CN113188556B - Intelligent network-connected automobile track planning method and device based on driving safety field - Google Patents
Intelligent network-connected automobile track planning method and device based on driving safety field Download PDFInfo
- Publication number
- CN113188556B CN113188556B CN202110459321.8A CN202110459321A CN113188556B CN 113188556 B CN113188556 B CN 113188556B CN 202110459321 A CN202110459321 A CN 202110459321A CN 113188556 B CN113188556 B CN 113188556B
- Authority
- CN
- China
- Prior art keywords
- planning
- track
- vehicle
- speed
- intelligent network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000005457 optimization Methods 0.000 claims abstract description 19
- 238000007670 refining Methods 0.000 claims abstract description 10
- 238000005381 potential energy Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000011478 gradient descent method Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 238000004088 simulation Methods 0.000 claims description 7
- 230000006855 networking Effects 0.000 claims 1
- 230000006399 behavior Effects 0.000 description 15
- 238000004422 calculation algorithm Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The application discloses an intelligent network-connected automobile track planning method and device based on a driving safety field, wherein the method comprises the following steps: calculating risks caused by traffic environments to intelligent network-connected automobiles based on driving safety fields; generating a sparse searchable area according to the surrounding road traffic environment of the automobile, the own speed of the automobile and a prediction time domain; and generating a lane selection track in the searchable area based on the risk, refining the sparse search area on the basis of the lane selection track to obtain an initial planning path and planning speed, and finally carrying out deduction in a prediction time domain to solve the problem of track tracking optimization to obtain a final planning track of the intelligent network-connected automobile in the prediction time domain. The track planning method can improve the accuracy, the effectiveness and the instantaneity of track planning, effectively ensure the driving safety and is more intelligent.
Description
Technical Field
The application relates to the technical field of driving safety, in particular to an intelligent network-connected automobile track planning method and device based on driving safety field.
Background
Currently, key technologies of intelligent network-connected vehicles include autonomous positioning, environmental awareness, behavior decision-making, trajectory planning, motion control, network communication, and the like, wherein the trajectory planning is an indispensable part that determines what trajectory the intelligent network-connected vehicle will travel. However, facing complex traffic environments, how to make safe, efficient, trajectory planning that accommodates complex traffic still faces serious challenges.
In the related art, the track planning method generally used mainly comprises the following steps: 1) graph search based methods, 2) optimization based methods, 3) machine learning based methods, and 4) artificial potential energy field based methods.
Specifically, the graph-based search method is a very commonly used method, and is applicable to various fields for searching an optimal track from the point A to the point B. The method first requires representing the areas for planning as a state space with an occupancy grid, and objects within each area are represented by the form of the occupancy grid. Then searching out an optimal track by adopting search algorithms such as an A-type algorithm, a Dijkstra algorithm, a fast-expansion random tree and the like; the optimization-based method is that a track planning problem is expressed as an optimization problem, and a solution which enables a cost function to be minimum is solved under a series of constraints by adopting a numerical solution mode aiming at the optimization problem, so that a planned track is obtained; the machine learning-based method is to train the neural network by learning the driving result of the human driver to obtain the training result. In the track planning process, environment information is used as input to be transmitted to a neural network, so that a track planning structure can be output; the method based on the artificial potential energy field is an advanced method. The method adopts an artificial potential energy field to model obstacles, road structures, lane lines, destinations and the like in a traffic environment, and then adopts a gradient descent method to find an optimal path.
However, the related art has a certain disadvantage that the track planning result obtained by the method based on graph search may not be a smooth curve, and the planned track and the actual running track of the vehicle are inconsistent because the dynamics of the vehicle are not considered, so that the movement of the vehicle is not smooth enough. In addition, the track obtained by the optimization-based method can ensure the optimality, but because the actual traffic environment is complex, the modeling and solving of the optimization problem are difficult, and therefore, the problem of overlarge computational complexity is often faced. Also, as for the method based on machine learning, the neural network cannot be positioned when a problem occurs due to the non-interpretability of the neural network, so that the method is still difficult to be adopted in practical application. Finally, existing methods based on artificial potential energy fields are generally difficult to comprehensively consider various risks in traffic environments, including vehicle behaviors, complex road conditions and interaction among traffic participants, and lack of application in complex scenes, and are only applied to simple scenes, such as lane keeping, vehicle following, lane changing and the like. Accordingly, the related art has yet to be improved.
