CN117953609A

CN117953609A - Accident data processing method, device, equipment and medium for automatic driving vehicle

Info

Publication number: CN117953609A
Application number: CN202311815877.1A
Authority: CN
Inventors: 罗泽文; 陈�光; 段锐; 王磊; 胡宁
Original assignee: Faw Nanjing Technology Development Co ltd; FAW Group Corp
Current assignee: Faw Nanjing Technology Development Co ltd; FAW Group Corp
Priority date: 2023-12-26
Filing date: 2023-12-26
Publication date: 2024-04-30

Abstract

The invention discloses an accident data processing method, device, equipment and medium for an automatic driving vehicle. The method is characterized by comprising the following steps: under the condition that the target vehicle is in an automatic driving mode, acquiring vehicle running data of the target vehicle at the current moment; carrying out accident prediction according to the vehicle operation data, and determining the running accident risk of the target vehicle; and if the driving accident risk is high-risk driving, the vehicle operation data are stored in the target vehicle in a lasting mode, and the vehicle operation data are uploaded to an automatic driving supervision platform. When the vehicle is automatically driven, the running state of the vehicle is monitored at any time, and when running risk exists, the running data of the vehicle can be recorded, the data support is carried out for the automatic driving accident of the vehicle, and the recovery and analysis of the vehicle accident are ensured.

Description

Accident data processing method, device, equipment and medium for automatic driving vehicle

Technical Field

The present invention relates to the field of autopilot, and in particular, to a method, apparatus, device, and medium for processing accident data of an autopilot vehicle.

Background

With the development of automatic driving technology, automatic driving has entered a test stage at present and gradually entered a market as a vehicle function, it is expected that an automatic driving vehicle will travel on various roads, with the increase of the number of vehicles with the automatic driving function, responsibility recognition of the automatic driving vehicle at the time of accident is very difficult, accurate vehicle driving data is required to restore and analyze an accident scene of the automatic driving vehicle when clear traffic responsibility is divided, in the prior art, the vehicle driving data is recorded through a driving recording device, a driver is usually required to judge whether the vehicle records the vehicle driving data or not by himself, and in the automatic driving vehicle, the driver usually does not drive the vehicle, and it is difficult to judge whether the vehicle has a driving risk or not.

Disclosure of Invention

The invention provides an accident data processing method, device, equipment and medium for an automatic driving vehicle, so as to realize the running risk identification of the automatic driving vehicle and the effective storage of vehicle running data.

According to an aspect of the present invention, there is provided an accident data processing method of an autonomous vehicle, including:

Under the condition that the target vehicle is in an automatic driving mode, acquiring vehicle running data of the target vehicle at the current moment; the vehicle running data comprise driving state information and vehicle end image data;

Carrying out accident prediction according to the vehicle operation data, and determining the running accident risk of the target vehicle;

And if the driving accident risk is high-risk driving, the vehicle operation data are stored in the target vehicle in a lasting mode, and the vehicle operation data are uploaded to an automatic driving supervision platform.

According to another aspect of the present invention, there is provided an accident data processing apparatus of an autonomous vehicle, including:

The automatic driving data sensing module is used for acquiring vehicle running data of the target vehicle at the current moment under the condition that the target vehicle is in an automatic driving mode; the vehicle running data comprise driving state information and vehicle end image data;

The automatic driving risk perception module is used for carrying out accident prediction according to the vehicle operation data and determining the driving accident risk of the target vehicle;

and the vehicle data storage module is used for storing the vehicle operation data in the target vehicle in a lasting mode and uploading the vehicle operation data to an automatic driving supervision platform if the driving accident risk is high-risk driving.

According to another aspect of the present invention, there is provided an electronic apparatus including:

at least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the accident data processing method of an autonomous vehicle according to any one of the embodiments of the present invention.

According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute an accident data processing method for an autonomous vehicle according to any of the embodiments of the present invention.

