Accident early warning method and device, electronic equipment and medium
By obtaining vehicle, road and natural environment data, and using accident prediction models to generate and send early warning information, the problem that accident warning in the existing technology depends on the driver's subjective judgment, and the effective warning is provided before a traffic accident occurs, which improves the Road traffic safety.
Patent Information
- Application Number
- CN202411165645.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the accident warning method relies on the driver's subjective judgment and lacks an effective warning mechanism before the accident occurs, resulting in the inability to take timely preventive measures before the traffic accident occurs.
By obtaining the vehicle's driving status data, road status data and natural environment data, inputting the accident prediction model, generating early warning information, and sending early warning information to relevant vehicles, real-time prediction and early warning of the accident can be achieved.
It significantly improves the timeliness and effectiveness of accident warnings, reduces the possibility of accidents, and improves road traffic safety.
Smart Images

Figure CN119028150A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of vehicle safety, and in particular to an accident warning method, device, electronic equipment and medium. Background Art
[0002] In modern traffic systems, with the increasing number of vehicles and the complexity of the traffic environment, road safety issues are becoming increasingly prominent. In related technologies, accident warning methods mostly rely on the driver's subjective judgment to detect potential dangers. The accident perception system in the vehicle will only issue an alarm when an accident has occurred, lacking an effective accident warning mechanism before the accident occurs. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide an accident warning method, device, electronic device and medium, with the aim of providing effective accident warning before a traffic accident occurs.
[0004] In a first aspect, an embodiment of the present application provides an accident warning method, which is applied to an electronic device, wherein the electronic device is communicatively connected to a plurality of vehicles, and the method includes: Acquire driving state data of a first vehicle, acquire road state data of a road on which the first vehicle is driving, and acquire natural environment data of a driving environment of the first vehicle; Inputting the driving state data, the road state data and the natural environment data into an accident prediction model to obtain possible accident information output by the accident prediction model and the accident probability of the possible accident represented by the possible accident information; Generate warning information according to the possible accident information and the probability of the accident occurring; The warning information is sent to the first vehicle and a second vehicle related to the possible accident.
[0005] In some embodiments, the method further comprises: receiving accident information sent by the first vehicle when an accident occurs to the first vehicle; generating an accident emergency response strategy according to the accident information; Emergency resources are deployed according to the accident emergency strategy.
[0006] In some implementations, the accident information includes the accident location, the accident type, and driving status data of the first vehicle before the accident occurs, and generating an accident emergency strategy based on the accident information includes: Acquiring road status data of the accident location; Determine the road impact range of the accident occurring to the first vehicle according to the road state data of the accident location; Determining the severity of the accident occurring to the first vehicle according to the road impact range, the accident type, and the driving state data of the first vehicle before the accident occurred; An accident emergency response strategy is generated according to the severity.
[0007] In some implementations, the accident information includes the accident location, the accident emergency plan includes the type and quantity of required emergency resources, and the allocation of emergency resources according to the accident emergency strategy includes: The available resource information of each resource distribution point obtained from the resource database; Acquire traffic status data of an associated geographical area from a geographic information system, wherein the associated geographical area includes the locations of each of the resource distribution points and the accident location; Determine the resource information and the allocation path of each resource distribution point according to the traffic status data and the available resource information of each resource distribution point; The resource information to be allocated and the allocation path of each resource issuance point are sent to each resource issuance point, so that each resource issuance point allocates resources according to its own resource information to be allocated and the allocation path.
[0008] In some implementations, the road state data includes road traffic flow, and obtaining the road state data of the road on which the first vehicle is traveling includes: Acquire a road image of the road on which the first vehicle is traveling by using an image acquisition device; Performing vehicle recognition on the road image to determine the number of vehicles in the road image; Determining the road traffic flow according to the number of vehicles; and / or, The natural environment data includes weather conditions, and the obtaining of the natural environment data of the first vehicle driving environment includes: Collecting the first vehicle driving environment parameters by an environmental parameter collection device; The environmental parameters are input into a weather recognition model, and the weather state corresponding to the environmental parameters output by the weather recognition model is obtained.
[0009] In some implementations, generating warning information according to the possible accident information and the accident occurrence probability includes: When the probability of the accident is greater than a preset probability threshold, generating first warning information according to the possible accident information and the probability of the accident, the first warning information including an accident avoidance instruction for controlling the first vehicle and a second vehicle related to the possible accident to perform an accident avoidance operation; When the probability of the accident occurrence is less than or equal to the preset probability threshold, a second warning information is generated based on the possible accident information and the probability of the accident occurrence, and the second warning information includes operation suggestions, which are used to prompt the driver of the first vehicle and the driver of the second vehicle related to the possible accident to operate according to the operation suggestions.
