Expressway risk early warning method, device and equipment and readable storage medium

Abstract

The application discloses a highway risk early warning method, device and equipment and a readable storage medium. The method comprises the following steps: obtaining a target area with a safety hidden danger on a highway, and obtaining driving information of a vehicle in the target area; based on the driving information, simulation data of the vehicle driving in the target area is obtained; if the simulation data has driving risks, warning information is output to a display unit which is not passed through by the vehicle, so that the display unit displays the warning information; the warning information is used for prompting a driver of the vehicle that the current driving state has driving risks; the display unit is arranged beside the highway, and the interval between every two adjacent display units is a preset distance. The application realizes monitoring of the vehicle in the target area with the safety hidden danger on the highway, and makes corresponding prediction and early warning prompt on the driving state of the vehicle, so that the driving safety is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of highway early warning technology, and in particular to a highway risk early warning method, device, equipment and readable storage medium. BACKGROUND

[0002] On a highway, there are wet and slippery sections due to natural causes such as wind, rain, ice and snow. When a vehicle travels at high speed on the section, the vehicle has certain safety hazards. In addition, there are congestions on the highway due to construction or traffic accidents. Vehicles traveling at high speed on the highway may not know the conditions of the road section in front, which may cause the driver to not timely control the vehicle to brake and avoid, thereby causing traffic accidents and other problems. Therefore, the road section also has safety hazards.

[0003] Therefore, in the above-mentioned road section, the vehicle has certain safety hazards, and the driver may misjudge or not timely control when driving to the road section, thereby causing the vehicle to have a high risk of driving in the above-mentioned road section. SUMMARY

[0004] Therefore, the present application provides a highway risk early warning method, device, equipment and readable storage medium, which aims to improve the safety of vehicles driving on the highway.

[0005] To achieve the above-mentioned purpose, the present application provides a highway risk early warning method, which comprises the following steps:

[0006] Obtain a target area on a highway that has safety hazards, and obtain driving information of a vehicle in the target area;

[0007] Based on the driving information, simulation data of the vehicle driving in the target area is simulated;

[0008] If the simulation data has driving risk, output warning information to a display unit that the vehicle has not passed, so that the display unit displays the warning information; the warning information is used to prompt the driver of the vehicle that the current driving state has driving risk; the display unit is arranged beside the highway, and the interval between every two adjacent display units is a preset distance.

[0009] Illustratively, the simulation data includes a predicted driving route and a driving state, and the simulation data of the vehicle driving in the target area is simulated based on the driving information, which includes:

[0010] Based on the driving information, determine the speed information, light information, attitude information and position information of the vehicle;

[0011] determining lane information in which the vehicle is currently located in the target area based on the position information;

[0012] simulating a predicted driving route of the vehicle in the target area based on the lane information and the light information;

[0013] simulating a driving state of the vehicle in the target area based on the speed information and the attitude information.

[0014] simulating a predicted driving route of the vehicle in the target area based on the lane information and the light information, including:

[0015] determining a traffic condition on a to-be-driven section in front of the vehicle based on the lane information;

[0016] simulating a plurality of driving routes in which the vehicle normally passes through the to-be-driven section based on the traffic condition;

[0017] determining a driving direction of the vehicle based on the light information;

[0018] if the driving direction coincides with any one of the plurality of driving routes, simulating a predicted driving route in which the vehicle has no driving risk in the target area.

[0019] simulating a driving state of the vehicle in the target area based on the speed information and the attitude information, including:

[0020] counting an attitude change condition of the vehicle in a preset time length based on the attitude information;

[0021] simulating an attitude change trend of the vehicle based on the attitude change condition;

[0022] simulating a spatial trajectory of the vehicle when driving in the target area based on the attitude change trend and the speed information;

[0023] if the spatial trajectory coincides with a spatial trajectory of another vehicle, simulating a driving state in which the vehicle has driving risk in the target area.

[0024] simulating a predicted driving route of the vehicle in the target area based on the lane information and the light information, including:

[0025] obtaining road condition information of the expressway;

[0026] inputting the road condition information into a detection model to obtain a classification result;

[0027] determine a target area with a safety hazard on the expressway based on the classification result; the detection model is obtained by iteratively training an image classification model based on a road condition information dataset; and the road condition information training dataset is obtained by image acquisition on the expressway.

[0028] For example, before the inputting of the road condition information into the detection model and the obtaining of the classification result, the method comprises:

[0029] obtaining training samples of different road conditions;

[0030] inputting the training samples into an image classification model, and obtaining the detection model after the training of the image classification model is completed.

[0031] For example, the inputting of the training samples into the image classification model and the obtaining of the detection model after the training of the image classification model is completed comprises:

[0032] inputting the training samples into the image classification model, classifying the training samples, and obtaining training classification labels;

[0033] calculating the gradient of the image classification model based on the training classification labels and preset true labels corresponding to the training samples;

[0034] determining whether the image classification model meets a preset iterative training end condition based on the gradient;

[0035] if yes, the image classification model is used as the detection model;

[0036] if no, the iterative training of the image classification model is continued until the image classification model meets the preset iterative training end condition.

[0037] For example, to achieve the above object, the application further provides an expressway risk early warning device, which comprises:

[0038] an acquisition module configured to acquire a target area with a safety hazard on an expressway and acquire driving information of a vehicle in the target area;

[0039] a simulation module configured to simulate simulation data of the vehicle driving in the target area based on the driving information;

[0040] a judgment module configured to output warning information to a display unit not passed by the vehicle if the simulation data has driving risk, so that the display unit displays the warning information; the warning information is used to prompt a driver of the vehicle that the current driving state has driving risk; and the display unit is arranged beside the expressway, and the interval between every two adjacent display units is a preset distance.