Content of the application
The application provides an intelligent network-connected automobile track planning method and device based on a driving safety field, which are used for solving the problem that the intelligent network-connected automobile track planning is poor in consistency with an actual track.
An embodiment of a first aspect of the present application provides an intelligent network-connected vehicle track planning method based on a driving safety field, including the following steps: calculating risks caused by traffic environments to intelligent network-connected automobiles based on driving safety fields; generating a sparse searchable area according to the surrounding road traffic environment of the automobile, the own speed of the automobile and a prediction time domain; and generating a lane selection track in the searchable area based on the risk, refining the sparse search area on the basis of the lane selection track to obtain an initial planning path and planning speed, and carrying out deduction in the prediction time domain to solve the problem of track tracking optimization, so as to obtain a final planning track of the intelligent network-connected automobile in the prediction time domain.
Optionally, in an embodiment of the present application, the calculating the risk of the traffic environment to the intelligent network-connected automobile based on the driving security field includes: calculating the risk by using a model obtained by modeling traffic environment elements by a driving safety field, wherein the calculation formula of the risk is as follows:
E D ＝E P +E K +E B ，
wherein ,ED 、E P 、E K and EB The field intensities of the driving safety field, the potential energy field, the kinetic energy field and the behavior field are respectively corresponding.
Optionally, in an embodiment of the present application, the refining the sparse search area based on the lane selection track to obtain an initial planned path and a planned speed includes: refining the sparse search region around the reference track; and obtaining an optimal path by using a gradient descent method, and simultaneously planning the speed by using a self-adaptive following method to plan a speed target so as to obtain the initial planned path and the planning speed.
Optionally, in an embodiment of the present application, the generating model of the planning speed is:
wherein ,amax Maximum speed of the own vehicle, v d For the target speed, d is the distance between the own vehicle and the front vehicle, d * Is the target minimum vehicle distance.
Optionally, in an embodiment of the present application, the performing deduction in the predicted time domain, solving a problem of track tracking optimization, to obtain a final planned track of the intelligent internet-connected vehicle in the predicted time domain, includes: and carrying out dynamics simulation on the prediction time domain at each sampling time by using a closed-loop dynamics method, and carrying out deduction by using a controller which is the same as an actual vehicle control module and a vehicle model to obtain a final planning track of the intelligent network-connected vehicle in the prediction time domain.
An embodiment of a second aspect of the present application provides an intelligent network-connected vehicle track planning device based on a driving security field, including: the calculation module is used for calculating risks caused by traffic environment to the intelligent network-connected automobile on the basis of the driving safety field; the generation module is used for generating a sparse searchable area according to the surrounding road traffic environment of the automobile, the own speed of the automobile and a prediction time domain; the planning module is used for generating a lane selection track in the searchable area based on the risk, refining the sparse search area on the basis of the lane selection track to obtain an initial planning path and planning speed, deducting in the prediction time domain, solving the problem of track tracking optimization, and obtaining a final planning track of the intelligent network-connected automobile in the prediction time domain.
Optionally, in an embodiment of the present application, the calculating module is specifically configured to calculate the risk by using a model obtained by modeling a traffic environment element by a driving safety field, where a calculation formula of the risk is:
E D ＝E P +E K +E B ，
wherein ,ED 、E P 、E K and EB The field intensities of the driving safety field, the potential energy field, the kinetic energy field and the behavior field are respectively corresponding.
Optionally, in one embodiment of the present application, the planning module includes: the thinning unit is used for thinning the sparse search area around the reference track; and the planning unit is used for planning the speed by using the self-adaptive vehicle following method while obtaining the optimal path by using the gradient descent method, and planning a speed target so as to obtain the initial planning path and the planning speed.
Optionally, in an embodiment of the present application, the generating model of the planning speed is:
wherein ,amax Maximum speed of the own vehicle, v d For the target speed, d is the distance between the own vehicle and the front vehicle, d * Is the target minimum vehicle distance.
Optionally, in an embodiment of the present application, the planning module is specifically configured to perform, at each sampling time, a dynamics simulation in the prediction horizon by using a closed-loop dynamics method, and perform deduction with a vehicle model by using the same controller as an actual vehicle control module, so as to obtain a final planned track of the intelligent network-connected vehicle in the prediction horizon.
