CN114090908A - Road network data processing method and device - Google Patents
Road network data processing method and device Download PDFInfo
- Publication number
- CN114090908A CN114090908A CN202111389644.0A CN202111389644A CN114090908A CN 114090908 A CN114090908 A CN 114090908A CN 202111389644 A CN202111389644 A CN 202111389644A CN 114090908 A CN114090908 A CN 114090908A
- Authority
- CN
- China
- Prior art keywords
- road network
- road
- data
- network data
- track
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/203—Drawing of straight lines or curves
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Remote Sensing (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention provides a road network data processing method and a road network data processing device, wherein the method comprises the following steps: the method comprises the steps of obtaining a vehicle motion track and road network perception data collected in real time, and determining track nodes based on the vehicle motion track; fitting according to the track node and the current vehicle position to obtain a road track curve, and fitting road network perception data to obtain a road edge curve; and adding corresponding logic characteristics to data points contained in the road track curve and the road edge curve to generate a road network data structure and obtain road network data so as to use the road network data after passing the inspection. The continuous and complete road network structure is generated, real-time rendering, path planning and the like are performed, real-time acquisition of road network data in the intelligent driving process is achieved, the processed road network data are presented to vehicle users, the vehicle users can sense the surrounding environment in time, and the road network structure has good expansibility and universality.
Description
Technical Field
The present invention relates to the field of vehicle technologies, and in particular, to a method and an apparatus for processing road network data.
Background
The road network data is a representation mode of road characteristics, and can include basic information such as positions, directions and boundaries of roads, and extension information such as movement speed limitation and movement direction limitation of objects on the roads. The construction of the road network usually needs huge data volume, and the data is generally acquired by various vehicle-mounted sensors and then obtained after processing such as training, fusion, alignment, calibration and the like, so that the acquisition of the road network needs certain time cost and labor cost, and the real-time acquisition and real-time use of the road network data in the intelligent driving process cannot be realized.
Disclosure of Invention
In view of the above problems, embodiments of the present invention are proposed to provide a road network data processing method and a road network data processing apparatus, which overcome or at least partially solve the above problems.
The embodiment of the invention discloses a road network data processing method, which comprises the following steps:
the method comprises the steps of obtaining a vehicle motion track and road network perception data collected in real time, and determining track nodes based on the vehicle motion track;
fitting according to the track node and the current vehicle position to obtain a road track curve, and fitting the road network perception data to obtain a road edge curve;
and adding corresponding logic characteristics to data points contained in the road track curve and the road edge curve to generate a road network data structure and obtain road network data so as to use the road network data after passing the inspection.
Optionally, the fitting the track node with the current vehicle position to obtain a road track curve includes:
acquiring a current vehicle position and a previous track node set by the current vehicle position in a vehicle motion track;
and fitting a track curve between the previous track node and the current vehicle position based on a least square method to obtain a road track curve.
Optionally, the fitting the road network perception data to obtain a road edge curve includes:
determining road network data points and interference data points from the road network perception data;
acquiring coordinates of the road network data, substituting the coordinates of the road network data points into the fitted road track curve, and determining the distribution positions of the road network data points relative to the vehicle motion track;
and fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trail to obtain a road edge curve.
Optionally, the determining a road network data point and an interference data point from the road network perception data includes:
acquiring a data point retrieval radius, and if other data points exist in a retrieval area with a certain data point distance as the data point retrieval radius, determining the data point and the other data points as road network data points;
and if no other data point exists in the searching area for the distance of another data point, which is the searching radius of the data point, determining the data point as an interference data point.
Optionally, the obtaining a data point retrieval radius includes:
and acquiring a plurality of adjacent data points from the road network perception data, calculating a mean value of the distances between the plurality of adjacent data points, and processing the calculated mean value by adopting a preset scale factor to obtain a data point retrieval radius.
Optionally, the fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trajectory to obtain a road edge curve includes:
obtaining edge data including road network data points positioned on a preset road boundary based on the distribution positions of the road network data points relative to the vehicle motion trail;
and fitting the edge data based on a least square method to obtain a linear regression equation aiming at a road edge curve, so as to realize the fitting of the road boundary.
Optionally, the response logic feature added by the contained data point comprises a track feature and a road edge feature; the generating a road network data structure and obtaining road network data by adding corresponding logical features to data points included in the road track curve and the road edge curve includes:
and adding track features to the track data points contained in the road track curve and adding road edge features to the road network data points contained in the road edge curve to form a road network data structure.