[0010] In some implementations, generating warning information according to the possible accident information and the accident occurrence probability includes: Generate warning information according to the possible accident information and the probability of the accident, and record the possible accident in the warning record; The method further comprises: Upon receiving the accident information sent by the first vehicle when the first vehicle has an accident, searching the early warning record for a possible accident whose accident type is the same as the accident type corresponding to the accident information; When a possible accident is found whose accident type is the same as the accident type corresponding to the accident information, the model parameters of the accident prediction model are adjusted.
[0011] In a second aspect, an embodiment of the present application provides an accident warning device, which is applied to an electronic device, wherein the electronic device is communicatively connected to a plurality of vehicles, and the device comprises: A data acquisition module, used to acquire driving state data of the first vehicle, acquire road state data of the road on which the first vehicle is driving, and acquire natural environment data of the driving environment of the first vehicle; A prediction module, used to input the driving state data, the road state data and the natural environment data into an accident prediction model, and obtain possible accident information output by the accident prediction model, and an accident probability of a possible accident represented by the possible accident information; An early warning generation module, used to generate early warning information according to the possible accident information and the probability of the accident; An information sending module is used to send the warning information to the first vehicle and a second vehicle related to the possible accident.
[0012] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the accident warning method provided in the first aspect of the embodiment of the present application are implemented.
[0013] In a fourth aspect, an embodiment of the present application provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by a processor, the processor enables the processor to implement the accident warning method provided in the first aspect of the embodiment of the present application.
[0014] In the embodiment of the present application, the electronic device first acquires and analyzes multi-dimensional data based on vehicle driving status data, road status data and natural environment data. The electronic device then introduces an accident prediction model, which can combine historical data with machine learning algorithms to achieve dynamic real-time prediction, obtain possible accident information and the probability of possible accidents represented by the possible accident information, and finally generate warning information based on the possible accident information and the probability of possible accidents represented by the possible accident information. The generated warning information is not only for the first vehicle where the accident risk occurs, but also can be sent to the second vehicle related to the possible accident. This multi-party warning mechanism significantly improves the timeliness and effectiveness of the warning and reduces the possibility of accidents. Especially in an environment with dense traffic flow, this extensive warning mechanism helps to spread dangerous information faster, prompting other drivers to respond in a timely manner, thereby improving overall road safety. In this way, the electronic device significantly enhances the effect of road traffic accident prevention through data diversification, model intelligence and the wide dissemination of warning information, and realizes the provision of effective accident warnings before traffic accidents occur. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a flowchart of the accident warning method provided in the embodiment of the present application; Figure 2 is a schematic diagram of the structure of an accident warning device provided in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0016] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.
[0017] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.
[0018] The accident warning method, device, electronic device and medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.
[0019] See also Figure 1 , is a flow chart of an accident warning method provided by an embodiment of the present application. In a first aspect, an accident warning method is provided in an electronic device, wherein the electronic device is communicatively connected with a plurality of vehicles, and includes the following steps S100 to S400: Step S100: Acquire driving state data of the first vehicle, acquire road state data of the road on which the first vehicle is driving, and acquire natural environment data of the driving environment of the first vehicle.
[0020] The subject that executes the accident warning method provided in the embodiment of the present application can be a cluster of electronic devices, each of which is connected to multiple vehicles in communication, and each of which can be connected to various data acquisition devices arranged on the road network in communication, and each electronic device can execute the accident warning method provided in the embodiment of the present application. Edge computing devices can be deployed on each data acquisition device arranged on the road network to perform preliminary processing and analysis on the collected data to reduce the amount of data transmission. The electronic device can first obtain a variety of real-time data about the first vehicle through each data acquisition device arranged on the road network, including the following three categories: The driving state data, such as the speed, acceleration, braking state, position, and distance from the vehicle in front of the first vehicle, reflect the current motion state of the first vehicle. For example, the speed of the first vehicle and the distance between the first vehicle and the vehicle in front can be collected and acquired by a radar or a laser radar arranged on the road network, and the position of the first vehicle can be collected and acquired by a global positioning system (GPS) on board the first vehicle.
[0021] The road state data is the road condition information of the road on which the first vehicle is traveling, such as whether the road surface is flat, road obstacle information, road slope, whether there are curves, and road traffic flow, etc. For example, the road traffic flow can be obtained by collecting, acquiring, and analyzing images with high-definition cameras arranged on the road network, and the road obstacle information can be obtained by collecting laser reflection signals with a laser radar arranged on the road network, calculating the reflection time of the laser signals, generating three-dimensional point cloud data of the surrounding road environment, analyzing the point cloud data, and then identifying obstacles on the road.
