CN116872923A

CN116872923A - Vehicle running risk prediction method, storage medium and computer device

Info

Publication number: CN116872923A
Application number: CN202311032699.5A
Authority: CN
Inventors: 李梦雨; 宋永康; 曾华荣; 黄萌; 田向远
Original assignee: Ningbo Lutes Robotics Co ltd
Current assignee: Ningbo Lutes Robotics Co ltd
Priority date: 2023-08-16
Filing date: 2023-08-16
Publication date: 2023-10-13

Abstract

The invention provides a vehicle running risk prediction method, a storage medium and computer equipment. Wherein the method comprises the following steps: acquiring obstacle information and environment perception information nearby a vehicle; predicting and obtaining a risk area near the vehicle in a preset time period in the future according to the obstacle information and the environment perception information; and marking the risk area on a map of a man-machine interaction interface in the vehicle. By the scheme, the risk area around the vehicle can be predicted in advance, meanwhile, a driver or a passenger can clearly know the risk area around the vehicle, and the driving scheme can be changed according to the risk area; under the condition of automatic driving of the vehicle, in-vehicle personnel can know the reason of the vehicle modification driving scheme through a human-computer interaction interface, so that the trust of the driver on the automatic driving vehicle is enhanced, the vehicle makes countermeasures for a risk area in advance, the driving comfort of the vehicle can be improved, and the probability of sudden braking and other conditions due to emergency is reduced.

Description

Vehicle running risk prediction method, storage medium and computer device

Technical Field

The present invention relates to intelligent driving vehicles, and more particularly, to a vehicle driving risk prediction method, a storage medium, and a computer device.

Background

Studies have shown that if an intelligent driving car makes an unexpected motion that is dangerous for the driver, the vehicle will take over control, and the driver will lose trust of the intelligent driving function for a long time later, even refuse to use it. Therefore, the potential risk is found in advance, and early warning information or planning control on auxiliary vehicles is provided for drivers, so that the accident hazard degree is reduced, the accident occurrence is avoided, and the method has important significance in supporting the vehicle planning control algorithm and even the driving safety of human beings.

Disclosure of Invention

It is an object of the present invention to predict a risk area around a vehicle in advance.

It is a further object of the invention to improve driver confidence in an autonomous vehicle.

A further object of the invention is to ensure the comfort of the vehicle driving.

In particular, the present invention provides a vehicle running risk prediction method, which includes:

acquiring obstacle information and environment perception information nearby a vehicle;

predicting and obtaining a risk area near the vehicle in a preset time period in the future according to the obstacle information and the environment perception information;

and marking the risk area on a map of a man-machine interaction interface in the vehicle.

Optionally, the step of predicting the risk area near the vehicle in the future preset time period according to the obstacle information and the environmental awareness information includes:

calculating a first track of the vehicle in a future preset time period through a control planning module;

predicting the obstacle information and the environment perception information through a track prediction algorithm to obtain a second track of the obstacle in a preset time period in the future;

and determining a strong interaction position between the first track and the second track, and taking the strong interaction position as a risk area near the vehicle.

Optionally, the step of predicting the risk area near the vehicle in the future preset time period according to the obstacle information and the environmental awareness information further includes:

judging whether the vehicle runs continuously according to the driving scheme and has risks or not through a risk prediction algorithm according to the obstacle information and the driving scheme of the vehicle in combination with the environment perception information;

and (5) mutually supplementing and verifying the judging result and the strong interaction position through a risk prediction algorithm.

Optionally, the step of identifying the risk area on the map of the man-machine interaction interface in the vehicle further comprises:

and calling a control planning module to plan a future driving scheme of the vehicle according to the map.

Optionally, the step of identifying the risk area on a map of the human-machine interaction interface inside the vehicle further comprises:

and carrying out risk classification on the risk areas, and representing different risk grades through different identifications.

Optionally, before the step of risk classification of the risk area, further includes:

road section information near a vehicle in a map is acquired, and the road section information is used for identifying the accident occurrence frequency of the road section, the construction condition of the road section and a special road section, wherein the special road section comprises a pedestrian crossing;

and adding the road section information into the risk area, and executing the step of classifying the risk areas.

and acquiring weather information, and marking the part which can influence normal driving of the vehicle in the weather information in a map.

uploading the map to a cloud end, and timely updating map information on the cloud end;

and when the following vehicles pass through the same position, a risk prompt is given through the cloud.

According to yet another aspect of the present invention, there is also provided a machine-readable storage medium having stored thereon a machine-executable program which, when executed by a processor, implements the vehicle running risk prediction method of any one of the above.

