A vehicle starting cooperative warning method and system based on intelligent projection and vehicle

Abstract

The application provides a vehicle starting cooperative warning method and system based on intelligent projection and a vehicle, and belongs to the technical field of intelligent traffic and vehicle active safety, and comprises the following steps: detecting vehicle state information and traffic signal lamp state, and acquiring position information and motion state information of a target crossing in front of the vehicle; calculating a conflict time of the target crossing to an adjacent lane and a position interval at a conflict moment, determining a collision risk level based on the conflict time and the position interval at the conflict moment; when the collision risk level reaches a preset high risk threshold, predicting a predicted position of the target crossing according to a speed vector of the target crossing, calculating a predicted angle of a projection device based on the predicted position, and introducing a lateral safety offset angle to obtain a final projection angle; and controlling the projection device to project a warning pattern to the ground in front of the vehicle in the adjacent lane according to the final projection angle, so as to send a warning to a driver in the adjacent lane. The application improves the effectiveness of cross-lane safety warning.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation and vehicle active safety technology, and particularly relates to a vehicle start-up collaborative warning method, system and vehicle based on intelligent projection. Background Technology

[0002] At urban road intersections, when the red light turns green and vehicles are preparing to start, pedestrians or non-motorized vehicles often rush across the road at the last second. At this time, the driver in the inner lane often has their view completely blocked by larger vehicles parked or starting alongside them on the right, making it impossible to see the suddenly appearing lateral movement target. This phenomenon is known as "ghost peek," and it can easily cause vehicles in the inner lane to accelerate quickly because they cannot see the pedestrian, leading to serious collisions.

[0003] Existing technologies can only detect obstacles directly in front of the vehicle through in-vehicle active safety systems. For targets that suddenly enter the vehicle's path from the side, there is a detection delay or blind spot, and it cannot prevent the vehicle from starting. Blind spot monitoring systems are mainly used for lane change warnings and monitor vehicles approaching from the side and rear. They have limited effect on warning of vertical targets that quickly cut in from the side, and the warning target is only the driver of the vehicle. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this invention proposes a vehicle start-up collaborative warning method, system, and vehicle based on intelligent projection, in order to solve the problems that traffic facility warnings cannot provide dynamic and accurate warnings for specific and sudden risk events, and that existing safety systems only target their own vehicles and cannot directly issue visual warnings to drivers in adjacent lanes whose vision is obstructed.

[0005] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention discloses a vehicle start-up cooperative warning method based on intelligent projection, comprising: Detect vehicle status information and traffic light status, and obtain the position and motion status information of the target crossing in front of the vehicle; Based on the motion state information of the crossing target and the estimated motion state of vehicles in adjacent lanes, the conflict time and positional distance at the moment of conflict of the crossing target to the adjacent lane are calculated, and the collision risk level is determined based on the conflict time and positional distance at the moment of conflict. When the collision risk level reaches a preset high-risk threshold, the predicted position of the target is predicted based on the velocity vector of the target, the predicted angle of the projection device is calculated based on the predicted position, and the lateral safety offset angle is introduced to obtain the final projection angle. According to the final projection angle, the projection device is controlled to project a warning pattern onto the ground in front of vehicles in adjacent lanes to warn drivers in adjacent lanes.

[0006] A further technical solution involves activating a risk scanning window of a preset duration when the vehicle speed is detected to be below a preset stationary speed threshold and the gear is in driving mode, and when the traffic light is detected to meet the start-up trigger conditions. Within the risk scanning window, the position and motion status information of the target crossing in front of the vehicle are perceived through the fusion of a camera and a ranging sensor.

[0007] A further technical solution involves calculating the conflict time and positional distance between the target and the adjacent lane based on the target's motion state information and the estimated motion state of vehicles in the adjacent lane. Specifically: Using the centerline position of the adjacent lane as a reference, calculate the collision time of the crossing target reaching the center of the adjacent lane based on the lateral velocity component of the crossing target:

[0008] in, The lateral coordinates of the center lines of adjacent lanes The current horizontal coordinates of the target are used for crossing. The lateral velocity component is the velocity component that crosses the target. Substituting the time of conflict into the equations for the trajectory crossing the target and the motion equations for vehicles in adjacent lanes, the longitudinal positions of both at the time of conflict are calculated:

[0009]

[0010] The positional spacing at the moment of conflict is:

[0011] in, The positional spacing at the moment of conflict. The longitudinal position of the target at the moment of conflict. The longitudinal position of vehicles in adjacent lanes at the moment of conflict. Let y be the starting acceleration of a vehicle in the adjacent lane, t be time, and y be the starting acceleration of the vehicle in the adjacent lane. t To traverse the target's current vertical coordinates, vy t This represents the longitudinal velocity component that crosses the target.

[0012] A further technical solution is to determine the collision risk level based on the conflict time and the distance between locations; when the distance between locations at the time of conflict is less than the safe distance threshold and the conflict time is less than the reaction time threshold, the collision risk level is determined to be high risk; when the distance between locations at the time of conflict is less than the safe distance threshold and the conflict time is not less than the reaction time threshold, the collision risk level is determined to be medium risk; in other cases, the collision risk level is determined to be no risk.

[0013] A further technical solution involves calculating the prediction angle of the projection device based on the predicted position as follows:

[0014] in, To predict the projection angle, The relative coordinates between the predicted location and the installation location of the projection device.