According to the technical scheme, under the condition that the target vehicle is in the automatic driving mode, vehicle operation data of the target vehicle at the current moment are obtained, and under the condition that the vehicle is in the automatic driving mode, the vehicle operation data of the target vehicle at each moment in the automatic driving process are aimed at, so that the data effectiveness is guaranteed; according to the vehicle running data, accident prediction is carried out, the running accident risk of the target vehicle is determined, the risk existing in the automatic driving process of the vehicle is judged according to the vehicle running data, the predicted accident risk can be extracted through active judgment of the risk, further the vehicle running data at a plurality of moments are stored, the data support is improved for judging the vehicle running accident, and the accident identification efficiency and accuracy are improved; if the running accident risk is high-risk running, the vehicle running data are stored in the target vehicle in a lasting mode, the vehicle running data are uploaded to an automatic driving supervision platform, when the automatic driving is predicted to be high-risk, the vehicle running data can be stored in a lasting mode automatically, the data are uploaded, the local data are guaranteed, meanwhile, the automatic driving supervision platform is reported, and the accident scene can be restored and analyzed. The technical problem that driving accidents of the automatic driving vehicle in the prior art are difficult to judge accident responsibility is solved, vehicle driving data of the automatic driving vehicle can be effectively monitored, and data support is carried out for the automatic driving accidents of the vehicle.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of an accident data processing method for an automatic driving vehicle according to an embodiment of the present invention;

FIG. 2 is a flow chart of another method for processing accident data of an autonomous vehicle according to a second embodiment of the present invention;

Fig. 3 is a schematic structural diagram of an accident data processing apparatus for an autopilot vehicle according to a third embodiment of the present invention;

Fig. 4 is a schematic structural view of an electronic device implementing an accident data processing method of an autonomous vehicle according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

Example 1

Fig. 1 is a flowchart of an accident data processing method of an automatic driving vehicle according to an embodiment of the present invention, where the method may be applied to a scenario where responsibility is determined after a traffic accident is sent to the automatic driving vehicle, and the method may be performed by an accident data processing apparatus of the automatic driving vehicle, where the accident data processing apparatus of the automatic driving vehicle may be implemented in a form of hardware and/or software, and the accident data processing apparatus of the automatic driving vehicle may be configured in an electronic device. As shown in fig. 1, the method includes:

With the development of vehicle intellectualization, vehicle assisted driving can help a vehicle driver to perform assisted driving, so as to reduce driving operations of the vehicle driver and improve driving feeling. At present, the auxiliary driving of the vehicle in the market is usually L2.9 level automatic driving, automatic lane changing and automatic parking can be carried out, but the monitoring and the intervention of the driver of the vehicle are still needed, under specific conditions, the vehicle can prompt to exit the auxiliary driving and prompt the driver of the vehicle to take over the vehicle, the L3 level automatic driving can be carried out on the vehicle usually, the vehicle can independently complete driving, the driver is not required to be in a state of taking over the vehicle, and the driver is still required to keep the control of the vehicle; the L4 level automatic driving is highly automatic, can independently complete all driving behaviors, and a driver does not need to keep attention.

S110, acquiring vehicle operation data of the target vehicle at the current moment under the condition that the target vehicle is in an automatic driving mode.

The target vehicle may be an autonomous vehicle capable of completing an autonomous mission. Alternatively, in a normal case, a vehicle capable of completing an automatic driving task needs to have sensing road information and vehicle position information, and the automatic driving vehicle senses the road information and obtains the road information through any one sensing sensor of a vehicle-mounted camera, a vehicle-mounted laser radar, a millimeter wave radar and ultrasonic waves. The vehicle-mounted camera acquires image information of a road, the vehicle-mounted laser radar acquires laser characteristic information of the road and the millimeter wave radar acquires millimeter wave signals of the road. Vehicle position information is typically automatically located by an on-board location sensor; exemplary in-vehicle positioning sensors include GPS (Global Positioning System) global positioning system, inertial navigation system, and visual positioning system.

Optionally, in the automatic driving vehicle, in order to prevent a situation that the driver cannot take over the automatic driving vehicle, an on-board camera and a sensor are arranged in the automatic driving vehicle, and image data in the vehicle is acquired through the on-board camera in the vehicle, so that the vehicle is monitored.

Optionally, the automatic driving mode may be a mode in which the vehicle is automatically controlled to run by using an artificial intelligent device in the automatic driving vehicle according to road information obtained by the vehicle, and the artificial intelligent device can automatically identify according to the obtained road information, so that the vehicle driving task is completed without intervention of a driver. In general, the automatic driving mode is set in the vehicle-mounted system, and can be manually started according to the setting of the controller in the vehicle, or the position and the road capable of supporting automatic driving in the vehicle-mounted system are preset, and the driver is automatically prompted whether to enter the automatic driving mode when the automatic driving vehicle enters the position and the road capable of supporting automatic driving.