The method has the advantages that the influence of different traffic participants, road structures, behavior predictions and other factors in a traffic scene on the intelligent network-connected automobile is subjected to unified quantitative modeling, so that a foundation is provided for the safety and the real-time performance of a track planning algorithm, the track planning of the intelligent network-connected automobile is performed on the basis, an optimal lane selection track is obtained in a macroscopic planning layer by adopting a layered framework, the motion track is obtained by refining in a microscopic planning layer, the accuracy, the effectiveness and the real-time performance of track planning are improved, the driving safety is effectively ensured, and the intelligent system is realized.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
fig. 1 is a flowchart of an intelligent network-connected vehicle track planning method based on a driving security field according to an embodiment of the present application;
FIG. 2 is a schematic top view of a typical scene ride safety field according to one embodiment of the present application;
FIG. 3 is a schematic diagram of an automotive layered trajectory planning method according to one embodiment of the present application;
FIG. 4 is a search area generation diagram according to one embodiment of the present application;
FIG. 5 is a schematic diagram of lane selection path generation according to one embodiment of the present application;
FIG. 6 is a schematic diagram of a generated trace according to one embodiment of the present application;
fig. 7 is an exemplary diagram of an intelligent network-connected vehicle trajectory planning device based on a driving safety farm according to an embodiment of the present application.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present application and are not to be construed as limiting the present application.
The following describes an intelligent network-connected automobile track planning method and device based on a driving safety field according to the embodiment of the application with reference to the accompanying drawings. The application provides an intelligent network-connected automobile track planning method based on a driving safety field, in the method, the influence of different traffic participants, road structures, behavior predictions and other factors in a traffic scene on the intelligent network-connected automobile is subjected to unified quantitative modeling, so that a foundation is provided for the safety and the real-time performance of a track planning algorithm, the track planning of the intelligent network-connected automobile is carried out on the foundation, a layered framework is adopted, the optimal lane selection track is obtained in a macroscopic planning layer, and then the motion track is obtained by refining in a microscopic planning layer, so that the accuracy, the effectiveness and the real-time performance of track planning are improved, the driving safety is effectively ensured, and the intelligent vehicle is realized. Therefore, the problem that the planned track and the actual track of the intelligent network-connected automobile are poor in consistency is solved.
Specifically, fig. 1 is a schematic flow chart of an intelligent network-connected vehicle track planning method based on a driving security field according to an embodiment of the present application.
As shown in fig. 1, the intelligent network-connected automobile track planning method based on the driving safety field comprises the following steps:
in step S101, a risk caused to an intelligent network-connected automobile by a traffic environment is calculated based on a driving security field.
Optionally, in one embodiment of the present application, calculating a risk to an intelligent network-connected automobile from a traffic environment based on a driving security farm includes: calculating risks by using a model obtained by modeling traffic environment elements by a driving safety field, wherein the calculation formula of the risks is as follows:
E D ＝E P +E K +E B ，
wherein ,ED 、E P 、E K and EB The field intensities of the driving safety field, the potential energy field, the kinetic energy field and the behavior field are respectively corresponding.
It can be understood that firstly, modeling is performed on the traffic environment elements by adopting a driving safety field method, so as to represent the risk of each element of the human-vehicle-road in the driving environment of the vehicle to the intelligent automobile:
E D ＝E P +E K +E B ， (1)
in the formula,ED 、E P 、E K and EB The field intensities of the driving safety field, the potential energy field, the kinetic energy field and the behavior field are respectively corresponding.
As shown in fig. 2, the potential energy field is used to represent the influence degree of the stationary object on the road on the running risk of the intelligent automobile, and can be specifically classified into two types. The first type is an object capable of actually colliding, such as a stopped vehicle, an obstacle and the like, and the object of the first type can generate risk field intensity relative to a vehicle for each position in a traffic scene; the second category of potential energy fields is objects that cannot practically collide but play an important role in driving safety, such as lane lines, traffic signs, and the like. The kinetic energy field is used for representing the influence degree of the moving object on the road on the risk of the self-vehicle, and the moving object on the road can generate the risk field intensity relative to the self-vehicle on each position in the traffic scene. The behavior field is used for representing the influence degree of the behavior characteristics of the intelligent automobile on the running risk, and compared with an intelligent automobile with a mature stability algorithm, the intelligent automobile has much smaller risk generated by the intelligent automobile in the testing stage.