Optionally, after generating the road network data structure and obtaining the road network data, the method further includes:
analyzing the data under the same logic characteristic, and if the fitted road track curve and the line segment in the road edge curve are not crossed or data point loss is not generated, checking the obtained road network data to pass;
and storing the road network data passing the inspection so as to use the road network data passing the inspection in real time by calling a preset rendering program or a preset interface.
The embodiment of the invention also discloses a road network data processing device, which comprises:
the system comprises a track node determining module, a road network sensing module and a road network sensing module, wherein the track node determining module is used for acquiring a vehicle motion track and road network sensing data acquired in real time and determining track nodes based on the vehicle motion track;
the curve fitting module is used for fitting the track node with the current vehicle position to obtain a road track curve and fitting the road network perception data to obtain a road edge curve;
and the road network data generation module is used for generating a road network data structure and obtaining road network data by adding corresponding logic characteristics to data points contained in the road track curve and the road edge curve so as to use the road network data after passing the inspection.
Optionally, the curve fitting module comprises:
the track node determination submodule is used for acquiring the current vehicle position and a previous track node set by the current vehicle position in the vehicle motion track;
and the road track curve fitting submodule is used for fitting the track curve between the previous track node and the current vehicle position based on a least square method to obtain the road track curve.
Optionally, the curve fitting module comprises:
the data point determining submodule is used for determining road network data points and interference data points from the road network perception data;
the distribution position determining submodule is used for acquiring the coordinates of the road network data, substituting the coordinates of the road network data points into the fitted road track curve and determining the distribution positions of the road network data points relative to the vehicle motion track;
and the road edge curve fitting submodule is used for fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trail to obtain a road edge curve.
Optionally, the data point determination submodule includes:
a retrieval radius obtaining unit for obtaining a data point retrieval radius;
the road network data point determining unit is used for determining that a certain data point is a road network data point when other data points exist in a retrieval area with a data point distance as a data point retrieval radius;
and the interference data point determining unit is used for determining the data point as an interference data point when no other data point exists in the searching area for the distance of another data point, wherein the distance of the other data point is the searching radius of the data point.
Optionally, the retrieval radius obtaining unit includes:
and the retrieval radius determining subunit is used for acquiring a plurality of adjacent data points from the road network perception data, calculating a mean value of the distances between the plurality of adjacent data points, and processing the calculated mean value by adopting a preset scale factor to obtain the data point retrieval radius.
Optionally, the road-edge curve fitting submodule includes:
the edge data acquisition unit is used for obtaining edge data containing road network data points positioned on a preset road boundary based on the distribution positions of the road network data points relative to the vehicle motion trail;
and the road edge curve fitting unit is used for fitting the edge data based on a least square method to obtain a linear regression equation aiming at the road edge curve and realize the fitting of the road boundary.
Optionally, the response logic feature added by the contained data point includes a track feature and a road edge feature; the road network data generation module comprises:
and the road network data structure forming submodule is used for adding a track characteristic to the track data points contained in the road track curve and adding a road edge characteristic to the road network data points contained in the road edge curve to form a road network data structure.
Optionally, after generating the road network data structure and obtaining the road network data, the apparatus further includes:
the road network data inspection module is used for analyzing data under the same logic characteristic, and if the fitted road track curve and the line segment in the road edge curve are not crossed or data point loss does not occur, the obtained road network data is inspected to pass;
and the road network data using module is used for storing the road network data passing the inspection so as to use the road network data passing the inspection in real time by calling a preset rendering program or a preset interface.
The embodiment of the invention also discloses a vehicle, which comprises: the road network data processing device, the processor, the memory and the computer program stored on the memory and capable of running on the processor realize the steps of any one of the road network data processing methods when the computer program is executed by the processor.
The embodiment of the invention also discloses a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of any road network data processing method are realized.
The embodiment of the invention has the following advantages:
in the embodiment of the invention, the track nodes are determined based on the motion track of the vehicle, the determined track nodes are fitted with the current vehicle position to obtain the road track curve, the road network informing data collected in real time are fitted to obtain the road edge curve, and then the corresponding logic characteristics are respectively added to the data points contained in the fitted road track curve and the road edge curve to form the road network data structure to obtain the road network data, so that the road network data passing the inspection can be used in the following process. The road network data collected by the sensor are filtered and smoothed from the vehicle pose, a continuous and complete road network structure is generated to perform real-time rendering, path planning and the like, the road network data are collected in real time in the intelligent driving process, the processed road network data are presented to vehicle users, the vehicle users can sense the surrounding environment in time, and the road network data processing system has good expansibility and universality.