[0022] Natural environment data: including external environmental factors that affect driving conditions, such as temperature, humidity, wind speed, and light intensity. For example, the temperature can be obtained by thermocouples or RTD sensors arranged on the road network, which measure the ambient temperature and convert it into a voltage or current signal, and then into a temperature value. The humidity can be obtained by a capacitive humidity sensor arranged on the road network. The wind speed can be measured and obtained by an anemometer (such as a hot wire anemometer or a vane anemometer) arranged on the road network. The rainfall can be measured and obtained by a rain gauge (such as a tipping bucket rain gauge) arranged on the road network.
[0023] Step S200: Input driving status data, road status data and natural environment data into the accident prediction model to obtain possible accident information output by the accident prediction model and the accident probability of the possible accident represented by the possible accident information.
[0024] The driving status data, road status data and natural environment data obtained by the electronic equipment are input into a trained accident prediction model. The model calculates the types of accidents that may occur and their probability of occurrence based on historical accident data and machine learning algorithms. The accident prediction model can take into account various complex situations, such as the risk of loss of control when road conditions deteriorate and the risk of collision in low visibility on rainy days.
[0025] Exemplarily, the electronic device inputs the acquired driving status data, road status data and natural environment data into the accident prediction model, and the accident prediction model outputs a result based on the input data: a skidding accident may occur at the sharp bend ahead due to the slippery road surface and the vehicle being accelerated, and the probability of an accident is 35%.
[0026] During the training process of the accident prediction model, the original data is first cleaned, outliers are removed and missing values are filled to ensure the integrity and accuracy of the data. Then, key features are extracted from the original data, such as speed change rate, vehicle spacing fluctuation, traffic flow peak, etc. These features will be used as model inputs. The numerical features are then standardized so that different features have the same scale to ensure stability during the model training process.
[0027] After data processing is completed, the data is divided into training sets, validation sets, and test sets to ensure the generalization ability of the model and reduce the risk of overfitting. Then, according to the characteristics of the data and the type of problem, a suitable deep learning model is selected. For example, a long short-term memory network can be selected for time series data, a convolutional neural network can be selected for image data, and a recurrent neural network can be selected for sequence prediction. Finally, the model is trained using the training set, and the model parameters and hyperparameters are adjusted to optimize the model performance so that it can make accurate predictions.
[0028] After the accident prediction model is trained, the validation set is used to evaluate the model, and the model effect is measured by indicators such as accuracy, recall, and F1 score to ensure the reliability of the model in practical applications. Through this complete training and implementation process, it can be ensured that the constructed accident prediction model has high accuracy, generalization, and robustness, so as to effectively cope with complex prediction tasks.
[0029] Step S300: Generate warning information based on possible accident information and accident probability.
[0030] The electronic device generates targeted warning information based on the output of the accident prediction model. The warning information can not only indicate the impending accident, i.e., the possible accident, but also remind the driver of the potential severity or urgency of the accident in combination with the probability of the possible accident. This warning information can be presented to the driver in a variety of ways, such as visual and auditory, so that the driver can respond in time.
[0031] For example, the warning message generated by the electronic device may be: "There is a sharp bend 200 meters ahead, the road is slippery, and the risk of skidding is 35%. Please slow down." This information will pop up through the dashboard display screen, accompanied by a warning sound to remind the driver to take preventive measures.
[0032] Step S400: Sending warning information to the first vehicle and a second vehicle related to the possible accident.
[0033] The electronic device not only sends the warning information to the first vehicle (i.e., the vehicle that may have an accident), but also sends the warning information to other vehicles that may be affected by the accident and are associated with the possible accident represented by the possible accident information, i.e., the second vehicle. This multi-vehicle communication allows the drivers of other vehicles to be informed of potential dangerous situations in advance and adjust their driving behavior according to the prompts, thereby reducing the possibility of secondary accidents.
[0034] For example, the warning message received by the first vehicle is "There is a sharp bend 200 meters ahead, the road is slippery, and the risk of skidding is 35%, please slow down." If there are other vehicles at the sharp bend (such as a rear vehicle 100 meters away from the sharp bend), the vehicle will also receive a warning message: "There may be a skidding accident at the sharp bend ahead, please keep a safe distance."
