Mobile traffic safety reminding device based on artificial intelligence

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to a mobile traffic safety reminding device based on artificial intelligence, which comprises a data acquisition module and an unmanned aerial vehicle, and further comprises a construction module, a simulation module, a monitoring module and a warning module, and judging whether the vehicle completes the guidance task according to the real-time track and the continuous image of the vehicle, if not, determining that the vehicle is abnormal, issuing a warning voice to the abnormal vehicle, and predicting the abnormal vehicle track of the abnormal vehicle according to the continuous image. And generating an avoidance track for vehicles around the abnormal vehicle according to the abnormal vehicle track through the virtual road model, continuously collecting images of the surrounding vehicles, and continuously issuing prompt voice to the surrounding vehicles until the surrounding vehicles enter the avoidance track, so as to help a traffic control department to judge the state of a driver. And the waste of human resources of the traffic management department in the traffic safety reminding process is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a mobile traffic safety reminder device based on artificial intelligence. Background Art

[0002] A traffic accident generally refers to an incident in which a vehicle on the road causes personal injury, death or property damage to others due to fault or accident. Generally speaking, traffic accidents are not only caused by unspecified people violating the Road Traffic Safety Law, but may also be caused by natural disasters such as earthquakes, typhoons, floods and other force majeure reasons.

[0003] In life, common causes of traffic accidents are the use of sedatives, sudden illness, drunk driving, fatigue driving, etc. These will cause drivers to experience adverse reactions such as drowsiness, distraction, reduced or blurred vision, and these reactions often reduce the driver's driving ability, thereby causing traffic accidents.

[0004] At the same time, in real life, roads under construction are also very likely to cause traffic accidents. This is because during the road construction process, the construction team needs to be stationed on the lane to work. In the existing technology, warning signs are often placed to warn passing vehicles, but this method requires the vehicle driver to actively perceive the warning. If the driver's reaction ability decreases due to various reasons at this time, the probability of an accident will be greatly increased.

[0005] The current solution is to set up several checkpoints through the traffic management department to conduct surprise inspections on drivers, and summarize the inspection results to the traffic management department's processing end for archiving.

[0006] Although the above solution can reduce the occurrence of such traffic accidents to a certain extent, it requires a large amount of manpower investment, which is a waste of human resources. In addition, the driver's status is judged and handled by manpower, which has a certain lag when facing a traffic accident that is about to occur or has already occurred on the road. It is difficult to promptly guide the surrounding vehicles that are driving normally, which may cause a chain accident. Summary of the Invention

[0007] To solve the above problems, the present invention provides a mobile traffic safety reminder device based on artificial intelligence, which is used to help traffic management departments judge the driver's status and reduce the waste of human resources of traffic management departments during traffic safety reminders.

[0008] To achieve the above-mentioned object, the technical solution of the present invention is as follows: a mobile traffic safety reminder device based on artificial intelligence, comprising a data acquisition module and a drone, further comprising a construction module, a simulation module, a monitoring module and a warning module, wherein the monitoring module is mounted on the drone; A construction module is used to receive a road map input by a user, construct a virtual road model based on the road map, and supplement the virtual road model in combination with traffic rules; The data acquisition module is used to divide the road into several acquisition segments, each of which contains the initial position of the road, and sequentially acquire image information of vehicles on the road; The simulation module is used to obtain the vehicle's driving state based on the image information of the initial position acquisition segment after the vehicle enters the road section to be tested, simulate the vehicle's driving state, and obtain the predicted trajectory of driving in a legal state; A monitoring module is used to sequentially compare the image information of each acquisition segment with the corresponding position of the predicted trajectory. When the vehicle's speed in the image information of the vehicle acquired in any acquisition segment does not match the speed data of the vehicle when it reaches the acquisition segment in the predicted trajectory, the vehicle's trajectory does not match the form trajectory of the vehicle when it reaches the acquisition segment in the predicted trajectory, or the vehicle acceleration increment in the image information acquired in any acquisition segment is greater than a preset increment, the vehicle is determined to have an abnormality. Starting from the acquisition segment, the vehicle image is continuously acquired, a real-time trajectory is generated based on the vehicle image, and the driver's behavior is acquired based on the vehicle image, and the driver's behavior is associated with the real-time trajectory. The warning module is used to issue warnings to the driver and classify the driver's dangerous driving level according to the driver's behavior. Within the preset time, when the driver's dangerous driving level decreases, the continuous image acquisition of the vehicle is stopped. When the driver's dangerous driving level remains unchanged or increases, a guiding task is issued to the driver of the abnormal vehicle, and it is judged based on the real-time trajectory and continuous images of the vehicle whether the vehicle has completed the guiding task. If completed, the vehicle is judged to be normal and the continuous image acquisition of the vehicle is stopped. If not completed, the vehicle is judged to be abnormal and a warning voice is issued to the abnormal vehicle.