In the embodiment of the application, the influence of different traffic participants, road structures, behavior predictions and other factors in a traffic scene on the intelligent automobile is subjected to unified quantitative modeling, so that the problem of high complexity of a decision-making planning algorithm caused by a plurality of traffic environment elements is solved, and a foundation is provided for the effectiveness and instantaneity of the decision-making planning algorithm.
In step S102, a sparse searchable area is generated from the surrounding road traffic environment of the car, the own vehicle speed of the car, and the prediction horizon.
As shown in fig. 3, in the embodiment of the present application, the track gauge of the intelligent network-connected automobile is divided into two layers, which are respectively macro-planning and micro-planning, and in the macro-planning layer, the track gauge simulates a human driver, and first, the lane on which the intelligent network-connected automobile is required to travel is obtained, and then in the micro-planning layer, the track on which the intelligent automobile is required to travel is refined.
It can be understood that in this step, in the macro planning layer, firstly, the searchable area is generated on the basis of the road traffic environment around the intelligent network-connected vehicle, and the sparse search range is generated on the basis of the vehicle speed and the prediction time domain. As shown in fig. 4, the nodes in the search area may be equally spaced on the horizontal axis and equally spaced on the vertical axis, one for each lane.
Further, an optimal road selection track is calculated in a search area by adopting a dynamic programming method and is used as a reference value of a final track programming to be output to a micro-programming, and an optimization target is to seek that the field intensity of a dynamic driving safety field on a path is minimum and meanwhile frequent lane changing of a vehicle is avoided, wherein the following formula is as follows:
in the formula,ωD Is a time-varying weight coefficient that increases with increasing k, ω C Is a constant weight coefficient, P C To penalize the lane change operation to avoid excessive unnecessary lane changes. Lane selection path generation as shown in fig. 5, the path describes where the intelligent car should be located in which lane.
In step S103, a lane selection track is generated in the searchable area based on the risk, and the sparse search area is refined on the basis of the lane selection track to obtain an initial planning path and a planning speed, and the initial planning path and the planning speed are deduced in a prediction time domain, so that the problem of track tracking optimization is solved, and a final planning track of the intelligent network-connected vehicle in the prediction time domain is obtained.
Optionally, in an embodiment of the present application, the thinning of the sparse search area based on the lane selection track to obtain the initial planned path and the planned speed includes: thinning the sparse search area around the reference track; and obtaining an optimal path by using a gradient descent method, and simultaneously planning the speed by using a self-adaptive following method to plan a speed target so as to obtain an initial planned path and a planned speed.
That is, in the micro planning layer, the sparse search range is first refined around the reference track to obtain a better track, and then an optimal path is obtained according to the gradient descent method. Meanwhile, speed planning is carried out according to the self-adaptive car following method, and a speed target is planned on the premise of ensuring safety.
In one embodiment of the present application, the generation model of the planning speed is:
wherein ,amax Maximum speed of the own vehicle, v d For the target speed, d is the distance between the own vehicle and the front vehicle, d * Is the target minimum vehicle distance.
It will be appreciated that embodiments of the present application may employ a self-cruising model, an intelligent driver model, to generate a speed trajectory. The intelligent driver model is shown as follows:
in the formula,amax Maximum speed of the own vehicle, v d For the target speed, d is the distance between the own vehicle and the front vehicle, d * As a target minimum vehicle distance, it is defined as:
in the formula,d0 To block the distance between vehicles, T g For the safety event interval, deltav is the difference between the speeds of the own vehicle and the preceding vehicle, a min Is the minimum acceleration of the bicycle.
Optionally, in an embodiment of the present application, deduction is performed in a predicted time domain, so as to solve a problem of track tracking optimization, and obtain a final planned track of the intelligent network-connected automobile in the predicted time domain, including: and carrying out dynamics simulation on the prediction time domain at each sampling time by using a closed-loop dynamics method, and carrying out deduction by using a controller which is the same as an actual vehicle control module and a vehicle model to obtain a final planning track of the intelligent network-connected vehicle in the prediction time domain.
In the actual execution process, the planned track obtained finally is not smooth enough, and the planned track is inconsistent with the actual path of the vehicle due to incomplete tracking in the actual running process of the vehicle, so that a closed-loop dynamics method can be adopted, dynamics simulation is carried out in a prediction time domain at each sampling time, the same controller as an actual vehicle control module is used for carrying out deduction with a vehicle model, a motion planning track in the prediction time domain is obtained, and an optimization problem to be solved in the prediction time domain can be described as:
solving the optimization problem can obtain the motion trail planned by the intelligent automobile in the prediction time domain, as shown in fig. 6.