[0015] A further technical solution, wherein the introduction of a lateral safety offset angle to obtain the final projection angle, specifically includes: Determine whether a lateral safety offset angle needs to be superimposed. When the lateral distance across the target is less than the lateral safety distance threshold, a lateral safety offset angle is superimposed on the predicted projection angle. The lateral safety offset angle is:

[0016] in, For the lateral safety offset angle, For the lateral safety boundary factor, The current horizontal coordinates of the target being traversed; The final projection angle is calculated based on the predicted projection angle and the lateral safety offset angle:

[0017] in, For the final projection angle, To predict the projection angle.

[0018] Secondly, this invention discloses a vehicle start-up coordination warning system based on intelligent projection, comprising: The data acquisition module is configured to detect vehicle status information and traffic light status, and acquire the position and motion status information of the target crossing in front of the vehicle. The risk assessment module is configured to: calculate the conflict time and positional distance at the moment of conflict of the crossing target to the adjacent lane based on the motion state information of the crossing target and the estimated motion state of vehicles in the adjacent lane; and determine the collision risk level based on the conflict time and positional distance at the moment of conflict. An angle calculation module is configured to: when the collision risk level reaches a preset high-risk threshold, predict the predicted position of the target based on the velocity vector of the target, calculate the predicted angle of the projection device based on the predicted position, and introduce a lateral safety offset angle to obtain the final projection angle. The projection control module is configured to control the projection device to project a warning pattern onto the ground in front of vehicles in adjacent lanes according to the final projection angle, so as to warn drivers in adjacent lanes.

[0019] Thirdly, the present invention discloses a vehicle, including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when run by the processor, complete the steps of the above-mentioned vehicle start-up cooperative warning method based on intelligent projection.

[0020] Fourthly, the present invention discloses a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-mentioned vehicle start-up coordination warning method based on intelligent projection.

[0021] Fifthly, the present invention discloses a computer program product comprising executable instructions stored in a computer-readable storage medium; wherein, when the vehicle's processor reads and executes the executable instructions from the computer-readable storage medium, it completes the steps of the aforementioned vehicle start-up coordination warning method based on intelligent projection.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a solution that can proactively identify risks in obstructed areas on the side of a vehicle at the moment of vehicle start-up, and provide intuitive and strong warnings to vehicles in adjacent lanes whose visibility is obstructed and which may directly cause accidents. Specifically, through a three-level risk classification algorithm based on a dual-dimensional quantitative index of conflict time and longitudinal distance between conflict points, the collision risk of each crossing target is accurately assessed in real time. Warnings are triggered only when high-risk conditions are met. By dynamically calculating the projection angle based on the predicted target position, the warning light wall is always accurately presented in the high-risk area in front of the pedestrian's movement path. This achieves dynamic and accurate warnings that are targeted to specific targets and updated in real time as the target moves. Compared with roadside warning facilities in fixed locations, the pertinence and timeliness of the warnings are significantly improved.

[0023] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0025] Figure 1 This is a flowchart of the vehicle start-up collaborative warning method based on intelligent projection described in Embodiment 1 of the present invention.

[0026] Figure 2 This is a flowchart of the collision risk determination algorithm described in Embodiment 1 of the present invention.

[0027] Figure 3 This is a flowchart of the calculation of the final projection angle as described in Embodiment 1 of the present invention.

[0028] Figure 4 This is a block diagram of the vehicle start-up cooperative warning system based on intelligent projection as described in Embodiment 2 of the present invention. Detailed Implementation

[0029] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0030] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0031] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0032] Example 1 In one or more embodiments, a vehicle start-up cooperative warning method based on intelligent projection is disclosed, such as... Figure 1 As shown, it includes the following steps: Step S101: Detect vehicle status information and traffic light status. When the vehicle speed is detected to be lower than the preset stationary speed threshold and the gear is in driving gear, and the traffic light is detected to meet the start trigger condition, start a risk scanning window of preset duration. Within the risk scanning window, obtain the position information and motion status information of the target crossing in front of the vehicle through the fusion perception of the camera and the ranging sensor.

[0033] As one implementation method, information collection mainly relies on the following approaches: The vision sensor is a wide-angle camera deployed at the front of the vehicle (such as inside the windshield or near the logo) to capture real-time images of the area in front of and to the sides of the vehicle.

[0034] The ranging sensor can be a millimeter-wave radar, lidar, or ultrasonic sensor, which works in conjunction with a camera to detect the position, speed, and trajectory of moving targets in front of and to the sides of the vehicle.

[0035] The vehicle status and signal acquisition unit communicates with the vehicle's CAN bus to acquire vehicle status, including speed and gear position; at the same time, it acquires the real-time status of traffic lights (such as the moment when the red light turns green) through V2X communication or image recognition technology.

[0036] Step S1011: Detect vehicle status information and traffic light status.

[0037] Specifically, the onboard processor reads vehicle speed and gear information in real time via the Controller Area Network (CAN) bus. When the vehicle speed is detected to be below a preset stationary speed threshold and the gear is in drive, it is determined that the vehicle is stationary and waiting to start. At the same time, the onboard processor obtains the phase status and remaining countdown information of traffic lights through the vehicle-to-everything (V2X) communication unit; when V2X communication is unavailable, the onboard processor acquires images of the road ahead through a wide-angle camera, performs color recognition on the traffic light area, and obtains the traffic light status.

[0038] Step S1012: Determine whether the traffic lights meet the start-up trigger conditions.

[0039] Specifically, when the traffic light is detected to change from red to green, or when the remaining countdown of the green light is no more than 2 seconds, it is determined that the traffic light meets the start trigger condition. If both the vehicle is stationary and the traffic light meets the start trigger condition, then step S1013 is executed; otherwise, the process returns to step S1011 to continue monitoring.

[0040] Step S1013: Start the risk scan window for a preset duration.