The vehicle running data comprise driving state information and vehicle end image data; the driving state information may include at least one of vehicle position information, vehicle state information, automatic driving information and path planning information corresponding to an automatic driving mode, vehicle perception information, and vehicle identification information of the vehicle. The vehicle-end image data can be obtained by acquiring image data from the inside and the outside of an automatic driving vehicle through a vehicle-mounted camera.

Alternatively, the vehicle position information may include: vehicle longitude, vehicle latitude, vehicle altitude, vehicle heading angle, GNSS speed, vehicle longitudinal acceleration, vehicle lateral acceleration, vehicle yaw rate, and vehicle roll rate are recorded based on a vehicle GNSS (Global Navigation SATELLITE SYSTEM ) time stamp.

Optionally, the vehicle status information generally includes high frequency status information and low frequency status information, and the high frequency status information includes recording accelerator opening, brake pedal flag, steering wheel angle, gear, vehicle lamp status, horn status, wiper status, seat pressure sensor pressure status, running mode of the vehicle, and remote control command to vehicle receiving delay based on the vehicle GNSS time stamp. The running modes of the vehicle comprise a manual driving mode, an automatic driving model and an automatic driving manual intervention mode. The low frequency status information includes recording steering wheel steering torque, engine output speed, engine torque, drive motor speed, drive motor torque, fuel remaining, hundred kilometers average fuel consumption, battery remaining, hundred kilometers average power consumption, total mileage, master cylinder pressure, chassis and autopilot system failure, collision and collision risk, and vehicle current failure code based on the vehicle GNSS time stamp.

Optionally, the automatic driving information includes recording an AD system requested heading angle, an AD system requested vehicle speed, an AD system requested steering wheel angle, an AD system requested steering wheel steering torque, an AD system requested gear, an AD system requested longitudinal acceleration, an AD system requested lateral acceleration, an AD system requested vehicle lamp status, an AD system requested engine speed, an AD system requested engine torque, an AD system requested drive motor speed, and an AD system requested drive motor torque based on the vehicle GNSS time stamp.

Optionally, the path planning information includes recording a path plan of the vehicle in the road based on the vehicle GNSS time stamp.

Optionally, the vehicle sensing information includes a number of frames recorded based on the vehicle GNSS time stamp, a target object, a type of the sensing target object, a heading angle of the sensing target object, a number of the sensing target objects, a lateral distance of the sensing target object, a longitudinal distance of the sensing target object, a speed of the sensing target object, a length of the sensing target object, a width of the sensing target object, and a height of the sensing target object. The perception target may be an obstacle or a device in the road, among others. By way of example, the obstacles may be other vehicles, pedestrians, and objects; the device may be a road infrastructure.

Optionally, the vehicle identification information includes recording traffic light phase and traffic light status based on the vehicle GNSS time stamp.

Optionally, the end-of-vehicle image data may include a front scene image, a rear scene image, a left Fang Changjing image, a right scene image, a driver and steering wheel image, an autopilot system perception interface image, and a remote console image.

Specifically, when the target vehicle can automatically drive, an automatic driving mode of the vehicle is started, and when the target vehicle is detected to be in the automatic driving mode, the target vehicle acquires vehicle running data at the current moment.

S120, accident prediction is carried out according to the vehicle operation data, and the running accident risk of the target vehicle is determined.

Optionally, when the automatic driving vehicle runs on the road, the running track can be automatically planned according to the automatic driving mode, and because of various emergency situations or irresistible situations in the road, the automatic driving vehicle has a risk of traffic accidents, in order to prevent the automatic driving vehicle from causing traffic accidents on the road, the automatic driving vehicle can predict the accident according to the running data of the vehicle after entering the automatic driving mode, and judge the probability and risk of the accident of the current running track on the road.

The driving accident risk may be an accident risk probability existing in a driving trajectory of the autonomous vehicle. It should be noted that, the vehicle operation data obtained by the automatic driving vehicle in the road includes road data and vehicle self data, and the accident probability calculation is performed according to the road data and the vehicle self data, so as to determine the running accident risk of the automatic driving vehicle.

Specifically, the target vehicle predicts accident probability of the obstacles in the running track one by one and in combination according to the vehicle running data, judges the accident probability in the running track, and further determines the running accident risk of the target vehicle according to the accident probability. Wherein the accident probability and the driving accident risk are dynamic and change in time. For example, a target vehicle travels in an urban overhead, accident probability prediction is performed on other vehicles and road facilities in a travel track in combination with vehicle operation data of the target vehicle, accident probability between each vehicle and the target vehicle is determined, the travel track of the other vehicles is monitored, and the accident probability of the other vehicles is adjusted at any time under the condition that the other vehicles are judged to have lane changes, overtaking or out of control and the like.