[0035] Through the above steps S100-S400, the electronic device first acquires and analyzes multi-dimensional data based on vehicle driving status data, road status data and natural environment data. The electronic device then introduces an accident prediction model, which can combine historical data with machine learning algorithms to achieve dynamic real-time prediction, obtain possible accident information and the probability of possible accidents represented by the possible accident information, and finally generate warning information based on the possible accident information and the probability of possible accidents represented by the possible accident information. The generated warning information is not only for the first vehicle where the accident risk occurs, but also can be sent to the second vehicle related to the possible accident. This multi-party warning mechanism significantly improves the timeliness and effectiveness of the warning and reduces the possibility of accidents. Especially in an environment with dense traffic flow, this extensive warning mechanism helps to spread dangerous information faster, prompting other drivers to respond in a timely manner, thereby improving overall road safety. In this way, the electronic device significantly enhances the effect of road traffic accident prevention through data diversification, model intelligence and the wide dissemination of warning information, and provides effective accident warnings before traffic accidents occur.
[0036] In some embodiments, the method may further include: receiving accident information sent by the first vehicle when an accident occurs to the first vehicle; Generate accident emergency response strategies based on accident information; Deploy emergency resources according to accident emergency strategies.
[0037] In this embodiment, the electronic device first receives accident information, which is information sent to the electronic device via a communication network when the accident sensing system of the first vehicle detects an accident when the first vehicle has an accident. The accident information may include the type of accident (such as collision, rollover), the time and location of the accident, the degree of damage to the vehicle, the status of the occupants, etc.
[0038] Exemplarily, the accident perception system mainly relies on the sensors and related algorithms built into the first vehicle to perceive the occurrence of an accident. For example, the accident perception system obtains the acceleration changes of the first vehicle in all directions through the acceleration sensor. When the first vehicle is subjected to a severe collision, there will be obvious acceleration changes. The accident perception system can have a built-in acceleration threshold (such as exceeding a certain g value). If the detected acceleration exceeds this threshold, it is determined that an accident may have occurred; the accident perception system detects the rotation and rollover of the first vehicle through the gyroscope. If the first vehicle rotates or rolls violently, the gyroscope will record abnormal angular velocity changes. The accident perception system can have a built-in angular velocity threshold. If the detected angular velocity exceeds this threshold, it can be determined that the first vehicle has a rollover accident; the accident perception system detects the physical collision of the first vehicle through the collision sensor. When the collision sensor is triggered, the accident perception system determines that the first vehicle has a collision accident. After determining that the first vehicle has an accident, the accident perception system generates accident information and sends it to the electronic device.
[0039] After receiving the accident information, the electronic device generates an emergency strategy based on the accident information. The emergency strategy may include: whether to immediately notify emergency services (such as ambulances and fire departments), how to guide other nearby vehicles to avoid the accident scene, and temporary re-planning of traffic, such as dispatching vehicles to detour or closing some lanes. The emergency strategy can be sent to various departments associated with the emergency strategy, and the departments can be linked to handle the accident emergency.
[0040] The emergency strategy may include emergency resource deployment, and the electronic device deploys resources to deal with the accident according to the generated emergency strategy. The emergency resource deployment may include dispatching rescue vehicles to the accident site, or dispatching rescue materials to the accident site.
[0041] In this way, after an accident occurs, electronic equipment can quickly and effectively mobilize resources to handle the accident emergency, reducing the damage caused by the accident and the risk of secondary accidents.
[0042] In some implementations, the accident information includes the accident location, the accident type, and the driving state data of the first vehicle before the accident. Generating an accident emergency strategy based on the accident information may include: Obtain road status data at the accident location; Determine the road impact range of the accident involving the first vehicle based on the road state data of the accident location; Determining the severity of the accident that occurred on the first vehicle based on the road impact range, the accident type, and the driving status data of the first vehicle before the accident occurred; Generate incident response strategies based on severity.
[0043] After receiving the accident information, the electronic equipment can further analyze the road status and the impact of the accident on surrounding traffic, so as to formulate more accurate accident emergency response strategies.
[0044] After obtaining the accident location, the electronic device collects road status data at the accident location. This data may come from various data acquisition devices arranged on the road network and may include road width, whether there are obstacles, the number of lanes, road traffic flow, etc.
[0045] The electronic device then analyzes the road impact range of the accident based on the road status data. For example, the accident may cause some lanes to be closed, traffic flow to be restricted, or there may be a risk of a secondary accident at the accident site. The electronic device will evaluate these factors to determine the extent and scope of the impact of the accident on surrounding road traffic.
[0046] After obtaining the accident type and the driving status data of the first vehicle before the accident, the electronic device further analyzes the severity of the accident based on the accident type (such as collision, rollover, vehicle fire, etc.) and the driving status data of the first vehicle before the accident (such as the vehicle's speed, acceleration, whether emergency braking, etc.). For example, when a high-speed vehicle collides severely and causes a chain reaction of multiple vehicles, the severity of the accident will be higher; while a low-speed minor collision may have a smaller impact. Finally, the electronic device formulates a corresponding accident emergency strategy based on the severity of the accident.
