention index, and environmental correction index pre-configured based on weather characteristics;
[0070] Based on the obtained risk index, the first formula is used to obtain the target risk index of the target pedestrian.
[0071] Based on the comparison results between the target risk index and the index threshold, it is determined whether there is a collision risk to the target pedestrian;
[0072] The distance index is configured to be obtained based on a comparison between a first distance between the target pedestrian and the target vehicle and a preset interval; wherein, the larger the first distance, the larger the corresponding distance index.
[0073] The time index is configured to be obtained by comparing the time difference between the target pedestrian and the target vehicle at the intersection of their trajectories with the corresponding time interval; wherein, the larger the time difference, the larger the corresponding time index.
[0074] The behavioral intent index is configured to be obtained based on the collected behavioral characteristics of the target pedestrian; wherein, the behavioral characteristics are key features of the pedestrian's intention to cross towards vehicles or roads; the more significant the key features, the larger the corresponding behavioral intent index.
[0075] The behavioral characteristics include at least: the pedestrian's body orientation, the pedestrian's pausing behavior, and the pedestrian's walking behavior.
[0076] Specifically, this application constructs a multi-dimensional risk index using three pedestrian-related indices: distance, time, and behavioral intent, along with a pre-configured environmental correction index based on weather characteristics. Each index is obtained according to preset rules based on pedestrian movement data and environmental characteristics. The distance and time indices are positively correlated with actual risk, while the behavioral intent index is positively correlated with crossing risk. Simultaneously, multiple indices are calculated using a first formula to obtain a target risk index, which is then compared with an index threshold to complete the collision risk assessment. This achieves a multi-dimensional and accurate assessment of pedestrian collision risk. Furthermore, the acquisition rules and assessment methods for each index are designed using preset methods, resulting in efficient algorithm execution that can adapt to the real-time response requirements of in-vehicle terminals, significantly improving the accuracy and timeliness of collision risk assessment.
[0077] In one specific embodiment, the step of obtaining the target risk index of the target pedestrian based on the acquired risk index using a first formula specifically includes:
[0078] The first formula is as follows:
[0079] Target Risk Index = Distance Index + Time Index + Behavioral Intent Index + Environmental Correction Index;
[0080] Correspondingly, if the target risk index is less than or equal to the index threshold, then it is determined that there is no risk of collision with the target pedestrian.
[0081] If the target risk index is greater than the index threshold, then the target pedestrian is determined to be at risk of collision.
[0082] Example
[0083] If the initial distance between the target pedestrian and the target vehicle is more than 25 meters, the corresponding distance index is 0; if the initial distance falls within the range of 10-15 meters, the corresponding distance index is 0.1; if the initial distance falls within the range of 5-10 meters, the corresponding distance index is 0.2; if the initial distance is less than 5 meters, the corresponding distance index is 0.25.
[0084] If the time difference is ≤3s, the corresponding time index is the highest value of 0.35. If the time difference falls between 3s and 6s, the corresponding time index is 0.25-0.35. If the time difference is >6s, the corresponding time index range is <0.25.
[0085] The behavioral intent index ranges from 0 to 0.2; for example, if the target pedestrian is standing still, the time index is 0, and if the pedestrian is walking, the time index is 0.2.
[0086] The environmental correction index ranges from 0 to 0.2; the lower the visibility or the lower the light intensity, the larger the corresponding environmental correction index.
[0087] In this embodiment, the target risk index ranges from 0 to 1.
[0088] In one specific embodiment, step S4 specifically includes:
[0089] S401: Obtain environmental features within a preset range; the environmental features include at least: intersection type, road type, and road width;
[0090] S402: Based on the acquired environmental features, input the pre-trained scene classification model and output the target scene of the target pedestrian;
[0091] S403: Based on the target scene of the target pedestrian, the acquired weather features, and the target risk index, obtain the target projection parameters corresponding to the target scene by looking up the scene and projection mapping table; wherein, the weather features include at least: visibility and light intensity;
[0092] S404: Based on the acquired target projection parameters, control the vehicle headlight projection module to project onto a preset part of the target vehicle to warn the target pedestrian in advance; the target projection parameters include at least: flashing frequency, color intensity, marking pattern, marking pattern size and visual complexity of the marking pattern; wherein, the marking pattern includes at least: warning signs (such as deceleration signs), distance signs and guidance signs.