[0041] Exemplarily, to achieve the above object, the application further provides a highway risk early warning device, which comprises a memory, a processor and a highway risk early warning program stored in the memory and executable on the processor, and the highway risk early warning program is configured to implement the steps of the highway risk early warning method.

[0042] Exemplarily, to achieve the above object, the application further provides a computer readable storage medium, which stores a highway risk early warning program, and the highway risk early warning program implements the steps of the highway risk early warning method when executed by a processor.

[0043] Compared with the prior art, in which there are some road sections with safety hazards on the highway, the risk of vehicle driving on the road sections is high, and the driver may be involved in a risk accident due to negligence, judgment error or untimely operation of the vehicle, etc. In the application, the target area with safety hazards on the highway is obtained, and the driving information of the vehicle in the target area is obtained. According to the driving information, the driving action of the vehicle driving in the target area is simulated to obtain corresponding simulation data, and the simulation data is analyzed and judged to determine whether the driving action of the vehicle corresponding to the simulation data has safety hazards. If the simulation data has safety hazards, warning information is output to the display unit not passed by the vehicle, so that the display unit displays the warning information. The warning information is used to prompt the driver of the vehicle that the current driving state has driving risk. At the same time, the display unit is arranged beside the highway, and the vehicle will pass a display unit every driving preset distance, so as to ensure that the driver of the vehicle can see the relevant warning information, thereby improving the safety of driving on the highway. That is, by focusing on the target area with safety hazards on the highway, the driving state of the vehicle in the target area is simulated, and when the driving state of the vehicle has safety hazards, the warning information is output to prompt the driver of the vehicle to avoid risks, thereby improving the safety of driving on the highway. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is a flowchart of a first embodiment of the highway risk early warning method of the application;

[0045] Figure 2 is a flowchart of a second embodiment of the highway risk early warning method of the application;

[0046] Figure 3 is a structural diagram of a hardware running environment related to the embodiment of the application.

[0047] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0048] It should be understood that the specific embodiments described herein are merely exemplary and are not intended to limit the present application.

[0049] The present application provides a highway risk early warning method, referring to Figure 1 , Figure 1 The flowchart of the first embodiment of the highway risk early warning method of the present application.

[0050] The present application provides an embodiment of the highway risk early warning method, and it should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in an order different from that described herein. In order to facilitate description, the following omits the description of the execution subject of each step of the highway risk early warning method, and the highway risk early warning method comprises:

[0051] Step S110: obtaining a target area with safety hazards on the highway, and obtaining driving information of a vehicle in the target area;

[0052] The highway may be affected by the environment, construction or traffic accidents on the highway, and there may be safety hazards in some sections of the area. This area is the target area.

[0053] For example, the safety hazards existing in the target area include: the road is slippery due to rain and snow, resulting in safety hazards of vehicle braking being affected, the road has poor visibility due to heavy fog, and the driver's attention is not concentrated during vehicle driving, resulting in collision, scratching and other situations with other lanes or vehicles in front, and the vehicle lane is occupied due to the construction area of the highway, resulting in the driver of the vehicle unknowingly driving into the construction area, causing vehicle damage or threatening the personal safety of the driver.

[0054] At the same time, the target area is further divided into a risk area and a detection area. The risk area is a specific road section area with safety hazards such as slippery road sections or construction road sections, wherein the risk area also includes accident-prone road sections, such as sharp turning road sections, road sections with visual blind area, etc. Traffic accidents often occur on the road section. The detection area is a road section that vehicles must pass through when entering the risk area. The driving information of the vehicle is detected on the road section, so as to facilitate warning the vehicle with driving risk when the vehicle enters the risk area.

[0055] The driving information is information related to the driving of the vehicle.

[0056] The driving information includes, for example, speed information, light information, attitude information, and position information of the vehicle.

[0057] The speed information includes, for example, current speed, acceleration, and the like of the vehicle.

[0058] The light information includes, for example, left turn signal, right turn signal, brake light, and the like of the vehicle.

[0059] The attitude information includes, for example, an attitude of the vehicle driving along a curve, an attitude of the vehicle keeping straight driving, and the like.

[0060] The position information includes, for example, a position of the vehicle, a distance between the vehicle and another vehicle, a distance between the vehicle and another obstacle, a lane in which the vehicle is located, and the like.

[0061] The target area with a safety hazard on the expressway is obtained, for example, by:

[0062] Step a: obtaining road condition information of the expressway.

[0063] The road condition information includes, for example, image information and data information, and is a detailed explanation of a current road condition of the expressway, such as a current traffic volume on the expressway, a predicted construction position of the expressway, and an abnormal event (congestion, traffic accident, and the like) on the expressway.

[0064] The road condition information is obtained, for example, by obtaining image information through a camera on the expressway and determining a construction position of the expressway according to a road construction prediction scheme.

[0065] The image information needs to be further determined to obtain the road condition information of the expressway.

[0066] The construction position is usually marked on a construction drawing, and thus the specific construction section of the expressway is determined by determining specific position parameters shown on the construction drawing.

[0067] Step b: inputting the road condition information into a detection model to obtain a classification result.

[0068] The detection model is a neural network classification model, which can recognize the image information and classify the image information according to the training content, so as to obtain the classification result.