According to still another aspect of the present invention, there is also provided a computer apparatus including a memory, a processor, and a machine executable program stored on the memory and running on the processor, and the processor implementing any one of the above-described vehicle running risk prediction methods when executing the machine executable program.

According to the vehicle running risk prediction method, in the running process of a vehicle, firstly, obstacle information near the vehicle is obtained, then, a risk area near the vehicle in a preset time period in the future is predicted according to the obstacle information, and the risk area is marked on a map of a human-computer interaction interface in the vehicle. By the method, the risk area around the vehicle can be predicted in advance, and meanwhile, a driver or a passenger can clearly know the risk area around the vehicle and can change the driving scheme according to the risk area; under the condition of automatic driving of the vehicle, in-vehicle personnel can know the reason of the vehicle modification driving scheme through a human-computer interaction interface, so that the trust of the driver on the automatic driving vehicle is enhanced, the vehicle makes countermeasures for a risk area in advance, the driving comfort of the vehicle can be improved, and the probability of sudden braking and other conditions due to emergency is reduced.

Further, the vehicle running risk prediction method of the invention can also acquire the road section information and weather information nearby in the map, mark the part which can influence the normal driving of the vehicle in the risk map, then carry out risk classification on the risk area, and represent different risk grades through different marks. Thereby more clearly and comprehensively showing the risk areas near the vehicle.

Further, according to the vehicle running risk prediction method, a first track of a vehicle in a future preset time period is calculated through a control planning module, obstacle information is predicted through a track prediction algorithm to obtain a second track of an obstacle in the future preset time period, a strong interaction position between the first track and the second track is determined, and the strong interaction position is used as a risk area near the vehicle; and judging whether the vehicle runs continuously according to the driving scheme or not through a risk prediction algorithm according to the obstacle information and the driving scheme of the vehicle, and mutually supplementing and verifying the judging result and the strong interaction position through the risk prediction algorithm. Therefore, the risk area near the vehicle is more comprehensively determined through various calculation modes, and the driving safety is ensured.

The above, as well as additional objectives, advantages, and features of the present invention will become apparent to those skilled in the art from the following detailed description of a specific embodiment of the present invention when read in conjunction with the accompanying drawings.

Drawings

Some specific embodiments of the invention will be described in detail hereinafter by way of example and not by way of limitation with reference to the accompanying drawings. The same reference numbers will be used throughout the drawings to refer to the same or like parts or portions. It will be appreciated by those skilled in the art that the drawings are not necessarily drawn to scale. In the accompanying drawings:

FIG. 1 is a flow chart of a method of predicting risk of driving a vehicle according to one embodiment of the invention;

FIG. 2 is a flow chart of a method for predicting risk of driving a vehicle according to another embodiment of the invention;

FIG. 3 is a schematic diagram of a vehicle strong interaction region of a vehicle travel risk prediction method according to one embodiment of the present invention;

FIG. 4 is a flow chart of a method of predicting risk of driving a vehicle according to yet another embodiment of the invention;

FIG. 5 is a flow chart of a method of predicting risk of driving a vehicle according to still another embodiment of the invention;

FIG. 6 is a schematic diagram of a machine-readable storage medium in a method of predicting risk of driving a vehicle in accordance with one embodiment of the invention; and

fig. 7 is a schematic diagram of a computer device in a vehicle running risk prediction method according to an embodiment of the present invention.

Detailed Description

It should be understood by those skilled in the art that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention, and the some embodiments are intended to explain the technical principles of the present invention and are not intended to limit the scope of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive effort, based on the embodiments provided by the present invention, shall still fall within the scope of protection of the present invention.

Fig. 1 is a flowchart of a vehicle running risk prediction method according to an embodiment of the present invention. The process generally includes:

step S101, obtaining obstacle information in the vicinity of the vehicle and environment awareness information. The obstacle information may include static obstacles and dynamic obstacles near the vehicle, the static obstacles may include obstacles such as road fences, the dynamic obstacles may be other vehicles near the vehicle, and the environment sensing information may generally include road condition information such as road traffic lights; the manner of acquiring the obstacle information and the environment awareness information may include: the acquisition is performed by a vehicle own sensor such as a vehicle body radar or the like. The person skilled in the art can determine the range of the obstacle information and the environment sensing information and the acquisition mode according to the actual situation.