[0041] Specifically, the onboard processor initiates a risk scanning window that lasts for 2 to 3 seconds, within which subsequent perception and judgment steps are executed cyclically at a preset sampling interval.

[0042] Step S1014: Obtain the position and motion status information of the target by fusing the camera and the ranging sensor.

[0043] Specifically, a wide-angle camera captures images of the front and side front of the vehicle. Based on a deep learning object detection model (such as YOLO, SSD, etc.), it identifies pedestrians and non-motorized vehicles in the images, obtaining the target category and image coordinates. A millimeter-wave radar acquires the target's relative distance, relative speed, and azimuth information. The onboard processor fuses the camera detection results with the millimeter-wave radar detection results to obtain the position coordinates (x, y, y) of the target traversing the vehicle's coordinate system. t y t ) and velocity vector (vx) t vy t ), where x t y is the horizontal coordinate. t vx is the vertical axis. t For the lateral velocity component, vy t This represents the longitudinal velocity component. When the angle between the target's direction of motion and the vehicle's direction of travel is greater than 45 degrees, the target is determined to have the intention to cross the road and is included in the list of targets that cross the road.

[0044] Among them, the position coordinates (x, y) of the target traversing the vehicle coordinate system t y t ), x t y represents the horizontal distance, with positive values ​​for the right side and negative values ​​for the left side; t This represents the vertical distance; positive values ​​indicate the front.

[0045] This embodiment sets dual trigger conditions: "vehicle speed is lower than a preset stationary speed threshold and the gear is in driving" and "traffic light meets the start trigger condition". It also activates a risk scanning window of 2 to 3 seconds to precisely lock the system activation time at the critical moment of starting at the intersection, which is the highest risk period. This avoids false alarms caused by continuous system operation and significantly improves the pertinence of the warning.

[0046] Step S102: Based on the motion state information of the crossing target and the estimated motion state of vehicles in adjacent lanes, calculate the conflict time and positional distance between the crossing target and the adjacent lane. Determine the collision risk level based on the conflict time and positional distance. Figure 2 As shown, the specific steps include: Step S1021: Obtain the motion state parameters of crossing the target, including the current lateral coordinate x of crossing the target. t Vertical coordinate y t Lateral velocity component vx t and longitudinal velocity component vy t .

[0047] Step S1022: Determine whether the lateral velocity component is valid.

[0048] When the absolute value of the lateral velocity component is less than the preset lateral velocity threshold, it is determined that the lateral movement speed across the target is too small, there is no risk of crossing, the collision risk level is determined to be no risk, and the judgment ends.

[0049] Step S1023: Calculate the conflict time t conflict Using the centerline position of the adjacent lane as a reference, and based on the lateral velocity component of the crossing target, the conflict time from the crossing target to the center of the adjacent lane is calculated according to the following formula:

[0050] in, The lateral coordinates of the center lines of adjacent lanes The current horizontal coordinates of the target are used for crossing. This represents the lateral velocity component crossing the target. When When the value is less than zero, it indicates that the target crossing the lane has already crossed the center line of the adjacent lane, the risk has been reduced, the collision risk level is determined to be no risk, and the judgment ends.

[0051] Step S1024: Establish a predicted motion model for vehicles in adjacent lanes.

[0052] The estimated motion state of vehicles in adjacent lanes is as follows: vehicles in adjacent lanes accelerate from a standstill with a starting acceleration. The starting acceleration ranges from 1.5 m / s² to 2.5 m / s² and can be dynamically adjusted according to the vehicle type recognition results. For example, 2.0 m / s² is used for SUVs and 1.8 m / s² is used for sedans.

[0053] The equation for the change of longitudinal position of vehicles in adjacent lanes over time is:

[0054] in, Let t be the starting acceleration of the vehicle in the adjacent lane, and t be time. This represents the longitudinal position of vehicles in adjacent lanes.

[0055] Step S1025: Calculate the positional distance at the moment of conflict.

[0056] Substituting the time of conflict into the equations for the trajectory crossing the target and the motion equations for vehicles in adjacent lanes, the longitudinal positions of both at the time of conflict are calculated:

[0057]

[0058] The positional spacing at the moment of conflict is:

[0059] in, The positional spacing at the moment of conflict. The longitudinal position of the target at the moment of conflict. This represents the longitudinal position of vehicles in adjacent lanes at the moment of conflict.

[0060] Step S1026: Determine the collision risk level based on the conflict time and location distance.

[0061] When the distance between the positions at the moment of conflict is less than the safe distance threshold and the time of conflict is less than the reaction time threshold, the collision risk level is determined to be high risk; when the distance between the positions at the moment of conflict is less than the safe distance threshold and the time of conflict is not less than the reaction time threshold, the collision risk level is determined to be medium risk; otherwise, the collision risk level is determined to be no risk.

[0062] Preferably, the safety distance threshold can be set to 2 meters, and the reaction time threshold can be set to 1.5 seconds.

[0063] This embodiment addresses the problem that traffic facility warnings cannot provide dynamic and precise warnings for specific, sudden risk events. Existing roadside warning facilities (such as intersection light strips and voice broadcast systems) can only make coarse-grained judgments based on the time difference between the arrival of vehicles and pedestrians at the intersection. They cannot distinguish whether a pedestrian is actually on a high-risk trajectory that could collide with a vehicle in an adjacent lane, nor can they dynamically adjust the warning content according to real-time changes in the target's position and speed. By using a three-level risk classification algorithm based on a dual-dimensional quantitative index of conflict time and longitudinal distance between conflict points, collision risks are accurately divided into three levels: high risk, medium risk, and no risk. Projection warnings are only triggered under high-risk conditions. Compared with the qualitative judgment methods of existing technologies, the accuracy of risk assessment is significantly improved, effectively reducing the false alarm rate. By establishing a predicted motion model of vehicles in adjacent lanes accelerating from a standstill, the collision risk assessment is made more consistent with the actual motion characteristics of starting scenarios at intersections.