[0069] The classification standard of the detection model is, for example, a classification judgment process of judging whether a traffic accident exists in each section of the expressway or whether a construction area exists in each section of the expressway, so as to obtain the corresponding classification result.

[0070] Exemplarily, the step of inputting the road condition information into the detection model and obtaining the classification result includes:

[0071] Step c: Obtain training samples of different environmental road conditions;

[0072] The training samples are data sets for training the neural network model. Using this data set, the neural network learns the classification and judgment method. For example, the training samples include image information when a traffic accident occurs on a highway (features such as road congestion and abnormal posture of the vehicle involved in the traffic accident), and image information when the highway is unobstructed (features such as normal vehicle driving posture and no congestion), so that the neural network model learns to classify the image information of these two contents.

[0073] Different environmental road conditions include road conditions under different weather conditions and road conditions under different traffic phenomena.

[0074] For example, information on road conditions in different environments includes situations that pose risks to vehicle driving, such as heavy fog (low visibility), rainy and snowy weather (slippery roads), as well as road condition information including traffic accidents, road condition information with smooth traffic, and road condition information including traffic congestion caused by construction areas on highways.

[0075] The above data is used as training samples to train the neural network model.

[0076] Step d: Input the training samples into the image classification model, and obtain the detection model after the image classification model completes training.

[0077] The training samples are input into the image classification model to train the image classification model so that the image classification model learns to classify images of different road conditions. For example, the classification results include risky road conditions (traffic accidents, slippery roads, low road visibility, etc.) and risk-free road conditions (smooth and unobstructed roads, and no external environment affecting vehicle driving, etc.).

[0078] After the image splitting model is trained, the detection model is obtained.

[0079] Exemplarily, inputting the training sample into the image classification model, and obtaining the detection model after the image classification model completes training, includes:

[0080] Step e: inputting the training samples into the image classification model, classifying the training samples, and obtaining training classification labels;

[0081] When the training samples are input into the image classification model for training, the image classification model will classify the corresponding image information. The initial classification result is based on the general properties of the image information.

[0082] Thus, after classification, the corresponding classification results are marked to obtain training classification labels, for example, in the training process, the number of vehicles contained in the image information is classified, or the similarity of the road segments in the image information is taken as the classification standard, and so on.

[0083] Step f: based on the training classification label and the preset true label corresponding to the training sample, the gradient of the image classification model is calculated;

[0084] According to the actual effect of accurately classifying the road condition information of the expressway, the related labels are set in advance, for example, the road condition information is divided into risk road segments and non-risk road segments, and so on. The type of label is a preset true label, which is used as a label evaluation standard for the image classification model. When the training classification label obtained by the image classification model in the training process is consistent with the preset true label, it is determined that the training is completed, and a detection model capable of detecting whether a road segment has a risk is obtained.

[0085] The difference between the evaluation of the training classification label and the preset true label is the gradient of the image classification model. The gradient needs to be calculated according to the training process parameters and classification result parameters of the image classification model. The calculation method is a general method, which will not be described here.

[0086] Step g: based on the gradient, it is determined whether the image classification model satisfies the preset iterative training end condition;

[0087] According to the gradient size, the similarity between the training classification label of the image classification model and the preset true label is determined. When the gradient satisfies the evaluation condition, it is proved that the training task of the training model has reached the expected goal, that is, the image classification model has obtained a detection model that meets the established classification requirements after iterative training.

[0088] The preset iterative training is the evaluation standard of the training image classification model set in advance. When the gradient meets the evaluation standard, it is determined that the current image classification model has been trained.

[0089] Step h: if it is satisfied, the image classification model is used as a detection model;

[0090] Step i: if it is not satisfied, the image classification model is iteratively trained until the image classification model satisfies the preset iterative training end condition.

[0091] When the gradient satisfies the preset iterative training end condition, the training action of the image classification model is completed. When the gradient does not satisfy the preset iterative training end condition, the image classification model is continuously trained until the gradient satisfies the preset iterative training end condition.

[0092] Step j: determining a target area with a safety hazard on the expressway based on the classification result; the detection model is obtained by iteratively training an image classification model based on a road condition information dataset; and the road condition information training dataset is obtained by image acquisition on the expressway.

[0093] According to the classification result, the classification result includes the classified target area with a safety hazard and the ordinary road section area without a safety hazard, so the target area with a safety hazard on the expressway can be determined from the classification result.

[0094] The road condition information training dataset is a training sample composed of road condition information.

[0095] Step S120: simulating to obtain simulation data of the vehicle driving in the target area based on the driving information;

[0096] According to the driving information, the simulation data of the vehicle in the target area is simulated, and the simulation data has multiple simulation results, for example, the vehicle keeps accelerating or decelerating, for example, the vehicle changes lanes, and the like. The vehicle may drive on a route, and the vehicle may drive in a state.

[0097] The simulation data is a simulation of the vehicle according to the driving information, which satisfies the possible route and driving state of the vehicle through the target area.

[0098] The exemplary simulation data includes the predicted driving route and driving state of the vehicle.

[0099] Among them, according to the driving information of the vehicle, the speed change of the vehicle in the target area is determined, and according to the speed change, the acceleration state or deceleration state or the state of the vehicle driving at a constant speed is determined in a simulated manner.

[0100] Among them, according to the driving information of the vehicle, the position of the vehicle and the action of the vehicle: keeping straight or entering a turning state, etc. Action, thereby simulating the predicted driving route of the vehicle.