[0047] In this way, electronic devices can dynamically assess the scope and severity of the accident and quickly generate the most appropriate accident emergency response strategy to effectively respond to the accident and ensure road safety.
[0048] In some implementations, the accident information includes the accident location, the accident emergency plan includes the type and quantity of emergency resources required, and the emergency resource allocation according to the accident emergency strategy may include: The available resource information of each resource distribution point obtained from the resource database; Obtaining traffic status data of an associated geographic area from a geographic information system, the associated geographic area including the location of each resource distribution point and the location of an accident; Determine the resource information and allocation path of each resource distribution point based on traffic status data and available resource information of each resource distribution point; The resource information to be allocated and the allocation path of each resource issuance point are sent to each resource issuance point, so that each resource issuance point allocates resources according to its own resource information to be allocated and the allocation path.
[0049] When allocating resources, electronic devices can first obtain available resource information of each resource distribution point, which may include emergency centers, fire stations, police stations, medical facilities, etc. These distribution points have specific emergency resources, such as ambulances, fire trucks, rescue equipment, etc. The electronic device will check the number and type of resources currently available at each distribution point; the electronic device will then obtain the accident location and traffic status data of the area where each resource distribution point is located from the geographic information system (GIS). This data may include current road traffic conditions, traffic flow, road closures, congestion information, etc., to ensure that the choice of resource allocation routes can avoid traffic jams and achieve rapid response.
[0050] After completing the acquisition of the above data, the electronic device determines the resource allocation and allocation path based on the above data. The electronic device first determines the resource information and allocation path to be allocated for each resource distribution point based on the traffic status data and available resource information. In this process, the electronic device not only considers the resources available at each resource distribution point, but also considers whether the allocation path of these resources is unobstructed, and selects the best path to ensure the shortest response time.
[0051] The electronic device can generate allocation instructions for each resource distribution point based on the determined resource information and allocation path of each resource distribution point, and send them to each resource distribution point. Each resource distribution point then performs actual resource allocation work according to the allocation instructions, and transports and responds according to the allocation path specified by the electronic device.
[0052] In this way, electronic devices can intelligently deploy emergency resources, combining available resources and traffic conditions to achieve fast and efficient accident response, reduce casualties and speed up rescue.
[0053] Exemplarily, the accident information obtained by the electronic device includes the accident location, and the emergency strategy determined to be completed indicates that the emergency resources required at the accident site include ambulances, firefighting equipment, and rescue personnel. The electronic device first obtains the available resource information of the resource distribution point: Distribution Point A - has 2 ambulances and 1 fire truck; Distribution Point B - has 5 ambulances but no fire truck; Distribution Point C - has 3 ambulances and 2 fire trucks. The electronic device then obtains the traffic status data of the accident location from the GIS system. The data shows that the traffic in the area where Distribution Point A is located is smooth, and there is traffic congestion on the road leading to the accident location from Distribution Point B. Although Distribution Point C is far away from the accident location, the traffic conditions are good.
[0054] The electronic device combines traffic status data with available resource information of each resource distribution point, and determines to dispatch 1 ambulance and 1 fire truck from distribution point A, and 2 ambulances and 1 fire truck from distribution point C. Although distribution point B has sufficient ambulances, due to traffic congestion, the electronic device determines not to dispatch resources from this distribution point. The electronic device sends the dispatch path and the information of the resources to be dispatched to distribution point A and distribution point C, instructing distribution point A and distribution point C to dispatch ambulances and fire trucks to the accident site, and navigate to the accident site through the best path provided by GIS.
[0055] In some implementations, the road state data includes road traffic flow, and obtaining the road state data of the road on which the first vehicle is traveling may include: Acquire a road image of the road on which the first vehicle is traveling by using an image acquisition device; Perform vehicle recognition on the road image to determine the number of vehicles in the road image; Determine road traffic volume based on the number of vehicles; and / or, The natural environment data includes weather conditions, and obtaining the natural environment data of the driving environment of the first vehicle may include: Collecting environmental parameters of the driving environment of the first vehicle by using an environmental parameter collection device; The environmental parameters are input into the weather recognition model, and the weather state corresponding to the environmental parameters output by the weather recognition model is obtained.
[0056] The electronic device may use an image acquisition device to obtain a real-time image of the road, and analyze the number of vehicles through image processing technology to estimate the traffic flow of the road. In this process, the electronic device may first use the image acquisition device to acquire a road image of the road where the first vehicle is located. The road image may include the overall situation of the road, such as vehicles, traffic signs, road conditions, etc. The image acquisition device may be a fixed road monitoring camera, a roadside unit, or the driving recorder of the first vehicle itself.
