CN113071515B

CN113071515B - Movable carrier control method, device, movable carrier and storage medium

Info

Publication number: CN113071515B
Application number: CN202110353481.4A
Authority: CN
Inventors: 吴荣华; 咸志伟; 韩旭
Original assignee: Dongfeng Motor Corp; Guangzhou Weride Technology Co Ltd
Current assignee: Dongfeng Motor Corp; Guangzhou Weride Technology Co Ltd
Priority date: 2021-03-31
Filing date: 2021-03-31
Publication date: 2022-07-05
Anticipated expiration: 2041-03-31
Also published as: CN113071515A

Abstract

The invention belongs to the technical field of automatic driving, and discloses a movable carrier control method and device, a movable carrier and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining the current road type where a target movable carrier is located, and obtaining the current driving information of the target movable carrier; determining a trend of change of movable objects around the target movable carrier; determining the type of the current takeover scene according to the variation trend, the current driving information and the current road type; generating a corresponding current takeover event based on the current takeover scene type; and determining a corresponding target execution instruction according to the current takeover event. Through the mode, the corresponding takeover scene is determined according to the current road type, the running information of the target movable carrier and the change trend of the peripheral movable objects, and the target movable carrier is controlled by adopting proper takeover operation, so that the safe running of the target movable carrier in the unmanned driving state is ensured.

Description

Movable carrier control method, device, movable carrier and storage medium

Technical Field

The invention relates to the technical field of automatic driving, in particular to a movable carrier control method and device, a movable carrier and a storage medium.

Background

With the development of science and technology, the automatic driving technology is mature, the automatic driving scheme for the movable carrier is diversified, but the taking over problem of the movable carrier capable of realizing automatic driving is followed, for example, for the automatic driving of an unmanned vehicle, if an event or a problem needing to be taken over is met in the driving process, manual taking over is required by a user or an administrator, and the automatic driving is converted into manual driving, so that the user experience is influenced.

Disclosure of Invention

The invention mainly aims to provide a movable carrier control method, a movable carrier control device, a movable carrier and a storage medium, and aims to solve the technical problem that manual pipe connection is still needed when the movable carrier meets the pipe connection problem in the prior art.

To achieve the above object, the present invention provides a movable carrier control method, comprising the steps of:

the method comprises the steps of obtaining the current road type where a target movable carrier is located, and obtaining the current driving information of the target movable carrier;

determining a trend of change of movable objects around the target movable carrier;

determining the type of the current takeover scene according to the change trend, the current driving information and the current road type;

generating a corresponding current takeover event based on the current takeover scene type;

and determining a corresponding target execution instruction according to the current takeover event.

Optionally, the determining a trend of change of the movable objects around the target movable carrier specifically includes:

acquiring current image information and current point cloud information around the target movable carrier;

fusing according to the current image information and the current point cloud information to obtain current three-dimensional image information;

and determining the change trend of movable objects around the target movable carrier according to the current three-dimensional image information and the historical three-dimensional image information.

Optionally, the trend includes a flow rate trend and a speed trend;

the determining the change trend of the movable objects around the target movable carrier according to the current three-dimensional image information and the historical three-dimensional image information specifically comprises:

obtaining the flow difference and position change information of the movable object according to the current three-dimensional image information and the historical three-dimensional image information, and obtaining the time interval information of the three-dimensional image information and the historical three-dimensional image information;

obtaining a flow change curve of the movable object by fitting a prediction model based on the flow difference and the time interval information;

obtaining the flow change trend of the movable object according to the slope of the flow change curve;

obtaining a speed change curve of the movable object by fitting a prediction model based on the position change information and the time interval information;

and obtaining the speed change trend of the movable object according to the slope of the speed change curve.

Optionally, the determining a current takeover scene type according to the change trend, the current driving information, and the current road type specifically includes:

extracting texture features of the current road type, and determining the complexity of the current scene through a road detection model based on the change trend and the texture features;

acquiring a first driving direction, a first driving position and a first driving speed in the current driving information;

acquiring a second driving direction, a second driving position and a second driving speed of a movable object around the target movable carrier according to the three-dimensional image information;

determining a relative position, a relative travel direction and a relative travel speed of the target movable carrier and the movable object based on the first travel direction, the first travel position, the first travel speed, the second travel direction, the second travel position and the second travel speed, and taking the relative position, the relative travel direction and the relative travel speed as relative travel data;

performing collision prediction simulation according to the complexity and the relative driving data to obtain a collision risk estimation result;

and determining the corresponding current takeover scene type according to the collision risk estimation result.