[0009] Furthermore, the warning module is also used to connect to the signal of the traffic management department processing end. While issuing the warning voice, it obtains the license plate information of the vehicle based on the image information of the vehicle, and sends the license plate information and real-time trajectory of the vehicle to the traffic management department processing end. The warning module is also used to predict the abnormal trajectory of the abnormal vehicle based on the continuous images of the vehicle after issuing the warning voice, and obtain the normal trajectory of vehicles around the abnormal vehicle based on the image of the abnormal vehicle. According to the road map, the trajectory of the abnormal vehicle and the normal trajectory, the avoidance lane, lane change position, lane change speed and deceleration position are determined for the vehicles around the abnormal vehicle. At the same time, prompt voice is issued to the corresponding surrounding vehicles according to the avoidance lane, lane change position, lane change speed and deceleration position, until the corresponding surrounding vehicles perform corresponding operations at the deceleration position and lane change position according to the lane change speed, and drive into the avoidance lane.

[0010] Furthermore, the construction module is further configured to receive the construction location and construction plan of the road construction team input by the user, and supplement the virtual road model according to the construction location and construction plan.

[0011] Furthermore, the warning module is also used to determine whether the abnormal vehicle coincides with the construction location of the road based on the abnormal vehicle trajectory. If the abnormal vehicle trajectory coincides with the road construction location, a prompt message is issued to the construction team.

[0012] Furthermore, the warning module is also used to calculate the time interval from the issuance of the guidance task to the completion of the guidance task by the vehicle after the vehicle completes the guidance task. When the time interval is less than or equal to the preset value, the vehicle is normal; if the time interval is greater than the preset value, the vehicle is abnormal.

[0013] Furthermore, the simulation module is also used to connect to the meteorological bureau network to obtain real-time weather information and adjust the forecast trajectory based on the weather information.

[0014] Furthermore, the data acquisition module is also used to remove noise from image information.

[0015] Furthermore, the warning module is also used to obtain the number of vehicles based on the image information collected in any collection segment within a unit time. When the number of vehicles is greater than a preset value, a prompt message is sent to the traffic management department processing end.

[0016] Furthermore, the road map includes the road section location, length, width, number of lanes, lane layout, ancillary facilities information and sidewalk information. The warning unit is also used to determine whether the abnormal vehicle trajectory coincides with the sidewalk based on the abnormal vehicle trajectory and the road map. When the two coincide, the warning module plays a prompt voice to pedestrians at the overlapping position.

[0017] Furthermore, the simulation module is also used to adjust the predicted trajectory of vehicles that subsequently enter the road based on the image information.

[0018] The technical principles and beneficial effects of the above scheme are as follows: 1. This solution issues directional tasks to abnormal vehicles and judges the driver's status based on the completion status of the abnormal vehicles. When the driver is in an abnormal state, such as fatigue driving, drunk driving, and drunk driving, the driver responds correctly. At this time, the driver is warned and a reasonable avoidance trajectory is planned for surrounding vehicles to guide the surrounding vehicles to avoid the abnormal vehicle, thereby reducing the probability of collision between abnormal vehicles and normal vehicles and causing traffic accidents. Compared with the existing technology, this solution has a lower degree of human participation and can effectively reduce the waste of human resources of the traffic management department. It can not only warn drivers who may be suspected of dangerous driving and reduce the risk of safety accidents caused by continuing to drive, but also plan reasonable routes for normal vehicles around to help such vehicles avoid abnormal vehicles and further reduce the occurrence of traffic accidents.

[0019] 2. This solution constructs a virtual road model and a virtual vehicle model, and uses these two virtual models to predict the driving status of real vehicles entering the road. If the vehicle information status in the subsequently collected image information does not match the predicted status, the vehicle may be driving abnormally. Compared with the existing technology, this solution uses predicted trajectories to screen vehicles, so that only vehicles with abnormal driving status are imaged in a mobile manner. This reduces the number of mobile image acquisitions such as drones, as well as the amount of calculation and computing time for subsequent abnormal vehicle determination, which helps to reduce the cost of using this solution.