[0101] Among them, the simulation result of the simulation data has multiple possibilities, for example, the current driving information satisfies that multiple driving routes of the vehicle can pass through the target area.

[0102] Step S130: if the simulation data has driving risk, output warning information to the display unit not passed by the vehicle, so that the display unit displays the warning information; the warning information is used to prompt the driver of the vehicle that the current driving state has driving risk; the display unit is arranged beside the expressway, and the interval between every two adjacent display units is a preset distance.

[0103] For example, the simulation data exists driving risk situations include driving too fast on the wet road, speeding into the curve, driving without reducing speed in front of the obstacle, low-speed driving in front of the vehicle in heavy fog, rear vehicle driving too fast exists rear-end collision driving track and so on.

[0104] In which, the vehicle will produce side slip, drift and other unstable braking conditions on the wet road, thereby affecting the vehicle braking effect, when the vehicle braking effect is affected, it will make the vehicle unable to produce emergency braking effect when the obstacle appears in front of the vehicle, and at this time the vehicle speed is too fast, exceeding the predetermined speed, for example, 60km / h or 70km / h, and the vehicle cannot produce emergency braking effect on the wet road.

[0105] In which, the vehicle enters the curve at a speed, which will cause the vehicle to drive into the curve at a speed that is too fast to avoid obstacles after entering the curve, or the vehicle will produce side slip and other conditions due to road conditions, affecting the stability of vehicle driving, at this time, the vehicle enters the curve at a speed, which exists corresponding safety hazards.

[0106] In which, when there is an obstacle in front of the vehicle, for example, traffic accidents or construction areas appear in front of the vehicle, the driver of the vehicle does not notice such phenomenon, and the driver of the vehicle still drives at high speed, which will produce collision with the obstacle, at this time, when the vehicle drives at high speed towards the obstacle, there is a certain safety hazard.

[0107] In which, in heavy fog weather, due to low visibility, the driver of the vehicle will be distracted or the driver will make a mistake, which will cause rear-end collision with the front vehicle, at this time, according to the speed and driving direction of the vehicle, it can be determined that if the vehicle keeps the same speed, it will collide with the front vehicle, in the usual sunny environment, the driver can directly see the distance between the front vehicle and the vehicle, but in heavy fog weather, the driver's vision is limited, if the front vehicle brakes in emergency, and the driver of the rear vehicle does not react in time, or the speed of the rear vehicle is too fast to directly brake in emergency to avoid collision with the front vehicle, at this time, in heavy fog weather, the vehicle speed is too fast, which exists safety hazards.

[0108] If the simulation data exists driving risk, it is determined that the vehicle will produce corresponding risk if it continues to drive at the current driving state, then the warning information will be output.

[0109] In which, the warning information is information for prompting the driver of the vehicle, for example, prompting the driver of the vehicle to slow down when entering the front area, slow down when the front area is under construction, or slow down when the front area has traffic accidents and the like.

[0110] The warning information is output to a display unit, the display unit is a plurality of units arranged beside the expressway, the unit can display prompt information, and a distance between every two adjacent display units is a preset distance, the distance is 200 meters or 500 meters, that is, a display unit is arranged every 200 meters or every 500 meters.

[0111] Compared with the prior art, in the prior art, there are some road sections with safety hazards on the expressway, the risk of vehicle driving on the road sections is high, and the driver will be in a risk accident due to negligence, judgment error or untimely operation of the vehicle. In the present application, the target region with safety hazards on the expressway is obtained, and the driving information of the vehicle in the target region is obtained. According to the driving information, the driving action of the vehicle driving in the target region is simulated to obtain corresponding simulation data, and the simulation data is analyzed and judged to determine whether the vehicle driving action corresponding to the simulation data has safety hazards. If the simulation data has safety hazards, warning information is output to the display unit not passed by the vehicle, so that the display unit displays the warning information. The warning information is used to prompt the driver of the vehicle that the current driving state has driving risk. At the same time, the display unit is arranged beside the expressway, and the vehicle will pass through a display unit every preset distance. Therefore, the driver of the vehicle can see the relevant warning information, thereby improving the safety of driving on the expressway. That is, by focusing on the target region with safety hazards on the expressway, the driving state of the vehicle in the target region is simulated, and when the driving state of the vehicle has safety hazards, the warning information is output to prompt the driver of the vehicle to avoid risks, thereby improving the safety of driving on the expressway.

[0112] For example, with reference to Figure 2 , Figure 2 is a flowchart of a second embodiment of the expressway risk early warning method of the present application. Based on the first embodiment of the expressway risk early warning method of the present application, the second embodiment is proposed. The method further comprises:

[0113] Step S210: determining the speed information, light information, attitude information and position information of the vehicle based on the driving information;

[0114] For example, the speed information, light information, attitude information and position information of the vehicle contained in the driving information are determined, and the four types of information are further determined.

[0115] Among them, the acceleration information in the vehicle speed information, the current vehicle speed information of the vehicle, the deceleration information of the vehicle, etc. are determined. According to the speed information of the vehicle, the farthest distance of the vehicle driving in a certain time in the future can be determined.

[0116] The vehicle light information is determined, so as to determine whether the vehicle uses the vehicle light at present, and if the vehicle light is used, the type of the vehicle light used by the vehicle is determined, which is the left turn light, the right turn light or the brake light. According to the light information of the vehicle, the driving direction or the driving state of the vehicle can be determined. For example, the brake light determines that the vehicle is currently braking, the turn light determines the driving direction of the vehicle, and no turn light indicates that the vehicle keeps straight driving.