Optionally, the acquiring a second driving direction, a second driving position, and a second driving speed of the movable object around the target movable carrier according to the three-dimensional image information specifically includes:

acquiring a second driving speed and a second driving position of the movable object according to the three-dimensional image;

extracting warning features of the movable object in the three-dimensional image;

and obtaining a second predicted driving direction of the movable object through a preset behavior prediction model according to the warning characteristics, the driving speed and the driving position.

Optionally, the determining a corresponding target execution instruction according to the current takeover event specifically includes:

sending the takeover event to a server so that the server feeds back a corresponding event solution instruction according to the takeover event;

predicting a command to be operated at the next moment of automatic driving when the event solving command is received;

and comparing the instruction to be operated with the event solving instruction to obtain a target execution instruction.

Optionally, the comparing the instruction to be operated with the event resolution instruction to obtain a target execution instruction specifically includes:

acquiring a preset road safety range corresponding to the current road type;

obtaining a first possibility value that the predicted running state of the target movable carrier exceeds the preset road safety range through a preset behavior prediction model according to the current running information of the target movable carrier and the instruction to be operated;

obtaining a second possibility value that the predicted driving state of the target movable carrier exceeds the preset road safety range through a preset behavior prediction model according to the current driving information of the target movable carrier and the event solving instruction;

setting a probability value with a smaller value of the first probability value and the second probability value as a target probability value;

and determining an instruction corresponding to the target possibility value, and taking the determined instruction as a target execution instruction.

Further, to achieve the above object, the present invention also proposes a movable carrier control apparatus including:

the information acquisition module is used for acquiring the current road type of a target movable carrier and acquiring the current driving information of the target movable carrier;

a trend determination module for determining a trend of change of movable objects around the target movable carrier;

the type determining module is further used for determining the type of the current takeover scene according to the change trend, the current driving information and the current road type;

an event generating module, configured to generate a corresponding current takeover event based on the current takeover scenario type;

and the instruction determining module is used for determining a corresponding target execution instruction according to the current takeover event.

In addition, to achieve the above object, the present invention provides a server, including: a memory, a processor, and a removable carrier control program stored on the memory and executable on the processor, the removable carrier control program configured to implement the steps of the removable carrier control method as described above.

Further, to achieve the above object, the present invention also provides a computer-readable storage medium having stored thereon a removable carrier control program which, when executed by a processor, implements the steps of the removable carrier control method as described above.

The method comprises the steps of obtaining the current road type of a target movable carrier and obtaining the current driving information of the target movable carrier; determining a trend of change of movable objects around the target movable carrier; determining the type of the current takeover scene according to the change trend, the current driving information and the current road type; generating a corresponding current takeover event based on the current takeover scene type; and determining a corresponding target execution instruction according to the current takeover event. Through the mode, the corresponding takeover scene type is determined according to the current road type, the running information of the target movable carrier and the change trend of the peripheral movable objects, the target movable carrier is controlled by adopting a proper takeover execution instruction according to the takeover scene type, the movable carrier can solve the current problem through the movable carrier and the server when meeting the takeover problem in the unmanned state, the intelligence of automatic driving is improved, meanwhile, the safe running of the movable carrier is guaranteed, and the user experience is improved.

Drawings

FIG. 1 is a schematic flow chart diagram of a method for controlling a movable carrier according to a first embodiment of the present invention;

fig. 2 is a schematic flow chart of step S30 in the second embodiment of the movable carrier control method according to the present invention;

FIG. 3 is a schematic diagram of the positional relationship of a target movable carrier and surrounding movable objects in an embodiment of the present invention;

fig. 4 is a schematic flow chart of step S50 of the movable carrier control method according to the third embodiment of the present invention;

FIG. 5 is a block diagram of a movable carrier control apparatus according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a removable carrier for a hardware operating environment according to an embodiment of the present invention;

the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Referring to fig. 1, fig. 1 is a schematic flow chart of a movable carrier control method according to a first embodiment of the present invention.

In this embodiment, the movable carrier control method includes the steps of:

step S10, acquiring the current road type where the target movable carrier is located, and acquiring the current driving information of the target movable carrier.

It should be noted that the execution subject of the present embodiment is a movable carrier, and the movable carrier may have various expressions, such as a carrier with a moving capability, such as an automobile, a robot, an aircraft, and the like, which is not limited in this embodiment.

It can be understood that the image acquisition is performed through a camera and a sensor carried on the target movable carrier, so as to obtain the information of the peripheral road where the target movable carrier is located, and the current road type is obtained through a recognition and detection algorithm.

It is understood that the current traveling information includes, but is not limited to, information of a traveling speed, a traveling direction, a traveling acceleration, and a roll angle during traveling of the target movable carrier.

Step S20, determining a trend of change of the movable objects around the target movable carrier.

It should be noted that, in the current environment where the target movable carrier is located, movable objects, such as people, trucks, bicycles, etc., may have certain influence on the traveling of the target movable carrier, and may collide or scratch, so that a change trend of the movable objects around the target movable carrier needs to be obtained in real time.