[0020] 3. This solution collects vehicle data in segments by setting up collection segments. This collection solution makes this solution compatible with the existing traffic monitoring network. When collecting image information, the images collected by the traffic monitoring network can be used without the need for additional cameras, thereby reducing the cost of this solution.

[0021] 4. This solution classifies the driver's dangerous driving level according to the driver's driving status. After the initial warning, if the driver's dangerous driving level decreases, the driver has a certain normal reaction ability and no subsequent verification link is performed. If the driver's dangerous driving level does not change or increases, the driver's status is obviously abnormal at this time. At this time, the driver will be subject to subsequent verification links. By screening abnormal vehicles, the number of vehicles that need to be verified and alleviated in the future is greatly reduced, which helps to reduce the amount of data processing during the use of this solution.

[0022] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1A schematic diagram of an embodiment of a mobile traffic safety reminder device based on artificial intelligence according to the present invention; Figure 2 This is a structural diagram of an embodiment of a mobile traffic safety reminder device based on artificial intelligence of the present invention; Figure 3 This is a flow chart of an embodiment of the mobile traffic safety reminder device based on artificial intelligence of the present invention.

[0024] The reference numerals in the drawings of the specification include: 1. UAV. DETAILED DESCRIPTION

[0025] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0026] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0027] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0028] The following is further described in detail through specific implementation methods: Example:

[0029] As attached Figure 1 -Attached Figure 3 As shown: A mobile traffic safety reminder device based on artificial intelligence includes a data acquisition module and a drone 1, and the drone 1 is provided with a construction module, a simulation module, a monitoring module and a warning module.

[0030] The construction module is used to receive the road map input by the user. The road map contains the road section location, length, width, number of lanes, lane layout, ancillary facilities information and sidewalk information of the monitored road section. A virtual road model is constructed based on the road map, and combined with the driving principles in the laws and regulations related to traffic safety, the speed and driving mode of each lane in the virtual road model are restricted to form a guiding task.

[0031] The construction module is also used to receive input of the location and construction plan of the construction team monitoring the road construction, and to supplement and adjust the virtual road model according to the construction location. At the same time, it makes directional task adjustments according to the degree of impact of the construction plan on the surrounding environment, so that the construction location on the virtual road model follows the changes in the actual construction progress. By adjusting the virtual road model according to the construction location and construction plan, the virtual road model is made to fit the real road more closely, thereby improving the accuracy of the subsequently generated predicted trajectory, and thus improving the accuracy of abnormal vehicle judgment in this scheme. At the same time, the construction plan is used to adjust the construction location on the virtual road model. While avoiding frequent collection of real construction team data, it can also make the virtual road model generally follow the changes in the real road, further improving the authenticity of the virtual road model and the accuracy of the subsequent predicted trajectory.

[0032] The data acquisition module includes several cameras (in this embodiment, the cameras are installed next to the road to be measured). Several acquisition sections are set up on the monitored section, and one of the acquisition sections is located at the starting position of the monitored section. Image information of vehicles on the road is collected in sequence starting from the starting position of the monitored section. The data acquisition module is also used to remove noise from the image information, thereby reducing the impact on subsequent recognition.

[0033] The simulation module obtains the vehicle's characteristics, such as vehicle color, signal, and license plate information, based on the image information collected at the initial position of the monitored road after the vehicle enters the monitored road. At the same time, it obtains the vehicle's driving status, such as vehicle speed and driving habits, based on the image information. According to the vehicle's driving status, a corresponding virtual vehicle model is constructed, and the virtual vehicle model and virtual road model are used to simulate the trajectory that the vehicle should travel under legal conditions, that is, the predicted trajectory.

[0034] The simulation module is also used to connect to the meteorological bureau network, obtain real-time weather information through the meteorological bureau network, and adjust the predicted trajectory according to the weather information. In severe weather (such as rainy days), normally driving cars will also experience drastic changes in speed. At this time, adjusting the predicted trajectory according to weather changes can make the predicted trajectory more consistent with the normal actual driving trajectory, further reducing the probability of misjudgment; the simulation module is also used according to the amount of image information. When the amount of image information is small, the traffic volume is small at this time, and subsequent vehicles entering the monitoring lane can travel at a normal speed. If there is a lot of image information, the traffic volume is large at this time, and subsequent vehicles entering the monitoring lane are difficult to travel at a normal speed. At this time, the predicted trajectory of subsequent vehicles entering is adjusted to adapt to the actual driving conditions of the road, further reducing the probability of misjudgment.