[0117] The attitude information of the vehicle is determined, so as to determine the attitude of the vehicle in space, such as the attitude state when driving into a curve, and the driving attitude of the vehicle in the lane, whether the vehicle is pressing the line or occupying multiple lanes.

[0118] The position information of the vehicle is determined, so as to determine the distance between the vehicle and the vehicle, the distance between the vehicle and the risk area in the target area, and the lane in which the vehicle is currently located, such as the speed limit of the leftmost lane and the rightmost lane.

[0119] Step S220: determining the lane information in which the vehicle is currently located in the target area based on the position information;

[0120] According to the position information, the lane information in which the vehicle is located can be determined, which includes the position of the lane and the speed limit of the lane, so that according to the speed limit information of different lanes, whether the current driving speed of the vehicle is overspeed can be determined.

[0121] At the same time, according to the position of the lane, the driving route of the vehicle is determined, such as keeping driving on the straight lane, or driving on the lane into the ramp, and the future driving route of the vehicle is different on different lanes.

[0122] Step S230: simulating the predicted driving route of the vehicle in the target area based on the lane information and the light information;

[0123] The predicted driving route is a simulation of the possible driving route of the vehicle in the future according to the driving information of the vehicle.

[0124] For example, the vehicle keeps straight driving, and the current speed is unchanged, and the acceleration is positive, so that a plurality of predicted driving routes can be simulated, such as keeping straight driving and accelerating through the front target area, or the vehicle changes lane during driving and passes through the front target area.

[0125] Therefore, by comprehensively considering the lane information and the light information, the expected driving route of the vehicle can be further determined. For example, if the vehicle is currently driving on a straight lane, it will be directly determined that the vehicle will keep straight if the light information of the vehicle is not considered. However, if the light information of the vehicle is considered, it will be determined that the vehicle will change lane to the left according to the left turn signal, thereby avoiding the situation that too many expected driving routes are simulated.

[0126] For example, the simulation of the expected driving route of the vehicle in the target area based on the lane information and the light information comprises:

[0127] Step k: determining the traffic condition on the to-be-traveled road section in front of the vehicle based on the lane information;

[0128] The lane information determines the lane position of the vehicle, and the number of lanes on the expressway is a fixed value, which is four or five lanes. Therefore, according to the lane information, the number of vehicles on the same lane can be determined.

[0129] For example, if the number of vehicles on the current lane is not more than 10 or not more than 15, it is determined that the current traffic condition is smooth, and if the number of vehicles on the current lane is more than 20 or 30, it is determined that the current traffic condition is congested. Similarly, according to the number of vehicles, the traffic condition is determined, and the specific evaluation standard of the number is determined according to the actual situation of the expressway.

[0130] The to-be-traveled road section is the road section to be traveled in front of the vehicle, that is, the road section that must be traveled by the vehicle.

[0131] For example, U-turn is not allowed on the expressway, and according to relevant regulations, the vehicle cannot exit the ramp after entering the ramp, that is, the direction of the vehicle driving on the expressway is usually one-way. Therefore, in order to reach the destination through the expressway, all road sections in the one-way driving direction of the vehicle are to-be-traveled road sections, and the road section behind the vehicle is not calculated.

[0132] Step l: simulating a plurality of driving routes of the vehicle normally passing through the to-be-traveled road section based on the traffic condition;

[0133] According to the traffic condition (smooth traffic or congested traffic), a plurality of driving routes of the vehicle normally passing through the to-be-traveled road section are simulated. For example, there are usually a plurality of lanes in one driving direction of the expressway, that is, the vehicle can drive through the to-be-traveled road section on the leftmost lane, the middle lane or the rightmost lane.

[0134] For example, when the traffic condition of the lane is smooth, the vehicle can smoothly pass through the lanes, and when the traffic condition of the lane is congested, the vehicle needs to avoid the congested lane and select a lane with fewer vehicles, for example, the number of vehicles in the leftmost lane is the largest (causing congestion and frequent braking), the number of vehicles in the middle lane is the second largest, and the number of vehicles in the rightmost lane is the smallest. At this time, when the vehicle normally passes through the to-be-traveled road section, the route of driving in the middle lane or the route of driving in the rightmost lane is the normal passing route of the to-be-traveled road section.

[0135] Step m: determining the driving direction of the vehicle based on the light information;

[0136] According to the light information of the vehicle, the driving direction of the vehicle is determined, that is, the direction of the above-mentioned turn signal is determined to determine the driving direction of the vehicle.

[0137] Step n: if the driving direction coincides with any of the plurality of driving routes, the expected driving route of the vehicle in the target area without driving risk is simulated.

[0138] The plurality of driving routes is the expected route when normally driving through the to-be-traveled road section, and according to the driving direction of the vehicle, there are coinciding and non-coinciding situations with the plurality of driving routes. For example, the plurality of driving routes includes driving in the leftmost lane and driving in the rightmost lane, and the vehicle is currently in the middle lane, and the vehicle does not turn on the left turn signal or the right turn signal. The driving direction of the vehicle is determined to be straight, that is, at this time, the driving direction does not coincide with the plurality of driving routes, and at this time, the vehicle drives in a straight manner, which may cause driving risk of the vehicle, for example, congestion in the middle lane, which is easy to cause rear-end phenomenon, or there is a construction area in front of the middle lane, and after the vehicle drives to the front, there is a situation that the vehicle cannot normally drive due to the construction area.