It should be understood that, for the traveling of the target movable carrier, the peripheral information needs to be acquired to further obtain the variation trend of the movable objects around the target movable carrier, and in this embodiment, determining the variation trend of the movable objects around the target movable carrier may specifically include:

acquiring current image information and current point cloud information around the target movable carrier; fusing according to the current image information and the current point cloud information to obtain current three-dimensional image information; and determining the change trend of movable objects around the target movable carrier according to the current three-dimensional image information and the historical three-dimensional image information, and determining the change trend of the movable objects around by comparing the current three-dimensional image information with the historical three-dimensional image information, so that the subsequent process of determining the type of the takeover scene is more accurate.

It should be noted that the point cloud information is data information obtained by lidar (light Detection And ranging).

In a specific implementation, corresponding time data exists in both the image information and the point cloud information, the acquired image information and the point cloud information are subjected to corresponding temporal matching fusion through image information acquired by a cloud end connected between a target mobile terminal and the target mobile terminal to obtain three-dimensional image information around a target mobile carrier, assuming that data included in the image information is { C1, C2, C3 … …, Cn }, and data included in the point cloud information is { Q1, Q2, Q3 … …, Qn }, where 1, 2, 3, … …, n respectively represent different times, at this time, the image data and the point cloud data are subjected to matching fusion according to corresponding times, for example, data included in historical three-dimensional image information corresponding to time 1 is { C1, Q1} and data included in historical three-dimensional image information corresponding to time 2 are taken as { C2, Q2}, ..., using { Cn, Qn } as data included in the current three-dimensional image information corresponding to time n.

It can be understood that, according to the comparison result between the current three-dimensional image information and the historical three-dimensional image information, the change trend of the movable object around the target movable carrier can be determined, but the change trend of the movable object includes a flow rate change trend and a speed change trend, and there is a specific analysis manner for the flow rate change trend and the speed change trend, in this embodiment, the determining the change trend of the movable object around the target movable carrier according to the current three-dimensional image information and the historical three-dimensional image information may specifically include:

obtaining the flow difference and position change information of the movable object according to the current three-dimensional image information and the historical three-dimensional image information, and obtaining the time interval information of the three-dimensional image information and the historical three-dimensional image information; obtaining a flow change curve of the movable object by fitting a prediction model based on the flow difference and the time interval information; obtaining the flow change trend of the movable object according to the slope of the flow change curve; obtaining a speed change curve of the movable object by fitting a prediction model based on the position change information and the time interval information; and obtaining the speed change trend of the movable object according to the slope of the speed change curve.

It should be noted that, the current three-dimensional image information and the historical three-dimensional image are compared to obtain time interval information between the current three-dimensional image information and the historical three-dimensional image information, the movable objects existing in the two pieces of image information can be obtained by performing recognition and classification through depth learning, the running speed of the movable objects is obtained according to the position change information of the movable objects in the three-dimensional image information, the running speed in a preset sampling period is fitted with a prediction model, and finally a speed change curve of the movable objects around the target movable carrier is obtained.

For example, according to the comparison between the current three-dimensional image and the historical three-dimensional image, if the number of vehicles around the target movable carrier in the current three-dimensional image is greater than the number of vehicles around the target movable carrier in the historical three-dimensional image, the trend of the flow rate change of the movable objects around the target movable carrier is in a growing state. Meanwhile, according to the position changes of the movable objects around the target movable carrier in the current three-dimensional image and the historical three-dimensional image in the identification interval time, the movable vehicle A is positioned behind the target movable carrier in the historical three-dimensional image, but the movable vehicle A is positioned in front of the target movable carrier in the current three-dimensional image after ten minutes, and the acceleration state of the speed change trend of the movable vehicle A around the target movable carrier is indicated.

In a specific implementation, according to the flow difference change of the movable object in the current three-dimensional image information and the historical three-dimensional image information and the time interval information, the flow change in a preset sampling period is subjected to a fitting prediction model, and finally a flow change curve of the movable object around the target movable carrier is obtained, for example, 22: 32: historical three-dimensional image at time 05 with 22: 32: historical three-dimensional image at time 15 22: 32: comparison at 15 hours shows that the number of people around the detected target mobile carrier is changed from 8 to 12, and 22: 32: 15 time history three-dimensional image and 22: 32: comparing the historical three-dimensional images at 25 hours, changing the number of the people around the detected target movable carrier from 12 to 15, and comparing the number of the people around the detected target movable carrier with the number of the people around the detected target movable carrier, namely 22: 32: 25 with 22: 32: 35, comparing the current three-dimensional images, detecting that the number of people around the target movable carrier is changed from 15 to 32, namely that the change trend of the human flow is gradually increased, obtaining a change curve of the human flow by fitting a prediction model, obtaining a predicted change trend through data and curves, namely the human flow is 0.75 times +1.75, and predicting that the ratio of the number of people around the target movable carrier is 22: 32: at 45, the number of people around the target mobile carrier is about 35.