[0093] Specifically, in this embodiment, environmental features such as intersection type, road type, and road width are collected, and a pre-trained scene classification model is used to achieve accurate matching of the target scene. Then, based on weather characteristics and target risk index, the corresponding target projection parameters are retrieved through the scene and projection mapping table. Scene features, environmental conditions, collision risk level and projection parameters are associated, realizing the scene-based, personalized and accurate adaptation of projection parameters. At the same time, the projection parameters include multi-dimensional configurable content such as flashing frequency, color intensity, sign pattern and visual complexity. The sign types include warning, distance and guidance signs. These control the vehicle headlight projection module to perform projection, so that the projection warning action can highly meet the actual needs of different scenes, weather and risk levels. Ultimately, it ensures that the target pedestrian can clearly identify the warning information and greatly reduce the probability of pedestrian-vehicle collision.
[0094] It is understood that the scenario classification model established in this embodiment, through the collection of environmental features such as intersection type, pedestrian-vehicle mixed traffic status, road width, turning characteristics, and road functional attributes, pre-sets six typical interaction scenarios for vulnerable road participants, including intersection crossing, roadside starting and stopping, narrow traffic around schools, turning on narrow roads, and pedestrian-vehicle mixed traffic in residential areas. Among them, road functions include: residential area functions, school functions, and urban and rural functions.
[0095] It should be noted that the vehicle headlight projection module in this embodiment adopts a full-color laser projection module, which supports 400,000 pixels and 1,000 lumens brightness output, and can project a combination of dynamic symbols, patterns, distance markers, guide paths and other content.
[0096] The module integrates an adaptive adjustment unit that adjusts the projection brightness according to the ambient light intensity and the projection size according to the target distance (enlarging the symbol ratio at long distances).
[0097] Furthermore, this application also includes the following steps:
[0098] The system uses a camera to monitor the reactions of pedestrians within a preset range in real time (such as slowing down or avoiding them), and dynamically adjusts the flicker frequency and color intensity of the projection parameters.
[0099] If the target does not respond, a secondary warning will be automatically triggered, adding a buzzer (inside the vehicle) and a flashing red border (projection) to enhance the warning effect.
[0100] illustrative
[0101] Pedestrian crossing warning scenario
[0102] A vehicle traveling at 20 km / h approaches an intersection without traffic lights. Cameras and lidar detect a pedestrian standing 15 meters to the right at a speed of 0.5 m / s, initially indicating an intention to cross. The prediction and decision-making module, based on the pedestrian's position and the vehicle's trajectory, calculates a collision risk coefficient of 0.7 (high risk) and matches the "crossing the intersection" scenario interaction strategy, generating a combined projection scheme of "dynamic deceleration symbol + distance countdown + guiding zebra crossing." The vehicle's headlight projection module activates the headlights and side mirrors: the headlights project a red dynamic deceleration arrow (flashing twice per second) onto the ground, simultaneously displaying a "3m" distance marker; the side mirrors project a white guiding zebra crossing towards the pedestrian, clearly defining the safe passage area. The feedback adjustment module detects the pedestrian starting to decelerate and observe, maintaining the current projection parameters; when the pedestrian steps onto the guiding zebra crossing, the projected arrow turns green, and the distance marker updates in real time until the pedestrian has safely crossed.