Drawings
FIG. 1 is a flow chart illustrating steps of a road network data processing method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a vehicle motion trajectory and a data sampling point cloud provided by an embodiment of the invention;
FIG. 3 is a flow chart illustrating steps of another embodiment of a method for processing road network data according to the present invention;
fig. 4 is a diagram of an implementation process of road network data processing according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of processing road network data according to an embodiment of the present invention;
fig. 6 is a block diagram showing an embodiment of a road network data processing device according to the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention more comprehensible, the present invention is described in detail with reference to the accompanying drawings and the detailed description thereof.
The currently common road network processing method (after processing such as training, fusion, alignment, calibration and the like) needs certain time cost and labor cost, and if road network data can be obtained in real time in intelligent driving, the road network data are processed and presented to a driver and passengers, and the method is favorable for helping the user to timely sense the surrounding environment.
The embodiment of the invention provides a road network generation method based on vehicle motion trail and perception information aiming at the environment with less dynamic interference and simple scene characteristics such as a parking garage, and the like. The road network data which can be conveniently rendered in real time and can be easily called by other interfaces are obtained by processing through means of filtering, classifying, smoothing, fitting and the like, the data collected by the sensor can be quickly processed in environments with less dynamic interference and simple scene characteristics, such as parking garages, shopping malls and the like, the processing steps are simple, the time complexity is low, and the processed data can be used in real time and is convenient for processing such as rendering in real time, path planning and the like; and the road width is not estimated and processed on the basis of the vehicle track when real-time map rendering is carried out by carrying out filtering smoothing and other processing on the basis of real collected data, so that the phenomena of inaccurate road data presentation, easy mold penetration and the like are avoided, and the obtained road network data has high accuracy.
Referring to fig. 1, a flowchart illustrating steps of an embodiment of a road network data processing method according to the present invention is shown, and specifically, the method may include the following steps:
in the embodiment of the invention, from the vehicle pose, the processing such as filtering, smoothing and the like is completed based on the road network data collected by the sensor, and a continuous and complete road network structure is generated for real-time rendering, path planning and the like, so that the method has better expansibility and universality. Specifically, the vehicle motion track capable of reflecting the vehicle pose and the road network perception data acquired by the sensor in real time can be acquired, so that the road network data can be processed from the starting point of the vehicle pose.
In order to ensure that the acquired road network information can be analyzed, rendered and the like in real time when the vehicle moves, the motion track of the vehicle needs to be segmented, so that the data volume of single processing can be reduced on one hand, and support can be provided for subsequent data classification on the other hand.
The segmentation of the vehicle motion trajectory can be realized by determining trajectory nodes based on the vehicle motion trajectory. The track nodes can be generally selected to be areas with dense information at intersections and the like, the motion track of the vehicle can be divided into a plurality of sections of straight lines for processing, and node information corresponding to the track nodes can be added to the track by monitoring whether the steering angle of the vehicle exceeds a threshold value, whether the acquired data amount exceeds an extreme value and the like.
in an embodiment of the present invention, in the process of processing the collected road network information, it is necessary to perform filtering, smoothing, and the like on the collected road network information to generate a continuous and complete road network structure, which is convenient for performing real-time rendering, path planning, and the like in the following. The vehicle may be as shown in the left diagram of fig. 2 when acquiring the road network information, and the road network data and the trajectory data collected by most of the vehicle-mounted sensors at present are point data, such as the road network data, the interference data and the trajectory data corresponding to the point data shown in the right diagram of fig. 2. The vehicle trajectory data may be obtained by an internal sensor, and the road network data may be detected by an external sensor such as a camera or a radar, which may include many interference factors, that is, the road network information collected by the vehicle may include the vehicle trajectory data and road network perception data including road network data points and interference data points.
In the process of processing road network information, actually, the trajectory data of a vehicle and the road network perception data comprising road network data points and interference data points are processed, at the moment, in order to facilitate the construction of a road network data structure, the relevant information of a road is determined based on the motion trajectory of the vehicle, a road trajectory curve is obtained by fitting the current vehicle position based on the determined trajectory nodes, and a road edge curve is obtained by fitting the road network perception data, so that the road can be planned by the road trajectory curve and the road edge curve obtained by fitting in the subsequent process.
