CN112130567A - Data processing method and device - Google Patents
Data processing method and device Download PDFInfo
- Publication number
- CN112130567A CN112130567A CN202011005875.2A CN202011005875A CN112130567A CN 112130567 A CN112130567 A CN 112130567A CN 202011005875 A CN202011005875 A CN 202011005875A CN 112130567 A CN112130567 A CN 112130567A
- Authority
- CN
- China
- Prior art keywords
- positioning matrix
- map data
- positioning
- information
- matrix
- 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
- 238000003672 processing method Methods 0.000 title abstract description 9
- 239000011159 matrix material Substances 0.000 claims abstract description 329
- 238000000034 method Methods 0.000 claims abstract description 48
- 238000012937 correction Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000004590 computer program Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 description 13
- 238000006243 chemical reaction Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000013519 translation Methods 0.000 description 10
- 230000004927 fusion Effects 0.000 description 9
- 230000001960 triggered effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
- G05D1/0274—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means using mapping information stored in a memory device
Abstract
The embodiment of the invention provides a data processing method and a data processing device, wherein the method comprises the following steps: acquiring preset first map data; acquiring second map data acquired in real time; acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix; generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data; updating the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix; and performing positioning correction by adopting the updated first positioning matrix. By the embodiment of the invention, the positioning correction is realized by adopting the updated first matrix, the updated first positioning matrix fully considers the first residual error information and the second residual error information, the error influence can be reduced, and the robustness of the positioning information is improved.
Description
Technical Field
The present invention relates to the field of vehicle technologies, and in particular, to a data processing method and apparatus.
Background
Along with the development of intelligent automobiles, more and more vehicles are provided with positioning systems, and the positioning systems of the vehicles can provide route driving guidance for drivers, so that the drivers can conveniently and quickly determine the current positions, and convenient service is provided for public travel.
When the intelligent automobile drives into the parking lot, the track map of the current driving track can be constructed on line, the track map is matched with the scene semantic map constructed in advance on line, the positioning of the current automobile on the scene semantic map is obtained, and the positioning information can serve for autonomous parking and navigation in the parking lot.
However, in the current matching and positioning algorithm, since the matching and positioning are performed by completely depending on the transformation matrix in the previous positioning information, when the error of the previous positioning information is large, a drift error is easily generated, which may cause the error of the obtained vehicle positioning information to be large, and the robustness to be low.
Disclosure of Invention
In view of the above, it is proposed to provide a data processing method and apparatus that overcomes or at least partially solves the above mentioned problems, comprising:
a method of data processing, the method comprising:
acquiring preset first map data;
acquiring second map data acquired in real time;
acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix;
generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data;
updating the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix;
and performing positioning correction by adopting the updated first positioning matrix.
Optionally, the method further comprises:
and updating the first residual error information according to the second residual error information.
Optionally, the generating a second positioning matrix and second residual information for the second positioning matrix according to the first map data and the second map data includes:
acquiring target track information acquired in real time;
when the target track information collected in real time meets a first preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data, the second map data and the first positioning matrix;
and when the real-time track information meets a second preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data.
Optionally, when the real-time trajectory information meets a first preset condition, generating a second positioning matrix according to the first map data, the second map data, and the first positioning matrix, including:
converting the second map data by using the first positioning matrix;
generating an intermediate positioning matrix according to the converted second map data and the first map data;
and combining the intermediate positioning matrix and the first positioning matrix to obtain a second positioning matrix.
Optionally, when the target trajectory information acquired in real time meets a second preset condition, generating a second positioning matrix according to the first map data and the second map data, including:
converting the second map data according to the first map data;
and generating a second positioning matrix according to the converted second map data and the first map data.
Optionally, the first preset condition is that a track distance in the real-time acquired target track information is greater than a first preset distance, and the second preset condition is that the track distance in the real-time track information is greater than a second preset distance;
wherein the second preset distance is greater than the first preset distance;
and/or the first preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a first preset number, and the second preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a second preset number;
wherein the second preset number is greater than the first preset number.
Optionally, before the obtaining the latest first positioning matrix and the first residual information for the first positioning matrix, the method further includes:
and carrying out initial matching on the first map data to obtain an initial first positioning matrix and first residual error information aiming at the initial first positioning matrix.