And step S102, predicting and obtaining a risk area near the vehicle in a preset time period in the future according to the obstacle information. In some embodiments, the present step may calculate the trajectory of the vehicle itself within a preset time period in the future by a prediction algorithm, and predict the future trajectory of the obstacle in step S101, so as to determine the area where the vehicle may risk in the future according to the interaction of the future trajectory.

Step S103, marking the risk area on a map of a man-machine interaction interface in the vehicle. In some embodiments, the risk areas around the vehicle may be identified in a navigation map of a vehicle central display screen for viewing by a driver or passenger. The man-machine interaction device for displaying the risk area can be determined by a person skilled in the art according to the actual situation.

By the method, the risk area near the vehicle can be predicted in time and displayed to the driver or the passenger, so that the driver can conveniently adjust the driving scheme in time or the passenger can understand the reason that the automatic driving vehicle modifies the driving scheme, the trust of the passenger on the automatic driving vehicle is further enhanced, the vehicle can make countermeasures for the risk area in advance, the driving comfort of the vehicle can be improved, and the probability of sudden braking and other conditions due to emergency is reduced.

In some alternative embodiments, the step of predicting the risk area near the vehicle within the future preset time period according to the obstacle information and the environmental awareness information may include: calculating a first track of the vehicle in a future preset time period through a control planning module; predicting the obstacle information and the environment perception information through a track prediction algorithm to obtain a second track of the obstacle in a preset time period in the future; and determining a strong interaction position between the first track and the second track, and taking the strong interaction position as a risk area near the vehicle. It should be noted that, the first track and the second track in this step are only for distinguishing the two tracks, and the sequence of the two tracks is not limited, and there is no strict sequence limitation between the step of generating the first track and the step of generating the second track, so those skilled in the art can set the execution sequence of the two steps according to the actual situation.

In other embodiments, the step of predicting the risk area near the vehicle within the future preset time period according to the obstacle information and the environmental awareness information may further include: judging whether the vehicle runs continuously according to the driving scheme and has risks or not through a risk prediction algorithm according to the obstacle information and the driving scheme of the vehicle in combination with the environment perception information; and (5) mutually supplementing and verifying the judging result and the strong interaction position through a risk prediction algorithm. By this step, the risk region in the vicinity of the vehicle can be comprehensively predicted.

In this embodiment, road section information near the vehicle in the map may also be obtained, where the road section information is used to identify the occurrence frequency of road section accidents, the construction condition of the road section, and a special road section, an example of which is a road section such as a crosswalk; the link information is then added to the risk area and identified together.

Alternatively, the method for identifying the risk area on the map of the man-machine interaction interface in the vehicle may include: and carrying out risk classification on the risk areas, and representing different risk grades through different identifications. An alternative embodiment is for example: the risk level is distinguished through colors, for example, the position where the possible collision in the future is predicted is regarded as a strong risk, and the risk level can be marked red; the accident-prone road section and the construction road section are regarded as medium risks, and orange can be marked; crosswalk locations are considered low risk and may be marked yellow. The risk level can be determined by a person skilled in the art according to the actual importance level of the road section.

In some embodiments, the step of identifying the risk area on a map of the vehicle interior human-machine interaction interface comprises: the method comprises the steps of obtaining weather information, marking a part of the weather information which can influence normal driving of the vehicle in a map, for example, displaying thunderstorm or heavy fog weather in the map through special animation effects, so that passengers or drivers can clearly know weather conditions near the vehicle, and accordingly, a driving scheme can be adjusted in a targeted mode.

After the step of identifying the risk area on the map of the vehicle interior human-computer interaction interface, the method further comprises: and calling a control planning module to plan a future driving scheme of the vehicle according to the risk map.

Optionally, after the step of identifying the risk area on the map of the man-machine interaction interface in the vehicle, the map can be uploaded to the cloud end, and the map information on the cloud end can be updated in time; and when the following vehicles pass through the same position, a risk prompt is given through the cloud. By the method, the risk areas can be shared in the cloud, so that the calculated amount of the system in the vehicle is reduced.

Fig. 2 is a flowchart illustrating a vehicle traveling risk prediction method according to another embodiment of the present invention. In this embodiment, the flow generally includes:

step S201, acquiring positioning information and map information.

Step S202, obstacle information and environment perception information are acquired. In some alternative embodiments, the step comprises: the positioning information in the root step S201 determines the position of the vehicle, so that the road information near the positioning point of the vehicle is obtained through the map information; in addition, the obstacle information and the environment sensing information near the vehicle can be acquired through the vehicle sensor such as vehicle-mounted radar and the like, wherein the obstacle information can comprise static obstacles and dynamic obstacles near the vehicle, the static obstacles can comprise obstacles such as road fences, the dynamic obstacles can be other vehicles near the vehicle, and the environment sensing information can generally comprise road condition information such as road traffic lights.