After the road track curve and the road edge curve are obtained through fitting, corresponding logic features such as road edge features, noise features, track features and the like can be added to data points contained in the fitted curve, a continuous and complete data structure for road network data can be formed based on the added logic features, and point cloud data without logic relations can be processed based on the added logic features, so that the road network data passing through the inspection can be applied more quickly and conveniently, and the method has better universality and expansibility.
In the embodiment of the invention, the track nodes are determined based on the motion track of the vehicle, the determined track nodes are fitted with the current vehicle position to obtain the road track curve, the road network informing data collected in real time are fitted to obtain the road edge curve, and then the corresponding logic characteristics are respectively added to the data points contained in the fitted road track curve and the road edge curve to form the road network data structure to obtain the road network data, so that the road network data passing the inspection can be used in the following process. The road network data collected by the sensor are filtered and smoothed from the vehicle pose, a continuous and complete road network structure is generated to perform real-time rendering, path planning and the like, the road network data are collected in real time in the intelligent driving process, the processed road network data are presented to vehicle users, the vehicle users can sense the surrounding environment in time, and the road network data processing system has good expansibility and universality.
Referring to fig. 3, a flowchart illustrating steps of another embodiment of a road network data processing method according to the present invention is shown, which may specifically include the following steps:
in the embodiment of the invention, the road network information acquired in the figure 2 is processed by means of filtering, classifying, smoothing, fitting and the like to obtain road network data which can be conveniently rendered in real time and can be easily called by other interfaces, and the rapid processing and the subsequent real-time use of the sensor are realized in environments with less dynamic interference and simple scene characteristics, such as parking garages, shopping malls and the like. Specifically, referring to fig. 4, which shows an implementation process diagram of road network data processing provided by the embodiment of the present invention, a processing process of road network information may be composed of the following parts: the method is realized by obtaining vehicle motion tracks and road network perception data, determining track nodes, fitting road tracks based on a least square method, processing data based on a two-dimensional ideal high-pass filter, classifying and sequencing road network data, linearly fitting road edges based on linear regression, forming a road network data structure, checking data logic and storing data to full-field data.
In an embodiment of the present invention, after determining the track node based on the vehicle motion track, i.e. segmenting the vehicle motion track, the road track curve may be fitted. At the moment, the current vehicle position and the last track node set in the vehicle motion track of the current vehicle position can be obtained, and then a track curve between the last track node and the current vehicle position is fitted based on a least square method to obtain a road track curve.
Specifically, a track node is set on the vehicle motion track, and in a track curve between the last set track node and the current vehicle position, the curvature change is not large, and at this time, a high-order polynomial of 3 to 5 orders can be used to perform approximate fitting on the track curve between the two nodes, which can be expressed as:
f(x)≈g(x)＝k0+k1(x-x0)+k2(x-x0)2+…+kn(x-x0)n
in the formula, f may be an actual curve of the road track, g may be a fitting curve of the road track, k is a parameter sequence, n is a fitting number, where x may be a coordinate point of the current vehicle position, and x is generally used in a subsequent application process0This value is negligible in the embodiment of the present invention.
Taylor expansion is performed on f, let x0Taking the sum of squares of f and g interpolation as an accumulated error as zero value, and recording as loss to judge the fitting accuracy, wherein an expression of the accumulated error is as follows:
wherein x for the fitted curve giAnd y for the actual curve fiIn order to acquire the acquired data points of the vehicle motion track, in order to optimize the fitting degree at this time, the partial derivative of loss on each parameter can be specifically made to be 0, namely
After the partial derivative results are collated, an equation set for the fitted road track curve can be obtained, and the fitted road track curve can be shown in fig. 5, where the equation set specifically can be:
K＝X\Y
wherein K is [ K ]0,k1,k2,…,kn]-1，
Where K denotes the weight of the argument X, XiAnd yiAnd (3) acquiring the nth or m data points represented by n and m for acquiring the acquired vehicle motion track data points.
302, fitting the distribution positions of road network data points in the road network perception data relative to the vehicle motion trail to obtain a road boundary curve;
in order to facilitate the construction of the road network data structure, the relevant information of the road needs to be determined, and besides the road track curve, a road boundary curve can be fitted. Specifically, the road network sensing data can be classified and sorted after the collected road network sensing data are denoised, and then a road boundary curve is fitted.
In an embodiment of the present invention, road network data points and interference data points may be determined from road network sensing data based on denoising processing of the road network sensing data, when classifying and sorting the road network data, coordinates of the road network data may be obtained, the coordinates of the road network data points are substituted into a road trajectory curve obtained by fitting, distribution positions of the road network data points relative to a vehicle motion trajectory are determined, and then road boundary curves are obtained by fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trajectory.
In practical applications, data other than trajectory data may be data obtained by fusing multiple sensors, and there are situations of data duplication, data interference, and the like, and in order to ensure reliability of the processed road network data, it is necessary to perform denoising processing on interference data contained in the road network data, that is, to distinguish road network data points and interference data points from road network perception data.
The way to distinguish the road network perception data can be determined by whether other data points exist in the search area which is near a certain data point and has the distance of the search radius of the data point.
Specifically, the data point search radius may be obtained at this time, and if there are other data points in the search area whose distance from a certain data point is the data point search radius, it is determined that the data point and the other data points are road network data points; and if no other data point exists in the searching area for the distance of another data point, which is the searching radius of the data point, determining the data point as an interference data point.
The data point retrieval radius can be determined by obtaining a plurality of adjacent data points from road network perception data, calculating a mean value of distances between the plurality of adjacent data points, and processing the calculated mean value by adopting a preset scale factor.
Illustratively, n adjacent road network data points can be taken from the data set, n-1 adjacent path point distances are calculated, and the following results are obtained by averaging the distances:
to increase the robustness of the algorithm and reduce the probability of missing a true value point, a scale factor k may be given at this time such that the data point search radius is
In this case, the distance in the vicinity of an arbitrary data point A is set to
Finding other data points in the retrieval area, and the system can regard the data point A and the retrieved data point as target points, namely road network data points; if no other data except the point A is found in the search area, the point is considered as an interference point. It should be noted that, because the track nodes are determined in advance, the denoising process can be quickly realized only by traversing the road network data obtained from the track nodes.
In the process of classifying and sequencing road network data, the reliability of road network data points subjected to denoising processing is high, and the road network data points can be used for representing the relative pose information of roads and vehicles. However, the point data information is still not suitable for direct use of interfaces such as rendering and path planning, and in order to better construct the road network information, the data set needs to be classified and sorted first.
Specifically, for any road network data point i, coordinates of the road network data are obtained, and the coordinates can be expressed as xi, yi (approximately, a road and a vehicle are located in one plane, and influence caused by a z-axis position difference is ignored), at this time, the coordinates of the road network data point can be substituted into the fitted road track curve, so as to obtain the following expression:
direction＝k0+k1xi+…+knxi n-yi
then, the distribution position of the road network data points relative to the motion trail of the vehicle can be determined through the expression, wherein if the direction is greater than 0, the road network data points are positioned on the upper side of the motion trail of the vehicle; if the direction is equal to 0, the road network data point is on the motion track of the vehicle; if the direction is less than 0, the road network data point is positioned at the lower side of the motion trail of the vehicle. By the method, the classification of road network data can be realized, the roads on the two sides where the vehicle moves forward are divided, and the data points are sequenced according to the time sequence of data collection so as to represent the forward direction of the roads.
In the process of fitting the road boundary curve, edge data including road network data points located on a preset road boundary can be obtained based on the distribution positions of the road network data points relative to the vehicle motion track, and then the edge data is fitted based on a least square method to obtain a linear regression equation for the road boundary curve, so that the fitting of the road boundary is realized.
Specifically, the least square method may also be used to fit two edge data of the road (which refers to the road network data sets located on both sides of the driving track). It should be noted that the accumulated error between the actual curve and the fitted curve of the road edge can be used as an evaluation index of the interference factor, and the fluctuation of the error value can be used as a quantization index of the denoising processing effect, which is not described in the embodiment of the present invention.
Similar to the derivation process in the road trajectory curve fitting process, which is not repeated herein, the road boundary curve obtained by fitting may be as shown in fig. 5, and at this time, the linear regression equation of the obtained road boundary (for a single road edge) is:
f(x)＝k0+k1x+k2x2+…+knxn+ε
g(x)＝k0+k1x+k2x2+…+knxn
in the formula, f is an actual track, g is a fitting track, an uncertain item epsilon can represent the interference degree, and the larger epsilon is, the stronger interference in a scene is; the amplitude of change Δ ε of the degree of interference between adjacent roadsi＝|εi|-|εi-1And l, representing the denoising effect, wherein the smaller the denoising effect, the more effective the denoising treatment is. The road edge is obtained based on the obtained fitting track g during planning, the actual track f can be an ideal road edge, and the uncertainty can be used for evaluating fluctuation between a real value and an ideal value.
after the road track curve and the road edge curve are obtained through fitting, corresponding logic features such as road edge features, noise features, track features and the like can be added to data points contained in the fitted curve, a continuous and complete data structure for road network data can be formed based on the added logic features, and point cloud data without logic relations can be processed based on the added logic features, so that the road network data passing through the inspection can be applied more quickly and conveniently, and the method has better universality and expansibility.
And step 304, analyzing the data under the same logic characteristics, and storing the road network data passing the inspection for subsequent use.
After the road network data structure is generated and the road network data is obtained, the obtained road network data can be subjected to a process of data logical check and data storage to full field data.
In the process of data logic inspection, data under the same logic characteristic can be analyzed, and if the fitted road track curve and the line segment in the road boundary curve are not crossed or data points are not missing, the obtained road network data is inspected to pass. As shown in fig. 4, data under the same characteristic is analyzed, if the fitted line segment and other line segments have a crossing phenomenon, it indicates that data fitting is wrong, secondary fitting is fed back, and the data is prompted to a user; if the fitted line segment has information loss, if the line segment is broken, returning to the denoising step possibly because of excessive denoising treatment, optimizing parameters, and carrying out secondary filtering; if the above situations do not occur and the processed data still has abnormity, the data collection part may have problems and the system prompts the user.
In the process from data storage to full field data, if the whole process is not abnormal, the road network data processing can be completed, and the road network data passing the inspection can be stored, so that the road network data passing the inspection can be used in real time by calling a preset rendering program or a preset interface.
In the embodiment of the invention, from the vehicle position and posture, the road network data collected by the sensor are filtered, smoothed and the like, and a continuous and complete road network structure is generated to perform real-time rendering, path planning and the like, so that the road network data are collected in real time in the intelligent driving process, the processed road network data are presented to the vehicle users, the vehicle users can timely perceive the surrounding environment, and the method has better expansibility and universality.
It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
Referring to fig. 6, a block diagram of a road network data processing apparatus according to an embodiment of the present invention is shown, and may specifically include the following modules:
the track node determining module 601 is configured to obtain a vehicle motion track and road network perception data acquired in real time, and determine a track node based on the vehicle motion track;
a curve fitting module 602, configured to fit the track node with the current vehicle position to obtain a road track curve, and fit the road network sensing data to obtain a road boundary curve;
a road network data generating module 603, configured to add corresponding logical features to data points included in the road track curve and the road boundary curve, generate a road network data structure, and obtain road network data, so as to use the road network data after passing the inspection.
In one embodiment of the invention, the curve fitting module 602 may include the following sub-modules:
the track node determination submodule is used for acquiring the current vehicle position and a previous track node set by the current vehicle position in the vehicle motion track;
and the road track curve fitting submodule is used for fitting the track curve between the previous track node and the current vehicle position based on a least square method to obtain the road track curve.
In one embodiment of the invention, the curve fitting module 602 may include the following sub-modules:
the data point determining submodule is used for determining road network data points and interference data points from the road network perception data;
the distribution position determining submodule is used for acquiring the coordinates of the road network data, substituting the coordinates of the road network data points into the fitted road track curve and determining the distribution positions of the road network data points relative to the vehicle motion track;
and the road edge curve fitting submodule is used for fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trail to obtain a road edge curve.
In one embodiment of the present invention, the data point determination submodule may include the following units:
a retrieval radius obtaining unit for obtaining a data point retrieval radius;
the road network data point determining unit is used for determining that a certain data point is a road network data point when other data points exist in a retrieval area with a data point distance as a data point retrieval radius;
and the interference data point determining unit is used for determining the data point as an interference data point when no other data point exists in the searching area for the distance of another data point, wherein the distance of the other data point is the searching radius of the data point.
In an embodiment of the present invention, the retrieval radius obtaining unit may include the following sub-units:
and the retrieval radius determining subunit is used for acquiring a plurality of adjacent data points from the road network perception data, calculating a mean value of the distances between the plurality of adjacent data points, and processing the calculated mean value by adopting a preset scale factor to obtain the data point retrieval radius.
In one embodiment of the invention, the road-edge curve fitting submodule may include the following elements:
the edge data acquisition unit is used for obtaining edge data containing road network data points positioned on a preset road boundary based on the distribution positions of the road network data points relative to the vehicle motion trail;
and the road edge curve fitting unit is used for fitting the edge data based on a least square method to obtain a linear regression equation aiming at the road edge curve and realize the fitting of the road boundary.
In one embodiment of the invention, the response logic features added by the contained data points comprise track features and road edge features; the road network data generation module 603 may include the following sub-modules:
and the road network data structure forming submodule is used for adding a track characteristic to the track data points contained in the road track curve and adding a road edge characteristic to the road network data points contained in the road edge curve to form a road network data structure.
In an embodiment of the present invention, after generating the road network data structure and obtaining the road network data, the apparatus may further include the following modules:
the road network data inspection module is used for analyzing data under the same logic characteristic, and if the fitted road track curve and the line segment in the road edge curve are not crossed or data point loss does not occur, the obtained road network data is inspected to pass;
and the road network data using module is used for storing the road network data passing the inspection so as to use the road network data passing the inspection in real time by calling a preset rendering program or a preset interface.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
An embodiment of the present invention further provides a vehicle, including:
the road network data processing method comprises a road network data processing device, a processor, a memory and a computer program which is stored on the memory and can run on the processor, wherein when the computer program is executed by the processor, each process of the road network data processing method embodiment is realized, the same technical effect can be achieved, and no repeated description is provided for avoiding repetition.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when being executed by a processor, the computer program implements each process of the road network data processing method embodiment, and can achieve the same technical effect, and in order to avoid repetition, details are not repeated here.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all changes and modifications that fall within the true scope of the embodiments of the present invention.
Finally, it should also be noted that, in this document, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal 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 terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The method for processing road network data and the device for processing road network data provided by the present invention are described in detail above, and a specific example is applied in the text to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (11)
1. A method for processing road network data, the method comprising:
the method comprises the steps of obtaining a vehicle motion track and road network perception data collected in real time, and determining track nodes based on the vehicle motion track;
fitting according to the track node and the current vehicle position to obtain a road track curve, and fitting the road network perception data to obtain a road edge curve;
and adding corresponding logic characteristics to data points contained in the road track curve and the road edge curve to generate a road network data structure and obtain road network data so as to use the road network data after passing the inspection.
2. The method of claim 1, wherein fitting the trajectory node to a current vehicle position to obtain a road trajectory curve comprises:
acquiring a current vehicle position and a last track node set by the current vehicle position in a vehicle motion track;
and fitting a track curve between the previous track node and the current vehicle position based on a least square method to obtain a road track curve.
3. The method according to claim 1, wherein said fitting said road network perception data to obtain a road edge curve comprises:
determining road network data points and interference data points from the road network perception data;
acquiring coordinates of the road network data, substituting the coordinates of the road network data points into the fitted road track curve, and determining the distribution positions of the road network data points relative to the vehicle motion track;
and fitting the road network data points according to the distribution positions of the road network data points relative to the vehicle motion trail to obtain a road edge curve.
4. The method of claim 3, wherein said determining road network data points and interference data points from said road network perceptual data comprises:
acquiring a data point retrieval radius, and if other data points exist in a retrieval area with a certain data point distance as the data point retrieval radius, determining the data point and the other data points as road network data points;
and if no other data point exists in the searching area for the distance of another data point, which is the searching radius of the data point, determining the data point as an interference data point.
5. The method of claim 4, wherein obtaining the data point search radius comprises:
and acquiring a plurality of adjacent data points from the road network perception data, calculating a mean value of the distances between the plurality of adjacent data points, and processing the calculated mean value by adopting a preset scale factor to obtain a data point retrieval radius.
6. The method according to claim 3, 4 or 5, wherein said fitting said road network data points to obtain a road edge curve according to distribution positions of said road network data points relative to said vehicle motion trajectory comprises:
obtaining edge data including road network data points positioned on a preset road boundary based on the distribution positions of the road network data points relative to the vehicle motion trail;
and fitting the edge data based on a least square method to obtain a linear regression equation aiming at a road edge curve, so as to realize the fitting of the road boundary.
7. The method of claim 1, wherein the response logic features added by the included data points include trajectory features and road edge features; the generating a road network data structure and obtaining road network data by adding corresponding logical features to data points included in the road track curve and the road edge curve includes:
and adding track features to the track data points contained in the road track curve and adding road edge features to the road network data points contained in the road edge curve to form a road network data structure.
8. The method of claim 1 or 7, further comprising, after generating the road network data structure and obtaining the road network data:
analyzing the data under the same logic characteristic, and if the fitted road track curve and the line segment in the road edge curve are not crossed or data point loss is not generated, checking the obtained road network data to pass;
and storing the road network data passing the inspection so as to use the road network data passing the inspection in real time by calling a preset rendering program or a preset interface.
9. An apparatus for processing road network data, the apparatus comprising:
the system comprises a track node determination module, a road network sensing module and a road network sensing module, wherein the track node determination module is used for acquiring a vehicle motion track and road network sensing data acquired in real time and determining track nodes based on the vehicle motion track;
the curve fitting module is used for fitting the track node with the current vehicle position to obtain a road track curve and fitting the road network perception data to obtain a road edge curve;
and the road network data generation module is used for generating a road network data structure and obtaining road network data by adding corresponding logic characteristics to data points contained in the road track curve and the road edge curve so as to use the road network data after passing the inspection.
10. A vehicle, characterized by comprising: processing device, processor, memory of road network data according to claim 9 and computer program stored on said memory and capable of running on said processor, said computer program when executed by said processor implementing the steps of the method for processing road network data according to any of claims 1-8.
11. A computer-readable storage medium, characterized in that said computer-readable storage medium stores thereon a computer program, which when executed by a processor implements the steps of the method for processing road network data according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111389644.0A CN114090908A (en) | 2021-11-19 | 2021-11-19 | Road network data processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111389644.0A CN114090908A (en) | 2021-11-19 | 2021-11-19 | Road network data processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114090908A true CN114090908A (en) | 2022-02-25 |
Family
ID=80302986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111389644.0A Pending CN114090908A (en) | 2021-11-19 | 2021-11-19 | Road network data processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114090908A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116182862A (en) * | 2022-12-30 | 2023-05-30 | 广州小鹏自动驾驶科技有限公司 | Road boundary determination method, device, electronic equipment and storage medium |
-
2021
- 2021-11-19 CN CN202111389644.0A patent/CN114090908A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116182862A (en) * | 2022-12-30 | 2023-05-30 | 广州小鹏自动驾驶科技有限公司 | Road boundary determination method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108399218B (en) | Automatic driving vehicle positioning based on Walsh kernel projection technology | |
| CN109813328B (en) | Driving path planning method and device and vehicle | |
| CN105957342B (en) | Track grade road plotting method and system based on crowdsourcing space-time big data | |
| US9076047B2 (en) | System and method for recognizing parking space line markings for vehicle | |
| CN108629231B (en) | Obstacle detection method, apparatus, device and storage medium | |
| WO2017203509A1 (en) | Method of predicting traffic collisions and system thereof | |
| WO2016086792A1 (en) | Driving behavior analysis method and device | |
| WO2006090735A1 (en) | Object recognizing apparatus and object recognizing method using epipolar geometry | |
| US20220398856A1 (en) | Method for reconstruction of a feature in an environmental scene of a road | |
| CN105373689A (en) | Method and device for determining road curvature in electronic map | |
| CN108645375B (en) | Rapid vehicle distance measurement optimization method for vehicle-mounted binocular system | |
| WO2019104866A1 (en) | Method and device for drawing region outline and computer readable storage medium | |
| CN113011285B (en) | Lane line detection method and device, automatic driving vehicle and readable storage medium | |
| US20200180646A1 (en) | Sensor fusion target prediction device and method for vehicles and vehicle including the device | |
| CN112880694A (en) | Method for determining the position of a vehicle | |
| CN111351499A (en) | Path identification method and device, computer equipment and computer readable storage medium | |
| Revilloud et al. | A lane marker estimation method for improving lane detection | |
| CN114090908A (en) | Road network data processing method and device | |
| CN110413942B (en) | Lane line equation screening method and screening module thereof | |
| CN116413740B (en) | Laser radar point cloud ground detection method and device | |
| JP2020006788A (en) | Construction limit determination device | |
| JP2013069045A (en) | Image recognition device, image recognition method, and image recognition program | |
| CN117409569A (en) | Intelligent traffic system and method based on big data | |
| CN117237911A (en) | Image-based dynamic obstacle rapid detection method and system | |
| CN116563341A (en) | Visual positioning and mapping method for processing dynamic object in complex environment |
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 |