[0064] Step S103: Determine whether the collision risk level has reached the preset high-risk threshold. If the collision risk level is high, execute the sub-steps in this step; if the collision risk level is medium or no risk, return to step S1014 to continue scanning until the risk scanning window ends.

[0065] When the collision risk level reaches a preset high-risk threshold, the predicted position of the target is predicted based on the velocity vector of the target, the predicted angle of the projection device is calculated based on the predicted position, and the lateral safety offset angle is introduced to obtain the final projection angle.

[0066] like Figure 3 As shown, the steps for calculating the final projection angle include the following sub-steps: Step S1031: Obtain the current position and velocity vector of the target crossing. The current position coordinates (x, y, y) of the target crossing are obtained from the fusion sensing results of step S104 in Example 1. t y t ) and velocity vector (vx) t vy t ).

[0067] The polar coordinates of the target relative to the projection device are calculated as follows:

[0068]

[0069]

[0070] in, The polar coordinates of the target relative to the projection device; This is the original angle.

[0071] Step S1032: Predict the predicted position of the target after a preset prediction time.

[0072]

[0073]

[0074] in, The predicted location for crossing the target; The preset prediction time can be set to 0.5 seconds, which allows the location to be determined 0.5 seconds in advance, enhancing the timeliness of the warning.

[0075] Preferably, the preset prediction time is 0.3 seconds to 0.8 seconds; in this embodiment, the prediction time is used. =0.5 seconds.

[0076] Step S1033: Calculate the predicted projection angle based on the relative coordinates between the predicted position and the installation position of the projection device.

[0077] Calculate the lateral and longitudinal differences of the predicted position relative to the projection device:

[0078]

[0079] in, The relative coordinates between the predicted location and the installation location of the projection device; This is the offset of the projection device. The projection device is installed inside the vehicle's headlight assembly, and its installation position is offset relative to the origin of the vehicle's coordinate system.

[0080] Furthermore, the predicted projection angle is calculated:

[0081] in, To predict the projection angle.

[0082] Step S1034: Determine whether a lateral safety offset angle needs to be superimposed. This applies when considering the lateral distance across the target. When the distance is less than the lateral safety distance threshold, a lateral safety offset angle is added to the predicted projection angle:

[0083] in, For the lateral safety offset angle, The lateral safety boundary factor ranges from 3 to 7 degrees; in this embodiment, it is set to 5 degrees. When the target is on the left, the bias is to the left; when the target is on the right, the bias is to the right.

[0084] Step S1035: Calculate the final projection angle based on the predicted projection angle and the lateral safety offset angle:

[0085] in, This is the final projection angle.

[0086] Preferably, the final projection angle is limited to the mechanical rotation range [MIN_PROJECTOR_ANGLE, MAX_PROJECTOR_ANGLE] of the projection device to prevent exceeding the mechanically permissible range.

[0087] It should be noted that when there are multiple targets, the target with the highest risk is used as the basis for angle calculation.

[0088] This embodiment predicts the target's position several seconds later based on its velocity vector, and then calculates the final projection angle by superimposing a lateral safety offset angle on top of this. This ensures that the warning light wall is always accurately presented in the high-risk area in front of the pedestrian's movement path, effectively compensating for the projection lag caused by system delay and target movement, and improving the foresight and effectiveness of the warning.

[0089] Step S104: According to the final projection angle, control the projection device to project a warning pattern onto the ground in front of the vehicle in the adjacent lane to warn the driver in the adjacent lane.

[0090] Specifically, the onboard processor sends control commands to the projection device, which then projects warning patterns onto the ground in front of vehicles in adjacent lanes according to the final projection angle. The warning patterns include flashing light wall patterns and pedestrian warning icons.

[0091] In one implementation, the projection device is integrated into the vehicle's headlight assembly or both sides of the front bumper, capable of projecting high-brightness, high-definition laser or LED light onto the ground in front of the vehicle (at approximately a 306-degree angle to the vehicle). The projection content generator generates specific warning patterns according to instructions from the control module. Preferably, the warning pattern is a virtual wall projected onto the ground between the vehicle and the adjacent lane, extending in front of vehicles in the adjacent lane.

[0092] Example of projected content: Basic version: A continuous, flashing wall of red or yellow light.

[0093] Enhanced version: Above or in the light wall, there can be eye-catching icons, arrows, or short text such as "Watch out for pedestrians!" or "STOP!".

[0094] Therefore, at the critical moment when the red light turns green and the vehicle is about to start moving, it not only focuses on the safety of its own vehicle, but also takes the initiative to act as the "eyes" of vehicles in adjacent lanes, visualizing the risk through projection technology and projecting it directly onto the area of ​​risk, forcibly attracting the attention of the relevant drivers.

[0095] This invention solves the problem that existing safety systems only target their own vehicles and cannot directly provide visual warnings to drivers in adjacent lanes whose vision is obstructed. Existing in-vehicle active safety systems respond by braking, decelerating, or alerting their own driver, failing to transmit information about pedestrians crossing obstacles to drivers in adjacent lanes. This invention, through projection direction control, dynamically adjusts the projection angle based on the location of the risk target, ensuring the virtual wall is accurately presented in high-risk areas. It creatively changes the projection warning direction from directly in front of the vehicle to the ground in front of the adjacent lane, projecting a flashing light wall and pedestrian warning icons directly in front of the adjacent lane driver's field of vision via an in-vehicle projection device. Utilizing the human eye's high sensitivity to moving light sources, it creates a forced attention effect, allowing drivers in adjacent lanes who are in a blind spot due to large vehicles to intuitively perceive the presence of pedestrians crossing. This overcomes the fundamental limitation of existing technologies in "ghost pedestrian" scenarios where risk information cannot be transmitted across lanes, achieving inter-vehicle collaborative safety warnings.