And S130, if the running accident risk is high-risk running, the vehicle running data are stored in the target vehicle in a lasting mode, and the vehicle running data are uploaded to an automatic driving supervision platform.

The high-risk driving may be that a traffic accident occurs in a high probability of a driving track of the target vehicle in the automatic driving mode. For example, if the probability of the target vehicle being 10% and the probability of the target vehicle being evaded by other vehicles are considered to be high-risk accidents, and if the probability of the target vehicle being evaded by other vehicles is considered to be 50%, the target vehicle is considered to avoid collision by adjusting the driving track, and if the risk of the target vehicle being evaded by other vehicles is considered to be low-risk accidents.

Optionally, the target vehicle performs persistent storage on the vehicle operation data, and the vehicle operation data is stored in a storage space of the target vehicle in a file storage mode, so that the vehicle operation data is prevented from being lost when an accident occurs in the target vehicle. When the target vehicle performs persistent storage on the vehicle running data, judging whether the data amount stored in the storage space in the vehicle meets the vehicle running data, and when the data amount in the storage space cannot meet the vehicle running data, selecting the vehicle running data with the latest storage time stamp as the storage time stamp by the target vehicle according to the vehicle GNSS time stamp of the vehicle running data for persistent storage; in another embodiment, the target vehicle may actively select to perform data compression on the vehicle operation data when the vehicle operation data is acquired, so as to reduce the data amount of the vehicle operation data and improve the storage capability of the target vehicle.

The autopilot supervisory platform may be a monitoring system set by a management authority handling autopilot incidents. The method is characterized in that a target vehicle can be connected with an automatic driving supervision platform, vehicle operation data is uploaded to the automatic driving supervision platform through a network when an accident occurs in the target vehicle, vehicle identification information is used as a data identification when the data is uploaded in the target vehicle, and the uploading time, license plate number and vehicle operation data are combined to generate accident uploading data of the target vehicle, so that the accident uploading data are uploaded. The vehicle identification information may be a vehicle identification code assigned to each vehicle when shipped.

Specifically, if it is predicted that the running accident risk of the target vehicle is high-risk running, it is stated that the target vehicle is difficult to avoid the occurrence of a traffic accident, in order to clarify the division responsibility of the traffic accident of the automatic driving vehicle, it is necessary to store the vehicle running data before the time of the traffic accident in a persistent manner, further store the vehicle running data in the target vehicle, and upload the vehicle running data to the automatic driving supervision platform to be supervised by the automatic driving supervision platform.

Optionally, in another optional embodiment of the present invention, the persisting the vehicle operation data in the target vehicle includes:

Acquiring an available storage space of the target vehicle; acquiring historical storage data of the target vehicle and a storage time stamp of the historical storage data under the condition that the data volume of the vehicle operation data is larger than the available storage space; according to the storage time stamp of the history storage data and the available storage space, carrying out storage calculation to determine a data space to be covered; and storing the vehicle operation data in the to-be-covered data space and the available storage space in a lasting manner, and recording a storage time stamp of the vehicle operation data.

Wherein the available storage space may be the remaining storage space in the data storage space of the target vehicle. Before the vehicle operation data is stored permanently, it is necessary to determine whether the data storage space of the target vehicle satisfies the data amount of the vehicle operation data.

The history storage data may be vehicle operation data that has been stored in a data storage space of the target vehicle. When the vehicle operation data is stored at the current time, it is recognized that the vehicle operation data at the previous time is stored in the target vehicle, and the vehicle operation data at the previous time is used as the history storage data in the data storage space of the target vehicle. The stored time stamp is typically a vehicle GNSS time stamp of the vehicle operation data, and the time information of the stored data can be effectively judged by recording the stored time stamp of the vehicle operation data.

The data space to be covered can be a data storage space needing to be replaced by data, and the data storage space to be covered can be a data storage space with historical storage data or without storage data. Optionally, when determining the data space to be covered, preferentially determining the available storage space as the data space to be covered, judging according to the storage time stamp of the historical storage data when the available storage space cannot store the vehicle running data at the current time, screening the storage time stamp with the largest time interval with the current time, taking the historical storage data corresponding to the screened storage time stamp as the data space to be covered, and taking the storage space corresponding to the data to be covered together as the data space to be covered.