[0057] The electronic device then processes and analyzes the collected road images, applies computer vision technology to identify vehicles, automatically identifies the vehicles that appear in the road images, and calculates the number of vehicles in the current image to determine the road traffic flow. In this way, the electronic device can monitor the road traffic flow in real time and use the road traffic flow to analyze and warn of possible accidents.
[0058] Exemplarily, the electronic device acquires a plurality of road images per second through a camera, uses an edge detection algorithm to identify the outline of vehicles in each road image, and marks the vehicles to determine the number of vehicles in each road image, and finally calculates the road traffic flow based on the number of vehicles in each road image.
[0059] The electronic device may also obtain the natural environment data of the driving environment of the first vehicle through the environmental parameter acquisition device, and use the weather recognition model to determine the specific weather conditions in the environment. The electronic device first acquires the environmental parameters of the driving environment of the first vehicle through the environmental parameter acquisition device. The environmental parameter acquisition device may include a variety of sensors, such as temperature sensors, humidity sensors, light sensors, air pressure sensors, anemometers, and rain sensors. These devices are installed on the vehicle or on the roadside, and measure the environmental parameters around the first vehicle in real time, such as air temperature, humidity, rainfall, light intensity, etc.; the first vehicle inputs the environmental parameters into the weather recognition model, and the weather recognition model may be a machine learning model or a statistical model, which can predict the weather conditions based on the input parameter data. The weather recognition model can be based on pre-trained meteorological data and analyze the combination of multiple parameters to determine the weather type, such as sunny, rainy, foggy, snowy, etc. In this way, based on the recognition results of the weather recognition model, the electronic device can use the weather conditions to analyze and warn of possible accidents.
[0060] Exemplarily, the first vehicle is equipped with multiple environmental parameter collection devices, including temperature sensors, humidity sensors, and light sensors, etc. The collected parameters include: temperature - 4°C, humidity - 85%, and light intensity - low. The electronic device inputs these collected environmental parameters into the weather recognition model. The weather recognition model analyzes these parameters and pre-trained meteorological data and outputs the weather state as "snowy", that is, the electronic device recognizes that the weather state on the current road is snowing.
[0061] In some implementations, generating warning information based on possible accident information and accident probability may include: When the probability of an accident occurring is greater than a preset probability threshold, generating first warning information according to the possible accident information and the probability of an accident occurring, the first warning information including an accident avoidance instruction for controlling the first vehicle and a second vehicle related to the possible accident to perform an accident avoidance operation; When the probability of an accident is less than or equal to a preset probability threshold, a second warning information is generated based on the possible accident information and the probability of an accident. The second warning information includes operation suggestions, which are used to prompt the driver of the first vehicle and the driver of the second vehicle related to the possible accident to operate according to the operation suggestions.
[0062] Electronic devices can generate different types of warning information based on different accident probabilities. Specifically, there are two situations: when the probability of an accident is higher than a preset probability threshold, the electronic device will generate an emergency avoidance instruction to directly control the driving system of the first vehicle and the driving system of the second vehicle related to the possible accident to perform avoidance operations; when the probability of an accident is lower than or equal to the preset probability threshold, the electronic device will generate prompt operation suggestions.
[0063] When the accident prediction model determines that the probability of a possible accident is greater than a preset probability threshold set by the electronic device, the electronic device will generate a first warning message. The first warning message can not only inform the driver that there is a high probability of a possible accident, but also provide specific accident avoidance instructions to control the driving system of the first vehicle and the driving system of the second vehicle related to the possible accident to perform emergency automatic driving operations, such as deceleration, steering, braking, etc., to avoid the possible accident.
[0064] For example, based on the driving state data, road state data, and natural environment data of the first vehicle, the electronic device predicts that the probability of the first vehicle having a rear-end collision has reached 90%, exceeding a preset probability threshold. At this time, the electronic device generates a first warning message and controls the driving system of the first vehicle to decelerate immediately.
[0065] When the accident prediction model determines that the probability of a possible accident is lower than or equal to the preset probability threshold, the electronic device generates a second warning message, which can provide operational suggestions but does not involve emergency evasive operations. The second warning message can prompt the driver to pay attention to the current driving environment, stay alert, and suggest taking appropriate preventive measures.
[0066] For example, the electronic device determines that the probability of a possible accident is 50% based on the accident prediction model, which is less than the preset probability threshold. At this time, the electronic device generates a second warning message to prompt the driver of the first vehicle to slow down and keep a safe distance, and at the same time remind the surrounding second vehicle to pay attention to possible emergencies ahead.
[0067] In this way, electronic devices can effectively respond to different degrees of potential accident risks and improve road safety.