In step S203, the obstacle trajectory is calculated by the trajectory prediction algorithm. In some alternative embodiments, the step comprises: and (3) calculating the obstacle information in the step S202 according to a trajectory prediction algorithm to obtain the trajectory of each obstacle in a certain time period in the future.

Step S204, a vehicle planning control module is invoked. In some alternative embodiments, the vehicle planning control module is configured to plan the driving scheme of the vehicle, so as to determine the travel track of the vehicle in the current state, so that in this step, the current driving scheme of the vehicle itself is invoked to output the future travel track of the vehicle.

Step S205, determine whether it is a strong interaction region. In some alternative embodiments, the step comprises: judging whether a strong interaction region occurs in the future according to the obstacle track calculated in the step S203 and the vehicle self-travelling track calculated in the step S204, wherein the strong interaction region represents a region in which dangerous states such as collision occur in the future, and the like, and a person skilled in the art can determine specific judging conditions of the strong interaction region according to the actual situation of the vehicle and the surrounding situation of the vehicle.

In step S206, a future risk of the vehicle is calculated by a risk prediction algorithm. In some alternative embodiments, the risk prediction algorithm may calculate information such as whether a collision will occur in the future and the time of occurrence of the collision according to information such as the speed and the distance of the vehicle and the obstacle. The calculation result of the step can mutually verify the strong interaction region in the step S205, so that the future risk region of the vehicle can be found more accurately and comprehensively.

Step S207, when step S205 determines yes, generates a risk map. The method comprises the following steps: acquiring the risk area calculated in the steps, and then acquiring information of a special road section near the vehicle in the map, wherein the special road section generally comprises: adding the special road sections into a risk area, such as accident multiple road sections, construction road sections, crosswalk and the like; and carrying out risk classification on the risk areas, and representing different risk grades through different identifications. In some alternative embodiments, the risk level may be differentiated by color, e.g., locations where future collisions are predicted to be likely are considered strong risks, may be reddish; the accident-prone road section and the construction road section are regarded as medium risks, and orange can be marked; crosswalk locations are considered low risk and may be marked yellow. The type of a particular road segment and its risk level can be determined by those skilled in the art based on the actual importance of the road segment.

Alternatively, other factors that can affect the driving safety of the vehicle may be noted in the risk map, for example, weather information, and when the vehicle obtains weather that can affect the driving safety of the vehicle, such as rainy days or foggy days, the weather information may be noted in the risk map, so as to remind the dangerous area in front. The type of information to be added in the risk map can be determined by a person skilled in the art according to actual conditions, so that the risk area is more comprehensively marked in the risk map.

After the automatic driving vehicle finishes executing step S207, the vehicle planning control module may be continuously invoked to adjust the driving scheme of the vehicle according to the risk area in the risk map, so as to ensure the safety and comfort of the vehicle driving, and avoid measures such as sudden braking and the like for avoiding obstacles.

And step S208, uploading the cloud. The method comprises the following steps: after the risk map is generated, uploading the risk map to the cloud, and arranging and splicing the risk map generated before the vehicle at the cloud, so that the risk map on the cloud end is updated, when any vehicle passes through an area contained in the risk map, the corresponding risk map can be obtained at the cloud, and the cloud can give out a corresponding risk prompt, so that the driving safety of the vehicle is guaranteed.

Step S209, displaying a man-machine interaction interface. The method comprises the following steps: the generated risk map is displayed on a human-computer interaction interface of the vehicle, so that a driver or a passenger can clearly know a dangerous area near the vehicle, and the driving scheme is adjusted in a targeted manner. The man-machine interaction interface can be a vehicle central control display screen or a position of an instrument panel and the like capable of displaying pictures, and a person skilled in the art can automatically adjust the position and the size of the risk map display according to the actual situation of the vehicle or the wish of the person in the vehicle.

Through the method, the dangerous area near the vehicle can be clearly identified in the risk map and displayed on the man-machine interaction interface, so that a driver can change the driving scheme in a targeted manner, and under the condition of automatic driving, the driving scheme can be automatically adjusted according to the risk map, and a passenger can clearly know the reason of changing the driving scheme, thereby enhancing the trust of the passenger on the automatic driving vehicle and simultaneously ensuring the safety and the comfort of the running of the vehicle.