[0035] The monitoring module is used to compare the image information with the vehicle driving status at the corresponding position in the predicted trajectory each time the image information of the vehicle is collected in the collection section. When the vehicle driving status in the image information corresponds to the vehicle driving status at the corresponding position in the predicted trajectory, the vehicle is in a normal driving state. When the vehicle driving speed in the image information obtained in any collection section does not match the corresponding data of the predicted trajectory (that is, the vehicle may be speeding or slower than the normal driving speed until the vehicle behind is blocked, etc.), the vehicle driving trajectory does not match the corresponding data of the predicted trajectory (for example, in the predicted trajectory, when the vehicle reaches the collection section, the vehicle should be in a straight driving state, and its predicted trajectory is a straight line, and at this time the image information collected in the collection section does not match the vehicle driving speed). , when the vehicle's driving trajectory is a curve or a wavy line, then the vehicle's driving trajectory does not match the predicted trajectory) or when the vehicle acceleration increment in the image information obtained in any acquisition segment is greater than the preset increment (when the acceleration increment is greater than the preset value, the vehicle may experience rapid deceleration or rapid acceleration), the vehicle is in an abnormal driving state. At this time, the drone 1 is started to follow the vehicle for shooting, continuously collect images of the vehicle, and generate a corresponding real-time trajectory based on the vehicle image. The driver's behavior actions are obtained based on the vehicle image, and the driver's behavior actions are associated with the real-time trajectory. For example, when the vehicle turns left, the driver needs to observe the road conditions, turn on the turn signal, and then turn the steering wheel.

[0036] For example, according to the predicted trajectory, when the vehicle passes through the second acquisition segment, the vehicle speed is A km / h, the vehicle is driving stably in the middle of the lane, and the time required to move from the acquisition segment of the initial position to the second acquisition segment is B min. However, in the image information actually collected in the second acquisition segment, if the difference between the vehicle speed and A is greater than the set value, the vehicle is not driving in the middle of the lane, crosses the lane, drives into another lane, and the time required to move from the initial acquisition segment to the second acquisition segment is much less than any one or more of B, then the vehicle is determined to be in an abnormal driving state.

[0037] The warning module is connected to the signal processing end of the traffic management department and is used to issue warnings to the driver. The driver's dangerous driving level is divided according to the driver's behavior and actions. The division is based on the degree of correlation with the occurrence of traffic accidents. The higher the correlation, the higher the level. For example, the danger level when the driver lowers his head to pick up an item is greater than the danger level when the direction of the driver's turn signal is inconsistent with the actual turning direction. The danger level when the direction of the driver's turn signal is inconsistent with the actual turning direction is greater than the danger level when the driver looks directly at the road conditions and speaks. Within the preset time, when the driver's dangerous driving level decreases, the continuous image acquisition of the vehicle is stopped. When the driver's dangerous driving level remains unchanged or increases, a guiding task is issued to the driver of the abnormal vehicle, such as through broadcasting. The loudspeaker plays "Driver of license plate number XXX, please perform the following actions..." At this time, it is judged whether the vehicle has completed the guidance task based on the real-time trajectory and the continuous images of the vehicle. For example, if the guidance task is to alternately turn on the high beam and low beam, the image information shows that the vehicle turns on the high beam and then turns off the high beam, and then turns on the low beam at the same time; or if the guidance task is to enter a certain lane, the real-time trajectory of the vehicle changes accordingly. If the vehicle is normal, the continuous image acquisition of the vehicle by the drone 1 is stopped. If the vehicle cannot complete the corresponding guidance task, or the completed action is inconsistent with the guidance task, the vehicle is abnormal. At this time, a warning voice is played to the vehicle through the loudspeaker, such as "Driver of vehicle with license plate number XXX, you may be suspected of dangerous driving. Please pull over." At the same time, based on the continuous images of the vehicle, the possible subsequent trajectory of the vehicle is predicted, that is, the abnormal vehicle trajectory is generated. If there are other vehicles around the vehicle at this time, the normal trajectory of the vehicles around the abnormal vehicle is obtained based on the image of the abnormal vehicle. According to the road map, the abnormal vehicle trajectory and the normal trajectory, the avoidance lane, lane change position, lane change speed and deceleration position of the vehicles around the abnormal vehicle are predicted, and the avoidance trajectory cannot overlap with the abnormal vehicle trajectory. Then, prompt voice is continuously played to the surrounding vehicles. The prompt voice includes the avoidance lane, lane change position, lane change speed and deceleration position, such as "Please pay attention, driver of license plate number ×××, the vehicle in front of you is driving dangerously, please adjust the speed by ×, and adjust to lane × after × meters. At the same time, based on the continuous images of other vehicles, it determines whether the corresponding vehicle has entered the avoidance trajectory, and stops playing the prompt voice only after the corresponding vehicle enters the avoidance trajectory. The warning module is also used to use image information to obtain the license plate information of the vehicle while issuing the prompt voice to the abnormal vehicle, and send the license plate information and real-time trajectory of the vehicle to the traffic management department processing end. By establishing a quick connection with the traffic management department processing end when an abnormal vehicle appears, the traffic management department's police response time is shortened, further reducing the probability of traffic accidents. At the same time, by sending the license plate and real-time trajectory to the traffic management department, it is convenient for the traffic management department to locate the vehicle. If the vehicle escapes, it can also be quickly tracked, shortening the speed of accident handling.