In the embodiment, the flow rate variation trend and the speed variation trend included in the variation trend are analyzed in different manners, and the variation trend of surrounding movable objects is comprehensively considered from the speed variation trend and the flow rate variation trend, so that a relatively comprehensive variation trend result is finally obtained.

And step S30, determining the type of the current takeover scene according to the change trend, the current driving information and the current road type.

It should be noted that the complexity of the current scene is obtained according to the flow and the speed variation trend of the movable carrier around the target carrier in the current environment and the current road type, for example, the current road type is a tunnel, no pedestrian exists around the target movable carrier, only vehicles running at a constant speed are used, the flow variation trend and the speed variation trend are stable, and the complexity of the current scene is a low degree; the current road type is a rugged mountain road, a few pedestrians are around the target movable carrier, and the number of vehicles is small, but the complexity of the current scene is high due to the rugged mountain road and the complex terrain; the current road type is a road near school, the traffic trend of surrounding vehicles and people is in increasing change, although the speed change trend of surrounding vehicles and people is stable, the traffic is large, the road condition of the current scene is complex, accidents are easy to happen, and the complexity of the current scene is high.

In specific implementation, after the complexity of the current scene is determined, whether the taking over is needed or not is determined based on the driving information of the target movable carrier, and if the taking over is determined to be needed, the type of the current taking over scene is determined based on the driving information of the target movable carrier. For example, when the complexity of the current scene is low, all data in the driving information of the current movable carrier are normal, and the variation trend of the movable object around the target movable carrier is a decreasing situation, the current scene does not need to take over, if any one of the complexity of the current scene, the current driving information and the variation trend is abnormal, take over is triggered, for example, when the complexity of the current scene is medium, the driving speed in the driving information of the current movable carrier is fast, and the flow and speed variation trend of the surrounding movable object is decreasing, take over is triggered, and the current take-over scene type is a medium-risk take-over scene type, when the complexity of the current scene is high, the driving speed in the driving information of the current movable carrier is still in a high-speed driving state, and the flow and speed variation trend of the surrounding movable object are increasing, the current takeover scene type is a high-risk takeover scene type, and the determination of the takeover scene type may also be according to the complexity and the driving information of the current scene under other conditions, which is not limited in this embodiment.

And step S40, generating a corresponding current takeover event based on the current takeover scene type.

It should be noted that after determining the current takeover scenario type, a corresponding takeover event may be generated, for example, if the current takeover scenario type is a high-risk collision takeover scenario type, the current takeover event is a collided takeover event, if the current takeover scenario is a medium-risk collision pedestrian takeover scenario type, the current takeover event is a collision takeover event, if the current takeover scenario type is a low-risk cut-off takeover scenario type, the current takeover event is a cut-off takeover event, and may also be another type of event, which is not limited in this embodiment.

And step S50, determining a corresponding target execution instruction according to the current takeover event.

It should be noted that, a corresponding target execution instruction is determined according to the takeover event, the takeover operation is performed on the target movable carrier according to the target execution instruction, and the target movable carrier is controlled to avoid the occurrence of the current takeover event, for example, when the target execution instruction is a collision takeover event, the target execution instruction may be a brake, a deceleration at a long distance, or another operation instruction that can avoid the occurrence of the current event, and when the target execution instruction is a collision takeover event, the target execution instruction may be a lane change or another operation instruction that can avoid the occurrence of the current event, so as to avoid the occurrence of the current collision event. For example, when the current takeover event is a scratch takeover event, the corresponding target execution instruction may be to change a travelling lane of the target movable carrier, as shown in fig. 3, and when the target movable carrier may scratch the movable object 3, the target movable carrier is travelling to a left lane.

The embodiment obtains the current road type of the target movable carrier and obtains the current driving information of the target movable carrier; determining a trend of change of movable objects around the target movable carrier; determining the type of the current takeover scene according to the change trend, the current driving information and the current road type; generating a corresponding current takeover event based on the current takeover scene type; and determining a corresponding target execution instruction according to the current takeover event, and taking over the target movable carrier according to the target execution instruction. Through the mode, the corresponding takeover scene type is determined according to the current road type, the running information of the target movable carrier and the change trend of the peripheral movable objects, the target movable carrier is controlled by adopting a proper takeover execution instruction according to the takeover scene type, the movable carrier can solve the current problem through the movable carrier and the server when meeting the takeover problem in the unmanned state, the intelligence of automatic driving is improved, meanwhile, the safe running of the movable carrier is guaranteed, and the user experience is improved.