[0096] Preferably, the system further includes step S105: determining whether the risk has been eliminated. When a target crossing the lane is detected leaving the adjacent lane, or the collision risk level is reduced to below medium risk, or the preset maximum warning time is exceeded, the system waits for the risk to be eliminated or the vehicle to start smoothly before controlling the projection device to automatically stop projection and the system returns to normal monitoring status.

[0097] Furthermore, the workflow of the present invention includes the following states: Idle State: After power-on, the system enters the idle state, continuously monitoring vehicle speed, gear position, and traffic light status. When the system detects that the vehicle speed is below a preset stationary speed threshold and the gear is in driving mode, the system transitions from the idle state to the scan preparation state.

[0098] Scanning Preparation State: The system continuously monitors traffic light status during the scanning preparation state. When a traffic light changes from red to green, or when the remaining countdown for the green light is less than 2 seconds, the system initiates a 2-3 second risk scan window, transitioning from the scanning preparation state to the risk assessment state. If the vehicle speed exceeds a preset stationary speed threshold during the scanning preparation state, the system returns to the idle state.

[0099] Risk Assessment Status: In the risk assessment status, the system cyclically performs sensor fusion sensing and collision risk quantification assessment at 50-millisecond sampling intervals. When a high-risk collision risk is detected, the system transitions from the risk assessment status to the projection warning status. When the risk scan window times out and no high-risk target is detected, the system transitions from the risk assessment status to the disabled status. Projection Warning State: In projection warning state, the system controls the projection device to continuously project warning patterns onto the ground in front of vehicles in adjacent lanes according to the final projection angle, and dynamically updates the projection angle and warning pattern content at 50-millisecond intervals. In projection warning state, the system simultaneously executes the following feedback monitoring mechanism: It detects speed changes of crossing targets. When the speed of the crossing target decreases below a preset deceleration threshold or the crossing target stops moving, the determined collision risk level is downgraded by one level (e.g., from high risk to medium risk). If the downgraded collision risk level is lower than the high risk threshold, the system switches from projection warning state to risk assessment state, suspends projection, and continues monitoring. When the crossing target leaves the adjacent lane, or the collision risk level decreases below medium risk, or the preset maximum warning duration is exceeded, the system switches from projection warning state to deactivated state.

[0100] Inactive State: In the inactive state, the system shuts down the projection device, releases sensor resources, and awaits the next trigger condition. When the system detects that the vehicle speed is below the preset stationary speed threshold and the gear is in driving mode, the system returns from the inactive state to the idle state, preparing for the next round of monitoring.

[0101] This invention is based entirely on the vehicle's own onboard sensors such as cameras, millimeter-wave radar, or lidar, as well as the vehicle's onboard processor. The projection device is integrated into the vehicle's headlights or front bumper, requiring no modification to the road infrastructure, making it easy to integrate and apply on vehicles.

[0102] This invention sets dual triggering conditions—a vehicle stationary waiting state and a traffic light meeting the start-up triggering condition—and activates a risk scanning window to precisely lock the system activation time at the critical moment of starting at an intersection, the highest-risk time period. Through the fusion perception of cameras and millimeter-wave radar, the position and movement status of crossing targets are reliably obtained. By projecting warning patterns onto the ground in front of vehicles in adjacent lanes, risk information is directly transmitted to drivers in adjacent lanes who cannot perceive the risk of pedestrians crossing due to obstruction by adjacent large vehicles, thus achieving inter-vehicle collaborative safety early warning.

[0103] Example 2 In one or more embodiments, a vehicle start-up cooperative warning system 400 based on intelligent projection is disclosed, such as... Figure 4 As shown, it specifically includes: The data acquisition module 401 is configured to: detect vehicle status information and traffic light status, and acquire the position information and motion status information of the target crossing in front of the vehicle; The risk assessment module 402 is configured to: calculate the conflict time and positional distance at the moment of conflict of the cross-traversing target to the adjacent lane based on the motion state information of the cross-traversing target and the estimated motion state of vehicles in the adjacent lane; and determine the collision risk level based on the conflict time and positional distance at the moment of conflict. Angle calculation module 403 is configured to: when the collision risk level reaches a preset high risk threshold, predict the predicted position of the target based on the velocity vector of the target, calculate the predicted angle of the projection device based on the predicted position, and introduce a lateral safety offset angle to obtain the final projection angle. The projection control module 404 is configured to control the projection device to project a warning pattern onto the ground in front of vehicles in adjacent lanes according to the final projection angle, so as to warn drivers in adjacent lanes.

[0104] In this embodiment, the data acquisition module 401 specifically executes the following process: Step S2011: Detect vehicle status information and traffic light status.

[0105] Specifically, the onboard processor reads vehicle speed and gear information in real time via the Controller Area Network (CAN) bus. When the vehicle speed is detected to be below a preset stationary speed threshold and the gear is in drive, it is determined that the vehicle is stationary and waiting to start. At the same time, the onboard processor obtains the phase status and remaining countdown information of traffic lights through the vehicle-to-everything (V2X) communication unit; when V2X communication is unavailable, the onboard processor acquires images of the road ahead through a wide-angle camera, performs color recognition on the traffic light area, and obtains the traffic light status.

[0106] Step S2012: Determine whether the traffic lights meet the start-up trigger conditions.