[0038] The warning module is also used to simulate the abnormal vehicle trajectory on the virtual road model based on the abnormal vehicle trajectory. When the abnormal vehicle trajectory coincides with the road construction location on the virtual road model, the abnormal vehicle may collide with the road construction location. At this time, a reminder message is issued to the construction team members through text messages or horns, prompting them to evacuate the dangerous area in time, or to implement corresponding protective measures to avoid casualties caused by the abnormal vehicle losing control and reduce public property losses caused by dangerous driving.

[0039] The warning module is also used to calculate the time interval from the issuance of the guidance task to the completion of the guidance task after the vehicle completes the guidance task. This time interval can reflect the driver's reaction speed to a certain extent. If the time interval is less than or equal to the preset value, the driver's reaction speed is normal and the vehicle is a normal driving vehicle. If the time interval is greater than the preset data, the driver's reaction speed is too slow and fatigue driving may occur. Although the vehicle has completed the guidance task, it is still an abnormal vehicle. At this time, the abnormal vehicle handling method is still followed, and a warning voice is issued to the vehicle, and corresponding avoidance trajectories are generated for surrounding vehicles.

[0040] The warning component is also used to calculate the number of vehicles in the acquisition segment per unit time based on image information. When the abnormal vehicle trajectory coincides with the sidewalk in the road map, the corresponding vehicle is very likely to impact the sidewalk and cause a safety accident. At this time, a prompt voice is played to pedestrians at the overlapping position, such as "Dangerous driving vehicles may appear here, please pay attention to avoid them." The voice is played in a loop until the real-time trajectory of the vehicle passes through the location, or it is confirmed through the real-time trajectory that the vehicle cannot reach the location. The warning module is also used to calculate the number of vehicles in the acquisition segment per unit time based on image information. When the number is greater than the set value, the acquisition segment is in a state of dense traffic flow. At this time, a prompt message is sent to the traffic management department to prompt the traffic management department to handle it, thereby further reducing the occurrence of traffic safety accidents.