Referring to fig. 2, fig. 2 is a flowchart illustrating a method for controlling a movable carrier according to a second embodiment of the present invention.

Based on the foregoing embodiment, the method for controlling a movable carrier in this embodiment specifically includes, in step S30:

and step S31, extracting the texture features of the current road type, and determining the complexity of the current scene through a road detection model based on the change trend and the texture features.

It should be noted that, extracting Haar texture features of a current road, constructing real texture parts and imaginary texture basic rectangular templates of the current road in four directions, obtaining a filtering response of the current road in any direction by using an orthogonal correction obtaining method, thereby realizing extraction of the texture features of the current road type, after obtaining the texture features of the current road type, realizing estimation of road vanishing points based on a road detection model, namely, a voting mechanism, realizing detection of a first main boundary based on vanishing point constraint by combining road direction consistency features and color differences of adjacent regions of the current road, and meanwhile updating initialized vanishing points based on the first main boundary, thereby realizing segmentation of drivable regions of the current road. And after the drivable area of the current road type is obtained, determining the complexity of the current scene according to the variation trend of the peripheral movable objects of the target movable carrier.

For example, according to the detection, it is found that the feasible region of the current road type is only two-way two-lane, the road width is only 6 meters, the two sides of the road are greenbelts and sidewalks, while the number of the peripheral movable objects of the target movable carrier in the current scene is gradually increased, the average driving speed is 40km/h, and the driving speed of the peripheral movable objects is still increased, and the complexity of the current scene is moderate.

Step S32, a first driving direction, a first driving position and a first driving speed in the current driving information are acquired.

It is understood that the first traveling direction, the first traveling position, and the first traveling speed refer to a traveling direction, a traveling position, and a traveling speed of the target movable carrier included in the current traveling information of the target movable carrier

Step S33, a second traveling direction, a second traveling position, and a second traveling speed of the movable object around the target movable carrier are acquired based on the three-dimensional image information.

The second traveling direction, the second traveling position, and the second traveling speed refer to a traveling direction, a traveling position, and a traveling speed of a movable object around the target movable carrier.

It can be understood that, only the traveling speed and the current position of the surrounding movable object can be sensed according to the three-dimensional image information, and the traveling direction and the behavior of the surrounding movable object at the next time cannot be obtained, so that the traveling direction of the surrounding movable object at the next time needs to be analyzed and predicted, in this embodiment, the obtaining of the second traveling direction, the second traveling position, and the second traveling speed of the movable object surrounding the target movable carrier according to the three-dimensional image information may specifically include:

acquiring a second driving speed and a second driving position of the movable object according to the three-dimensional image; extracting warning features of the movable object in the three-dimensional image; and obtaining a second predicted driving direction of the movable object through a preset behavior prediction model according to the warning characteristics, the driving speed and the driving position.

It should be understood that after the driving speed and the current position of the peripheral movable object are obtained, the predicted driving direction of the movable object is obtained through a preset behavior prediction model according to the direction of the movable object and warning information given by the movable carrier, such as a steering lamp, a brake lamp, double flashing lights or other warning information of a vehicle, wherein the preset behavior prediction model is obtained by training an initial neural network model through a large amount of sample data (namely warning information given by a plurality of movable objects at different speeds and different direction positions) in a road.

In a specific implementation, by predicting the traveling direction of the peripheral movable object according to the embodiment, the relative traveling direction of the target movable carrier and the peripheral movable object is obtained, and a more accurate collision prediction result can be obtained.

Step S34, determining a relative position, a relative travel direction, and a relative travel speed of the target movable carrier and the movable object based on the first travel direction, the first travel position, the first travel speed, the second travel direction, the second travel position, and the second travel speed, and taking the relative position, the relative travel direction, and the relative travel speed as relative travel data.

In the specific implementation, the current three-dimensional image information is acquired, the driving speed, the driving direction and the current position of the movable object around the target movable carrier are acquired through the three-dimensional image information and the sensor of the target movable carrier, and the relative driving data, namely the relative position, the relative driving direction and the relative driving speed, between the driving speed, the driving direction and the current position of the movable object around the target movable carrier is determined according to the driving information of the target carrier and the information of the movable object around the target movable carrier.

For example, if the first traveling speed of the target movable carrier is 30km/h, the first traveling direction is forward traveling, the first traveling position is at the position shown in fig. 3, the second traveling speed of the movable object 1 is 25km/h, the second traveling direction is forward traveling, and the second traveling position is at the position shown in fig. 3, the relative position of the target movable carrier and the movable object is 5 meters ahead of the target movable carrier, the relative traveling direction is forward traveling, and the relative traveling speed is-5 km/h, and the relative traveling data is { 5 meters ahead of the relative position, forward of the relative traveling direction, and-5 km/h } relative traveling speed.