Fig. 3 is a schematic view of a vehicle strong interaction region of a vehicle running risk prediction method according to an embodiment of the present invention. In some optional embodiments, the schematic diagram includes a host vehicle 301 and a host vehicle 302, where the host vehicle 301 represents the target vehicle itself, the host vehicle 302 represents other vehicles near the target vehicle, where the host vehicle 301 calculates a driving track 303 in a future preset time period through a system, and the host vehicle 302 calculates a driving track 304 in the future preset time period through the system, and obtains a strong interaction area 305 where the host vehicle 301 and the driving track 304 may occur in the future according to the driving track 303 and the driving track 304, so as to mark the strong interaction area 305, and display the strong interaction area on a man-machine interaction interface, so that a driver or a passenger can clearly see the possible situations of the host vehicle 301 and the host vehicle 302 in the future, and further change a driving scheme, and ensure driving safety; when the vehicle is an autonomous vehicle, the passenger can clearly know the reason that the autonomous vehicle changes the driving scheme, thereby enhancing the trust feeling of the passenger on the autonomous vehicle.

Fig. 4 is a flowchart of a vehicle running risk prediction method according to still another embodiment of the present invention. In this embodiment, the flow generally includes:

step S401, acquiring positioning information and map information.

Step S402, obtaining obstacle information and environment perception information. In some alternative embodiments, the step comprises: the positioning information in the root step S401 determines the position of the vehicle, so that the road information near the positioning point of the vehicle is obtained through the map information; in addition, the obstacle information and the environment sensing information near the vehicle can be acquired through the vehicle sensor such as vehicle-mounted radar and the like, wherein the obstacle information can comprise static obstacles and dynamic obstacles near the vehicle, the static obstacles can comprise obstacles such as road fences, the dynamic obstacles can be other vehicles near the vehicle, and the environment sensing information can generally comprise road condition information such as road traffic lights.

Step S403, calculating an obstacle trajectory by the autopilot motion prediction model. In some alternative embodiments, the step comprises: the calculation of the obstacle information in step S402 according to the autopilot motion prediction model (langcn model) to obtain the track of each obstacle in a certain time period in the future specifically includes: firstly, the vehicle and the obstacle information are put into an actronet (actor network module) for feature extraction, the map information is put into a MapNet (map network module) for feature extraction, then the features are fused in series, and finally the predicted track of the obstacle is output. Examples of the above fusion process are as follows: A2L (actor to lane obstacle and road) expresses real-time traffic flow, and can provide real-time road information for the obstacle, such as traffic jam or road occupancy; L2L (lane to lane roads and roads) expresses whether the traffic information of the road can affect other roads, so that the traffic information is transmitted on the lane diagrams and used for updating the node characteristics of each lane; the L2A (lane to action road and obstacle) expresses that updated traffic influences vehicle behavior, so that updated map features and real-time traffic information can be fused, and fed back to road participants; A2A (action to action) expresses interaction between obstacles, is used for processing interaction between the obstacles, and finally outputs characteristics of road participants, and then obtains a predicted track of the obstacle through a residual network. The person skilled in the art can decide the use method of the automatic driving movement prediction model according to the actual situation.

Step S404, a vehicle planning control module is invoked. In some alternative embodiments, the vehicle planning control module is configured to plan the driving schemes of the vehicle and may predict future travel trajectories for each driving scheme, so in this step, the current driving scheme of the vehicle is output as the future travel trajectories by invoking the vehicle planning control module.

Step S405, determining whether the area is a strong interaction area. In some alternative embodiments, the step comprises: judging whether a strong interaction region appears in the future according to the obstacle track calculated in the step S403 and the vehicle self-travelling track calculated in the step S404, wherein the strong interaction region represents a region where the two tracks overlap or are about to overlap in the future, namely a collision region, and a specific judging condition of the strong interaction region can be determined by a person skilled in the art according to the actual condition of the vehicle and the surrounding condition of the vehicle.

In step S406, a future risk of the vehicle is calculated by a TTC (Time-To-Collision Time) algorithm. In some alternative embodiments, the calculation formula of the TTC algorithm may be:

according to the formula, the continuous running speed and direction according to the current vehicle driving scheme can be calculated, the time for collision between the vehicle and the obstacle is generated, and when the collision is actually smaller than a preset threshold value, the area is marked as a risk area. The calculation result of the step can mutually verify the strong interaction region in the step S405, so that the future risk region of the vehicle can be found more accurately and comprehensively.