[0139] Therefore, according to the coincidence of the driving direction and the plurality of driving routes, it is determined whether the expected driving route of the vehicle has driving risk.

[0140] If the driving direction of the vehicle coincides with the plurality of driving routes, the expected driving route of the vehicle in the target area without driving risk is simulated, and if the driving direction of the vehicle does not coincide with the plurality of driving routes, the expected driving route of the vehicle in the target area with driving risk is simulated.

[0141] Step S240: simulating the driving state of the vehicle in the target area based on the speed information and the attitude information.

[0142] According to the speed information and the attitude information, the driving state of the vehicle in the target area is simulated, which includes the driving speed state of the vehicle and the action state of the vehicle driving.

[0143] For example, according to the speed information, the acceleration driving state of the vehicle or the deceleration driving state of the vehicle is determined, and for another example, in combination with the speed information and the attitude information of the vehicle, the acceleration turning-in driving state or the deceleration turning-in driving state of the vehicle is determined.

[0144] For example, the simulation of the driving state of the vehicle in the target area based on the speed information and the attitude information includes:

[0145] Step o: based on the attitude information, the attitude change of the vehicle in a preset time length is counted;

[0146] The attitude information is the action attitude of the vehicle in space, for example, the vehicle body is parallel to the driving direction of the lane, or the vehicle body intersects with the lane, and so on, wherein there is a situation that the road condition affects the attitude of the vehicle, for example, the ice layer is formed on the highway due to the ice and snow environment, and the vehicle will produce side slip or drift phenomenon when driving on the ice layer.

[0147] Therefore, when analyzing and simulating the attitude information of the vehicle, the attitude change of the vehicle in a preset time length needs to be collected and analyzed to count the continuous attitude change of the vehicle, for example, continuous lane changing, continuous turning of the vehicle, continuous sliding phenomenon of the vehicle on the ice layer, and so on.

[0148] The preset time length is a fixed detection time, for example, ten minutes or twenty minutes, and the attitude of the vehicle is continuously monitored in the preset time length, that is, the continuous attitude change is obtained when the attitude information of the vehicle is obtained.

[0149] The attitude change is the change of the continuous action of the attitude of the vehicle, for example, the vehicle has multiple lane changing driving modes in the preset time length, changes from the leftmost lane to the rightmost lane, or the vehicle continuously slides due to the ice layer on the lane in the preset time length, and produces fluctuating attitude change, left and right micro shaking, and so on.

[0150] Step p: based on the attitude change, the attitude change trend of the vehicle is simulated;

[0151] According to the attitude change, a motion trajectory of the vehicle in the preset time length can be determined, so as to determine the motion trend of the vehicle, for example, the vehicle continuously changes lanes from the leftmost lane to the rightmost lane, and it is detected that the vehicle changes from the leftmost lane to the middle lane in the preset time length, and the right turn signal of the vehicle is not turned off, so it can be determined that the vehicle still has the trend of changing lanes to the right side, that is, the attitude change trend of the vehicle is simulated.

[0152] Therefore, the posture change trend of the vehicle is similar to the predicted driving route of the vehicle, and the posture of the vehicle is taken as a main simulation feature to simulate the motion trend of the vehicle, wherein the simulation of the motion trend of the vehicle includes the posture action of the vehicle in the same lane, for example, when the road is slippery, the vehicle will slightly slide, and for another example, when there is an obstacle on the road, the vehicle needs to avoid the obstacle in front, that is, the vehicle needs to drive to the other lane, bypass the obstacle, and then the vehicle returns to the original lane.

[0153] Step q: simulating the spatial trajectory of the vehicle when driving in the target area based on the posture change trend and the speed information;

[0154] According to the posture change trend and the speed information, the spatial trajectory of the vehicle when driving in the target area is determined, for example, the posture of the vehicle when the vehicle is in the same lane and shakes, or the posture of the vehicle when the vehicle enters a curve and catches up with a vehicle in front of the curve.

[0155] For example, when the vehicle drives in the lane of the vehicle, the vehicle will be in contact with other vehicles due to the left or right deviation of the vehicle in different lanes, therefore, the posture change trend and the speed information of the vehicle are simulated to simulate the spatial trajectory of the vehicle when driving in the target area.

[0156] According to the spatial trajectory, it can be determined whether the vehicle will be in contact with other vehicles, or whether the vehicle will be in contact with obstacles, construction area barriers, etc. on the route passed by the vehicle.

[0157] Step r: if the spatial trajectory coincides with the spatial trajectory of the other vehicle, a driving state in which the vehicle has a driving risk in the target area is simulated.

[0158] If the spatial trajectory coincides with the spatial trajectory of the other vehicle, it is determined that when the vehicle drives at the current speed and posture, the vehicle and the other vehicle will collide or contact, thereby simulating a driving state in which the vehicle has a driving risk in the target area.

[0159] If the spatial trajectory does not coincide with the spatial trajectory of the other vehicle, it is determined that when the vehicle drives at the current speed and posture, the vehicle and the other vehicle will not collide or contact, thereby simulating a driving state in which the vehicle does not have a driving risk in the target area.

[0160] In the embodiment, the speed information, the light information, the attitude information and the position information of the vehicle are determined according to the driving information of the vehicle, the lane information in which the vehicle is currently located in the target area is determined according to the position information, the expected driving route of the vehicle in the target area is simulated further according to the lane information and the light information, the driving state of the vehicle in the target area is simulated further according to the speed information and the attitude information, and whether there is a safety hazard in driving of the vehicle in the target area is determined by comprehensively considering the expected driving route and the driving state.