According to the embodiment of the application, the influence of different traffic participants, road structures, behavior predictions and other factors in a traffic scene on the intelligent network-connected automobile is subjected to unified quantitative modeling, the problem of high complexity of a decision-making planning algorithm caused by a plurality of traffic environment elements is solved, and a foundation is provided for the effectiveness and instantaneity of the decision-making planning algorithm.
In addition, the problem of poor consistency between the intelligent automobile planning path and the actual track is solved, the intelligent network-connected automobile track planning method has high safety, instantaneity and acceptability, can adapt to the intelligent network-connected automobile track planning of complex traffic flow, and has good calculation efficiency.
In conclusion, the implementation of the routine vehicle safety field can solve the problem of high complexity of a decision-making planning algorithm caused by a plurality of traffic environment elements, provides a basis for the effectiveness and instantaneity of the decision-making planning algorithm, obtains an optimal lane selection track at a macroscopic planning layer, and refines the optimal lane selection track at a microscopic planning layer to obtain a motion track.
According to the intelligent network-connected automobile track planning method based on the driving safety field, which is provided by the embodiment of the application, the influence of different traffic participants, road structures, behavior predictions and the like in a traffic scene on the intelligent network-connected automobile is subjected to unified quantitative modeling, so that a foundation is provided for the safety and the instantaneity of a track planning algorithm, the track planning of the intelligent network-connected automobile is performed on the foundation, an optimal lane selection track is obtained in a macroscopic planning layer by adopting a layered framework, and then the motion track is refined in a microscopic planning layer, so that the accuracy, the effectiveness and the instantaneity of track planning are improved, the driving safety is effectively ensured, and the intelligent vehicle is realized.
Next, an intelligent internet-connected vehicle track planning device based on a driving safety field according to an embodiment of the application is described with reference to the accompanying drawings.
Fig. 7 is a block schematic diagram of an intelligent network-connected vehicle track planning device based on a driving safety field according to an embodiment of the application.
As shown in fig. 7, the intelligent network-connected vehicle track planning apparatus 10 based on a driving safety field includes: a calculation module 100, a generation module 200 and a planning module 300.
Specifically, the computing module 100 is configured to calculate, based on a driving security field, a risk caused to an intelligent network-connected vehicle by itself.
The generating module 200 is configured to generate a sparse searchable area according to a surrounding road traffic environment of the automobile, an own vehicle speed of the automobile and a prediction horizon.
The planning module 300 is configured to generate a lane selection track in a searchable area based on risk, refine the sparse search area based on the lane selection track to obtain an initial planning path and a planning speed, and perform deduction in a prediction time domain, so as to solve the problem of track tracking optimization, and obtain a final planning track of the intelligent network-connected vehicle in the prediction time domain.
Optionally, in an embodiment of the present application, the calculation module is specifically configured to calculate the risk by using a model obtained by modeling the traffic environment element by the driving safety field, where a calculation formula of the risk is:
E D ＝E P +E K +E B ，
wherein ,ED 、E P 、E K and EB The field intensities of the driving safety field, the potential energy field, the kinetic energy field and the behavior field are respectively corresponding.
Optionally, in one embodiment of the present application, the planning module includes: the thinning unit is used for thinning the sparse search area around the reference track; and the planning unit is used for planning the speed by using the self-adaptive vehicle following method while obtaining the optimal path by using the gradient descent method, and planning a speed target so as to obtain an initial planned path and a planned speed.
Optionally, in an embodiment of the present application, the generation model of the planning speed is:
wherein ,amax Maximum speed of the own vehicle, v d For the target speed, d is the distance between the own vehicle and the front vehicle, d * Is the target minimum vehicle distance.
Optionally, in an embodiment of the present application, the planning module is specifically configured to utilize a closed-loop dynamics method, perform dynamics simulation on a prediction time domain at each sampling time, and use the same controller as an actual vehicle control module to perform deduction with a vehicle model, so as to obtain a final planned track of the intelligent network-connected vehicle in the prediction time domain.