[0068] In some implementations, generating warning information based on possible accident information and accident probability may include: Generate early warning information based on possible accident information and accident probability, and record possible accidents in the early warning record; The method may also include: Upon receiving the accident information sent by the first vehicle in the event of an accident occurring to the first vehicle, searching the early warning record for a possible accident having the same accident type as the accident type corresponding to the accident information; When a possible accident is found whose accident type is the same as the accident type corresponding to the accident information, the model parameters of the accident prediction model are adjusted.
[0069] When the electronic device detects a potential accident, it generates warning information based on the possible accident information (such as accident type, location, weather, etc.) and the probability of the accident, and the information is sent to the first vehicle and the second vehicle so that the drivers can take appropriate preventive measures. At the same time, the electronic device records the potential accident in the warning record, and the content of the record may include detailed information of the possible accident, such as the accident type, location, probability, involved vehicles, etc.
[0070] If an actual accident occurs to the first vehicle, the electronic device can receive the accident information sent by the first vehicle. Based on the received accident information, the electronic device searches the previous warning records for possible accidents of the same type as the accident. If a possible accident of the same type is found, it means that the electronic device has previously predicted the occurrence of the accident but failed to completely avoid it. In this case, the electronic device will use this feedback information to adjust the parameters of the accident prediction model. This adjustment may involve recalibrating the model weights, correcting the model errors, etc., to improve the prediction accuracy of the accident prediction model.
[0071] For example, the electronic device predicts that the first vehicle has a 70% chance of rear-ending a collision on the road ahead, and generates a warning message based on the prediction to remind the first vehicle to slow down, and the electronic device also records this possible accident in the warning record. However, the first vehicle still rear-ends the vehicle ahead during the subsequent driving process. After the accident, the first vehicle sends the accident information to the electronic device, and the electronic device searches the warning record for the previously predicted similar possible accident based on the received accident information.
[0072] Therefore, the electronic device can determine that the accident is consistent with the previously predicted type, location and other information, indicating that the accident prediction model predicted the risk but underestimated the probability of a possible accident. Therefore, the electronic device adjusts the accident prediction model based on this feedback information, such as increasing the weight of the vehicle braking time and enhancing the impact analysis of slippery roads, so as to more accurately predict similar situations in the future and reduce the possibility of accidents.
[0073] See also Figure 2 , Figure 2 : is a structural diagram of an accident warning device provided in an embodiment of the present application. In a second aspect, an embodiment of the present application provides an accident warning device 200, which is applied to an electronic device, and the electronic device is communicatively connected with a plurality of vehicles. The device 200 includes: The data acquisition module 210 is used to acquire the driving state data of the first vehicle, acquire the road state data of the road on which the first vehicle is driving, and acquire the natural environment data of the driving environment of the first vehicle; The prediction module 220 is used to input the driving state data, the road state data and the natural environment data into the accident prediction model to obtain the possible accident information output by the accident prediction model and the accident probability of the possible accident represented by the possible accident information; The warning generation module 230 is used to generate warning information based on possible accident information and the probability of accident occurrence; The information sending module 240 is used to send warning information to the first vehicle and the second vehicle related to the possible accident.
[0074] The accident warning device 200 provided in the second aspect of the embodiment of the present application can implement each process implemented by the above-mentioned method embodiment and achieve the same beneficial effects. To avoid repetition, it will not be described here.
[0075] See also Figure 3 , is a structural diagram of an electronic device provided in an embodiment of the present application. The third aspect of the embodiment of the present application provides an electronic device 300, including a processor 310 and a memory 320, the memory 320 stores machine executable instructions that can be executed by the processor 310, and the processor 310 can execute the machine executable instructions to implement the above-mentioned accident warning method.
[0076] A fourth aspect of an embodiment of the present application provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by a processor, the processor implements the above-mentioned accident warning method.
[0077] In one embodiment of the present application, a computer program product is also provided, including a computer program, and when the computer program is executed by a processor, the accident warning method according to the above embodiment is implemented.
[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0079] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0080] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0081] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0082] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0083] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0084] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
[0085] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.
Claims
1. An accident warning method, characterized in that: Applied to an electronic device, the electronic device is communicatively connected with a plurality of vehicles, the method comprising: Acquire driving state data of a first vehicle, acquire road state data of a road on which the first vehicle is driving, and acquire natural environment data of a driving environment of the first vehicle; Inputting the driving state data, the road state data and the natural environment data into an accident prediction model to obtain possible accident information output by the accident prediction model and the accident probability of the possible accident represented by the possible accident information; Generate warning information according to the possible accident information and the probability of the accident occurring; The warning information is sent to the first vehicle and a second vehicle related to the possible accident.