[0103] Another embodiment provided in this application offers a novel design approach, such as the following example for low-visibility, same-carriage scenarios:
[0104] In rainy or foggy weather, when the vehicle is traveling at 25 km / h, the ambient light sensor detects visibility of less than 200 meters, and the perception module detects a cyclist following 5 meters behind. In this low-visibility scenario, where there is no direct risk of collision but a safe distance must be maintained, a "yellow outline marker + distance indication projection" scheme is generated. The rear bumper projection module projects a dynamic yellow border around the cyclist's outline (adjusting synchronously with the cyclist's movement), simultaneously displaying a "5m" safe distance marker, with increased brightness compared to sunny days. When the vehicle slows down, the projected marker automatically flashes to remind cyclists behind to slow down in advance and avoid rear-end collisions.
[0105] In one specific embodiment, step S4 further includes:
[0106] The facial image of the target pedestrian is acquired and input into the facial recognition model to obtain key facial features; the key facial features include at least: facial contour features, hair features, eye features, and wrinkle features;
[0107] Based on the acquired key facial features, the data is input into a pre-trained age classification model, which outputs the age category of the target pedestrian. The age category includes at least: children, the elderly, and middle-aged people. The age classification model is pre-trained based on key visual features of facial age.
[0108] Based on the age level of the target pedestrian, a corresponding projection strategy is adopted to control the projection of the vehicle headlights projection module.
[0109] In one specific embodiment, the step of controlling the vehicle headlight projection module to project according to the age level of the target pedestrian using a corresponding projection strategy specifically includes:
[0110] If the target pedestrian's age category is child, the corresponding projection auxiliary parameters are retrieved and transmitted to the vehicle headlight projection module to perform auxiliary projection; the projection auxiliary parameters include at least: warning cartoon pattern, number of cartoon patterns, display rhythm and display speed of the cartoon pattern;
[0111] If the target pedestrian's age category is elderly, then the corresponding projection auxiliary parameters are retrieved, and the visual complexity of the target projection parameter's sign pattern is adjusted to the lowest level. Based on the projection auxiliary parameters, the color vibrancy is adjusted to a high level, and the color saturation is adjusted to a low level. The projection auxiliary parameters include at least: the vibrancy and color saturation of the sign pattern.
[0112] If the target pedestrian's age category is middle-aged, then maintain the target projection parameters and control the vehicle headlight projection module to perform projection.
[0113] It is understandable that this embodiment specifically designs projection strategies for pedestrians of different age groups, achieving matching of the age characteristics of the warning projection. The advantage of this design is that it configures corresponding projection schemes according to the visual deviations and physiological differences of children, the elderly, and middle-aged people. The strategy execution is simple, and the parameter linkage adjustment process is efficient, enabling the projected warnings to better suit the recognition abilities of pedestrians of different ages. Specifically, for children, warning cartoon patterns and dynamic projections displaying rhythm and speed are configured to match children's visual interests and cognitive characteristics, thereby improving children's attention to and recognition rate of warning information. For the elderly, the visual complexity of the sign patterns is adjusted to the minimum, and the color parameters are adjusted to be highly vivid and low saturation to cater to the characteristics of the elderly's declining vision and low color recognition, while avoiding the visual stimulation and fatigue caused by high saturation colors, allowing the elderly to clearly and comfortably recognize the warning signs. For middle-aged people whose visual tolerance and cognitive abilities are in good condition, the basic target projection parameters are maintained without additional adjustments. The advantage of this design is that it ensures that the projected warnings can achieve efficient information transmission for pedestrians of different ages in every scenario, thereby improving the effectiveness of warnings for various groups.
[0114] It should be noted that this embodiment is mainly applied in low-speed, narrow-sight scenarios, such as when the vehicle speed limit is ≤30km / h.
[0115] It should be further noted that when the target pedestrian is a child, the visual complexity of the target projection parameters and the display speed of the cartoon pattern should be adjusted to the lowest level. This is to attract the child's curious attention and use the cartoon pattern to show the child the risk warning, so as to guide the child away. This avoids the situation where the cartoon pattern design is too complex, the movement is too fast, or the colors are too bright, which may cause the child's attention to be completely attracted by the picture, ignoring the driver's shouts and the sound of the vehicle, thus losing the instinctive reaction of self-protection.