It should be noted that, the foregoing explanation of the embodiment of the intelligent network-connected vehicle track planning method based on the driving safety field is also applicable to the intelligent network-connected vehicle track planning device based on the driving safety field of the embodiment, which is not repeated herein.
According to the intelligent network-connected automobile track planning device based on the driving safety field, which is provided by the embodiment of the application, the influence of different traffic participants, road structures, behavior prediction and other factors in a traffic scene on the intelligent network-connected automobile is subjected to unified quantitative modeling, so that a foundation is provided for the safety and the instantaneity of a track planning algorithm, the track planning of the intelligent network-connected automobile is carried out on the foundation, an optimal lane selection track is obtained in a macroscopic planning layer by adopting a layered framework, and then the motion track is refined in a microscopic planning layer, so that the accuracy, the effectiveness and the instantaneity of track planning are improved, the driving safety is effectively ensured, and the intelligent vehicle is realized.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless explicitly defined otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more N executable instructions for implementing specific logical functions or steps of the process, and further implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
It is to be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the N steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As with the other embodiments, if implemented in hardware, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.
Claims (4)
1. An intelligent network-connected automobile track planning method based on a driving safety field is characterized by comprising the following steps of:
calculating risks caused by traffic environments to intelligent network-connected automobiles based on driving safety fields;
generating a sparse searchable area according to the surrounding road traffic environment of the automobile, the own speed of the automobile and a prediction time domain; and
generating a lane selection track in the searchable area based on the risk, refining the sparse search area on the basis of the lane selection track to obtain an initial planning path and planning speed, and carrying out deduction in the prediction time domain to solve the problem of track tracking optimization, so as to obtain a final planning track of the intelligent network-connected automobile in the prediction time domain;
the risk of the intelligent network-connected automobile caused by the traffic environment is calculated based on the driving safety field, and the intelligent network-connected automobile comprises the following components:
calculating the risk by using a model obtained by modeling traffic environment elements by a driving safety field, wherein the calculation formula of the risk is as follows:
wherein ,
、/>
、/>
and />
The field intensities respectively correspond to a driving safety field, a potential energy field, a kinetic energy field and a behavior field;
the thinning of the sparse search area based on the lane selection track to obtain an initial planning path and a planning speed comprises the following steps:
refining the sparse search region around the reference track;
obtaining an optimal path by using a gradient descent method, and simultaneously planning a speed by using a self-adaptive following method to plan a speed target so as to obtain the initial planned path and the planning speed;
the generation model of the planning speed is as follows:
wherein ,
is the maximum speed of the own vehicle, +.>
For the target vehicle speed +.>
Is the distance between the own vehicle and the front vehicle, < >>
Is the target minimum vehicle distance.
2. The method of claim 1, wherein the performing deduction in the predicted time domain to solve a track tracking optimization problem, to obtain a final planned track of the intelligent network-connected vehicle in the predicted time domain, includes:
and carrying out dynamics simulation on the prediction time domain at each sampling time by using a closed-loop dynamics method, and carrying out deduction by using a controller which is the same as an actual vehicle control module and a vehicle model to obtain a final planning track of the intelligent network-connected vehicle in the prediction time domain.
3. Intelligent networking car track planning device based on driving safety field, characterized by comprising:
the calculation module is used for calculating risks caused by traffic environment to the intelligent network-connected automobile on the basis of the driving safety field;
the generation module is used for generating a sparse searchable area according to the surrounding road traffic environment of the automobile, the own speed of the automobile and a prediction time domain; and
the planning module is used for generating a lane selection track in the searchable area based on the risk, refining the sparse search area on the basis of the lane selection track to obtain an initial planning path and planning speed, carrying out deduction in the prediction time domain, solving the problem of track tracking optimization, and obtaining a final planning track of the intelligent network-connected vehicle in the prediction time domain;
the calculation module is specifically configured to calculate the risk by using a model obtained by modeling a traffic environment element by a driving safety field, where a calculation formula of the risk is:
wherein ,
、/>
、/>
and />
The field intensities respectively correspond to a driving safety field, a potential energy field, a kinetic energy field and a behavior field;
the planning module comprises:
the thinning unit is used for thinning the sparse search area around the reference track;
the planning unit is used for planning the speed by using the self-adaptive vehicle following method while obtaining the optimal path by using the gradient descent method, and planning a speed target so as to obtain the initial planning path and the planning speed;
the generation model of the planning speed is as follows:
wherein ,
is the maximum speed of the own vehicle, +.>
For the target vehicle speed +.>
Is the distance between the own vehicle and the front vehicle, < >>
Is the target minimum vehicle distance.