And step S35, performing collision prediction simulation according to the complexity and the relative driving data to obtain a collision risk estimation result.

The collision prediction model is based on the driving information, road type and environmental factors of a collided movable object, a Bayesian network model is used for constructing the collision prediction model, simulation data are obtained through Uc-road traffic scene modeling, the prior probability and the conditional probability of the parameters are calculated by combining traffic historical data and road measured data of the current road, and the reasonable effectiveness of the model is verified by utilizing GeNIe Bayesian simulation software, so that the collision prediction model is obtained. And simulating based on the relative driving data through a collision prediction model to obtain a collision risk estimation result.

And step S36, determining the corresponding current takeover scene type according to the collision risk estimation result.

As shown in fig. 3, a plurality of movable objects are present around the target movable carrier, the collision type of the target movable carrier and the surrounding movable objects is obtained through a collision prediction model, for example, the target movable carrier is in a state of simultaneously driving forward with the movable objects 1 and 2 during driving, but the driving speed of the target movable carrier is higher than that of the movable object 1, the driving speed of the target movable carrier is the same as that of the movable object 2, the possibility of rear-end collision between the target movable carrier and the movable object 1 is found through collision prediction, that is, the collision risk prediction result is a medium risk, and the takeover scene type is a medium risk rear-end collision scene type, that is, the possibility of simultaneous collision between the target movable carrier and the movable objects 1 and 2 is found through collision prediction, that is, that the collision risk prediction result is a high risk, and meanwhile, acquiring that the takeover scene type is a high-risk collision takeover scene type, or the target movable carrier runs forwards, the running direction of the movable object 3 is ready for lane changing, the running speed of the target movable carrier is the same as that of the movable object 3, the possibility that the target movable carrier and the movable object 3 are scratched is found through collision prediction, namely, the collision risk estimation result is low risk, and meanwhile, the takeover scene type is a low-risk scratch takeover scene type.

It should be noted that, after the takeover scene type is obtained, the corresponding current takeover scene type is further determined according to the degree of the collision risk, when the current takeover scene type is a low-risk collision type, the current takeover scene type is further planned to be a full-automatic driving takeover type, that is, the takeover operation is completed through the target movable carrier, and when the current takeover scene is a medium-risk and high-risk collision type, the current takeover scene type is further planned to be a semi-automatic driving takeover type, that is, the target movable carrier needs to run according to an operation instruction issued by a road terminal or a cloud (server) connected with the target movable carrier, so that the takeover operation is completed.

In the embodiment, by the mode, the complexity of the scene is determined according to the current road type and the variation trend of the peripheral movable objects, then the collision simulation is performed according to the relative driving data and the complexity of the target movable carrier and the peripheral movable objects, and the collision risk type is estimated, so that the type of the scene taking over is obtained, the taking over operation is more targeted, the probability of accidents is reduced, and the driving safety of the target movable carrier is improved.

Referring to fig. 4, fig. 4 is a flowchart illustrating a movable carrier control method according to a third embodiment of the present invention.

Based on the foregoing embodiment, the method for controlling a movable carrier in this embodiment specifically includes, in step S50:

and step S51, sending the takeover event to a server so that the server feeds back a corresponding event solution instruction according to the takeover event.

It should be noted that the takeover event is sent to the server, so that the server performs driving simulation under different operation instructions according to the type of the current takeover scene, so that a driving simulation result that the takeover event can be solved is obtained through a preset generation model, for example, when the takeover event is collided, the simulation is performed to decelerate or brake at a distance of 50 meters from the collided object, and the effect that the brake or deceleration can achieve at a distance of 100 meters from the collided object is selected, the simulation result that the takeover event can be avoided is selected, a corresponding event solving instruction is generated, and the solving instruction is fed back to the target movable carrier.

In a specific implementation, before sending the takeover event to the server, the takeover event is sent to a road terminal closest to a target movable carrier, so that the road terminal searches for and feeds back an event solution instruction corresponding to the current takeover event, and when the event solution instruction fed back by the road terminal is not received, the current takeover event is sent to the server, so that the server feeds back the corresponding event solution instruction according to the takeover event.

In step S52, when the event resolution command is received, a command to be operated at the next time of the automated driving is predicted.

It should be noted that, when the target mobile carrier receives the event resolution instruction sent by the server, the controller of the target mobile carrier may also predict the instruction to be operated at the next time based on the autopilot algorithm according to the current takeover event, and the instruction to be operated may also resolve the current takeover event.

Step S53, comparing the instruction to be operated with the event resolution instruction to obtain a target execution instruction.

It can be understood that, based on the current driving information of the target movable carrier, the to-be-operated instruction and the event resolution instruction are preferentially selected, and an operation instruction which meets the current driving state in the to-be-operated instruction and the event resolution instruction is obtained and used as a target instruction.