[0116] As can be seen from the above, this application has the following advantages:
[0117] Improve the timeliness of interaction: upgrade information transmission from "responding after operation" to "predicting before operation", provide more reaction time for vulnerable participants, and reduce the risk of pedestrian collisions at night.
[0118] Enhanced scene adaptability: Provides differentiated interaction solutions for high-risk scenarios such as low visibility and complex intersections. The brightness and color of the projected content can adapt to the ambient light intensity, improving recognition.
[0119] Reduce communication costs: Replace single warnings with intuitive and dynamic symbol combinations to reduce the difficulty of information interpretation for road users and reduce the rate of traffic conflicts at intersections.
[0120] Furthermore, in terms of hardware design, a cross-domain data interface is built between ADAS and the vehicle lighting system to achieve real-time synchronization of detection data and prediction results, solve the problem of collaboration barriers, and reduce interaction response latency.
[0121] On the other hand, this application provides a vehicle warning and control system for pedestrians, the system comprising:
[0122] The data acquisition module is configured to detect in real time whether there are target pedestrians within a preset range of the target vehicle;
[0123] The first judgment module is configured to collect the movement data of the target pedestrian if the target pedestrian exists; the movement data includes at least: pedestrian position, movement speed and movement direction;
[0124] The second judgment module is configured to determine whether there is a collision risk to the pedestrian based on the acquired movement data and using the first preset algorithm.
[0125] The processing module is configured to use a second preset algorithm to obtain the vehicle's projection parameters if there is a collision risk, so as to warn the target pedestrian in advance.
[0126] On the other hand, this application provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method.
[0127] On the other hand, this application provides a computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method described herein.
[0128] On the other hand, this application provides a simulation platform, including:
[0129] An electronic device for implementing the steps of the method described herein;
[0130] A processor that runs a program, and when the program runs, it executes the steps of the method from data output by the electronic device.
[0131] A storage medium for storing a program that, when run, executes the steps of the method on data output from an electronic device.
[0132] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A vehicle early warning and control method for pedestrians, characterized in that, The method includes the following steps: S1: Real-time detection of whether there are target pedestrians within a preset range of the target vehicle; S2: If a target pedestrian is present, collect the pedestrian's movement data; the movement data includes at least: pedestrian position, movement speed, and movement direction; S3: Based on the acquired movement data, the first preset algorithm is used to determine whether there is a collision risk to the pedestrian; S4: If there is a risk of collision, the second preset algorithm is used to obtain the vehicle's projection parameters in order to warn the target pedestrian in advance.
2. The method according to claim 1, characterized in that, Step S3, based on the acquired movement data, uses a first preset algorithm to determine whether there is a collision risk with the pedestrian, specifically including: Based on the acquired mobile data, a risk index is obtained according to preset rules; wherein, the risk index includes at least: distance index, time index, behavioral intention index, and environmental correction index pre-configured based on weather characteristics; Based on the obtained risk index, the first formula is used to obtain the target risk index of the target pedestrian. Based on the comparison results between the target risk index and the index threshold, it is determined whether there is a collision risk to the target pedestrian; The distance index is configured to be obtained based on a comparison between a first distance between the target pedestrian and the target vehicle and a preset interval; The time index is configured to be obtained by comparing the time difference between the target pedestrian and the target vehicle at the intersection of the trajectory with the corresponding time interval. The behavioral intent index is configured to be obtained based on the collected behavioral characteristics of the target pedestrian; wherein, the behavioral characteristics are key features of the pedestrian's intention to cross towards a vehicle or road; The behavioral characteristics include at least: the pedestrian's body orientation, the pedestrian's pausing behavior, and the pedestrian's walking behavior.