[0041] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A mobile traffic safety reminder device based on artificial intelligence, comprising a data acquisition module and a drone (1), characterized in that: It also includes a construction module, a simulation module, a monitoring module and a warning module, wherein the monitoring module is installed on the drone (1); A construction module is used to receive a road map input by a user, construct a virtual road model based on the road map, and supplement the virtual road model in combination with traffic rules; The data acquisition module is used to divide the road into several acquisition segments, each of which contains the initial position of the road, and sequentially acquire image information of vehicles on the road; The simulation module is used to obtain the vehicle's driving state based on the image information of the initial position acquisition section of the vehicle entering the road after the vehicle enters the road section to be tested, simulate the vehicle's driving state, and obtain the vehicle's predicted driving trajectory; A monitoring module is used to sequentially compare the image information of each acquisition segment with the corresponding position of the predicted trajectory. When the vehicle's speed in the image information of the vehicle acquired in any acquisition segment does not match the speed data of the vehicle when it reaches the acquisition segment in the predicted trajectory, the vehicle's trajectory does not match the form trajectory of the vehicle when it reaches the acquisition segment in the predicted trajectory, or the vehicle acceleration increment in the image information acquired in any acquisition segment is greater than a preset increment, the vehicle is determined to have an abnormality. Starting from the acquisition segment, the vehicle image is continuously acquired, a real-time trajectory is generated based on the vehicle image, and the driver's behavior is acquired based on the vehicle image, and the driver's behavior is associated with the real-time trajectory. The warning module is used to issue warnings to the driver and classify the driver's dangerous driving level according to the driver's behavior. Within the preset time, when the driver's dangerous driving level decreases, the continuous image acquisition of the vehicle is stopped. When the driver's dangerous driving level remains unchanged or increases, a guiding task is issued to the driver of the abnormal vehicle, and it is judged based on the real-time trajectory and continuous images of the vehicle whether the vehicle has completed the guiding task. If completed, the vehicle is judged to be normal and the continuous image acquisition of the vehicle is stopped. If not completed, the vehicle is judged to be abnormal and a warning voice is issued to the abnormal vehicle.

2. The mobile traffic safety reminder device based on artificial intelligence according to claim 1 is characterized in that: The warning module is also used to connect to the signal of the traffic management department processing end. While issuing the warning voice, it obtains the license plate information of the vehicle based on the image information of the vehicle, and sends the license plate information and real-time trajectory of the vehicle to the traffic management department processing end. The warning module is also used to predict the abnormal trajectory of the abnormal vehicle based on the continuous images of the vehicle after issuing the warning voice, and obtain the normal trajectory of vehicles around the abnormal vehicle based on the image of the abnormal vehicle. According to the road map, the trajectory of the abnormal vehicle and the normal trajectory, the avoidance lane, lane change position, lane change speed and deceleration position are determined for the vehicles around the abnormal vehicle. At the same time, prompt voice is issued to the corresponding surrounding vehicles according to the avoidance lane, lane change position, lane change speed and deceleration position until the corresponding surrounding vehicles perform corresponding operations at the deceleration position and lane change position according to the lane change speed, and drive into the avoidance lane.

3. The mobile traffic safety reminder device based on artificial intelligence according to claim 2 is characterized in that: The construction module is further used to receive the construction position and construction plan of the road construction team input by the user, and supplement the virtual road model according to the construction position and construction plan.

4. The mobile traffic safety reminder device based on artificial intelligence according to claim 3 is characterized in that: The warning module is also used to determine whether the abnormal vehicle coincides with the construction location of the road based on the abnormal vehicle trajectory. If the abnormal vehicle trajectory coincides with the road construction location, a prompt message is issued to the construction team.

5. The mobile traffic safety reminder device based on artificial intelligence according to claim 4 is characterized in that: The warning module is also used to calculate the time interval from the issuance of the guidance task to the completion of the guidance task by the vehicle after the vehicle completes the guidance task. When the time interval is less than or equal to the preset value, the vehicle is normal; if the time interval is greater than the preset value, the vehicle is abnormal.

6. The mobile traffic safety reminder device based on artificial intelligence according to claim 5 is characterized in that: The simulation module is also used to connect to the meteorological bureau network to obtain real-time weather information and adjust the forecast trajectory based on the weather information.

7. The mobile traffic safety reminder device based on artificial intelligence according to claim 6 is characterized in that: The data acquisition module is also used to remove noise from image information.

8. The mobile traffic safety reminder device based on artificial intelligence according to claim 7 is characterized in that: The warning module is also used to obtain the number of vehicles based on the image information collected in any collection segment within a unit time. When the number of vehicles is greater than a preset value, a prompt message is sent to the traffic management department processing end.

9. The mobile traffic safety reminder device based on artificial intelligence according to claim 8, characterized in that: The road map includes the road section location, length, width, number of lanes, lane layout, ancillary facilities information and sidewalk information. The warning unit is also used to determine whether the abnormal vehicle trajectory coincides with the sidewalk based on the abnormal vehicle trajectory and the road map. When the two coincide, the warning module plays a prompt voice to pedestrians at the overlapping position.

10. The mobile traffic safety reminder device based on artificial intelligence according to claim 9, characterized in that: The simulation module is also used to adjust the predicted trajectory of vehicles that subsequently enter the road based on the image information.

Images