In the specific implementation, after a target execution instruction corresponding to a current takeover event is acquired, the acquired target execution instruction is sent to a server, so that the server updates a preset generation model and synchronizes the target execution instruction to a road surface terminal.

It should be noted that, in order to obtain a better execution instruction to enable the target movable carrier to safely travel, the acquired instruction needs to be compared and analyzed, and finally a better execution instruction is selected, in this embodiment, the comparing the instruction to be operated with the event resolution instruction to obtain the target execution instruction may specifically include:

acquiring a preset road safety range corresponding to the current road type; obtaining a first possibility value that the predicted running state of the target movable carrier exceeds the preset road safety range through a preset behavior prediction model according to the current running information of the target movable carrier and the instruction to be operated; obtaining a second possibility value that the predicted driving state of the target movable carrier exceeds the preset road safety range through a preset behavior prediction model according to the current driving information of the target movable carrier and the event solving instruction; setting a probability value with a smaller value of the first probability value and the second probability value as a target probability value; and determining an instruction corresponding to the target possibility value, and taking the determined instruction as a target execution instruction.

It should be noted that, the obtaining of the preset road safety range of the current road type refers to obtaining the traffic rules of the current road, such as whether the current road can turn around, the positions of a straight lane and a turning lane, whether a whistle can be sounded, the speed limit, and the like.

It is understood that the traveling state of the target movable carrier at the next time under the command to be operated and the event resolution command is predicted based on the current traveling information of the target movable carrier, and the possibility that the predicted traveling state of the target movable carrier exceeds the preset road safety range refers to the possibility that the target movable carrier violates the current road traffic regulation in the predicted traveling state.

In a specific implementation, the instruction corresponding to the probability value smaller than the preset road safety range is selected as the target execution instruction. For example, if the target movable carrier incorporates an instruction to be operated in the state of the current travel information, an eighty percent possibility may cause an illegal lane change, and if only a thirty percent possibility may cause speeding on the current road in combination with an event resolution instruction, an event resolution instruction with a thirty percent possibility of speeding is selected as the target resolution instruction.

It should be noted that after the target execution instruction is obtained, the target execution instruction is sent to the server, so that the server updates the preset generation model, and synchronizes the target execution instruction to the road surface terminal.

For example, when the target movable carrier is an unmanned vehicle, when the unmanned vehicle encounters a takeover scene, initiating an application task needing takeover, and seeking help of the unmanned vehicle close to a road terminal, so that the road terminal searches whether a corresponding event solution instruction capable of solving the current takeover scene exists, if so, directly sending the event solution instruction to the unmanned vehicle terminal, if not, sending the takeover scene to a remote driving cloud platform, namely a server terminal, synchronizing scene information of the unmanned vehicle terminal by the server terminal, inquiring whether the corresponding event solution instruction exists in historical scene data based on the current scene information, if so, feeding back the corresponding event solution instruction to the unmanned vehicle terminal by the server, and if not, simulating based on the current driving state to obtain an event solution instruction specified by the server terminal, feeding back a corresponding event solving instruction to the unmanned vehicle end, when the unmanned vehicle end receives the event solving instruction, predicting the behavior of the unmanned vehicle end at the next moment based on an automatic driving algorithm to obtain an instruction to be operated, carrying out preferential selection in the instruction to be operated and the event solving instruction to obtain a final target execution instruction, finishing the takeover operation by the unmanned vehicle end, feeding back the obtained target execution instruction to the server end, and synchronously updating the target execution instruction to the road surface terminal

In specific implementation, the embodiment selects the instruction which is less prone to violating road safety by predicting the behavior of the instruction to be operated and the event solving instruction, so that the taking-over operation is safer and more accurate, and the safe driving of the target movable carrier is ensured.

In this embodiment, the takeover event is sent to a server, so that the server feeds back a corresponding event resolution instruction according to the takeover event; predicting a command to be operated at the next moment of automatic driving when the event solving command is received; comparing the instruction to be operated with the event solving instruction to obtain a target execution instruction; and taking over the target movable carrier according to the target execution instruction. Through the preferred selection of the two takeover instructions, a final target solution instruction is obtained, the takeover operation of the target movable carrier is ensured to be more fit with the actual situation, and meanwhile, the safety of the target movable carrier in the running process is ensured.

Further, referring to fig. 5, an embodiment of the present invention further provides a movable carrier control apparatus, including:

the information acquisition module 10 is configured to acquire a current road type where a target movable carrier is located, and acquire current driving information of the target movable carrier;

a trend determination module 20 for determining a trend of change of movable objects around the target movable carrier;

the type determining module 30 is further configured to determine a current takeover scene type according to the change trend, the current driving information, and the current road type;

an event generating module 40, configured to generate a corresponding current takeover event based on the current takeover scenario type;

the instruction determining module 50 is further configured to determine a corresponding target execution instruction according to the current takeover event.