3. The method according to claim 2, characterized in that, Based on the acquired risk index, the target risk index of the target pedestrian is obtained using a first formula, specifically including: The first formula is as follows: Target Risk Index = Distance Index + Time Index + Behavioral Intent Index + Environmental Correction Index; Correspondingly, if the target risk index is less than or equal to the index threshold, then it is determined that there is no risk of collision with the target pedestrian. If the target risk index is greater than the index threshold, then the target pedestrian is determined to be at risk of collision.
4. The method according to claim 3, characterized in that, Step S4 specifically includes: S401: Obtain environmental features within a preset range; the environmental features include at least: intersection type, road type, and road width; S402: Based on the acquired environmental features, input the pre-trained scene classification model and output the target scene of the target pedestrian; S403: Based on the target scene of the target pedestrian, the acquired weather features, and the target risk index, obtain the target projection parameters corresponding to the target scene by looking up the scene and projection mapping table; wherein, the weather features include at least: visibility and light intensity; S404: Based on the acquired target projection parameters, control the vehicle headlight projection module to project onto a preset part of the target vehicle to warn the target pedestrian in advance; the target projection parameters include at least: flashing frequency, color intensity, marking pattern, marking pattern size and visual complexity of the marking pattern; wherein, the marking pattern includes at least: warning signs, distance signs and guidance signs.
5. The method according to claim 4, characterized in that, Step S4 further includes: The facial image of the target pedestrian is acquired and input into the facial recognition model to obtain key facial features; the key facial features include at least: facial contour features, hair features, eye features, and wrinkle features; Based on the acquired key facial features, the data is input into a pre-trained age classification model, which outputs the age category of the target pedestrian. The age category includes at least: children, the elderly, and middle-aged people. The age classification model is pre-trained based on key visual features of facial age. Based on the age level of the target pedestrian, a corresponding projection strategy is adopted to control the projection of the vehicle headlights projection module.
6. The method according to claim 5, characterized in that, The step of controlling the vehicle headlight projection module to project based on the age level of the target pedestrian using a corresponding projection strategy includes: If the target pedestrian's age category is child, the corresponding projection auxiliary parameters are retrieved and transmitted to the vehicle headlight projection module to perform auxiliary projection; the projection auxiliary parameters include at least: warning cartoon pattern, number of cartoon patterns, display rhythm and display speed of the cartoon pattern; If the target pedestrian's age category is elderly, then the corresponding projection auxiliary parameters are retrieved, and the visual complexity of the target projection parameter's sign pattern is adjusted to the lowest level. Based on the projection auxiliary parameters, the color vibrancy is adjusted to a high level, and the color saturation is adjusted to a low level. The projection auxiliary parameters include at least: the vibrancy and color saturation of the sign pattern. If the target pedestrian's age category is middle-aged, then maintain the target projection parameters and control the vehicle headlight projection module to perform projection.
7. A vehicle warning and control system for pedestrians, characterized in that, The system includes: The data acquisition module is configured to detect in real time whether there are target pedestrians within a preset range of the target vehicle; The first judgment module is configured to collect the movement data of the target pedestrian if the target pedestrian exists; the movement data includes at least: pedestrian position, movement speed and movement direction; The second judgment module is configured to determine whether there is a collision risk to the pedestrian based on the acquired movement data and using the first preset algorithm. The processing module is configured to use a second preset algorithm to obtain the vehicle's projection parameters if there is a collision risk, so as to warn the target pedestrian in advance.
8. An electronic device, comprising: The system comprises a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; characterized in that the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method described in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The device stores a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method as described in any one of claims 1 to 6.
10. A simulation platform, characterized in that, include: An electronic device for implementing the steps of the method according to any one of claims 1-6; A processor that runs a program, which, when running, performs the steps of the method according to any one of claims 1-6 from data output by an electronic device. A storage medium for storing a program that, when run, performs the steps of the method according to any one of claims 1-6 on data output from an electronic device.