[0107] Specifically, when the traffic light is detected to change from red to green, or when the remaining countdown of the green light is no more than 2 seconds, it is determined that the traffic light meets the start trigger condition. If both the vehicle is stationary and the traffic light meets the start trigger condition, then step S1013 is executed; otherwise, the process returns to step S1011 to continue monitoring.

[0108] Step S2013: Start the risk scan window for a preset duration.

[0109] Specifically, the onboard processor initiates a risk scanning window that lasts for 2 to 3 seconds, within which subsequent perception and judgment steps are executed cyclically at a preset sampling interval.

[0110] Step S2014: Obtain the position and motion status information of the target by fusing the camera and the ranging sensor.

[0111] Specifically, a wide-angle camera captures images of the front and side front of the vehicle. Based on a deep learning object detection model (such as YOLO, SSD, etc.), it identifies pedestrians and non-motorized vehicles in the images, obtaining the target category and image coordinates. A millimeter-wave radar acquires the target's relative distance, relative speed, and azimuth information. The onboard processor fuses the camera detection results with the millimeter-wave radar detection results to obtain the position coordinates (x, y, y) of the target traversing the vehicle's coordinate system. t y t ) and velocity vector (vx) t vy t ), where x t y is the horizontal coordinate. t vx is the vertical axis. t For the lateral velocity component, vy t This represents the longitudinal velocity component. When the angle between the target's direction of motion and the vehicle's direction of travel is greater than 45 degrees, the target is determined to have the intention to cross the road and is included in the list of targets that cross the road.

[0112] Among them, the position coordinates (x, y) of the target traversing the vehicle coordinate system t y t ), x t y represents the horizontal distance, with positive values ​​for the right side and negative values ​​for the left side; t This represents the vertical distance; positive values ​​indicate the front.

[0113] In this embodiment, the risk assessment module 402 specifically executes the following process: Step S2021: Obtain the motion state parameters of crossing the target, including the current lateral coordinate x of crossing the target. t Vertical coordinate y t Lateral velocity component vx t and longitudinal velocity component vy t .

[0114] Step S2022: Determine whether the lateral velocity component is valid.

[0115] When the absolute value of the lateral velocity component is less than the preset lateral velocity threshold, it is determined that the lateral movement speed across the target is too small, there is no risk of crossing, the collision risk level is determined to be no risk, and the judgment ends.

[0116] Step S2023: Calculate the conflict time t conflict Using the centerline position of the adjacent lane as a reference, and based on the lateral velocity component of the crossing target, the conflict time from the crossing target to the center of the adjacent lane is calculated according to the following formula:

[0117] in, The lateral coordinates of the center lines of adjacent lanes The current horizontal coordinates of the target are used for crossing. This represents the lateral velocity component crossing the target. When When the value is less than zero, it indicates that the target crossing the lane has already crossed the center line of the adjacent lane, the risk has been reduced, the collision risk level is determined to be no risk, and the judgment ends.

[0118] Step S2024: Establish a predicted motion model for vehicles in adjacent lanes.

[0119] The estimated motion state of vehicles in adjacent lanes is as follows: vehicles in adjacent lanes accelerate from a standstill with a starting acceleration. The starting acceleration ranges from 1.5 m / s² to 2.5 m / s² and can be dynamically adjusted according to the vehicle type recognition results. For example, 2.0 m / s² is used for SUVs and 1.8 m / s² is used for sedans.

[0120] The equation for the change of longitudinal position of vehicles in adjacent lanes over time is:

[0121] in, Let t be the starting acceleration of the vehicle in the adjacent lane, and t be time. This represents the longitudinal position of vehicles in adjacent lanes.

[0122] Step S2025: Calculate the positional distance at the moment of conflict.

[0123] Substituting the time of conflict into the equations for the trajectory crossing the target and the motion equations for vehicles in adjacent lanes, the longitudinal positions of both at the time of conflict are calculated:

[0124]

[0125] The positional spacing at the moment of conflict is:

[0126] in, The positional spacing at the moment of conflict. The longitudinal position of the target at the moment of conflict. This represents the longitudinal position of vehicles in adjacent lanes at the moment of conflict.

[0127] Step S2026: Determine the collision risk level based on the conflict time and location spacing.

[0128] When the distance between the positions at the moment of conflict is less than the safe distance threshold and the time of conflict is less than the reaction time threshold, the collision risk level is determined to be high risk; when the distance between the positions at the moment of conflict is less than the safe distance threshold and the time of conflict is not less than the reaction time threshold, the collision risk level is determined to be medium risk; otherwise, the collision risk level is determined to be no risk.

[0129] In this embodiment, the angle calculation module 403 specifically executes the following process: Step S2031: Obtain the current position and velocity vector of the target crossing. The current position coordinates (x, y, y) of the target crossing are obtained from the fusion sensing results of step S104 in Example 1. t y t ) and velocity vector (vx) t vy t ).

[0130] The polar coordinates of the target relative to the projection device are calculated as follows:

[0131]

[0132]

[0133] in, The polar coordinates of the target relative to the projection device; This is the original angle.

[0134] Step S2032: Predict the predicted position of the target after a preset prediction time.

[0135]

[0136]

[0137] in, The predicted location for crossing the target; The preset prediction time can be set to 0.5 seconds, which allows the location to be determined 0.5 seconds in advance, enhancing the timeliness of the warning.

[0138] Preferably, the preset prediction time is 0.3 seconds to 0.8 seconds; in this embodiment, the prediction time is used. =0.5 seconds.

[0139] Step S2033: Calculate the predicted projection angle based on the relative coordinates between the predicted position and the installation position of the projection device.