The embodiment obtains the current road type of the target movable carrier and obtains the current driving information of the target movable carrier; determining a trend of change of movable objects around the target movable carrier; determining the type of the current takeover scene according to the change trend, the current driving information and the current road type; generating a corresponding current takeover event based on the current takeover scene type; determining a corresponding target execution instruction according to the current takeover event; and taking over the target movable carrier according to the target execution instruction. Through the mode, the corresponding takeover scene type is determined according to the current road type, the running information of the target movable carrier and the change trend of the peripheral movable objects, the target movable carrier is controlled by adopting a proper takeover execution instruction according to the takeover scene type, the movable carrier can solve the current problem through the movable carrier and the server when meeting the takeover problem in the unmanned state, the intelligence of automatic driving is improved, meanwhile, the safe running of the movable carrier is guaranteed, and the user experience is improved.

It should be noted that each module in the apparatus may be configured to implement each step in the method, and achieve the corresponding technical effect, which is not described herein again.

Referring to fig. 6, fig. 6 is a schematic diagram of a movable carrier structure of a hardware operating environment according to an embodiment of the present invention.

As shown in fig. 6, the movable carrier may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001 described previously.

Those skilled in the art will appreciate that the configuration shown in fig. 6 does not constitute a limitation of the movable carrier, and may include more or fewer components than shown, or some components may be combined, or a different arrangement of components.

As shown in fig. 6, a memory 1005, which is a storage medium, may include therein an operating system, a network communication module, a user interface module, and a removable carrier control program.

In the removable carrier shown in fig. 6, the network interface 1004 is primarily used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the removable carrier of the present invention may be provided in a removable carrier, which calls the removable carrier control program stored in the memory 1005 through the processor 1001 and performs the following operations:

Further, the processor 1001 may call a removable carrier control program stored in the memory 1005, and also perform the following operations:

acquiring a preset road safety range corresponding to the current road type;

Furthermore, an embodiment of the present invention further provides a computer-readable storage medium, where a removable carrier control program is stored on the computer-readable storage medium, and when executed by a processor, the removable carrier control program implements the following operations:

It should be noted that, when being executed by a processor, the computer-readable storage medium may also implement the steps in the method, and achieve the corresponding technical effects, which is not described herein again.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (e.g. Read Only Memory (ROM)/RAM, magnetic disk, optical disk), and includes several instructions for enabling a terminal device (e.g. a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. A movable carrier control method characterized by comprising:

determining a corresponding target execution instruction according to the current takeover event;

wherein, the determining a corresponding target execution instruction according to the current takeover event specifically includes:

when the event solving instruction is received, predicting an instruction to be operated at the next moment of automatic driving;

2. The movable carrier control method according to claim 1, wherein the determining a trend of change of the movable objects around the target movable carrier specifically includes:

3. The movable carrier control method as claimed in claim 2, wherein the variation tendency includes a flow rate variation tendency and a speed variation tendency;

4. The method for controlling a movable carrier according to claim 1, wherein the determining a current takeover scene type according to the variation trend, the current driving information, and the current road type specifically includes:

5. The movable carrier control method according to claim 4, wherein the obtaining of the second traveling direction, the second traveling position, and the second traveling speed of the movable object around the target movable carrier based on the three-dimensional image information specifically comprises:

and obtaining a second driving direction of the movable object through a preset behavior prediction model according to the warning characteristic, the second driving speed and the second driving position.

6. The method according to claim 1, wherein the comparing the command to be operated with the event resolution command to obtain a target execution command specifically comprises:

acquiring a preset road safety range corresponding to the current road type;

obtaining a first possibility value that the predicted driving state of the target movable carrier exceeds the preset road safety range through a preset behavior prediction model according to the current driving information of the target movable carrier and the command to be operated;

7. A movable carrier control apparatus, characterized by comprising:

the type determining module is used for determining the type of the current takeover scene according to the change trend, the current driving information and the current road type;

the instruction determining module is used for determining a corresponding target execution instruction according to the current takeover event;

the instruction determining module is further configured to send the takeover event to a server, so that the server feeds back a corresponding event solution instruction according to the takeover event;

8. A movable carrier, characterized in that the movable carrier comprises: memory, a processor and a removable carrier control program stored on the memory and executable on the processor, the removable carrier control program being configured to implement the steps of the removable carrier control method according to any of claims 1 to 6.

9. A storage medium having a removable carrier control program stored thereon, wherein the removable carrier control method when executed by a processor implements the steps of the removable carrier control method according to any one of claims 1 to 6.