[0161] In addition, the application further provides a highway risk early warning device, which comprises:

[0162] The acquisition module is configured to acquire a target area with a safety hazard on the highway and acquire driving information of a vehicle in the target area.

[0163] The simulation module is configured to simulate simulation data of driving of the vehicle in the target area based on the driving information.

[0164] The judgment module is configured to output warning information to a display unit not passed through by the vehicle if the simulation data has a driving risk, so that the display unit displays the warning information; the warning information is used to prompt a driver of the vehicle that the current driving state has a driving risk; and the display unit is arranged beside the highway, and a distance between every two adjacent display units is a preset distance.

[0165] The simulation module comprises:

[0166] The first determination sub-module is configured to determine speed information, light information, attitude information and position information of the vehicle based on the driving information.

[0167] The second determination sub-module is configured to determine lane information in which the vehicle is currently located in the target area based on the position information.

[0168] The first simulation sub-module is configured to simulate an expected driving route of the vehicle in the target area based on the lane information and the light information.

[0169] The second simulation sub-module is configured to simulate a driving state of the vehicle in the target area based on the speed information and the attitude information.

[0170] The first simulation sub-module comprises:

[0171] The first determination unit is configured to determine a traffic condition on a to-be-driven section in front of the vehicle based on the lane information.

[0172] The first simulation unit is configured to simulate a plurality of driving routes of the vehicle normally passing through the to-be-traveled road section based on the traffic condition.

[0173] The second determination unit is configured to determine a driving direction of the vehicle based on the light information.

[0174] The second simulation unit is configured to simulate a predicted driving route of the vehicle without driving risk in the target area if the driving direction coincides with any one of the plurality of driving routes.

[0175] The second simulation sub-module includes, for example:

[0176] The statistical module is configured to statistically determine a posture change condition of the vehicle within a preset time length based on the posture information.

[0177] The third simulation unit is configured to simulate a posture change trend of the vehicle based on the posture change condition.

[0178] The fourth simulation unit is configured to simulate a spatial trajectory of the vehicle when driving in the target area based on the posture change trend and the speed information.

[0179] The fifth simulation unit is configured to simulate a driving state of the vehicle with driving risk in the target area if the spatial trajectory coincides with a spatial trajectory of another vehicle.

[0180] The acquisition module includes, for example:

[0181] The acquisition sub-module is configured to acquire road condition information of the expressway.

[0182] The input sub-module is configured to input the road condition information to a detection model to obtain a classification result.

[0183] The third determination sub-module is configured to determine a target area with a safety hazard on the expressway based on the classification result; the detection model is obtained by iteratively training an image classification model based on a road condition information dataset; and the road condition information training dataset is obtained by image acquisition on the expressway.

[0184] The input sub-module includes, for example:

[0185] The acquisition unit is configured to acquire training samples of different environmental road conditions.

[0186] The input unit is configured to input the training samples to an image classification model, and obtain a detection model after the image classification model is trained.

[0187] The input unit includes, for example:

[0188] An input subunit, configured to input the training samples into an image classification model, classify the training samples, and obtain training classification labels;

[0189] A calculation subunit, configured to calculate a gradient of the image classification model based on the training classification labels and preset true labels corresponding to the training samples;

[0190] A determination subunit, configured to determine, based on the gradient, whether the image classification model satisfies a preset iterative training end condition;

[0191] A first judgment subunit is configured to use the image classification model as a detection model if the conditions are met;

[0192] The second judgment subunit is used to continue iterative training of the image classification model if the condition is not met until the image classification model meets the preset iterative training end condition.

[0193] The specific implementation of the highway risk warning device of the present application is basically the same as the various embodiments of the above-mentioned highway risk warning method, and will not be repeated here.

[0194] In addition, this application also provides a highway risk warning device. Figure 3 As shown, Figure 3 It is a structural diagram of the hardware operating environment involved in the embodiment of the present application.

[0195] For example, Figure 3 This is a structural diagram of the hardware operating environment of the highway risk warning equipment.

[0196] like Figure 3 As shown, the highway risk warning device may include a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302 and the memory 303 communicate with each other via the communication bus 304, the memory 303 is used to store computer programs; the processor 301 is used to implement the steps of the highway risk warning method when executing the program stored in the memory 303.

[0197] The communication bus 304 mentioned in the highway risk warning device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus 304 can be divided into an address bus, a data bus, and a control bus. For ease of illustration, the figure shows only one thick line, but this does not mean that there is only one bus or only one type of bus.

[0198] The communication interface 302 is configured to communicate between the expressway risk early warning device and other devices.

[0199] The memory 303 can include a random access memory (RMD), and can also include a non-volatile memory (NM), such as at least one disk memory. Optionally, the memory 303 can also be at least one storage device located away from the aforementioned processor 301.

[0200] The aforementioned processor 301 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0201] The expressway risk early warning device embodiment of the present application is basically the same as the aforementioned expressway risk early warning method embodiments, and will not be repeated here.

[0202] In addition, the present application also proposes a computer readable storage medium, and the computer readable storage medium stores an expressway risk early warning program. When the expressway risk early warning program is executed by a processor, the steps of the expressway risk early warning method described above are implemented.

[0203] The computer readable storage medium embodiment of the present application is basically the same as the aforementioned expressway risk early warning method embodiments, and will not be repeated here.

[0204] It should be noted that, in this document, the term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or system. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or system including the element.