[0140] Calculate the lateral and longitudinal differences of the predicted position relative to the projection device:

[0141]

[0142] in, The relative coordinates between the predicted location and the installation location of the projection device; This is the offset of the projection device. The projection device is installed inside the vehicle's headlight assembly, and its installation position is offset relative to the origin of the vehicle's coordinate system.

[0143] Furthermore, the predicted projection angle is calculated:

[0144] in, To predict the projection angle.

[0145] Step S2034: Determine whether a lateral safety offset angle needs to be superimposed. This applies when considering the lateral distance across the target. When the distance is less than the lateral safety distance threshold, a lateral safety offset angle is added to the predicted projection angle:

[0146] in, For the lateral safety offset angle, The lateral safety boundary factor ranges from 3 to 7 degrees; in this embodiment, it is set to 5 degrees. When the target is on the left, the bias is to the left; when the target is on the right, the bias is to the right.

[0147] Step S2035: Calculate the final projection angle based on the predicted projection angle and the lateral safety offset angle.

[0148] in, This is the final projection angle.

[0149] Preferably, the final projection angle is limited to the mechanical rotation range [MIN_PROJECTOR_ANGLE, MAX_PROJECTOR_ANGLE] of the projection device to prevent exceeding the mechanically permissible range.

[0150] In this embodiment, the projection control module 404 specifically involves the on-board processor sending control commands to the projection device, which then projects warning patterns onto the ground in front of vehicles in adjacent lanes according to the final projection angle. The warning patterns include a flashing light wall pattern and pedestrian warning icons.

[0151] Example 3 This embodiment provides a vehicle, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps of the above-described vehicle start-up coordination warning method based on intelligent projection.

[0152] In one implementation, its hardware architecture includes a vision sensor unit, a ranging sensor unit, a traffic light sensing unit, and a V2X communication unit.

[0153] The vision sensor unit includes a wide-angle camera, which is installed on the inside of the vehicle's windshield or at the vehicle logo, with a horizontal field of view of not less than 120 degrees, and is used to acquire images of the front and side front of the vehicle; the image resolution of the wide-angle camera is 1920 pixels × 1080 pixels, the frame rate is not less than 30 frames per second, and the output image is a digital image.

[0154] The ranging sensor unit includes millimeter-wave radar or lidar, which is installed inside the front bumper of the vehicle to acquire distance and speed information of targets crossing the road. The millimeter-wave radar operates at a frequency of 77 GHz, with a ranging range of 0.5 meters to 150 meters, a speed range of -30 meters per second to 30 meters per second, and an angular resolution of not less than 1 degree.

[0155] The traffic light sensing unit is connected to the vision sensor unit and obtains the traffic light status through V2X communication or image recognition by the vision sensor unit. The traffic light sensing unit first obtains the traffic light phase status and remaining countdown information through the V2X communication unit. When V2X communication is unavailable, the traffic light status is obtained by color recognition of the image collected by the vision sensor unit.

[0156] The V2X communication unit is used to obtain the phase status and remaining countdown information of traffic lights via wireless communication.

[0157] It also includes a projection device, integrated into the vehicle's headlight assembly or on both sides of the front bumper, used to project warning patterns onto the ground in front of vehicles in adjacent lanes according to control commands. The projection device is a laser projection device, a digital light processing (DLP) projection device, or a liquid crystal on silicon (LCOS) projection device; the installation position of the projection device allows it to project onto the ground in front of the vehicle at a tilt angle of 30 to 60 degrees, with a projection brightness of no less than 2000 lumens to ensure clear visibility of the projected pattern under strong daylight conditions; the projection device has a built-in electronically controlled steering mechanism, which can adjust the projection direction according to the final projection angle under the control of the projection control module, with a horizontal rotation range of ±30 degrees and a tilt rotation range of 30 to 60 degrees.

[0158] Existing technologies rely on driving cloud subsystems and vehicle-to-everything (V2X) communication to obtain camera images of obstructed vehicles, resulting in system failure in areas without network coverage or with high communication latency. Existing technologies either require installing radar on intersection gantries and embedding light strips in the road surface, leading to high modification costs and coverage only at already modified intersections. This embodiment employs a multi-sensor fusion solution combining a wide-angle camera and millimeter-wave radar. All perception, computation, and execution functions are completed within the vehicle itself. V2X communication is used as a preferred method, not a necessary condition, for traffic light perception, with visual recognition as a backup, achieving complete vehicle autonomy. The projection device is integrated into the headlight assembly or front bumper, utilizing existing front-end installation space without requiring additional structural modifications. The beneficial effects achieved by this embodiment are as follows: the vehicle's autonomous perception capability enables it to operate normally in areas with insufficient V2X infrastructure coverage, and its applicable scope covers all intersections; the projection device is installed at a tilt angle of 30 to 60 degrees, and at a typical intersection distance of 5 to 10 meters, the ground projection area is approximately 0.5 to 2 square meters, which is sufficient to produce a visual warning effect that can be perceived by the driver, while the projection brightness is not less than 2000 lumens to ensure daytime visibility; the modular four-layer architecture, including information acquisition, risk assessment, projection control, and projection device, facilitates integration with existing vehicle domain controllers, and each module can be reused with the vehicle's existing sensor system via CAN bus or Ethernet, reducing hardware costs.

[0159] Example 4 This embodiment provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the above-described vehicle start-up coordinated warning method based on intelligent projection.

[0160] Example 5 This embodiment provides a computer program product including executable instructions, which are computer instructions; the executable instructions are stored in a computer-readable storage medium. When the processor of an electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, the steps of the above-described vehicle start-up cooperative warning method based on intelligent projection are completed.

