CN112462368B - Obstacle detection method and device, vehicle and storage medium - Google Patents
Obstacle detection method and device, vehicle and storage medium Download PDFInfo
- Publication number
- CN112462368B CN112462368B CN202011339996.0A CN202011339996A CN112462368B CN 112462368 B CN112462368 B CN 112462368B CN 202011339996 A CN202011339996 A CN 202011339996A CN 112462368 B CN112462368 B CN 112462368B
- Authority
- CN
- China
- Prior art keywords
- attribute
- obstacle
- data
- determining
- accurate
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
- G01S2013/9322—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles using additional data, e.g. driver condition, road state or weather data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses an obstacle detection method, an obstacle detection device, a vehicle and a storage medium. The method comprises the following steps: acquiring original radar data of a candidate obstacle, and determining each attribute data of the candidate obstacle by combining vehicle state data; determining attribute accurate values corresponding to the attribute data, and determining the detection reliability of the candidate obstacle according to the attribute accurate values; and when the detection reliability is greater than a preset reliability threshold value, determining the candidate obstacle as a target obstacle. The method solves the problem of low accuracy of the obstacle information returned by part of radars, comprehensively analyzes whether the detected obstacle information is real and reliable according to different attribute data of the obstacles detected by the radars, and achieves the effect of improving the accuracy of the obstacle information detected by the radars.
Description
Technical Field
The embodiment of the invention relates to the technical field of data processing, in particular to a method and a device for detecting obstacles, a vehicle and a storage medium.
Background
In the field of unmanned driving, accurate and comprehensive detection of obstacles around a vehicle is crucial to safe driving of the vehicle, and collection of environmental information around the vehicle mainly depends on various vehicle sensors.
Laser radar is very outstanding at measuring accuracy and the ability of surveying the aspect such as human body, and position, size, the gesture etc. of the near object of vehicle of the accurate determination can be, however, because laser radar surveys through the transmission beam, receives environmental impact great, and the light beam just can not normal use after sheltering from, consequently can't open in bad weather such as sleet haze or sand and dust storm. In terms of interference immunity, millimeter wave radar is superior. The millimeter wave radar is not discarded by the market due to the fact that the millimeter wave radar can penetrate dust fog and rain and snow, is not influenced by severe weather, can work all weather, is long in detection distance, is not sensitive to static objects and non-metal objects, and can influence subsequent processing of an automatic driving sensing system due to the fact that barrier information returned by the millimeter wave radar is not very accurate under many conditions. Therefore, it is necessary to recognize the reliability of the obstacle information collected by the millimeter wave radar.
Disclosure of Invention
The invention provides an obstacle detection method, an obstacle detection device, a vehicle and a storage medium, which are used for realizing accurate detection of obstacles around the vehicle.
In a first aspect, an embodiment of the present invention provides an obstacle detection method, including:
acquiring original radar data of a candidate obstacle, and determining each attribute data of the candidate obstacle by combining vehicle state data;
determining attribute accurate values corresponding to the attribute data, and determining the detection reliability of the candidate obstacle according to the attribute accurate values;
and when the detection reliability is larger than a preset reliability threshold value, determining the candidate obstacle as a target obstacle.
Optionally, the determining an accurate attribute value corresponding to each attribute data includes:
and aiming at each attribute data, acquiring a corresponding attribute accurate value mapping table, searching a target accurate value matched with the attribute data in the attribute accurate value mapping table, and determining the target accurate value as the attribute accurate value corresponding to the attribute data.
Optionally, the step of determining the attribute accurate value mapping table includes:
dividing values of the target attribute of the obstacle to form at least two attribute value intervals;
acquiring a preset number of training radar data, and dividing each training radar data into corresponding radar data sets according to the attribute value interval;
for each radar data set, determining an accurate data amount and a total data amount of the training radar data, and determining a ratio of the accurate data amount to the total data amount as a training accurate value;
and establishing a mapping relation between each attribute value interval and the corresponding training accurate value to obtain an attribute accurate value mapping table of the obstacle target attribute.
Optionally, before establishing a mapping relationship between each attribute value interval and the corresponding training accurate value, the method further includes: and fitting each training accurate value.
Optionally, the determining the detection reliability of the candidate obstacle according to each of the attribute accuracy values includes:
and acquiring preset attribute weight, and performing weighted summation on each attribute accurate value to obtain the detection reliability of the candidate obstacle.
Optionally, the attribute data includes obstacle position, obstacle distance, obstacle speed, obstacle acceleration, obstacle direction, obstacle size and/or obstacle category.
Optionally, the vehicle state data includes vehicle speed data, direction of movement data and/or yaw angle data.
In a second aspect, an embodiment of the present invention further provides an obstacle detection apparatus, including:
the attribute data determining module is used for acquiring original radar data of the candidate obstacle and determining each attribute data of the candidate obstacle by combining vehicle state data;
the detection reliability determining module is used for determining attribute accurate values corresponding to the attribute data and determining the detection reliability of the candidate obstacle according to the attribute accurate values;
and the obstacle determining module is used for determining the candidate obstacle as a target obstacle when the detection reliability is greater than a preset reliability threshold value.
In a third aspect, an embodiment of the present invention further provides a vehicle, including:
one or more controllers;
a memory for storing one or more programs;
the radar is used for acquiring barrier information;
when executed by the one or more controllers, cause the one or more controllers to implement the method of obstacle detection according to any embodiment of the invention.
In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the method for obstacle detection according to any of the embodiments of the present invention.
According to the method, original radar data of the candidate obstacles are obtained, each attribute data of the candidate obstacles is determined by combining vehicle state data, the detection reliability of the candidate obstacles is determined according to the attribute accuracy value corresponding to each attribute data, when the detection reliability is larger than a preset reliability threshold value, the candidate obstacles are determined as target obstacles, the problem that the accuracy of obstacle information returned by partial radars is low is solved, whether the detected obstacle information is real and reliable is comprehensively analyzed according to different attribute data of the obstacles detected by the radars, and the effect of improving the accuracy of the obstacle information detected by the radars is achieved.
Drawings
Fig. 1 is a flowchart of an obstacle detection method according to an embodiment of the present invention;
fig. 2 is a flowchart of an obstacle detection method according to a second embodiment of the present invention;
fig. 3 is a schematic diagram illustrating attribute data determined in an obstacle detection method according to a second embodiment of the present invention;
fig. 4 is a flowchart of determining an attribute accurate value mapping table in an obstacle detection method according to a second embodiment of the present invention;
fig. 5 is a schematic diagram illustrating the accuracy of radar data determined in an obstacle detection method according to a second embodiment of the present invention;
fig. 6 is a schematic diagram of a layout of a radar apparatus in an obstacle detection method according to a second embodiment of the present invention;
fig. 7 is a schematic diagram illustrating a mapping relationship between a determined attribute value and an accurate value in the obstacle detection method according to the second embodiment of the present invention;
fig. 8 is a schematic diagram illustrating a mapping relationship between an attribute value and an accurate value determined in the obstacle detection method according to the second embodiment of the present invention;
fig. 9 is a schematic diagram illustrating a mapping relationship between a determined attribute value and an accurate value in an obstacle detection method according to a second embodiment of the present invention;
fig. 10 is a block diagram of an obstacle detection device according to a third embodiment of the present invention;
fig. 11 is a block diagram of a vehicle according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only a part of the structures related to the present invention, not all of the structures, are shown in the drawings, and furthermore, embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
Example one
Fig. 1 is a flowchart of an obstacle detection method according to an embodiment of the present invention, where the embodiment is applicable to a case of detecting an obstacle around a vehicle, and the method may be executed by an obstacle detection device, and the device may be implemented by software and/or hardware.
As shown in fig. 1, the method specifically includes the following steps:
and step 110, acquiring original radar data of the candidate obstacles, and determining each attribute data of the candidate obstacles by combining the vehicle state data.
Raw radar data is understood to be raw data information of surrounding objects detected by the radar device. The vehicle state data can be understood as current driving state data of the vehicle. Candidate obstacles may be understood as obstacles that may be present as a preliminary determination based on raw radar data collected by the radar device. Attribute data may be understood as data values that are used to characterize various attributes of an obstacle.
Optionally, the vehicle state data may include vehicle speed data, direction of motion data, and/or yaw angle data.
Optionally, the attribute data may include obstacle position, obstacle distance, obstacle velocity, obstacle acceleration, obstacle direction, obstacle size, and/or obstacle category.
At present, sensors for autonomous vehicles are mainly laser radar, millimeter wave radar, cameras, and the like. The laser radar has high ranging precision, strong directivity and quick response time, but is easily influenced by weather and cannot work in rain, fog and sand weather; the millimeter wave radar can work all weather, has long detection distance, but has high difficulty in identifying targets and is not sensitive to static objects and non-metal objects; the camera can obtain image information, but is influenced by the field of view, and the accuracy of obtaining distance information is far lower than that of a radar. The detection range, penetrability, resolution, field angle, environmental applicability and the like of the sensors are different from each other, and generally, the use condition and the detection capability of a single sensor are obviously insufficient. In this embodiment, a millimeter wave radar is mainly taken as an example, but the method in each embodiment of the present invention is not only suitable for detecting an obstacle by the millimeter wave radar, but also can be used for other radar devices, such as a laser radar, and of course, the method can also be used for a visual sensor, such as a camera.
Specifically, the original radar data collected by the millimeter wave radar can be continuously read from the corresponding millimeter wave radar interface through the millimeter wave radar drive, and the original radar data in a period of time is packed into a frame according to the requirement and sent to the corresponding obstacle detection device. The obstacle detection device can analyze the received original radar data frame to obtain information such as the ID of a candidate obstacle, the horizontal and vertical coordinate values of the obstacle, the horizontal and vertical speed of the obstacle, a coordinate conversion matrix, the type of the obstacle and the like according to the data protocol of the millimeter wave radar. Meanwhile, the obstacle detection device can acquire vehicle state data such as vehicle speed data, movement direction data and yaw angle data from other sensors of the vehicle. In the process of detecting the obstacle by the radar, the vehicle also moves in real time, so that each attribute data of the candidate obstacle needs to be determined according to original radar data and vehicle state data, the actual space position of the obstacle can be obtained through coordinate conversion, the size of an obstacle detection frame is adjusted according to the type and the speed of the obstacle, and the like, and the attribute data of the obstacle position, the obstacle distance, the obstacle speed, the obstacle acceleration, the obstacle direction, the obstacle size, the obstacle category and the like can be finally obtained through related calculation. For example, if the radar-detected lateral-to-longitudinal velocity of the candidate obstacle is relative to the velocity of the vehicle, then the raw obstacle-candidate lateral-to-longitudinal velocity may be converted in conjunction with the vehicle speed data into the actual velocity of the candidate obstacle relative to the ground.
And step 120, determining an attribute accurate value corresponding to each attribute data, and determining the detection reliability of the candidate obstacle according to each attribute accurate value.
The attribute accuracy value may be understood as an accuracy rate of the candidate obstacle information determined according to certain attribute data of the candidate obstacle. The detection reliability may be understood as a reliability degree of the candidate obstacle information detected by the radar device.
Specifically, a corresponding relation table of attribute values corresponding to different attributes of the obstacle and detection accuracy may be preset, and after attribute data of each attribute of the candidate obstacle is determined, different relation tables may be searched to determine an attribute accurate value corresponding to each attribute data. After each attribute of the candidate obstacle has a corresponding attribute accuracy value, the attribute accuracy values may be combined to calculate the detection reliability of the candidate obstacle, for example, an average value of the attribute accuracy values is calculated, and the average value is used as the detection reliability of the candidate obstacle. It can be understood that the preset corresponding relation table of the attribute value of the obstacle and the detection accuracy may be formed through a real vehicle test for radar devices of different models, and the weight values of different attributes may also be obtained through a large number of tests, for example, a certain radar device is sensitive to the movement of an object, so that when the detection reliability of a candidate obstacle is calculated, the attribute accuracy value corresponding to the speed attribute of the candidate obstacle may be given a larger weight value.
And step 130, when the detection reliability is greater than a preset reliability threshold value, determining the candidate obstacle as the target obstacle.
The preset confidence threshold value can be understood as a preset calibration value for judging whether the obstacle information is accurate or not.
Specifically, a preset confidence threshold may be obtained, after the detection confidence of the candidate obstacle is determined, the detection confidence is compared with the preset confidence threshold, and if the detection confidence is greater than the preset confidence threshold, it may be determined that the detected information of the candidate obstacle is accurate and may be used for subsequent processing of the automatic driving sensing system, so that the candidate obstacle is determined as the target obstacle. If the detection reliability is less than or equal to the preset reliability threshold, the detected information of the candidate obstacle can be considered to be inaccurate, and the candidate obstacle information can be ignored at the moment.
According to the technical scheme of the embodiment, the original radar data of the candidate obstacle is obtained, each attribute data of the candidate obstacle is determined by combining the vehicle state data, the detection reliability of the candidate obstacle is determined according to the attribute accurate value corresponding to each attribute data, when the detection reliability is larger than the preset reliability threshold value, the candidate obstacle is determined as the target obstacle, the problem that the accuracy of obstacle information returned by partial radars is low is solved, whether the detected obstacle information is real and reliable or not is comprehensively analyzed according to different attribute data of the obstacle detected by the radars, and the effect of improving the accuracy of the obstacle information detected by the radars is achieved.
Example two
Fig. 2 is a flowchart of an obstacle detection method according to a second embodiment of the present invention. In this embodiment, the obstacle detection method is further optimized based on the above embodiment.
As shown in fig. 2, the method specifically includes:
Specifically, the obstacle detection device may analyze the received raw radar data, and calculate attribute data corresponding to each attribute of the candidate obstacle in combination with vehicle state data acquired from other sensors of the vehicle.
Fig. 3 is a schematic diagram illustrating attribute data determined in an obstacle detection method according to a second embodiment of the present invention. As shown in fig. 3, taking the extraction of the candidate obstacle position attribute as an example, a polar coordinate system is set up with the vehicle center as the pole O and the vehicle central axis directly in front as the polar axis forward direction, the position data of the candidate obstacle in the raw radar data is represented by the polar angle θ from the candidate obstacle center point P to the vehicle centers O and P, and the actual position attribute data of the candidate obstacle can be calculated by combining the vehicle motion direction data and the like.
The attribute accurate value mapping table can be understood as a corresponding relation table of attribute values corresponding to various attributes of the obstacle and detection accuracy.
Specifically, an attribute accurate value mapping table corresponding to different attributes of the obstacle may be preset, the attribute accurate value mapping table may be obtained in advance according to a real vehicle test, after determining attribute data of each attribute of the candidate obstacle, the attribute accurate value mapping table corresponding to each attribute of the candidate obstacle may be searched, and a target accurate value having a mapping relationship with the attribute data may be determined as an attribute accurate value corresponding to the attribute data.
Optionally, fig. 4 is a flowchart of determining an attribute accurate value mapping table in the obstacle detection method according to the second embodiment of the present invention. As shown in fig. 4, the step of determining the attribute exact value mapping table may include:
Wherein, the target attribute can be understood as the obstacle attribute which will establish the attribute accurate value mapping table.
Specifically, the obstacle target attribute of the attribute accurate value mapping table to be established may be selected, and a possible value range of the target attribute may be divided into a plurality of attribute value intervals. For example, the detection distance of the vehicle radar is generally between 0.2 m and 3.0 m, then the distance of the candidate obstacle detected by the radar can be considered to be between 0.2 m and 3.0 m, and when the distance attribute of the obstacle is determined to be the target attribute and the accurate attribute value mapping table corresponding to the distance attribute is established, the distance attribute can be divided into 28 attribute value intervals according to the interval of 0.1 m. Particularly, as for the type attribute of the obstacle, the obstacle type can be divided in advance, and different flag bits are set for different types, as shown in the following table, different flag bit bits in the acquired radar data represent that different obstacle types are detected, each obstacle type can be determined into an attribute value interval, and the radar can detect how many obstacle types and can divide the value of the target attribute into how many attribute value intervals.
The training radar data can be understood as sample data used for determining the attribute accurate value mapping table, and the training radar data contains obstacle information detected by the radar device. It can be understood that training radar data can be obtained as much as possible, and the greater the number of training radar data is, the higher the accuracy of the determined attribute accuracy value mapping table is.
Specifically, according to the division of the attribute value intervals, training radar data with the value of the target attribute in an attribute value interval are divided into a radar data set, and if the training radar data are enough in quantity and the data are distributed comprehensively, each attribute value interval correspondingly forms a radar data set.
The accurate data amount can be understood as the amount of correct radar data in the training radar data, and the correct radar data means that the obstacle information detected by the radar device is basically consistent with the real information of the obstacle. The total amount of data may be understood as the total amount of training radar data contained in the radar data set.
Specifically, for each radar data set, the ratio of accurate training radar data therein may be calculated, and the calculated ratio may be determined as the training accurate value corresponding to the radar data set, that is, the training accurate value of the attribute value interval corresponding to the radar data set.
Fig. 5 is a schematic diagram illustrating the accuracy of radar data determined in an obstacle detection method according to a second embodiment of the present invention. Because the laser radar has high ranging precision and the defects that the laser radar is easily influenced by weather and the millimeter wave radar has low detection precision, in the embodiment, when the accurate value mapping table of each attribute of the millimeter wave radar for detecting the obstacle is determined, radar data collected by the laser radar can be used as standard data when the weather condition is good, and the accuracy of the radar data collected by the millimeter wave radar can be measured. As shown in fig. 5, the 5 × 5 grid in the figure is the obstacle map determined by the laser radar data, wherein the grids 51, 52 and 53 are the positions where the obstacles detected by the laser radar are located. The obstacle information detected by the millimeter wave radar may be projected onto a grid map of the laser radar, where the solid line area 54 represents the position of the millimeter wave radar obstacle. It can be seen that the total number of the grids occupied by the millimeter wave radar obstacle 54 is 6, and of the 6 grids, 3 grids are detected by the laser radar to have obstacles, so that it can be determined that the proportion of the accurate grids is 50%. If the accuracy threshold value is set to be 40% in advance, and the millimeter wave radar obstacle information is considered to be correct when the proportion of the accuracy grid exceeds the accuracy threshold value, the millimeter wave radar obstacle can be considered to be actually present at the moment.
Specifically, for each attribute value interval of the target attribute of the obstacle, a corresponding training accurate value can be determined through the steps, and after a mapping relation is established between each attribute value interval and the corresponding training accurate value, an attribute accurate value mapping table of the target attribute can be obtained.
Optionally, before establishing the mapping relationship between each attribute value interval and the corresponding training accurate value, the method may further include: and fitting each training accurate value.
Specifically, because some data may be inaccurate in the training radar data, a mapping relationship between each attribute value interval and the corresponding training accurate value may be established after each training accurate value is fitted.
For example, fig. 6 is a schematic diagram of a layout of a radar apparatus in an obstacle detection method according to a second embodiment of the present invention; fig. 7, fig. 8, and fig. 9 are schematic diagrams illustrating determining a mapping relationship between an attribute value and an accurate value in an obstacle detection method according to a second embodiment of the present invention. As shown in fig. 6, autonomous vehicles typically deploy millimeter wave radars directly in front of and directly behind the vehicle, with forward millimeter wave radars indicated at 61 and rearward millimeter wave radars indicated at 62. Taking the position attribute of the obstacle as an example, when a general obstacle is in the field angle range of the millimeter wave radar, the detection accuracy is high, and therefore the relationship between the detection accuracy of the millimeter wave radar and the position attribute of the obstacle may obey normal distribution, as shown in fig. 7, and therefore each training accurate value may be fitted by using a normal distribution function. For other attributes of the obstacle, different data distributions may exist, as shown in fig. 8, it may be more appropriate to use a clustering method to fit the training accurate values in the graph; similarly, if the representation of the data is linear, a linear fit can be used to find the relationship between the training accuracy value and the attribute value, as in the data distribution shown in fig. 9.
And step 230, acquiring preset attribute weights, and performing weighted summation on the accurate values of the attributes to obtain the detection reliability of the candidate obstacle.
The attribute weight can be understood as the importance degree corresponding to the attribute accuracy values of different attributes of the obstacle when determining the detection reliability.
Specifically, an attribute weight can be predetermined for different attributes of the obstacle through real vehicle testing, and when the detection reliability is determined, the attribute accurate values corresponding to the attributes are subjected to weighted summation to obtain the detection reliability of the candidate obstacle. Of course, the attribute weight corresponding to each attribute may also be all 1, that is, the accurate value of each attribute may be averaged, for example, there are n attributes in the candidate obstacle, and the corresponding accurate values of the attributes are P1，P2，P3，……，PnThen the confidence level of detection P of the candidate obstacle is equal to (P)1+P+…+Pn)/n。
And step 240, when the detection reliability is greater than a preset reliability threshold value, determining the candidate obstacle as a target obstacle.
The technical scheme of the embodiment includes that original radar data of candidate obstacles are obtained, each attribute data of the candidate obstacles is determined by combining vehicle state data, a corresponding attribute accurate value mapping table is obtained aiming at each attribute data, a target accurate value matched with the attribute data in the attribute accurate value mapping table is searched, the target accurate value is determined as the attribute accurate value corresponding to the attribute data, preset attribute weight is obtained, weighting and summing the accurate values of the attributes to obtain the detection reliability of the candidate obstacles, and when the detection reliability is greater than a preset reliability threshold, the candidate obstacles are determined as the target obstacles, the problem of low accuracy of part of obstacle information returned by the radar is solved, according to different attribute data of the obstacles detected by the radar, whether the detected obstacle information is real and reliable is comprehensively analyzed, and the effect of improving the accuracy of the obstacle information detected by the radar is achieved.
EXAMPLE III
The obstacle detection device provided by the embodiment of the invention can execute the obstacle detection method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. Fig. 10 is a block diagram of a structure of an obstacle detection apparatus according to a third embodiment of the present invention, and as shown in fig. 10, the apparatus includes: an attribute data determination module 310, a detection confidence determination module 320, and an obstacle determination module 330.
The attribute data determining module 310 is configured to obtain original radar data of the candidate obstacle, and determine attribute data of the candidate obstacle by combining vehicle state data.
And a detection reliability determining module 320, configured to determine an attribute accuracy value corresponding to each attribute data, and determine the detection reliability of the candidate obstacle according to each attribute accuracy value.
An obstacle determination module 330, configured to determine the candidate obstacle as a target obstacle when the detection reliability is greater than a preset reliability threshold.
According to the technical scheme of the embodiment, the original radar data of the candidate obstacle is obtained, each attribute data of the candidate obstacle is determined by combining the vehicle state data, the detection reliability of the candidate obstacle is determined according to the attribute accurate value corresponding to each attribute data, when the detection reliability is larger than the preset reliability threshold value, the candidate obstacle is determined as the target obstacle, the problem that the accuracy of obstacle information returned by partial radars is low is solved, whether the detected obstacle information is real and reliable or not is comprehensively analyzed according to different attribute data of the obstacle detected by the radars, and the effect of improving the accuracy of the obstacle information detected by the radars is achieved.
Optionally, the determining an accurate attribute value corresponding to each attribute data includes:
and aiming at each attribute data, acquiring a corresponding attribute accurate value mapping table, searching a target accurate value matched with the attribute data in the attribute accurate value mapping table, and determining the target accurate value as the attribute accurate value corresponding to the attribute data.
Optionally, the step of determining the attribute accurate value mapping table includes:
dividing values of the target attribute of the obstacle to form at least two attribute value intervals;
acquiring a preset number of training radar data, and dividing each training radar data into corresponding radar data sets according to the attribute value interval;
for each radar data set, determining an accurate data amount and a total data amount of the training radar data, and determining a ratio of the accurate data amount to the total data amount as a training accurate value;
and establishing a mapping relation between each attribute value interval and the corresponding training accurate value to obtain an attribute accurate value mapping table of the obstacle target attribute.
Optionally, before the mapping relationship between each attribute value interval and the corresponding training accurate value is established, the method further includes: and fitting each training accurate value.
Optionally, the determining the detection reliability of the candidate obstacle according to each of the attribute accuracy values includes:
and acquiring preset attribute weight, and performing weighted summation on each attribute accurate value to obtain the detection reliability of the candidate obstacle.
Optionally, the attribute data includes obstacle position, obstacle distance, obstacle speed, obstacle acceleration, obstacle direction, obstacle size, and/or obstacle category.
Optionally, the vehicle state data includes vehicle speed data, direction of movement data and/or yaw angle data.
The technical scheme of the embodiment includes that original radar data of candidate obstacles are obtained, each attribute data of the candidate obstacles is determined by combining vehicle state data, a corresponding attribute accurate value mapping table is obtained aiming at each attribute data, a target accurate value matched with the attribute data in the attribute accurate value mapping table is searched, the target accurate value is determined as the attribute accurate value corresponding to the attribute data, preset attribute weight is obtained, weighting and summing the accurate values of the attributes to obtain the detection reliability of the candidate obstacles, and when the detection reliability is greater than a preset reliability threshold, the candidate obstacles are determined as the target obstacles, the problem of low accuracy of the obstacle information returned by partial radar is solved, according to different attribute data of the obstacles detected by the radar, whether the detected obstacle information is real and reliable is comprehensively analyzed, and the effect of improving the accuracy of the obstacle information detected by the radar is achieved.
Example four
Fig. 11 is a block diagram of a vehicle according to a fourth embodiment of the present invention, as shown in fig. 11, the vehicle includes a controller 410, a memory 420, and a radar 430; the number of the controllers 410 may be one or more, and one controller 410 is illustrated in fig. 11; the controller 410, memory 420, and radar 430 in the vehicle may be connected by a bus or other means, as exemplified by the bus connection in fig. 11.
The memory 420 serves as a computer-readable storage medium, and may be used to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the obstacle detection method in the embodiment of the present invention (for example, the attribute data determination module 310, the detection reliability determination module 320, and the obstacle determination module 330 in the obstacle detection apparatus). The controller 410 executes various functional applications and data processing of the vehicle, that is, implements the above-described obstacle detection method, by executing software programs, instructions, and modules stored in the memory 420.
The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 420 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 420 may further include memory located remotely from the controller 410, which may be connected to the vehicle over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Radar 430 may be used to collect obstacle information, and the number of radars 430 in the vehicle may be one or more.
EXAMPLE five
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for obstacle detection, the method including:
acquiring original radar data of a candidate obstacle, and determining each attribute data of the candidate obstacle by combining vehicle state data;
determining attribute accurate values corresponding to the attribute data, and determining the detection reliability of the candidate obstacle according to the attribute accurate values;
and when the detection reliability is greater than a preset reliability threshold value, determining the candidate obstacle as a target obstacle.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the obstacle detection method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the obstacle detection device, the included units and modules are merely divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (8)
1. An obstacle detection method, comprising:
acquiring original radar data of a candidate obstacle, and determining each attribute data of the candidate obstacle by combining vehicle state data;
determining attribute accurate values corresponding to the attribute data, and determining the detection reliability of the candidate obstacle according to the attribute accurate values;
when the detection reliability is larger than a preset reliability threshold value, determining the candidate obstacle as a target obstacle;
the determining the attribute accurate value corresponding to each attribute data includes:
aiming at each attribute data, acquiring a corresponding attribute accurate value mapping table, searching a target accurate value matched with the attribute data in the attribute accurate value mapping table, and determining the target accurate value as the attribute accurate value corresponding to the attribute data;
the step of determining the attribute accurate value mapping table comprises the following steps:
dividing values of the target attribute of the obstacle to form at least two attribute value intervals;
acquiring a preset number of training radar data, and dividing each training radar data into corresponding radar data sets according to the attribute value interval;
for each radar data set, determining an accurate data amount and a total data amount of the training radar data, and determining a ratio of the accurate data amount to the total data amount as a training accurate value;
and establishing a mapping relation between each attribute value interval and the corresponding training accurate value to obtain an attribute accurate value mapping table of the obstacle target attribute.
2. The obstacle detection method according to claim 1, before mapping each of the attribute value intervals to a corresponding training accurate value, further comprising: and fitting each training accurate value.
3. The obstacle detection method according to claim 1, wherein the determining the detection reliability of the candidate obstacle from each of the attribute accuracy values includes:
and acquiring preset attribute weight, and performing weighted summation on each attribute accurate value to obtain the detection reliability of the candidate obstacle.
4. The obstacle detection method according to claim 1, characterized in that the attribute data comprises obstacle position, obstacle distance, obstacle speed, obstacle acceleration, obstacle direction, obstacle size and/or obstacle category.
5. Obstacle detection method according to claim 1, characterized in that the vehicle state data comprise vehicle speed data, direction of movement data and/or yaw angle data.
6. An obstacle detection device, characterized by comprising:
the attribute data determining module is used for acquiring original radar data of the candidate obstacle and determining each attribute data of the candidate obstacle by combining vehicle state data;
the detection reliability determining module is used for determining attribute accurate values corresponding to the attribute data and determining the detection reliability of the candidate obstacle according to the attribute accurate values;
an obstacle determination module, configured to determine the candidate obstacle as a target obstacle when the detection reliability is greater than a preset reliability threshold;
the determining the attribute accurate value corresponding to each attribute data includes:
aiming at each attribute data, acquiring a corresponding attribute accurate value mapping table, searching a target accurate value matched with the attribute data in the attribute accurate value mapping table, and determining the target accurate value as the attribute accurate value corresponding to the attribute data;
the step of determining the attribute accurate value mapping table comprises the following steps:
dividing values of the target attribute of the obstacle to form at least two attribute value intervals;
acquiring a preset number of training radar data, and dividing each training radar data into corresponding radar data sets according to the attribute value interval;
for each radar data set, determining an accurate data amount and a total data amount of the training radar data, and determining a ratio of the accurate data amount to the total data amount as a training accurate value;
and establishing a mapping relation between each attribute value interval and the corresponding accurate training value to obtain an attribute accurate value mapping table of the obstacle target attribute.
7. A vehicle, characterized in that the vehicle comprises:
one or more controllers;
a memory for storing one or more programs;
the radar is used for acquiring barrier information;
when executed by the one or more controllers, cause the one or more controllers to implement the obstacle detection method of any one of claims 1-5.
8. A storage medium containing computer executable instructions for performing the obstacle detection method of any one of claims 1-5 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011339996.0A CN112462368B (en) | 2020-11-25 | 2020-11-25 | Obstacle detection method and device, vehicle and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011339996.0A CN112462368B (en) | 2020-11-25 | 2020-11-25 | Obstacle detection method and device, vehicle and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112462368A CN112462368A (en) | 2021-03-09 |
| CN112462368B true CN112462368B (en) | 2022-07-12 |
Family
ID=74808375
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011339996.0A Active CN112462368B (en) | 2020-11-25 | 2020-11-25 | Obstacle detection method and device, vehicle and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112462368B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112883909A (en) * | 2021-03-16 | 2021-06-01 | 东软睿驰汽车技术（沈阳）有限公司 | Surrounding box-based obstacle position detection method and device and electronic equipment |
| CN113060131B (en) * | 2021-04-16 | 2022-08-05 | 中国第一汽车股份有限公司 | Obstacle map generation method and device, vehicle and storage medium |
| CN113619600B (en) * | 2021-08-17 | 2022-11-15 | 广州文远知行科技有限公司 | Obstacle data diagnosis method, obstacle data diagnosis device, movable carrier, and storage medium |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508246A (en) * | 2011-10-13 | 2012-06-20 | 吉林大学 | Method for detecting and tracking obstacles in front of vehicle |
| CN104965202A (en) * | 2015-06-18 | 2015-10-07 | 奇瑞汽车股份有限公司 | Barrier detection method and device |
| CN105372654A (en) * | 2015-12-14 | 2016-03-02 | 财团法人车辆研究测试中心 | Obstacle classification reliability quantification method |
| CN108596058A (en) * | 2018-04-11 | 2018-09-28 | 西安电子科技大学 | Running disorder object distance measuring method based on computer vision |
| CN109002041A (en) * | 2018-08-09 | 2018-12-14 | 北京智行者科技有限公司 | A kind of vehicle obstacle-avoidance method |
| CN109901567A (en) * | 2017-12-08 | 2019-06-18 | 百度在线网络技术（北京）有限公司 | Method and apparatus for exporting obstacle information |
| CN110136295A (en) * | 2018-02-09 | 2019-08-16 | 驭势（上海）汽车科技有限公司 | System and method for detecting the confidence level of automatic Pilot |
| CN110395251A (en) * | 2019-06-27 | 2019-11-01 | 驭势（上海）汽车科技有限公司 | A kind of automatic emergency brake decision-making technique based on Multi-sensor Fusion data |
| CN110502983A (en) * | 2019-07-11 | 2019-11-26 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of barrier in a kind of detection highway |
| CN111038500A (en) * | 2019-12-30 | 2020-04-21 | 宜宾凯翼汽车有限公司 | Automatic emergency braking execution method and system for automobile |
| CN111231982A (en) * | 2020-01-08 | 2020-06-05 | 中国第一汽车股份有限公司 | Obstacle identification method and device for intelligent driving, vehicle and storage medium |
| CN111316128A (en) * | 2018-12-28 | 2020-06-19 | 深圳市大疆创新科技有限公司 | Continuous obstacle detection method, device, system and storage medium |
| CN111427369A (en) * | 2020-06-08 | 2020-07-17 | 北京三快在线科技有限公司 | Unmanned vehicle control method and device |
| CN111452723A (en) * | 2019-01-18 | 2020-07-28 | 长沙智能驾驶研究院有限公司 | Intelligent vehicle and early warning control method, device, system and storage medium thereof |
| CN111505623A (en) * | 2020-04-24 | 2020-08-07 | 中南大学 | Method and system for detecting obstacle in driving process of unmanned vehicle and vehicle |
| CN111856417A (en) * | 2020-07-20 | 2020-10-30 | 中国第一汽车股份有限公司 | Performance analysis method and device for vehicle-mounted millimeter wave radar, terminal and storage medium |
| CN111886597A (en) * | 2019-06-28 | 2020-11-03 | 深圳市大疆创新科技有限公司 | Obstacle detection method and device for movable platform and movable platform |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7239277B2 (en) * | 2004-04-12 | 2007-07-03 | Time Domain Corporation | Method and system for extensible position location |
| US10642275B2 (en) * | 2018-06-18 | 2020-05-05 | Zoox, Inc. | Occulsion aware planning and control |
| CN111638520A (en) * | 2020-05-15 | 2020-09-08 | 安徽江淮汽车集团股份有限公司 | Obstacle recognition method, obstacle recognition device, electronic device and storage medium |
| CN111798698B (en) * | 2020-06-24 | 2022-04-08 | 中国第一汽车股份有限公司 | Method and device for determining front target vehicle and vehicle |
| CN111913183A (en) * | 2020-07-27 | 2020-11-10 | 中国第一汽车股份有限公司 | Vehicle lateral obstacle avoidance method, device and equipment and vehicle |
-
2020
- 2020-11-25 CN CN202011339996.0A patent/CN112462368B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102508246A (en) * | 2011-10-13 | 2012-06-20 | 吉林大学 | Method for detecting and tracking obstacles in front of vehicle |
| CN104965202A (en) * | 2015-06-18 | 2015-10-07 | 奇瑞汽车股份有限公司 | Barrier detection method and device |
| CN105372654A (en) * | 2015-12-14 | 2016-03-02 | 财团法人车辆研究测试中心 | Obstacle classification reliability quantification method |
| CN109901567A (en) * | 2017-12-08 | 2019-06-18 | 百度在线网络技术（北京）有限公司 | Method and apparatus for exporting obstacle information |
| CN110136295A (en) * | 2018-02-09 | 2019-08-16 | 驭势（上海）汽车科技有限公司 | System and method for detecting the confidence level of automatic Pilot |
| CN108596058A (en) * | 2018-04-11 | 2018-09-28 | 西安电子科技大学 | Running disorder object distance measuring method based on computer vision |
| CN109002041A (en) * | 2018-08-09 | 2018-12-14 | 北京智行者科技有限公司 | A kind of vehicle obstacle-avoidance method |
| CN111316128A (en) * | 2018-12-28 | 2020-06-19 | 深圳市大疆创新科技有限公司 | Continuous obstacle detection method, device, system and storage medium |
| CN111452723A (en) * | 2019-01-18 | 2020-07-28 | 长沙智能驾驶研究院有限公司 | Intelligent vehicle and early warning control method, device, system and storage medium thereof |
| CN110395251A (en) * | 2019-06-27 | 2019-11-01 | 驭势（上海）汽车科技有限公司 | A kind of automatic emergency brake decision-making technique based on Multi-sensor Fusion data |
| CN111886597A (en) * | 2019-06-28 | 2020-11-03 | 深圳市大疆创新科技有限公司 | Obstacle detection method and device for movable platform and movable platform |
| CN110502983A (en) * | 2019-07-11 | 2019-11-26 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of barrier in a kind of detection highway |
| CN111038500A (en) * | 2019-12-30 | 2020-04-21 | 宜宾凯翼汽车有限公司 | Automatic emergency braking execution method and system for automobile |
| CN111231982A (en) * | 2020-01-08 | 2020-06-05 | 中国第一汽车股份有限公司 | Obstacle identification method and device for intelligent driving, vehicle and storage medium |
| CN111505623A (en) * | 2020-04-24 | 2020-08-07 | 中南大学 | Method and system for detecting obstacle in driving process of unmanned vehicle and vehicle |
| CN111427369A (en) * | 2020-06-08 | 2020-07-17 | 北京三快在线科技有限公司 | Unmanned vehicle control method and device |
| CN111856417A (en) * | 2020-07-20 | 2020-10-30 | 中国第一汽车股份有限公司 | Performance analysis method and device for vehicle-mounted millimeter wave radar, terminal and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Algorithm Development of Level Crossing Obstacle Detection using Laser Radar Sensor;Young-June Kim等;《The transactions of The Korean Institute of Electrical Engineers》;20131201;第62卷(第12期);1813-1819 * |
| 基于三维激光雷达的动态障碍实时检测与跟踪;杨飞 等;《浙江大学学报(工学版)》;20120915(第09期);18-24 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112462368A (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112462368B (en) | Obstacle detection method and device, vehicle and storage medium | |
| CN109459734B (en) | Laser radar positioning effect evaluation method, device, equipment and storage medium | |
| WO2021104497A1 (en) | Positioning method and system based on laser radar, and storage medium and processor | |
| JP2019215853A (en) | Method for positioning, device for positioning, device, and computer readable storage medium | |
| CN110705458B (en) | Boundary detection method and device | |
| CN110889808B (en) | Positioning method, device, equipment and storage medium | |
| CN112693466A (en) | System and method for evaluating performance of vehicle environment perception sensor | |
| CN112083441B (en) | Obstacle detection method and system for depth fusion of laser radar and millimeter wave radar | |
| CN112116031B (en) | Target fusion method, system, vehicle and storage medium based on road side equipment | |
| CN111257892A (en) | Obstacle detection method for automatic driving of vehicle | |
| CN110516621B (en) | Method and device for detecting barrier-free driving area, vehicle and storage medium | |
| CN112949782A (en) | Target detection method, device, equipment and storage medium | |
| CN113189975A (en) | Method for determining position data and/or movement data of a vehicle | |
| CN109583416A (en) | Pseudo- Lane detection method and system | |
| CN112781599A (en) | Method for determining the position of a vehicle | |
| JP7386136B2 (en) | Cloud height measurement device, measurement point determination method, and cloud type determination method | |
| CN110411435B (en) | Robot positioning method and device and robot | |
| CN115034324B (en) | Multi-sensor fusion perception efficiency enhancement method | |
| CN113959457A (en) | Positioning method and device for automatic driving vehicle, vehicle and medium | |
| CN113296120B (en) | Obstacle detection method and terminal | |
| CN115151954A (en) | Method and device for detecting a drivable region | |
| CN117269940B (en) | Point cloud data generation method and perception capability verification method of laser radar | |
| CN113838125A (en) | Target position determining method and device, electronic equipment and storage medium | |
| CN117029840A (en) | Mobile vehicle positioning method and system | |
| CN114419573A (en) | Dynamic occupancy grid estimation method and device |
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 |