[0205] The above application embodiment serial numbers are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0206] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the methods described in various embodiments of the present application.

[0207] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A highway risk early warning method, characterized in that: The highway risk early warning method comprises the following steps: Obtaining road condition information of a highway, inputting the road condition information into a detection model to obtain a classification result, and determining target areas on the highway where safety hazards exist based on the classification result, wherein training samples of the detection model include different environmental road conditions, including road conditions under different weather conditions and road conditions under different traffic phenomena; Obtaining a target area on the highway where potential safety hazards exist, and obtaining driving information of vehicles in the target area; Based on the driving information, simulate and obtain simulation data of the vehicle driving in the target area, the simulation data including an estimated driving route and driving status; Determining the speed information, lighting information, posture information, and position information of the vehicle based on the driving information, and determining the posture of the vehicle in space through the posture information; Based on the position information, determining lane information where the vehicle is currently located within the target area; Simulating an expected driving route of the vehicle in the target area based on the lane information and the light information; The lighting information includes the use of vehicle lights and the type of vehicle lights used, and the driving direction or driving state of the vehicle is determined based on the lighting information; simulating a driving state of the vehicle in the target area based on the speed information and the posture information; If there is a driving risk in the simulated data, a warning message is output to a display unit that the vehicle has not passed through, so that the display unit displays the warning message; the warning message is used to remind the driver of the vehicle that there is a driving risk in the current driving state; the display unit is set next to the highway, and the interval between each two adjacent display units is a preset distance.

2. The highway risk early warning method according to claim 1, characterized in that: The simulating, based on the lane information and the light information, an expected driving route of the vehicle in the target area includes: Determining a traffic condition on a road section ahead of the vehicle based on the lane information; Based on the traffic conditions, simulating multiple driving routes of the vehicle normally passing through the road section to be driven; determining a driving direction of the vehicle based on the light information; If the driving direction coincides with any one of the plurality of driving routes, an estimated driving route of the vehicle in the target area without driving risk is obtained by simulation.

3. The highway risk early warning method according to claim 1, characterized in that: The simulating the driving state of the vehicle in the target area based on the speed information and the posture information includes: Based on the posture information, statistics are collected on the posture changes of the vehicle within a preset time period; Based on the posture change situation, simulating the posture change trend of the vehicle; Based on the posture change trend and the speed information, simulating a spatial trajectory of the vehicle when traveling in the target area; If the spatial trajectory coincides with the spatial trajectory of another vehicle, a driving state of the vehicle in which there is a driving risk in the target area is obtained through simulation.

4. The highway risk early warning method according to claim 1, characterized in that: The step of obtaining target areas on the highway with potential safety hazards includes: Obtain highway traffic information; Inputting the road condition information into the detection model to obtain a classification result; Based on the classification results, the target area on the highway where safety hazards exist is determined; the detection model is obtained by iteratively training the image classification model based on the road condition information dataset; the road condition information training dataset is obtained by collecting images of the highway.

5. The highway risk early warning method according to claim 4, characterized in that: Before inputting the traffic condition information into the detection model and obtaining the classification result, the method includes: Obtain training samples of different environmental road conditions; The training samples are input into the image classification model, and after the image classification model completes training, a detection model is obtained.

6. The highway risk early warning method according to claim 5, characterized in that: The step of inputting the training sample into the image classification model and obtaining the detection model after the image classification model completes training comprises: Input the training samples into the image classification model, classify the training samples, and obtain training classification labels; Calculating the gradient of the image classification model based on the training classification label and the preset true label corresponding to the training sample; Based on the gradient, determining whether the image classification model meets a preset iterative training end condition; If satisfied, the image classification model is used as the detection model; If not, continue to iteratively train the image classification model until the image classification model meets the preset iterative training end condition.

7. A highway risk warning device, characterized in that: The highway risk early warning device includes: an acquisition module, configured to acquire road condition information of a highway, input the road condition information into a detection model, obtain a classification result, and determine a target area on the highway where potential safety hazards exist based on the classification result, wherein the training samples of the detection model include different environmental road conditions, including road conditions under different weather conditions and road conditions under different traffic phenomena; Obtaining a target area on the highway where potential safety hazards exist, and obtaining driving information of vehicles in the target area; a simulation module, configured to simulate and obtain simulation data of the vehicle traveling in the target area based on the driving information, wherein the simulation data includes an estimated driving route and a driving status; Determining the speed information, lighting information, posture information, and position information of the vehicle based on the driving information, and determining the posture of the vehicle in space through the posture information; Based on the position information, determining lane information where the vehicle is currently located within the target area; Simulating an expected driving route of the vehicle in the target area based on the lane information and the light information; The lighting information includes the use of vehicle lights and the type of vehicle lights used, and the driving direction or driving state of the vehicle is determined based on the lighting information; simulating a driving state of the vehicle in the target area based on the speed information and the posture information; A judgment module is configured to output a warning message to a display unit that the vehicle has not passed through if the simulated data indicates a driving risk, so that the display unit displays the warning message; the warning message is configured to prompt the driver of the vehicle that the current driving state indicates a driving risk; the display units are disposed adjacent to the highway, and a preset distance is provided between each two adjacent display units.

8. A highway risk warning device, characterized in that: The device includes: a memory, a processor, and a highway risk warning program stored in the memory and executable on the processor, wherein the highway risk warning program is configured to implement the steps of the highway risk warning method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a highway risk warning program, which, when executed by a processor, implements the steps of the highway risk warning method according to any one of claims 1 to 6.

Images