Specifically, when the vehicle running data is subjected to persistent storage, an available storage space of the target vehicle is obtained, whether the data amount of the vehicle running data can be stored in the available storage space is judged, if the data amount of the vehicle running data is larger than the available storage space, the vehicle running data needs to cover historical storage data, storage calculation is performed through a storage timestamp of the historical storage data and the available storage space, the data to be covered is determined, the data space to be covered is determined, the vehicle running data is stored in the data space to be covered and the available storage space in a persistent mode, and the storage timestamp of the vehicle running data is recorded.

Example two

Fig. 2 is a flowchart of another accident data processing method for an automatic driving vehicle according to the second embodiment of the present invention, where the relationship between the present embodiment and the above embodiments is a specific method for implementing driving accident prediction for a target vehicle in an automatic driving mode. As shown in fig. 2, the accident data processing method of the autonomous vehicle includes:

S210, acquiring vehicle operation data of the target vehicle at the current moment under the condition that the target vehicle is in an automatic driving mode.

S220, acquiring path planning information of the target vehicle under the condition that the vehicle position information meets the preset vehicle position condition corresponding to the automatic driving mode.

The vehicle position condition may be a road position where automatic driving is allowed to be set in advance. Optionally, in general, in order to ensure driving safety of the autopilot vehicle, an automatic driving capability is preset in the road, so when the autopilot mode is started by the target vehicle, whether the target vehicle can perform the autopilot is determined according to the vehicle position information, and when the vehicle position of the target vehicle is determined to meet the vehicle position condition, the target vehicle is allowed to perform the autopilot through the autopilot model.

The path planning information may be an automatic driving mode of the target vehicle planning a track of travel in the road. Optionally, when the target vehicle enters an automatic driving mode and performs an automatic driving state, the automatic driving mode of the target vehicle plans a driving track of the automatic driving according to the vehicle running data, and generates path planning information.

Specifically, when the target vehicle selects to automatically drive, whether the vehicle position information of the target vehicle meets the vehicle position condition corresponding to the automatic driving mode is judged, and when the target vehicle meets the vehicle position condition, the target vehicle is taken over by the automatic driving mode to automatically drive. And when the target vehicle enters automatic driving, acquiring path planning information of the target vehicle.

S230, determining obstacle path information according to the vehicle perception information and the vehicle end image data.

The obstacle path information may be a trajectory of an obstacle in the road in the automatic driving mode of the target vehicle.

Specifically, when the target vehicle is automatically driven, the target vehicle obtains vehicle perception information and vehicle end image data from a road, identifies an obstacle in the road, calculates an obstacle track of the obstacle, and uses the obstacle track as obstacle path information. Here, the obstacle trajectory is trajectory information of the obstacle with respect to the target vehicle.

S240, accident prediction is carried out according to the obstacle path information, the whole vehicle state information and the path planning information, and the running accident risk of the target vehicle is determined.

Specifically, the target vehicle automatically runs according to the path planning information, predicts the accident probability according to the whole vehicle state information and the obstacle path information of each obstacle, judges the accident probability in the running track, and further determines the running accident risk of the target vehicle according to the accident probability. For example, a target vehicle travels in an urban overhead, other vehicles and road facilities in the travel track are used as barriers, the track of the barriers is calculated, accident probability prediction is carried out according to the whole vehicle state information and the path planning information and the track of the barriers, the accident probability between each barrier and the target vehicle is determined, the travel track of the barriers is monitored at all times, and when the travel track of the barriers and the path planning information of the target vehicle are judged to have conflict, the accident probability of the barriers and the target vehicle is adjusted.

Optionally, in another optional embodiment of the present invention, the accident prediction according to the obstacle path information, the whole vehicle state information and the path planning information, determining a running accident risk of the target vehicle includes:

Performing automatic avoidance prediction according to the obstacle path information, the path planning information and the whole vehicle state information through a preset vehicle avoidance model to obtain a target avoidance path and vehicle avoidance probability corresponding to the target avoidance path; and if the vehicle avoidance probability is smaller than or equal to a preset avoidance probability threshold value, setting the running accident risk as a high risk.

The vehicle avoidance model may be a neural network model that is pre-trained for avoidance predictions. It should be noted that, the vehicle avoidance model may be obtained based on preset model training data and an initial avoidance model training, where the model training data includes obstacle path training data, path planning training data, and vehicle state training data; generating an avoidance training set and an avoidance test set based on model training data, inputting the avoidance training set into an initial avoidance model to perform model training to obtain a target avoidance model, testing the target avoidance model according to the avoidance test set, performing reverse training according to a test result, and finally obtaining a vehicle avoidance model.

Optionally, the vehicle avoidance model is composed of an avoidance path planning network and an avoidance probability calculation network, the avoidance path planning network is used for planning a vehicle avoidance path for the target vehicle, the vehicle avoidance path can be used for the target vehicle to avoid the obstacle, the avoidance probability calculation network is used for calculating the vehicle avoidance probability of the avoidance path, and the vehicle avoidance probability can judge the success rate of the target vehicle adopting the vehicle to avoid the obstacle by adopting the vehicle avoidance path.

Wherein, the target avoidance path may be a travel track of the target vehicle avoidance obstacle; it should be noted that, a plurality of avoidance routes may be obtained for an obstacle, and a plurality of obstacles are usually present on the road, and then a target avoidance route is determined according to the plurality of avoidance routes corresponding to the plurality of obstacles.

The preset avoidance probability threshold value can be probability parameter information which can judge that the target vehicle can safely run; the preset avoidance probability threshold may be used to determine whether the target avoidance path has a collision risk. For example, the preset avoidance probability threshold may be 90%.

Specifically, after obstacle path information, path planning information and whole vehicle state information are obtained, the obstacle path information, the path planning information and the whole vehicle state information are input into a preset vehicle avoidance model to carry out automatic avoidance prediction, the target avoidance path output by the vehicle avoidance model and the vehicle avoidance probability corresponding to the target avoidance path are obtained, the vehicle avoidance probability and a preset avoidance probability threshold are compared, and if the vehicle avoidance probability is smaller than or equal to the preset avoidance probability threshold, the target vehicle is considered to have high-risk driving accident risk.

Optionally, in another optional embodiment of the present invention, the performing, by using a preset vehicle avoidance model, automatic avoidance prediction according to the obstacle path information, the path planning information, and the overall vehicle state information, to obtain a target avoidance path and a vehicle avoidance probability corresponding to the target avoidance path includes:

Planning at least one vehicle avoidance path according to the obstacle path information and the path planning information based on the avoidance path planning network; and carrying out evasion probability calculation on each vehicle evasion path according to the whole vehicle state information based on the evasion probability calculation network, taking the vehicle evasion path with the highest evasion probability as a target evasion path, and determining the evasion probability of the target evasion path as the vehicle evasion probability.

Optionally, when the target vehicle avoids the obstacle, a plurality of vehicle avoidance paths may exist, but due to the influence of the running state of the restricted target vehicle, the target vehicle is difficult to complete the running path corresponding to each vehicle avoidance path, so as to calculate the vehicle avoidance probability corresponding to each vehicle avoidance path according to the whole vehicle state information of the target vehicle. Illustratively, when the target vehicle avoids a plurality of obstacles, a vehicle avoidance path A is selected, wherein the target vehicle accelerates and turns, and the vehicle avoidance path B is: the straight-line acceleration non-turning and vehicle avoidance path C is a sharp turning lane change after straight-line braking; due to the influences of vehicle speed, vehicle acceleration, vehicle turning capability and vehicle braking capability, the probability of rear rollover of the vehicle in the avoidance of an obstacle is 60%, the acceleration of the vehicle in the avoidance path B is low, full collision cannot be avoided, the probability of overlarge high-speed collision damage is 90%, the braking capability of the vehicle in the avoidance path C is too low but influenced by a road surface, and the collision probability is 10%.

Specifically, an avoidance path planning network of a vehicle avoidance model performs avoidance path planning on a target vehicle according to obstacle path information and path planning information to obtain at least one vehicle avoidance path, an avoidance probability calculation network calculates the avoidance probability corresponding to each vehicle avoidance path, the vehicle avoidance path with the highest avoidance probability is used as a target avoidance path, and the avoidance probability of the target avoidance path is determined as the vehicle avoidance probability. For example, when the target vehicle avoids a plurality of obstacles, three vehicle avoidance paths, namely a vehicle avoidance path A, a target vehicle accelerates and turns, and a vehicle avoidance path B, are calculated: the straight-line acceleration non-turning and vehicle avoidance path C is a sharp turning lane change after straight-line braking; the probability of rear rollover of the vehicle avoiding path A, which is subjected to the influence of the vehicle speed, the vehicle acceleration, the vehicle turning capacity and the vehicle braking capacity, is 60%, and the corresponding probability of rear rollover of the vehicle avoiding path A is 40%; the acceleration of the vehicle avoidance path B is low, the probability of overlarge high-speed collision injury caused by the unavoidable complete collision is 90%, and the corresponding avoidance probability of the vehicle avoidance path B is 10%; when the braking capability of the vehicle avoidance path C is too low but is influenced by the road surface, and the collision probability is still 10%, the corresponding avoidance probability of the vehicle avoidance path C is 90%, the vehicle avoidance path C is further selected as a target avoidance path, and the vehicle avoidance probability of the target avoidance path is set to be 90%.

And S250, if the running accident risk is high-risk running, the vehicle running data are stored in the target vehicle in a lasting mode, and the vehicle running data are uploaded to an automatic driving supervision platform.

Optionally, in another optional embodiment of the present invention, before if the driving accident risk is high risk driving, the method further includes:

Under the condition that the driver operation exists in the detected vehicle-end image data, acquiring the running state of the automatic driving mode; setting the running accident risk to a high risk in a case where the automatic driving mode is in operation; the running accident risk is set to be low risk in the case where the automatic driving mode is off.

Wherein the driver operation may be an action in which the driver operates the target vehicle to travel; optionally, the in-vehicle data in the vehicle-end image data is detected, when the driving operation of the driver of the vehicle on the target vehicle is detected, the running state of the automatic driving mode of the vehicle needs to be judged, whether the automatic driving mode of the vehicle is running is judged, if the automatic driving mode of the vehicle is running, the driver can interfere with the running track of the vehicle in the automatic driving mode, the running accident risk is set as high risk, and if the automatic driving mode of the vehicle is closed, the driver takes over the vehicle to run normally, and the running accident risk is set as low risk. The driving operation may be manual operation of a brake, a throttle and a steering wheel.

Optionally, in another optional embodiment of the present invention, after setting the running accident risk to be a high risk, the method further includes:

And generating risk prompt information and driving prompt information, wherein the driving prompt information is used for prompting that the target vehicle is switched from the automatic driving mode to the manual driving mode.

The risk prompt information may be information that the target vehicle prompts that the automatic driving mode has driving risk; the risk prompt information is used for prompting the driver that accident risk exists. Optionally, the risk prompt information may be voice information, text information, or other prompt information; the voice information can be voice information broadcast through an artificial intelligent assistant of the target vehicle, the text information can be that the target vehicle is displayed on a vehicle-mounted display device and an instrument panel, and other prompt information can be steering wheel vibration, seat vibration and/or lamplight prompt.

Optionally, after the target vehicle generates the risk prompt information and the driving prompt information, the risk prompt is performed on the driver, the driver is prompted to take over the target vehicle, and after the target vehicle detects that the driver takes over the target vehicle, the automatic driving mode is switched to the manual driving mode.

Example III

Fig. 3 is a schematic structural diagram of an accident data processing apparatus for an automatic driving vehicle according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes: an autopilot data awareness module 310, an autopilot risk awareness module 320, and a vehicle data store module 330, wherein,

The automatic driving data sensing module 310 is configured to obtain vehicle operation data of the target vehicle at a current time when the target vehicle is in an automatic driving mode; the vehicle running data comprise driving state information and vehicle end image data;

The automatic driving risk perception module 320 is configured to predict an accident according to the vehicle operation data, and determine a running accident risk of the target vehicle;

The vehicle data storage module 330 is configured to, if the driving accident risk is high-risk driving, persist the vehicle operation data in the target vehicle and upload the vehicle operation data to an autopilot supervisory platform.

Optionally, the automatic driving risk sensing module is specifically configured to:

Acquiring path planning information of the target vehicle under the condition that the vehicle position information meets the preset vehicle position condition corresponding to the automatic driving mode;

determining obstacle path information according to the vehicle perception information and the vehicle end image data;

And predicting accidents according to the obstacle path information, the whole vehicle state information and the path planning information, and determining the running accident risk of the target vehicle.

Optionally, the automatic driving risk perception module is specifically further configured to:

Performing automatic avoidance prediction according to the obstacle path information, the path planning information and the whole vehicle state information through a preset vehicle avoidance model to obtain a target avoidance path and vehicle avoidance probability corresponding to the target avoidance path;

And if the vehicle avoidance probability is smaller than or equal to a preset avoidance probability threshold value, setting the running accident risk as a high risk.

planning at least one vehicle avoidance path according to the obstacle path information and the path planning information based on the avoidance path planning network;

And carrying out evasion probability calculation on each vehicle evasion path according to the whole vehicle state information based on the evasion probability calculation network, taking the vehicle evasion path with the highest evasion probability as a target evasion path, and determining the evasion probability of the target evasion path as the vehicle evasion probability.

Optionally, the device further comprises an image detection module, a first risk evaluation module and a second risk evaluation module; wherein:

The image detection module is used for acquiring the running state of the automatic driving mode under the condition that the driver operation exists in the detected vehicle-end image data;

the first risk evaluation module is used for setting the running accident risk to be high risk under the condition that the automatic driving mode is running;

The second risk evaluation module is configured to set the driving accident risk to be a low risk when the automatic driving mode is in a closed state.

Optionally, the device further comprises a risk prompting module; wherein:

The risk prompt module is used for generating risk prompt information and driving prompt information, wherein the driving prompt information is used for prompting that the target vehicle is switched from the automatic driving mode to the manual driving mode.

Optionally, the vehicle data storage module is specifically configured to:

Acquiring an available storage space of the target vehicle;

acquiring historical storage data of the target vehicle and a storage time stamp of the historical storage data under the condition that the data volume of the vehicle operation data is larger than the available storage space;

According to the storage time stamp of the history storage data and the available storage space, carrying out storage calculation to determine a data space to be covered;

And storing the vehicle operation data in the to-be-covered data space and the available storage space in a lasting manner, and recording a storage time stamp of the vehicle operation data.

The accident data processing device for the automatic driving vehicle provided by the embodiment of the invention can execute the accident data processing method for the automatic driving vehicle provided by any embodiment of the invention, and has the mode module and the beneficial effects corresponding to the execution method.

Example IV

Fig. 4 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their patterns are meant to be exemplary only, and are not meant to limit implementations of the invention described and/or claimed herein.

As shown in fig. 4, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.

Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as an accident data processing method of an autonomous vehicle.

In some embodiments, the accident data processing method of an autonomous vehicle may be implemented as a computer program tangibly embodied on a computer readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the accident data processing method of an autonomous vehicle described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the accident data processing method of the autonomous vehicle by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the modes/operations specified in the flowchart and/or block diagram to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.

Example five

The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the accident data processing method of an autonomous vehicle as provided by any embodiment of the present invention, the method comprising:

The computer storage media of embodiments of the invention may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present invention may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

It will be appreciated by those of ordinary skill in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be centralized on a single computing device, or distributed over a network of computing devices, or they may alternatively be implemented in program code executable by a computer device, such that they are stored in a memory device and executed by the computing device, or they may be separately fabricated as individual integrated circuit modules, or multiple modules or steps within them may be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method of accident data processing for an autonomous vehicle, comprising:

2. The method of claim 1, wherein the driving status information includes vehicle position information, vehicle status information, and vehicle perception information; the accident prediction according to the vehicle operation data, determining the running accident risk of the target vehicle, includes:

3. The method of claim 2, wherein the accident prediction based on the obstacle path information, the vehicle status information, and the path planning information, determining a driving accident risk of the target vehicle, comprises:

4. A method according to claim 3, wherein the vehicle avoidance model comprises an avoidance path planning network and an avoidance probability calculation network; the automatic avoidance prediction is performed according to the obstacle path information, the path planning information and the whole vehicle state information through a preset vehicle avoidance model to obtain a target avoidance path and a vehicle avoidance probability corresponding to the target avoidance path, and the method comprises the following steps:

5. The method according to claim 1, further comprising, before if the running accident risk is high risk running:

Under the condition that the driver operation exists in the detected vehicle-end image data, acquiring the running state of the automatic driving mode;

Setting the running accident risk to a high risk in a case where the automatic driving mode is in operation;

The running accident risk is set to be low risk in the case where the automatic driving mode is off.

6. The method according to claim 5, further comprising, after said setting said running accident risk to be high risk:

7. The method of claim 1, wherein the persisting the vehicle operation data in the target vehicle comprises:

Acquiring an available storage space of the target vehicle;

8. An accident data processing apparatus for an autonomous vehicle, comprising:

9. An electronic device, the electronic device comprising:

at least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the accident data processing method of an autonomous vehicle according to any one of claims 1 to 7.

10. A computer readable storage medium storing computer instructions for causing a processor to perform the method of accident data processing for an autonomous vehicle according to any one of claims 1 to 7.