2. The method according to claim 1, characterized in that: The method further comprises: receiving accident information sent by the first vehicle when an accident occurs to the first vehicle; generating an accident emergency response strategy according to the accident information; Emergency resources are deployed according to the accident emergency strategy.
3. The method according to claim 2, characterized in that: The accident information includes the accident location, the accident type, and the driving state data of the first vehicle before the accident. The generating of the accident emergency strategy according to the accident information includes: Acquiring road status data of the accident location; Determine the road impact range of the accident occurring to the first vehicle according to the road state data of the accident location; Determining the severity of the accident occurring to the first vehicle according to the road impact range, the accident type, and the driving state data of the first vehicle before the accident occurred; An accident emergency response strategy is generated according to the severity.
4. The method according to claim 2, characterized in that: The accident information includes the accident location, the accident emergency plan includes the type and quantity of emergency resources required, and the emergency resource allocation according to the accident emergency strategy includes: The available resource information of each resource distribution point obtained from the resource database; Acquire traffic status data of an associated geographical area from a geographic information system, wherein the associated geographical area includes the locations of each of the resource distribution points and the accident location; Determine the resource information and the allocation path of each resource distribution point according to the traffic status data and the available resource information of each resource distribution point; The resource information to be allocated and the allocation path of each resource issuance point are sent to each resource issuance point, so that each resource issuance point allocates resources according to its own resource information to be allocated and the allocation path.
5. The method according to claim 1, characterized in that: The road state data includes road traffic flow, and the step of obtaining the road state data of the road on which the first vehicle is traveling includes: Acquire a road image of the road on which the first vehicle is traveling by using an image acquisition device; Performing vehicle recognition on the road image to determine the number of vehicles in the road image; Determining the road traffic flow according to the number of vehicles; and / or, The natural environment data includes weather conditions, and the obtaining of the natural environment data of the first vehicle driving environment includes: Collecting the first vehicle driving environment parameters by an environmental parameter collection device; The environmental parameters are input into a weather recognition model, and the weather state corresponding to the environmental parameters output by the weather recognition model is obtained.
6. The method according to claim 1, characterized in that The generating of warning information according to the possible accident information and the probability of the accident occurrence includes: When the probability of the accident is greater than a preset probability threshold, generating first warning information according to the possible accident information and the probability of the accident, the first warning information including an accident avoidance instruction for controlling the first vehicle and a second vehicle related to the possible accident to perform an accident avoidance operation; When the probability of the accident occurrence is less than or equal to the preset probability threshold, a second warning information is generated based on the possible accident information and the probability of the accident occurrence, and the second warning information includes operation suggestions, which are used to prompt the driver of the first vehicle and the driver of the second vehicle related to the possible accident to operate according to the operation suggestions.
7. The method according to claim 1, characterized in that The generating of warning information according to the possible accident information and the probability of the accident occurrence includes: Generate warning information according to the possible accident information and the probability of the accident, and record the possible accident in the warning record; The method further comprises: Upon receiving the accident information sent by the first vehicle when the first vehicle has an accident, searching the early warning record for a possible accident whose accident type is the same as the accident type corresponding to the accident information; When a possible accident is found whose accident type is the same as the accident type corresponding to the accident information, the model parameters of the accident prediction model are adjusted.
8. An accident warning device, characterized in that: Applied to an electronic device, the electronic device is communicatively connected with a plurality of vehicles, the device comprising: A data acquisition module, used to acquire driving state data of the first vehicle, acquire road state data of the road on which the first vehicle is driving, and acquire natural environment data of the driving environment of the first vehicle; A prediction module, used to input the driving state data, the road state data and the natural environment data into an accident prediction model, and obtain possible accident information output by the accident prediction model, and an accident probability of a possible accident represented by the possible accident information; An early warning generation module, used to generate early warning information according to the possible accident information and the probability of the accident; An information sending module is used to send the warning information to the first vehicle and a second vehicle related to the possible accident.
9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the accident warning method as described in any one of claims 1 to 7 are implemented.
10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions, which, when executed by a processor, enable the processor to implement the accident warning method according to any one of claims 1-7.
Citation Information
Patent Citations
Traffic accident monitoring method and system
CN106781458A
Vehicle-mounted traffic accident early warning device based on cloud edge combination
CN111081020A
Method and device for identifying vehicle accident, electronic equipment and storage medium
CN117163046A
Vehicle safety early warning method, system and equipment and storage medium
CN117456714A
Electronic device, warning message providing method therefor, and non-transitory computer-readable recording medium
WO2018186625A1
Cited By
Vehicle accident emergency response method, electronic equipment and storage medium
CN122050195A
Vehicle accident emergency response method, electronic device, and storage medium
CN122050195B