4. The apparatus of claim 3, wherein the planning module is specifically configured to perform a dynamics simulation in the prediction horizon at each sampling time by using a closed-loop dynamics method, and perform deduction with a vehicle model using a controller identical to an actual vehicle control module, to obtain a final planned trajectory of the intelligent network-connected vehicle in the prediction horizon.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110459321.8A CN113188556B (en) | 2021-04-27 | 2021-04-27 | Intelligent network-connected automobile track planning method and device based on driving safety field |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110459321.8A CN113188556B (en) | 2021-04-27 | 2021-04-27 | Intelligent network-connected automobile track planning method and device based on driving safety field |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113188556A CN113188556A (en) | 2021-07-30 |
| CN113188556B true CN113188556B (en) | 2023-06-13 |
Family
ID=76979308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110459321.8A Active CN113188556B (en) | 2021-04-27 | 2021-04-27 | Intelligent network-connected automobile track planning method and device based on driving safety field |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113188556B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106114507B (en) * | 2016-06-21 | 2018-04-03 | 百度在线网络技术（北京）有限公司 | Local path planning method and device for intelligent vehicle |
| CN108648447B (en) * | 2018-05-08 | 2019-04-05 | 清华大学 | Vehicular intelligent security decision method based on traffic safety field |
| CN111079834B (en) * | 2019-12-16 | 2020-10-02 | 清华大学 | Intelligent vehicle safety situation assessment method considering multi-vehicle interaction |
| CN112644498B (en) * | 2021-01-05 | 2022-04-08 | 南京航空航天大学 | Intelligent vehicle safety decision-making method based on driving safety field |
-
2021
- 2021-04-27 CN CN202110459321.8A patent/CN113188556B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113188556A (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bai et al. | Hybrid reinforcement learning-based eco-driving strategy for connected and automated vehicles at signalized intersections | |
| CN111898211B (en) | Intelligent vehicle speed decision method based on deep reinforcement learning and simulation method thereof | |
| Wegener et al. | Automated eco-driving in urban scenarios using deep reinforcement learning | |
| He et al. | Rule-based safety-critical control design using control barrier functions with application to autonomous lane change | |
| CN111222630B (en) | Autonomous driving rule learning method based on deep reinforcement learning | |
| CN114495527B (en) | Internet-connected intersection vehicle road collaborative optimization method and system in mixed traffic environment | |
| CN111267830B (en) | Hybrid power bus energy management method, device and storage medium | |
| Chalaki et al. | Experimental validation of a real-time optimal controller for coordination of cavs in a multi-lane roundabout | |
| Kim et al. | A modeling framework for connectivity and automation co-simulation | |
| Pariota et al. | Integrating tools for an effective testing of connected and automated vehicles technologies | |
| Gupta et al. | Eco-driving of connected and autonomous vehicles with sequence-to-sequence prediction of target vehicle velocity | |
| CN112784867A (en) | Training deep neural networks using synthetic images | |
| CN116564095A (en) | CPS-based key vehicle expressway tunnel prediction cruising cloud control method | |
| Jiang et al. | Learning the policy for mixed electric platoon control of automated and human-driven vehicles at signalized intersection: A random search approach | |
| Li et al. | Development and testing of advanced driver assistance systems through scenario-based system engineering | |
| CN113188556B (en) | Intelligent network-connected automobile track planning method and device based on driving safety field | |
| US20230192118A1 (en) | Automated driving system with desired level of driving aggressiveness | |
| CN113511215B (en) | Hybrid automatic driving decision method, device and computer storage medium | |
| Liang et al. | Shared steering control with predictive risk field enabled by digital twin | |
| Xiong et al. | Noncooperative-game-based intelligent vehicle decision method for lane-changing interactive behavior | |
| Torris | Eco-Driving of Connected and Automated Vehicles (CAVs) | |
| Kavas Torris | Eco-Driving of Connected and Automated Vehicles (CAVs) | |
| Wang et al. | Modeling driver behavior on urban streets | |
| US20230398997A1 (en) | Control of vehicle automated driving operation with independent planning model and cognitive learning model | |
| CN113581182B (en) | Automatic driving vehicle lane change track planning method and system based on reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |