CN111998864B - Unmanned vehicle local path planning method, device, equipment and storage medium - Google Patents
Unmanned vehicle local path planning method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111998864B CN111998864B CN202010804007.4A CN202010804007A CN111998864B CN 111998864 B CN111998864 B CN 111998864B CN 202010804007 A CN202010804007 A CN 202010804007A CN 111998864 B CN111998864 B CN 111998864B
- Authority
- CN
- China
- Prior art keywords
- path
- local
- preset
- local path
- curve
- 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 48
- 230000007613 environmental effect Effects 0.000 claims abstract description 49
- 238000003062 neural network model Methods 0.000 claims abstract description 42
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 20
- 230000011218 segmentation Effects 0.000 claims description 18
- 238000013527 convolutional neural network Methods 0.000 claims description 12
- 230000004888 barrier function Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
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
- G01C21/3407—Route searching; Route guidance specially adapted for specific applications
- G01C21/343—Calculating itineraries, i.e. routes leading from a starting point to a series of categorical destinations using a global route restraint, round trips, touristic trips
-
- 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
-
- 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
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3461—Preferred or disfavoured areas, e.g. dangerous zones, toll or emission zones, intersections, manoeuvre types, segments such as motorways, toll roads, ferries
-
- 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
- G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
- G01C21/3492—Special cost functions, i.e. other than distance or default speed limit of road segments employing speed data or traffic data, e.g. real-time or historical
-
- 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 relates to the technical field of unmanned, and discloses a method, a device, equipment and a storage medium for planning a local path of an unmanned vehicle, wherein the method comprises the following steps: obtaining target environment information in a preset range of a target vehicle, determining an environment congestion level according to the target environment information, generating discrete path points based on a preset global reference path and the environment congestion level, generating a local path curve according to the discrete path points, and inputting the local path curve into a preset deep neural network model to obtain an optimal local path curve. According to the application, the local path curve determined according to the preset global reference path and the environmental congestion level is input into the preset deep neural network model to obtain the optimal local path curve so as to realize comprehensive evaluation of the target environmental information.
Description
Technical Field
The application relates to the technical field of unmanned aerial vehicles, in particular to a method, a device, equipment and a storage medium for planning a local path of an unmanned aerial vehicle.
Background
With the development of unmanned technology, unmanned vehicles may become a mainstream travel mode of the masses in the future, but due to the improvement of the living standard of the masses, the number of vehicles manually driven in large cities with more vehicles is huge. Accordingly, an unpredictable complex scene may occur in the running process of the unmanned vehicle, so that traffic accidents occur, and therefore, how to improve the local path planning efficiency and accuracy of the unmanned vehicle and improve the unmanned safety becomes a problem to be solved urgently.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The application mainly aims to provide a method, a device, equipment and a storage medium for planning a local path of an unmanned vehicle, and aims to solve the technical problem of how to improve the local path planning efficiency and accuracy of the unmanned vehicle and improve the safety of unmanned vehicles.
In order to achieve the above purpose, the present application provides a method for planning a local path of an unmanned vehicle, which comprises the following steps:
when a target vehicle is detected to start an unmanned mode, acquiring target environment information in a preset range of the target vehicle, and determining an environment crowdedness level according to the target environment information;
generating discrete path points based on a preset global reference path and the environmental congestion level, and generating a local path curve according to the discrete path points;
and inputting the local path curve into a preset deep neural network model to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve.
Preferably, the step of acquiring the target environmental information within the preset range of the target vehicle when the target vehicle is detected to start the unmanned mode specifically includes:
when detecting that a target vehicle starts an unmanned mode, acquiring environmental information in a preset range of the target vehicle;
and carrying out coordinate system conversion processing on the environment information to obtain the target environment information under the Frenet coordinate system.
Preferably, the step of determining an environmental congestion level according to the target environmental information specifically includes:
extracting barrier information from the target environment information, and determining the number of barriers according to the barrier information;
and determining the environmental congestion degree level according to the number of the barriers.
Preferably, before the step of generating the discrete path points based on the preset global reference path and the environmental congestion level and generating the local path curve according to the discrete path points, the method further includes:
extracting current coordinate information of the target vehicle from the target environment information, and acquiring destination coordinate information;
and generating a preset global reference path according to the current coordinate information and the destination coordinate information.
Preferably, the step of generating a discrete path point based on a preset global reference path and the environmental congestion level, and generating a local path curve according to the discrete path point specifically includes:
intercepting a preset global reference path with preset length in real time as a local reference path of the target vehicle, and carrying out segmentation processing on the local reference path to obtain the local reference path after segmentation processing;
determining discrete path points according to the environmental congestion level and the local reference path after the segmentation processing;
and generating a discrete path point set based on the discrete path points, and fitting the discrete path point set to obtain a local path curve.
Preferably, before the step of inputting the local path curve to a preset deep neural network model to obtain an optimal local path curve and controlling the running of the target vehicle according to the optimal local path curve, the method further includes:
acquiring convolution kernel weights of a Bayesian convolution neural network model, and training the convolution kernel weights through a preset obstacle avoidance path curve training set to obtain weight distribution results;
obtaining a loss function based on the weight distribution result, and judging whether the loss function accords with a preset loss condition or not;
and when the loss function accords with the preset loss condition, judging that the model training is completed, and taking the Bayesian convolutional neural network model after the training is completed as a preset deep neural network model.
Preferably, the step of inputting the local path curve to a preset deep neural network model to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve specifically includes:
inputting the local path curve into a preset deep neural network model to obtain path evaluation parameters corresponding to the local path curve;
and determining an optimal local path curve according to the path evaluation parameters, and controlling the running of the target vehicle according to the optimal local path curve.
In addition, in order to achieve the above purpose, the application also provides a device for planning a local path of an unmanned vehicle, which comprises the following steps:
the system comprises a congestion degree determining module, a control module and a control module, wherein the congestion degree determining module is used for acquiring target environment information in a preset range of a target vehicle when the target vehicle is detected to start an unmanned mode, and determining an environment congestion degree level according to the target environment information;
the path planning module is used for generating discrete path points based on a preset global reference path and the environmental congestion level and generating a local path curve according to the discrete path points;
the path optimizing module is used for inputting the local path curve into a preset depth neural network model so as to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve.
In addition, in order to achieve the above object, the present application also proposes an unmanned vehicle local path planning apparatus, the apparatus comprising: the system comprises a memory, a processor and an unmanned aerial vehicle local path planning program stored on the memory and capable of running on the processor, wherein the unmanned aerial vehicle local path planning program is configured to realize the steps of the unmanned aerial vehicle local path planning method.
In addition, in order to achieve the above object, the present application also proposes a storage medium having stored thereon an unmanned aerial vehicle local path planning program which, when executed by a processor, implements the steps of the unmanned aerial vehicle local path planning method as described above.
When a target vehicle is detected to start an unmanned mode, target environment information in a preset range of the target vehicle is obtained, an environment crowding degree level is determined according to the target environment information, then a discrete path point is generated based on a preset global reference path and the environment crowding degree level, a local path curve is generated according to the discrete path point, the local path curve is input into a preset depth neural network model, an optimal local path curve is obtained, and running of the target vehicle is controlled according to the optimal local path curve. Compared with the prior art, the local reference path is obtained through a single sensor and the vehicle wireless communication technology, the environment congestion level is determined according to the target environment information, the local path curve is generated based on the preset global reference path and the environment congestion level, the operation amount is reduced, the operation efficiency is improved, further, the local path planning efficiency of the unmanned vehicle is improved, the optimal local path curve is obtained by inputting the local path curve into the preset deep neural network model, the comprehensive evaluation of the target environment information is realized, the accuracy of the local path planning of the unmanned vehicle is improved, the cost of the local path planning of the unmanned vehicle is reduced, and the safety of unmanned vehicles is also improved.
Drawings
FIG. 1 is a schematic structural diagram of an unmanned vehicle local path planning device in a hardware operating environment according to an embodiment of the present application;
FIG. 2 is a flow chart of a first embodiment of a method for planning a local path of an unmanned vehicle according to the present application;
FIG. 3 is a flow chart of a second embodiment of the method for planning a local path of an unmanned vehicle according to the present application;
fig. 4 is a block diagram of a first embodiment of the present application of a local path planning apparatus for an unmanned vehicle.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an unmanned vehicle local path planning device in a hardware running environment according to an embodiment of the present application.
As shown in fig. 1, the unmanned vehicle local path planning apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further 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 high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the configuration shown in fig. 1 does not constitute a limitation of the drone local path planning apparatus, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components.
As shown in fig. 1, an operating system, a data storage module, a network communication module, a user interface module, and an unmanned vehicle local path planning program may be included in the memory 1005 as one type of storage medium.
In the unmanned vehicle local path planning device shown in fig. 1, the network interface 1004 is mainly 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 unmanned aerial vehicle local path planning device can be arranged in the unmanned aerial vehicle local path planning device, and the unmanned aerial vehicle local path planning device calls the unmanned aerial vehicle local path planning program stored in the memory 1005 through the processor 1001 and executes the unmanned aerial vehicle local path planning method provided by the embodiment of the application.
The embodiment of the application provides a method for planning a local path of an unmanned vehicle, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the method for planning the local path of the unmanned vehicle.
In this embodiment, the method for planning a local path of an unmanned vehicle includes the following steps:
step S10: when a target vehicle is detected to start an unmanned mode, acquiring target environment information in a preset range of the target vehicle, and determining an environment crowdedness level according to the target environment information;
it is to be understood that the execution body of the embodiment is the above-mentioned unmanned vehicle local path planning device, which has functions of data processing, data communication, program running, and the like, and may be mounted on the target vehicle, or may be a built-in device of the target vehicle, and the implementation is not limited thereto. When the unmanned vehicle local path planning device detects that the target vehicle starts an unmanned mode, environmental information of the target vehicle in a preset range can be obtained through a preset camera device and a millimeter wave radar, the environmental information can comprise obstacle information and current coordinate information of the target vehicle, the current coordinate information can be geographic coordinates of the target vehicle corresponding to a world coordinate system (World coordinate system) and image information of the target vehicle, the obstacle information can be image information of obstacles, such as pedestrians, trees, other vehicles except the target vehicle, and the like, which can obstruct a driving path of the target vehicle and geographic coordinates corresponding to the world coordinate system, the preset range can be a circle center of the current geographic coordinates of the target vehicle, a circle with a first preset distance as a radius, or a second preset distance in front of the target vehicle, such as a fan-shaped range with a front angle of 30 degrees and a radius of 100 meters, the first preset distance and the second preset distance are not limited in this embodiment, and the preset range can be set according to practical requirements.
In a specific implementation, in order to improve the generation efficiency of the local path curve, the environment information may be subjected to coordinate system conversion processing to obtain the target environment information under the Frenet coordinate system (also referred to as Frenet-Serset formula). In a specific implementation, the geographic coordinates (X W ，Y W ，Z W ) Wherein an X-axis is established based on a traveling direction of the target vehicle (which may be a forward direction or a head direction of the target vehicle), a Y-axis is established based on a direction parallel to a road surface and perpendicular to the X-axis, and a Z-axis is established based on a direction perpendicular to a plane formed by the X-axis and the Y-axis. Then by (X) W ，Y W ，Z W ) Transforming the origin into a coordinate system to obtain a new origin (X) in the camera coordinate system (Camera coordinate system) C ，Y C ，Z C ) Then (X) C ，Y C ，Z C ) Performing coordinate system conversion processing for the origin to obtain a new origin (X, Y) under an image coordinate system (Pixel coordinate system), then performing coordinate system conversion processing by taking the (X, Y) as the origin to obtain target environment information under a Frenet coordinate system, wherein the Frenet coordinate system enables the target environment information to be more easily identified by a preset depth neural network model, and can more intuitively embody the motion relationship among different objects on a road, and furtherThe generation efficiency of the local path curve is also improved.
It is easy to understand that after the target environmental information is obtained, the obstacle information may be extracted from the target environmental information, the number of obstacles may be determined according to the obstacle information, and then the environmental congestion level may be determined according to the number of obstacles, where the environmental congestion level may be classified into three levels, i.e., a high, a medium, and a low, where different numbers of obstacles correspond to different environmental congestion levels, e.g., when the number of obstacles is greater than or equal to 4, the corresponding environmental congestion level is high, and when the number of obstacles is greater than 0 and less than 4, the corresponding environmental congestion level is low, and when no obstacle exists, the corresponding relationship between the number of obstacles and the environmental congestion level may be obtained according to the corresponding relationship between the number of obstacles and the environmental congestion level in the preset congestion relationship map, and may be adjusted according to the historical corresponding relationship.
It should be noted that, after the target environment information is obtained, the current coordinate information of the target vehicle may be extracted from the target environment information, the destination coordinate information may be obtained, then a preset global reference path may be generated according to the current coordinate information and the destination coordinate information, the destination coordinate information may be the destination coordinate information (including the position coordinate of the destination and the image information of the destination) input by the user, the preset global reference path may be a reference path which is planned based on the position coordinate of the target vehicle in the current coordinate information and the geographical coordinate of the destination in the destination coordinate information and is convenient for previewing the global, after the preset global reference path is obtained, further refinement is further required to be performed based on the preset global reference path to obtain a local reference path, and then a local path curve is generated based on the local reference path to improve the precision of local path planning.
Step S20: generating discrete path points based on a preset global reference path and the environmental congestion level, and generating a local path curve according to the discrete path points;
after the preset global reference path and the environmental congestion level are obtained, the preset global reference path with a preset length (for example, 100 meters) can be intercepted in real time to serve as a local reference path of the target vehicle, segmentation processing is performed on the local reference path to obtain the local reference path after segmentation processing (here, the local reference path can be equally divided into 5 segments, if each segment is 20 meters), discrete path points are then determined according to the environmental congestion level and the local reference path after segmentation processing, different environmental congestion levels can be understood to correspond to different discrete path points, for example, when the environmental congestion level is high, the number of corresponding discrete path points is 5 (namely, 5 discrete path points are set in the tangential direction of the segmentation points of the local reference path after segmentation processing with a length of 20 meters), when the environmental congestion level is medium, the number of equally divided discrete points is 3 (namely, 3 discrete path points are set in the tangential direction of the segmentation points of the local reference path after segmentation processing with a length of 20 meters), when the environmental congestion level is low, the number of corresponding discrete path points can be mapped to the preset discrete point with the current reference point according to the preset reference level, and the practical relation can be obtained according to the fact that the preset relation between the corresponding discrete point and the current reference level is 1, and the current relation can be determined according to the practical relation between the preset discrete point and the current relation is obtained.
It is easy to understand that after the discrete path points are obtained, one discrete path point can be selected from different discrete path points corresponding to the tangential direction of the segment points of the local reference path after the segmentation process in real time to form a discrete path point set, and when the discrete path point set is divided into 5 segments, the discrete path point set can be { P } 1i ，P 2i ，P 3i ，P 4i ，P 5i }，P 1i 、P 2i 、P 3i 、P 4i 、P 5i Discrete path points in the tangential direction of the five segment points respectively, i representing the order of their positions in the discrete path points in the direction of the location, and then using a unitary triple equation for the set of discrete path pointsFitting is performed together to obtain a local path curve under the Frenet coordinate system, and the unitary cubic equation can be determined according to actual requirements, which is not limited in the embodiment.
Step S30: and inputting the local path curve into a preset deep neural network model to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve.
It is easy to understand that after the local path curve is obtained, the local path curve may be input to a preset depth neural network model, and path evaluation parameters corresponding to the local path curve are obtained, where the path evaluation parameters may be a rating or a rating sum made by the preset depth neural network model on a slope, a curvature, a distance, a time, a traffic light number, a driving stability, and the like corresponding to the local path curve, where the traffic light number, the driving stability, and the driving stability, on a road corresponding to the local path curve, and then an optimal local path curve is determined according to the path evaluation parameters, and in a specific implementation, a local path curve with the highest rating sum or the highest rating may be selected as the optimal local path curve, and then driving of the target vehicle is controlled according to the optimal local path curve.
It should be understood that the foregoing is illustrative only and is not limiting, and that in specific applications, those skilled in the art may set the application as desired, and the application is not limited thereto.
When the unmanned mode of the target vehicle is started, the method acquires target environment information in a preset range of the target vehicle, determines an environment congestion level according to the target environment information, generates discrete path points based on a preset global reference path and the environment congestion level, generates a local path curve according to the discrete path points, inputs the local path curve into a preset depth neural network model to obtain an optimal local path curve, and controls running of the target vehicle according to the optimal local path curve. Compared with the prior art, the local reference path is obtained through a single sensor and the vehicle wireless communication technology, the environment congestion level is determined according to the target environment information, the local path curve is generated based on the preset global reference path and the environment congestion level, the operation amount is reduced, the operation efficiency is improved, further, the local path planning efficiency of the unmanned vehicle is improved, the optimal local path curve is obtained by inputting the local path curve into the preset deep neural network model, the comprehensive evaluation of the target environment information is realized, the accuracy of the local path planning of the unmanned vehicle is improved, the cost of the local path planning of the unmanned vehicle is reduced, and the safety of unmanned vehicles is also improved.
Referring to fig. 3, fig. 3 is a flow chart of a third embodiment of the method for planning a local path of an unmanned vehicle according to the present application.
Based on the above embodiments, in this embodiment, before step S30, the method further includes:
step S001: acquiring convolution kernel weights of a Bayesian convolution neural network model, and training the convolution kernel weights through a preset obstacle avoidance path curve training set to obtain weight distribution results;
step S002: obtaining a loss function based on the weight distribution result, and judging whether the loss function accords with a preset loss condition or not;
step S003: and when the loss function accords with the preset loss condition, judging that the model training is completed, and taking the Bayesian convolutional neural network model after the training is completed as a preset deep neural network model.
It should be noted that, the bayesian convolutional neural network model is built based on a bayesian convolutional neural network, and is adaptively limited based on the unmanned vehicle local path planning method of the present application, in the bayesian convolutional neural network model, a convolution layer, an activation layer and a pooling layer are in a series relationship, that is, a convolution layer, an activation layer, a pooling layer, a loss function (loss function), only a convolution kernel weight of the bayesian convolutional neural network model needs to be obtained, then the convolution kernel weight is trained through a preset obstacle avoidance path curve training set (which can be understood as a training set for performing local path planning training on the bayesian convolutional neural network model), a weight distribution result (which can be understood as a set of a priori distribution and posterior distribution corresponding to the convolution kernel weight) is obtained, and based on the weight distribution result, a variable decibel leaf-state estimation (variational Bayesian inference) is performed on the weight distribution result according to an optimal control theory (optimal control theory), the variable decibel leaf-state estimation (variational Bayesian inference) can be performed in an iterative manner, that is given in a preset model, namely, when the constraint loss is satisfied with a preset rule, the constraint rule is determined, the fuzzy rule is satisfied, and then the fuzzy rule is determined, and taking the trained Bayesian convolutional neural network model as a preset deep neural network model. In a specific implementation, after the Loss function Loss is obtained, whether the model is trained is judged by the following formula, namely when the Loss function Loss approaches 0 or is 0, the model is judged to be trained, and the trained Bayesian convolutional neural network model is used as a preset deep neural network model.
Loss＝E Q [logQ(w i |α)-logP(w i )]-E Q (logP(S|w i )]
Wherein w is i Represents the ith convolution kernel weight, E Q The method is characterized in that the method comprises the steps of taking expected values, Q represents variation posterior distribution, P represents variation prior distribution, alpha represents preset parameters conforming to Gaussian distribution, and S represents a preset obstacle avoidance path curve training set.
According to the embodiment, convolution kernel weights of a Bayesian convolution neural network model are obtained, the convolution kernel weights are trained through a preset obstacle avoidance path curve training set, a weight distribution result is obtained, a loss function is obtained based on the weight distribution result, whether the loss function meets preset loss conditions or not is judged, when the loss function meets the preset loss conditions, model training is judged to be completed, and the trained Bayesian convolution neural network model is used as a preset deep neural network model. And training the convolution kernel weight of the Bayesian convolution neural network model to obtain a weight distribution result. In contrast to the prior art that the optimal local path curve is obtained only by overlapping the obtained road condition characteristics when the local path planning is performed, the method and the device obtain the preset depth neural network model required by the method based on the Bayesian convolution neural network model, and perform the local path planning through the preset depth neural network model so as to improve the model precision and the model depth of the preset depth neural network model, improve the planning accuracy when the local path planning is performed based on the preset depth neural network model, and further improve the unmanned safety.
In addition, the embodiment of the application also provides a storage medium, wherein the storage medium is stored with an unmanned aerial vehicle local path planning program, and the unmanned aerial vehicle local path planning program realizes the steps of the unmanned aerial vehicle local path planning method when being executed by a processor.
Referring to fig. 4, fig. 4 is a block diagram illustrating a first embodiment of a local path planning apparatus for an unmanned vehicle according to the present application.
As shown in fig. 4, the device for planning a local path of an unmanned vehicle according to the embodiment of the present application includes:
the congestion degree determining module 10 is configured to obtain target environmental information within a preset range of a target vehicle when detecting that the target vehicle starts an unmanned mode, and determine an environmental congestion degree level according to the target environmental information;
the path planning module 20 is configured to generate discrete path points based on a preset global reference path and the environmental congestion level, and generate a local path curve according to the discrete path points;
the path optimizing module 30 is configured to input the local path curve to a preset deep neural network model, obtain an optimal local path curve, and control the driving of the target vehicle according to the optimal local path curve.
When a target vehicle is detected to start an unmanned mode, target environment information in a preset range of the target vehicle is obtained, an environment crowding degree level is determined according to the target environment information, then a discrete path point is generated based on a preset global reference path and the environment crowding degree level, a local path curve is generated according to the discrete path point, the local path curve is input into a preset depth neural network model, an optimal local path curve is obtained, and running of the target vehicle is controlled according to the optimal local path curve. Compared with the prior art, the local reference path is obtained through a single sensor and the vehicle wireless communication technology, the environment congestion level is determined according to the target environment information, the local path curve is generated based on the preset global reference path and the environment congestion level, the operation amount is reduced, the operation efficiency is improved, further, the local path planning efficiency of the unmanned vehicle is improved, the optimal local path curve is obtained by inputting the local path curve into the preset deep neural network model, the comprehensive evaluation of the target environment information is realized, the accuracy of the local path planning of the unmanned vehicle is improved, the cost of the local path planning of the unmanned vehicle is reduced, and the safety of unmanned vehicles is also improved.
Based on the first embodiment of the unmanned aerial vehicle local path planning device, a second embodiment of the unmanned aerial vehicle local path planning device is provided.
In this embodiment, the congestion degree determining module 10 is further configured to obtain environmental information within a preset range of the target vehicle when detecting that the target vehicle starts the unmanned mode;
the crowding degree determining module 10 is further configured to perform coordinate system conversion processing on the environmental information, and obtain target environmental information under a Frenet coordinate system.
The congestion degree determining module 10 is further configured to obtain target environmental information within a preset range of the target vehicle when detecting that the target vehicle starts an unmanned mode;
the congestion degree determining module 10 is further configured to extract obstacle information from the target environment information, and determine the number of obstacles according to the obstacle information;
the congestion level determination module 10 is further configured to determine an environmental congestion level according to the number of obstacles.
The path planning module 20 is further configured to extract current coordinate information of the target vehicle from the target environment information, and obtain destination coordinate information;
the path planning module 20 is further configured to generate a preset global reference path according to the current coordinate information and the destination coordinate information.
The path planning module 20 is further configured to intercept a preset global reference path with a preset length in real time as a local reference path of the target vehicle, and segment the local reference path to obtain the segmented local reference path;
the path planning module 20 is further configured to determine a discrete path point according to the environmental congestion level and the local reference path after the segmentation process;
the path planning module 20 is further configured to generate a discrete path point set based on the discrete path points, and fit the discrete path point set to obtain a local path curve.
The path optimizing module 30 is further configured to obtain a convolution kernel weight of the bayesian convolutional neural network model, and train the convolution kernel weight through a preset obstacle avoidance path curve training set to obtain a weight distribution result;
the path optimizing module 30 is further configured to obtain a loss function based on the weight distribution result, and determine whether the loss function meets a preset loss condition;
the path optimizing module 30 is further configured to determine that model training is completed when the loss function meets the preset loss condition, and take the bayesian convolutional neural network model after training as a preset deep neural network model.
The path optimizing module 30 is further configured to input the local path curve to a preset deep neural network model, and obtain a path evaluation parameter corresponding to the local path curve;
the path optimizing module 30 is further configured to determine an optimal local path curve according to the path evaluation parameter, and control the driving of the target vehicle according to the optimal local path curve.
Other embodiments or specific implementation manners of the unmanned aerial vehicle local path planning device can refer to the above method embodiments, and are not repeated here.
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 one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read-only memory/random-access memory, magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the application, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (9)
1. A method for planning a local path of an unmanned vehicle, the method comprising the steps of:
when a target vehicle is detected to start an unmanned mode, acquiring target environment information in a preset range of the target vehicle, and determining an environment crowdedness level according to the target environment information;
generating discrete path points based on a preset global reference path and the environmental congestion level, and generating a local path curve according to the discrete path points;
inputting the local path curve into a preset depth neural network model to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve;
the step of generating a discrete path point based on a preset global reference path and the environmental congestion level and generating a local path curve according to the discrete path point specifically includes:
intercepting a preset global reference path with preset length in real time as a local reference path of the target vehicle, and carrying out segmentation processing on the local reference path to obtain the local reference path after segmentation processing;
determining a plurality of discrete path points according to the environment congestion level and the local reference path after the segmentation processing, wherein the number of the discrete path points is increased as the environment congestion level is higher;
and generating a plurality of discrete path point sets based on the plurality of discrete path points, and fitting the plurality of discrete path point sets to obtain a plurality of local path curves.
2. The method according to claim 1, wherein the step of acquiring the target environmental information within a preset range of the target vehicle when the target vehicle is detected to turn on the unmanned mode, specifically comprises:
when detecting that a target vehicle starts an unmanned mode, acquiring environmental information in a preset range of the target vehicle;
and carrying out coordinate system conversion processing on the environment information to obtain the target environment information under the Frenet coordinate system.
3. The method of claim 1, wherein the step of determining an environmental congestion level based on the target environmental information comprises:
extracting barrier information from the target environment information, and determining the number of barriers according to the barrier information;
and determining the environmental congestion degree level according to the number of the barriers.
4. The method of claim 1, wherein prior to the step of generating discrete path points based on the preset global reference path and the environmental congestion level and generating a local path curve from the discrete path points, further comprising:
extracting current coordinate information of the target vehicle from the target environment information, and acquiring destination coordinate information;
and generating a preset global reference path according to the current coordinate information and the destination coordinate information.
5. The method of claim 1, wherein the step of inputting the local path curve to a predetermined depth neural network model to obtain an optimal local path curve and controlling the travel of the target vehicle according to the optimal local path curve further comprises:
acquiring convolution kernel weights of a Bayesian convolution neural network model, and training the convolution kernel weights through a preset obstacle avoidance path curve training set to obtain weight distribution results;
obtaining a loss function based on the weight distribution result, and judging whether the loss function accords with a preset loss condition or not;
and when the loss function accords with the preset loss condition, judging that the model training is completed, and taking the Bayesian convolutional neural network model after the training is completed as a preset deep neural network model.
6. The method according to claim 1, wherein the step of inputting the local path curve to a preset depth neural network model to obtain an optimal local path curve, and controlling the driving of the target vehicle according to the optimal local path curve specifically comprises:
inputting the local path curve into a preset deep neural network model to obtain path evaluation parameters corresponding to the local path curve;
and determining an optimal local path curve according to the path evaluation parameters, and controlling the running of the target vehicle according to the optimal local path curve.
7. An unmanned vehicle local path planning device, characterized in that the device comprises the following steps:
the system comprises a congestion degree determining module, a control module and a control module, wherein the congestion degree determining module is used for acquiring target environment information in a preset range of a target vehicle when the target vehicle is detected to start an unmanned mode, and determining an environment congestion degree level according to the target environment information;
the path planning module is used for generating discrete path points based on a preset global reference path and the environmental congestion level and generating a local path curve according to the discrete path points;
the path optimizing module is used for inputting the local path curve into a preset depth neural network model so as to obtain an optimal local path curve, and controlling the running of the target vehicle according to the optimal local path curve;
the path planning module is further used for intercepting a preset global reference path with preset length in real time as a local reference path of the target vehicle, and carrying out segmentation processing on the local reference path to obtain the local reference path after segmentation processing; determining a plurality of discrete path points according to the environment congestion level and the local reference path after the segmentation processing, wherein the number of the discrete path points is increased as the environment congestion level is higher; and generating a plurality of discrete path point sets based on the plurality of discrete path points, and fitting the plurality of discrete path point sets to obtain a plurality of local path curves.
8. An unmanned vehicle local path planning apparatus, the apparatus comprising: a memory, a processor and an unmanned vehicle local path planning program stored on the memory and operable on the processor, the unmanned vehicle local path planning program configured to implement the steps of the unmanned vehicle local path planning method of any one of claims 1 to 6.
9. A storage medium having stored thereon an unmanned vehicle local path planning program which, when executed by a processor, implements the steps of the unmanned vehicle local path planning method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010804007.4A CN111998864B (en) | 2020-08-11 | 2020-08-11 | Unmanned vehicle local path planning method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010804007.4A CN111998864B (en) | 2020-08-11 | 2020-08-11 | Unmanned vehicle local path planning method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111998864A CN111998864A (en) | 2020-11-27 |
| CN111998864B true CN111998864B (en) | 2023-11-07 |
Family
ID=73463831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010804007.4A Active CN111998864B (en) | 2020-08-11 | 2020-08-11 | Unmanned vehicle local path planning method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111998864B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112519797A (en) * | 2020-12-10 | 2021-03-19 | 广州小鹏自动驾驶科技有限公司 | Vehicle safety distance early warning method, early warning system, automobile and storage medium |
| CN112964267B (en) * | 2021-02-07 | 2022-12-27 | 舜宇光学（浙江）研究院有限公司 | Autonomous obstacle avoidance navigation method, system and equipment thereof |
| CN113189989B (en) * | 2021-04-21 | 2022-07-01 | 东风柳州汽车有限公司 | Vehicle intention prediction method, device, equipment and storage medium |
| CN112991744B (en) * | 2021-04-28 | 2021-08-13 | 北京科技大学 | Automatic driving decision-making method and system suitable for long-distance urban road |
| CN113268061A (en) * | 2021-05-14 | 2021-08-17 | 深圳中智永浩机器人有限公司 | Robot chassis multipoint navigation method and device, computer equipment and storage medium |
| CN113390431B (en) * | 2021-06-17 | 2022-09-30 | 广东工业大学 | Method and device for dynamically generating reference line, computer equipment and storage medium |
| CN114115216B (en) * | 2021-07-30 | 2023-07-25 | 东风柳州汽车有限公司 | Vehicle formation control method, device, equipment and storage medium |
| CN113625724B (en) * | 2021-08-31 | 2023-07-21 | 东风柳州汽车有限公司 | Unmanned vehicle path planning method, device, equipment and storage medium |
| CN113568416B (en) * | 2021-09-26 | 2021-12-24 | 智道网联科技(北京)有限公司 | Unmanned vehicle trajectory planning method, device and computer readable storage medium |
| CN114237229B (en) * | 2021-11-26 | 2023-06-13 | 青岛德智汽车科技有限公司 | Unstructured road work vehicle path planning method based on empirical path fitting |
| CN115326099B (en) * | 2022-10-11 | 2023-03-07 | 禾多科技（北京）有限公司 | Local path planning method and device, electronic equipment and computer readable medium |
Citations (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4790402A (en) * | 1987-09-28 | 1988-12-13 | Tennant Company | Automated guided vehicle |
| US4996468A (en) * | 1987-09-28 | 1991-02-26 | Tennant Company | Automated guided vehicle |
| KR20090123475A (en) * | 2008-05-28 | 2009-12-02 | 삼성전자주식회사 | Method for searching routes of potable terminal |
| KR101063302B1 (en) * | 2010-10-05 | 2011-09-07 | 국방과학연구소 | Control apparatus and method for autonomous navigation of unmanned ground vehicle |
| JP2012056063A (en) * | 2010-09-13 | 2012-03-22 | Institute Of National Colleges Of Technology Japan | Smooth motion path generating device and smooth motion path generating method |
| KR20150104484A (en) * | 2014-03-05 | 2015-09-15 | 국방과학연구소 | Method and apparatus for generating pathe of autonomous vehicle |
| CN105788330A (en) * | 2016-04-29 | 2016-07-20 | 百度在线网络技术（北京）有限公司 | Real-time road-condition early warning method of unmanned vehicle and apparatus thereof |
| KR20170043946A (en) * | 2015-10-14 | 2017-04-24 | 국방과학연구소 | An apparatus for planing a route of a mobile terminal and method therof |
| WO2018036197A1 (en) * | 2016-08-26 | 2018-03-01 | 中兴通讯股份有限公司 | Driverless vehicle control method, terminal, server, and system |
| US9910443B1 (en) * | 2016-10-14 | 2018-03-06 | Hyundai Motor Company | Drive control apparatus and method for autonomous vehicle |
| CN107992050A (en) * | 2017-12-20 | 2018-05-04 | 广州汽车集团股份有限公司 | Pilotless automobile local path motion planning method and device |
| CN108088456A (en) * | 2017-12-21 | 2018-05-29 | 北京工业大学 | A kind of automatic driving vehicle local paths planning method with time consistency |
| CN108489502A (en) * | 2018-03-23 | 2018-09-04 | 四川飞通系统集成有限公司 | Wisdom traffic route optimization method |
| CN108775902A (en) * | 2018-07-25 | 2018-11-09 | 齐鲁工业大学 | The adjoint robot path planning method and system virtually expanded based on barrier |
| CN108875998A (en) * | 2018-04-20 | 2018-11-23 | 北京智行者科技有限公司 | A kind of automatic driving vehicle method and system for planning |
| CN108898866A (en) * | 2018-07-04 | 2018-11-27 | 深圳万智联合科技有限公司 | A kind of effective intelligent vehicle control system |
| CN109697881A (en) * | 2019-01-15 | 2019-04-30 | 南通大学 | A kind of detour guidance method based on Real-time Road congestion information |
| CN109799822A (en) * | 2019-01-30 | 2019-05-24 | 中国石油大学(华东) | Mobile robot global smooth paths planing method |
| CN110147105A (en) * | 2019-05-27 | 2019-08-20 | 安徽江淮汽车集团股份有限公司 | Controlling of path thereof, equipment, storage medium and the device of automatic driving vehicle |
| CN110174893A (en) * | 2019-05-07 | 2019-08-27 | 重庆工程职业技术学院 | A kind of unmanned control method, system and vehicle |
| US10442082B1 (en) * | 2014-07-18 | 2019-10-15 | Al Incorporated | Methods and systems for automated robotic movement |
| CN110954122A (en) * | 2019-12-09 | 2020-04-03 | 东风商用车有限公司 | Automatic driving track generation method under high-speed scene |
| CN111289008A (en) * | 2020-04-28 | 2020-06-16 | 南京维思科汽车科技有限公司 | Local path planning algorithm for unmanned vehicle |
| CN111289978A (en) * | 2020-02-13 | 2020-06-16 | 北京享云智汇科技有限公司 | Method and system for making decision on unmanned driving behavior of vehicle |
| CN111338361A (en) * | 2020-05-22 | 2020-06-26 | 浙江远传信息技术股份有限公司 | Obstacle avoidance method, device, equipment and medium for low-speed unmanned vehicle |
| CN111459159A (en) * | 2020-03-16 | 2020-07-28 | 江苏理工学院 | Path following control system and control method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6380274B2 (en) * | 2015-07-28 | 2018-08-29 | トヨタ自動車株式会社 | Navigation device for autonomous vehicles |
| US10883844B2 (en) * | 2017-07-27 | 2021-01-05 | Waymo Llc | Neural networks for vehicle trajectory planning |
| JP7032295B2 (en) * | 2018-12-26 | 2022-03-08 | 本田技研工業株式会社 | Vehicle control systems, vehicle control methods, and programs |
-
2020
- 2020-08-11 CN CN202010804007.4A patent/CN111998864B/en active Active
Patent Citations (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4790402A (en) * | 1987-09-28 | 1988-12-13 | Tennant Company | Automated guided vehicle |
| US4996468A (en) * | 1987-09-28 | 1991-02-26 | Tennant Company | Automated guided vehicle |
| KR20090123475A (en) * | 2008-05-28 | 2009-12-02 | 삼성전자주식회사 | Method for searching routes of potable terminal |
| JP2012056063A (en) * | 2010-09-13 | 2012-03-22 | Institute Of National Colleges Of Technology Japan | Smooth motion path generating device and smooth motion path generating method |
| KR101063302B1 (en) * | 2010-10-05 | 2011-09-07 | 국방과학연구소 | Control apparatus and method for autonomous navigation of unmanned ground vehicle |
| KR20150104484A (en) * | 2014-03-05 | 2015-09-15 | 국방과학연구소 | Method and apparatus for generating pathe of autonomous vehicle |
| US10442082B1 (en) * | 2014-07-18 | 2019-10-15 | Al Incorporated | Methods and systems for automated robotic movement |
| KR20170043946A (en) * | 2015-10-14 | 2017-04-24 | 국방과학연구소 | An apparatus for planing a route of a mobile terminal and method therof |
| CN105788330A (en) * | 2016-04-29 | 2016-07-20 | 百度在线网络技术（北京）有限公司 | Real-time road-condition early warning method of unmanned vehicle and apparatus thereof |
| WO2018036197A1 (en) * | 2016-08-26 | 2018-03-01 | 中兴通讯股份有限公司 | Driverless vehicle control method, terminal, server, and system |
| US9910443B1 (en) * | 2016-10-14 | 2018-03-06 | Hyundai Motor Company | Drive control apparatus and method for autonomous vehicle |
| CN107992050A (en) * | 2017-12-20 | 2018-05-04 | 广州汽车集团股份有限公司 | Pilotless automobile local path motion planning method and device |
| CN108088456A (en) * | 2017-12-21 | 2018-05-29 | 北京工业大学 | A kind of automatic driving vehicle local paths planning method with time consistency |
| CN108489502A (en) * | 2018-03-23 | 2018-09-04 | 四川飞通系统集成有限公司 | Wisdom traffic route optimization method |
| CN108875998A (en) * | 2018-04-20 | 2018-11-23 | 北京智行者科技有限公司 | A kind of automatic driving vehicle method and system for planning |
| CN108898866A (en) * | 2018-07-04 | 2018-11-27 | 深圳万智联合科技有限公司 | A kind of effective intelligent vehicle control system |
| CN108775902A (en) * | 2018-07-25 | 2018-11-09 | 齐鲁工业大学 | The adjoint robot path planning method and system virtually expanded based on barrier |
| CN109697881A (en) * | 2019-01-15 | 2019-04-30 | 南通大学 | A kind of detour guidance method based on Real-time Road congestion information |
| CN109799822A (en) * | 2019-01-30 | 2019-05-24 | 中国石油大学(华东) | Mobile robot global smooth paths planing method |
| CN110174893A (en) * | 2019-05-07 | 2019-08-27 | 重庆工程职业技术学院 | A kind of unmanned control method, system and vehicle |
| CN110147105A (en) * | 2019-05-27 | 2019-08-20 | 安徽江淮汽车集团股份有限公司 | Controlling of path thereof, equipment, storage medium and the device of automatic driving vehicle |
| CN110954122A (en) * | 2019-12-09 | 2020-04-03 | 东风商用车有限公司 | Automatic driving track generation method under high-speed scene |
| CN111289978A (en) * | 2020-02-13 | 2020-06-16 | 北京享云智汇科技有限公司 | Method and system for making decision on unmanned driving behavior of vehicle |
| CN111459159A (en) * | 2020-03-16 | 2020-07-28 | 江苏理工学院 | Path following control system and control method |
| CN111289008A (en) * | 2020-04-28 | 2020-06-16 | 南京维思科汽车科技有限公司 | Local path planning algorithm for unmanned vehicle |
| CN111338361A (en) * | 2020-05-22 | 2020-06-26 | 浙江远传信息技术股份有限公司 | Obstacle avoidance method, device, equipment and medium for low-speed unmanned vehicle |
Non-Patent Citations (4)
| Title |
|---|
| Not science fiction: driverless car technologies are an effective way of helping to cut road fatalities and congestion, according to Anthony Baxendale;Anthony，Baxendale;Engineer;全文 * |
| 动态确定基因数的遗传算法路径规划;肖晓明;陈志兴;高平安;;计算机应用研究;26(7);2469-2470+2486 * |
| 基于Haar特征的车辆识别方法;韦俏杏, 等;大众科技;第21卷(第7期);31-33+30 * |
| 肖晓明 ; 陈志兴 ; 高平安 ; .动态确定基因数的遗传算法路径规划.计算机应用研究.2009,26(7),2469-2470+2486. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111998864A (en) | 2020-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111998864B (en) | Unmanned vehicle local path planning method, device, equipment and storage medium | |
| EP3714290B1 (en) | Lidar localization using 3d cnn network for solution inference in autonomous driving vehicles | |
| CN108089572B (en) | Method and device for vehicle positioning | |
| US10823575B2 (en) | Reference line smoothing method using piecewise spiral curves with weighted geometry costs | |
| US10466058B2 (en) | Navigation for vehicles | |
| EP3714285B1 (en) | Lidar localization using rnn and lstm for temporal smoothness in autonomous driving vehicles | |
| CN111971574B (en) | Deep learning based feature extraction for LIDAR localization of autonomous vehicles | |
| EP1733287B1 (en) | System and method for adaptive path planning | |
| CN111694356B (en) | Driving control method and device, electronic equipment and storage medium | |
| CN110576847A (en) | Focus-based tagging of sensor data | |
| CN108255170A (en) | The method for dynamically adjusting the speed control rate of automatic driving vehicle | |
| US11373328B2 (en) | Method, device and storage medium for positioning object | |
| CN112506219A (en) | Intelligent traffic supervision unmanned aerial vehicle track planning method and system and readable storage medium | |
| CN116225066A (en) | Unmanned aerial vehicle path optimization method based on chaotic mapping pely optimization algorithm | |
| CN112066976A (en) | Self-adaptive expansion processing method and system, robot and storage medium | |
| CN113189989B (en) | Vehicle intention prediction method, device, equipment and storage medium | |
| CN110721470A (en) | Control method and device for virtual object movement behavior and electronic equipment | |
| WO2021016920A1 (en) | Method, system and device for identifying accessibility, and computer-readable storage medium | |
| US11555928B2 (en) | Three-dimensional object detection with ground removal intelligence | |
| WO2023060586A1 (en) | Automatic driving instruction generation model optimization method and apparatus, device, and storage medium | |
| CN114425774B (en) | Robot walking road recognition method, robot walking road recognition device, and storage medium | |
| CN113119999B (en) | Method, device, equipment, medium and program product for determining automatic driving characteristics | |
| CN115235490A (en) | Driving route generation method, device, equipment and storage medium | |
| CN113885522B (en) | Obstacle avoidance method, device and equipment for water surface robot and readable storage medium | |
| EP3944137A1 (en) | Positioning method and positioning apparatus |
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 |