Step S407, when step S405 determines yes or step S406 is smaller than a preset threshold, generates a risk map. The method comprises the following steps: acquiring the risk area calculated in the steps, and then acquiring information of a special road section near the vehicle in the map, wherein the special road section generally comprises: adding the special road sections into a risk area, such as accident multiple road sections, construction road sections, crosswalk and the like; and carrying out risk classification on the risk areas, and representing different risk grades through different identifications. In some alternative embodiments, the risk level may be differentiated by color, e.g., locations where future collisions are predicted to be likely are considered strong risks, may be reddish; the accident-prone road section and the construction road section are regarded as medium risks, and orange can be marked; crosswalk locations are considered low risk and may be marked yellow. The type of a particular road segment and its risk level can be determined by those skilled in the art based on the actual importance of the road segment.

After the automatic driving vehicle finishes executing step S407, the vehicle planning control module can be continuously invoked to adjust the driving scheme of the vehicle according to the risk area in the risk map, so that the driving safety and comfort of the vehicle are ensured, and measures for avoiding obstacles such as sudden braking are avoided.

Step S408, uploading the cloud. The method comprises the following steps: after the risk map is generated, uploading the risk map to the cloud, and arranging and splicing the risk map generated before the vehicle at the cloud, so that the risk map on the cloud end is updated, when any vehicle passes through an area contained in the risk map, the corresponding risk map can be obtained at the cloud, and the cloud can give out a corresponding risk prompt, so that the driving safety of the vehicle is guaranteed.

Step S409, displaying a man-machine interaction interface. The method comprises the following steps: the generated risk map is displayed on a human-computer interaction interface of the vehicle, so that a driver or a passenger can clearly know a dangerous area near the vehicle, and the driving scheme is adjusted in a targeted manner. The man-machine interaction interface can be a vehicle central control display screen or a position of an instrument panel and the like capable of displaying pictures, and a person skilled in the art can automatically adjust the position and the size of the risk map display according to the actual situation of the vehicle or the wish of the person in the vehicle.

By the method, the track of the obstacle nearby the vehicle and the vehicle track calculated by the vehicle planning control module can be obtained through calculation of the automatic driving movement prediction model to judge, so that a risk area is found; meanwhile, calculating the time of collision between vehicles through a TTC algorithm, and judging the collision time as a risk area when the collision time is smaller than a preset threshold value; the two risk areas are mutually supplemented and verified, so that the risk area near the vehicle is more comprehensively found, the risk area near the vehicle is clearly identified in the risk map and displayed on the human-computer interaction interface, a driver can change the driving scheme in a targeted manner, under the condition of automatic driving, the driving scheme can be automatically adjusted according to the risk map, and a passenger can clearly know the reason of changing the driving scheme, so that the trust feeling of the passenger on the automatic driving vehicle is enhanced, and meanwhile, the safety and the comfort of the vehicle driving are ensured.

Fig. 5 is a flowchart of a vehicle running risk prediction method according to still another embodiment of the present invention. In this embodiment, the flow generally includes:

step S501, acquiring positioning information and map information.

Step S502, obtaining obstacle information and environment awareness information. In some alternative embodiments, the step comprises: the positioning information in the root step S501 determines the position of the vehicle, so that the road information near the positioning point of the vehicle is obtained through the map information; in addition, the obstacle information and the environment sensing information near the vehicle can be acquired through the vehicle sensor such as vehicle-mounted radar and the like, wherein the obstacle information can comprise static obstacles and dynamic obstacles near the vehicle, the static obstacles can comprise obstacles such as road fences, the dynamic obstacles can be other vehicles near the vehicle, and the environment sensing information can generally comprise road condition information such as road traffic lights.

Step S503, predicting the obstacle track by mmtransducer method. In some alternative embodiments, the step comprises: the track prediction of the obstacle information in step S502 is performed according to mmtransducer (multi-modal motion prediction framework), specifically including: and inputting the map information and the obstacle information into a map extractor and a motion extractor, and obtaining a predicted track after feature integration and selection. According to the method, a training strategy based on the region is provided, tracks at different positions are used as candidate inputs, the correlation among different tracks is reduced, the output multi-mode characteristics can be guaranteed, and the tracks of the obstacle can be predicted more accurately.

Step S504, a vehicle planning control module is invoked. In some alternative embodiments, the vehicle planning control module is used to plan the driving scheme of the vehicle itself, so in this step, the current driving scheme of the vehicle itself is output in the future travel track by invoking the vehicle planning control module.

Step S505, determine whether it is a strong interaction region. In some alternative embodiments, the step comprises: judging whether a strong interaction region appears in the future according to the obstacle track calculated in the step S503 and the vehicle self-travelling track calculated in the step S504, wherein the strong interaction region represents a region where the two tracks overlap or are about to overlap in the future, namely a collision region, and a specific judging condition of the strong interaction region can be determined by a person skilled in the art according to the actual condition of the vehicle and the surrounding condition of the vehicle.

In step S506, the future risk of the vehicle is calculated by a DRAC (Deceleration Rate to Avoid a Crash collision avoidance deceleration rate) algorithm. In some alternative embodiments, the calculation formula of the DRAC algorithm may be:

according to the formula, the speed and the direction according to the current driving scheme of the vehicle can be calculated, the time for collision between the vehicle and the front vehicle is calculated, and when the collision is actually smaller than a preset threshold value, the area is marked as a risk area. The calculation result of the step can mutually verify the strong interaction region in the step S505, so that the future risk region of the vehicle can be found more accurately and comprehensively.

In step S507, if it is determined in step S505 that the risk map is yes or if step S506 is smaller than a preset threshold value, a risk map is generated. The method comprises the following steps: acquiring the risk area calculated in the steps, and then acquiring information of a special road section near the vehicle in the map, wherein the special road section generally comprises: adding the special road sections into a risk area, such as accident multiple road sections, construction road sections, crosswalk and the like; and carrying out risk classification on the risk areas, and representing different risk grades through different identifications. In some alternative embodiments, the risk level may be differentiated by color, e.g., locations where future collisions are predicted to be likely are considered strong risks, may be reddish; the accident-prone road section and the construction road section are regarded as medium risks, and orange can be marked; crosswalk locations are considered low risk and may be marked yellow. The type of a particular road segment and its risk level can be determined by those skilled in the art based on the actual importance of the road segment.

After the automatic driving vehicle finishes executing step S507, the vehicle planning control module can be continuously invoked to adjust the driving scheme of the vehicle according to the risk area in the risk map, so that the driving safety and comfort of the vehicle are ensured, and measures for avoiding obstacles such as sudden braking and the like are reduced.

Step S508, uploading the cloud. The method comprises the following steps: after the risk map is generated, uploading the risk map to the cloud, and arranging and splicing the risk map generated before the vehicle at the cloud, so that the risk map on the cloud end is updated, when any vehicle passes through an area contained in the risk map, the corresponding risk map can be obtained at the cloud, and the cloud can give out a corresponding risk prompt, so that the driving safety of the vehicle is guaranteed.

Step S509, displaying a man-machine interaction interface. The method comprises the following steps: the generated risk map is displayed on a human-computer interaction interface of the vehicle, so that a driver or a passenger can clearly know a dangerous area near the vehicle, and the driving scheme is adjusted in a targeted manner. The man-machine interaction interface can be a vehicle central control display screen or a position of an instrument panel and the like capable of displaying pictures, and a person skilled in the art can automatically adjust the position and the size of the risk map display according to the actual situation of the vehicle or the wish of the person in the vehicle.

By the method, the track of the obstacle nearby the vehicle and the vehicle track calculated by the vehicle planning control module can be calculated by the mmTransformer method to judge, so that a risk area is found; the time of collision between the vehicle and the front vehicle is calculated through the DRAC algorithm, when the time of collision is smaller than a preset threshold value, the risk area is judged, and then the two risk areas are mutually supplemented and verified, so that the risk area near the vehicle is more comprehensively found, the risk area near the vehicle is clearly marked in a risk map, and a man-machine interaction interface is displayed, so that a driver can pertinently change a driving scheme.

The present implementation also provides a machine-readable storage medium and a computer device. Fig. 6 is a schematic diagram of a machine-readable storage medium 601 according to one embodiment of the invention, and fig. 7 is a schematic diagram of a computer device 703 according to one embodiment of the invention.

The machine-readable storage medium 601 has stored thereon a machine-executable program 602, which when executed by a processor, implements the vehicle running risk prediction method of any of the above embodiments.

The computer device 703 may include a memory 701, a processor 702, and a machine executable program 602 stored on the memory 701 and running on the processor 702, and the processor 702 implements the vehicle running risk prediction method of any of the embodiments described above when executing the machine executable program 602.

It should be noted that the logic and/or steps represented in the flow diagrams or otherwise described herein, e.g., calculating an obstacle trajectory, may be embodied in any machine-readable storage medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.

It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.

The computer device 703 may comprise a processor 702 adapted to execute stored instructions, a memory 701 providing temporary storage for the operation of said instructions during operation. Processor 702 may be a single-core processor, a multi-core processor, a computing cluster, or any number of other configurations. Memory 701 may include Random Access Memory (RAM), read only memory, flash memory, or any other suitable storage system.

The processor 702 may be connected through a system interconnect (e.g., PCI-Express, etc.) to an I/O interface (input/output interface) adapted to connect the computer device 703 to one or more I/O devices (input/output devices). The I/O devices may include, for example, a keyboard and a pointing device, which may include a touch pad or touch screen, among others. The I/O device may be a built-in component of the computer device 703 or may be a device externally connected to the computing device.

The processor 702 may also be linked through a system interconnect to a display interface suitable for connecting the computer device 703 to a display device. The display device may include a display screen as a built-in component of the computer device 703. The display device may also include a computer monitor, television, projector, or the like, which is externally connected to the computer device 703. In addition, a network interface controller (network interface controller, NIC) may be adapted to connect the computer device 703 to a network through a system interconnect. In some embodiments, the NIC may use any suitable interface or protocol (such as an internet small computer system interface, etc.) to transfer data. The network may be a cellular network, a radio network, a Wide Area Network (WAN), a Local Area Network (LAN), or the internet, among others. The remote device may be connected to the computing device through a network.

The flowcharts provided by this embodiment are not intended to indicate that the operations of the method are to be performed in any particular order, or that all of the operations of the method are included in all of each case. Furthermore, the method may include additional operations. Additional variations may be made to the above-described methods within the scope of the technical ideas provided by the methods of the present embodiments.

By now it should be appreciated by those skilled in the art that while a number of exemplary embodiments of the invention have been shown and described herein in detail, many other variations or modifications of the invention consistent with the principles of the invention may be directly ascertained or inferred from the present disclosure without departing from the spirit and scope of the invention. Accordingly, the scope of the present invention should be understood and deemed to cover all such other variations or modifications.

Claims

1. A vehicle travel risk prediction method, comprising:

acquiring the information of the obstacle nearby the vehicle and the environment perception information;

predicting and obtaining a risk area near the vehicle in a preset future time period according to the obstacle information and the environment perception information;

and marking the risk area on a map of the man-machine interaction interface in the vehicle.

2. The vehicle running risk prediction method according to claim 1, wherein,

the step of predicting the risk area near the vehicle within a preset future time period according to the obstacle information and the environment perception information comprises the following steps:

calculating a first track of the vehicle in the future preset time period through a control planning module;

predicting the obstacle information and the environment perception information through a track prediction algorithm to obtain a second track of the obstacle in the future preset time period;

3. The vehicle running risk prediction method according to claim 2, wherein,

the step of predicting the risk area near the vehicle within the preset future time period according to the obstacle information and the environment perception information further comprises the following steps:

judging whether the vehicle continues to run according to the driving scheme with risk or not through a risk prediction algorithm according to the obstacle information and the environment perception information and the driving scheme of the vehicle;

and mutually supplementing and verifying the judging result of the risk prediction algorithm and the strong interaction position.

4. The vehicle running risk prediction method according to claim 2, wherein,

the step of identifying the risk area on the map of the vehicle interior man-machine interaction interface further comprises:

and calling a control planning module to plan the future driving scheme of the vehicle according to the map.

5. The vehicle running risk prediction method according to claim 1, wherein,

the step of identifying the risk area on a map of the vehicle interior human-computer interaction interface further comprises:

6. The vehicle running risk prediction method according to claim 5, wherein,

before the step of classifying the risk areas, the method further comprises:

acquiring road section information near the vehicle in a map, wherein the road section information is used for identifying the occurrence frequency of road section accidents, road section construction conditions and special road sections, and the special road sections comprise crosswalk;

7. The vehicle running risk prediction method according to claim 1, wherein,

and acquiring weather information, and marking the part which can influence the normal driving of the vehicle in the weather information in the map.

8. The vehicle running risk prediction method according to claim 1, wherein,

uploading the map to a cloud end, and timely updating the map information on the cloud end;

9. A machine-readable storage medium having stored thereon a machine-executable program which when executed by a processor implements the vehicle running risk prediction method according to any one of claims 1 to 8.

10. A computer device comprising a memory, a processor and a machine executable program stored on the memory and running on the processor, and the processor implementing the vehicle running risk prediction method according to any one of claims 1 to 8 when executing the machine executable program.