[0161] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0162] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0163] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0164] The steps and methods involved in Embodiments 2 to 5 above correspond to those in Embodiment 1. For detailed implementation methods, please refer to the relevant description section of Embodiment 1. The descriptions of each embodiment in the above embodiments have different emphases. Parts not described in detail in a certain embodiment can be referred to the relevant descriptions of other embodiments.

[0165] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0166] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A vehicle start-up coordinated warning method based on intelligent projection, characterized in that, include: Detect vehicle status information and traffic light status, and obtain the position and motion status information of the target crossing in front of the vehicle; Based on the motion state information of the crossing target and the estimated motion state of vehicles in adjacent lanes, the conflict time and positional distance at the moment of conflict of the crossing target to the adjacent lane are calculated, and the collision risk level is determined based on the conflict time and positional distance at the moment of conflict. When the collision risk level reaches a preset high-risk threshold, the predicted position of the target is predicted based on the velocity vector of the target, the predicted angle of the projection device is calculated based on the predicted position, and the lateral safety offset angle is introduced to obtain the final projection angle. According to the final projection angle, the projection device is controlled to project a warning pattern onto the ground in front of vehicles in adjacent lanes to warn drivers in adjacent lanes.

2. The vehicle start-up coordinated warning method based on intelligent projection as described in claim 1, characterized in that, When the vehicle speed is detected to be lower than the preset stationary speed threshold and the gear is in driving gear, and the traffic light is detected to meet the start trigger conditions, a risk scanning window of preset duration is activated; within the risk scanning window, the position information and motion status information of the target crossing in front of the vehicle are perceived through the fusion of the camera and the ranging sensor.

3. The vehicle start-up coordinated warning method based on intelligent projection as described in claim 1, characterized in that, Based on the motion state information of the crossing target and the estimated motion state of vehicles in adjacent lanes, the conflict time of the crossing target reaching the adjacent lane and the positional distance at the time of conflict are calculated, specifically as follows: Using the centerline position of the adjacent lane as a reference, calculate the collision time of the crossing target reaching the center of the adjacent lane based on the lateral velocity component of the crossing target: in, The lateral coordinates of the center lines of adjacent lanes The current horizontal coordinates of the target are used for crossing. The lateral velocity component is the velocity component that crosses the target. Substituting the time of conflict into the equations for the trajectory crossing the target and the motion equations for vehicles in adjacent lanes, the longitudinal positions of both at the time of conflict are calculated: The positional spacing at the moment of conflict is: in, The positional spacing at the moment of conflict. The longitudinal position of the target at the moment of conflict. The longitudinal position of vehicles in adjacent lanes at the moment of conflict. Let y be the starting acceleration of a vehicle in the adjacent lane, t be time, and y be the acceleration of the vehicle in the adjacent lane. t To traverse the target's current vertical coordinates, vy t This represents the longitudinal velocity component that traverses the target.

4. The vehicle start-up coordinated warning method based on intelligent projection as described in claim 1, characterized in that, The collision risk level is determined based on the conflict time and the distance between the locations. When the distance between the locations at the time of the conflict is less than the safe distance threshold and the conflict time is less than the reaction time threshold, the collision risk level is determined to be high risk. When the distance between the locations at the time of the conflict is less than the safe distance threshold and the conflict time is not less than the reaction time threshold, the collision risk level is determined to be medium risk. In all other cases, the collision risk level is determined to be no risk.

5. The vehicle start-up coordinated warning method based on intelligent projection as described in claim 1, characterized in that, Based on the predicted position, the predicted angle of the projection device is calculated as follows: in, To predict the projection angle, The relative coordinates between the predicted location and the installation location of the projection device.

6. The vehicle start-up coordinated warning method based on intelligent projection as described in claim 1, characterized in that, The process of introducing a lateral safety offset angle to obtain the final projection angle specifically includes: Determine whether a lateral safety offset angle needs to be superimposed. When the lateral distance across the target is less than the lateral safety distance threshold, a lateral safety offset angle is superimposed on the predicted projection angle. The lateral safety offset angle is: in, For the lateral safety offset angle, For the lateral safety boundary factor, The current horizontal coordinates of the target being traversed; The final projection angle is calculated based on the predicted projection angle and the lateral safety offset angle: in, For the final projection angle, To predict the projection angle.

7. A vehicle start-up cooperative warning system based on intelligent projection, characterized in that, include: The data acquisition module is configured to detect vehicle status information and traffic light status, and acquire the position and motion status information of the target crossing in front of the vehicle. The risk assessment module is configured to: calculate the conflict time and positional distance at the moment of conflict of the crossing target to the adjacent lane based on the motion state information of the crossing target and the estimated motion state of vehicles in the adjacent lane; and determine the collision risk level based on the conflict time and positional distance at the moment of conflict. An angle calculation module is configured to: when the collision risk level reaches a preset high-risk threshold, predict the predicted position of the target based on the velocity vector of the target, calculate the predicted angle of the projection device based on the predicted position, and introduce a lateral safety offset angle to obtain the final projection angle. The projection control module is configured to control the projection device to project a warning pattern onto the ground in front of vehicles in adjacent lanes according to the final projection angle, so as to warn drivers in adjacent lanes.

8. A vehicle, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the vehicle start-up cooperative warning method based on intelligent projection as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, complete the vehicle start-up cooperative warning method based on intelligent projection as described in any one of claims 1-6.

10. A computer program product comprising executable instructions stored in a computer-readable storage medium; wherein, When the vehicle's processor reads executable instructions from a computer-readable storage medium and executes the executable instructions, it completes the vehicle start-up cooperative warning method based on intelligent projection as described in any one of claims 1-6.

Images