An apparatus for data processing, the apparatus comprising:
the first acquisition module is used for acquiring preset first map data;
the second acquisition module is used for acquiring second map data acquired in real time;
a third obtaining module, configured to obtain a first positioning matrix and first residual information for the first positioning matrix;
a generating module, configured to generate a second positioning matrix and second residual information for the second positioning matrix according to the first map data and the second map data;
a first positioning matrix updating module, configured to update the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix;
and the positioning correction module is used for performing positioning correction by adopting the updated first positioning matrix.
A vehicle comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, the computer program, when executed by the processor, implementing a method of data processing as described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of data processing as described above.
The embodiment of the invention has the following advantages:
the embodiment of the invention obtains the preset first map data, obtains the second map data collected in real time, obtains the first positioning matrix and the first residual error information aiming at the first positioning matrix, generates the second positioning matrix and the second residual error information aiming at the second positioning matrix according to the first map data and the second map data, updates the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix, adopts the updated first positioning matrix to carry out positioning correction, realizes that the positioning correction is carried out by adopting the updated first matrix, on one hand, the positioning correction speed can be improved, on the other hand, the updated first positioning matrix fully considers the first residual error information and the second residual error information, the error influence can be reduced, and the robustness of the positioning information is improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the description of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flow chart illustrating steps of a method for data processing according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating steps of another method for data processing according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating steps of a method for processing data according to another embodiment of the present invention;
fig. 4 is a flowchart illustrating a front-end and back-end matching positioning updating process according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a flowchart illustrating steps of a data processing method according to an embodiment of the present invention is shown, which may specifically include the following steps:
in the process of parking when a vehicle drives into a parking lot, the data acquisition starting point is the data acquisition starting point, the data acquisition is started, and a real-time track map is constructed through the acquired data.
The vehicle can acquire the current positioning position of the vehicle through the positioning system, and in order to acquire map positioning information with higher precision, a constructed parking lot scene map can be preset in the system, so that first map data can be acquired.
In an example, the first map data may include positioning information, pose information, and the like of a plurality of semantic landmark information associated with a road waypoint, such as a parking lot entrance, a deceleration strip, a curve exit point, a curve entry point, a slope exit point, a slope entry point, a road node, parking space information, and the like in the scene map.
after the first map data is acquired, the second map data may be acquired in the newly constructed real-time trajectory map. The second map data may include positioning information, pose information, and the like of various semantic landmark information in which parking lot entry, deceleration strip, curve exit point, curve entry point, slope exit point, slope entry point, road node, and parking space information are associated with road waypoints in the real-time trajectory map.
103, acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix;
after the second map data is acquired, a first positioning matrix and first residual information for the first positioning matrix may be acquired.
Wherein, the first positioning matrix and the first residual error information can be positioning matrix information adopted in the last positioning updating process, and can perform conversion such as rotation and translation and the like on the real-time map data through the positioning matrix information, because the converted map data and the first map data may have deviation, when the deviation is larger, the accuracy of obtaining the positioning information according to the positioning matrix conversion is lower, therefore, the residual error information corresponding to the positioning matrix information can be obtained by performing residual error calculation on the converted map data and the first map data,
in an embodiment of the present invention, before the obtaining the latest first positioning matrix and the first residual information for the first positioning matrix, the method further includes:
and carrying out initial matching on the first map data to obtain an initial first positioning matrix and first residual error information aiming at the initial first positioning matrix.
In practical application, a parking lot entrance can be used as a starting point, data acquisition is started, initial matching can be carried out on the acquired data and first map data, semantic landmark information capable of being matched is obtained, cloud computing is carried out according to positioning coordinates of the semantic landmark information, and therefore a first positioning matrix and first residual error information after conversion according to the first positioning matrix can be obtained.
In an example, a track map window which is increased along with a track from a starting point can be selected, semantic landmark information in the window is initially matched with semantic landmark information in first map data according to positions and mutual relevance, the number of matched semantic landmarks and the positioning coordinates of the semantic landmarks can be determined according to an initial matching result, when the number of the semantic landmarks is larger than a preset number, the initial matching can be determined to be successful, and a real-time track map and a scene map can be matched, so that a first positioning matrix and first residual error information can be obtained through calculation according to the positioning coordinates of the matched semantic landmarks.
after the first positioning matrix and the first residual error information are obtained, a second positioning matrix and second residual error information for the second positioning matrix can be generated by combining the second map data.
after the second positioning matrix and the second residual information, the first positioning matrix may be updated according to the first residual information, the second residual information, and the second positioning matrix.
In an example, the updating the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix includes:
determining first weight information corresponding to the first residual error information and second weight information corresponding to the second residual error information; and updating the first positioning matrix according to the first weight information, the second weight information and the second positioning matrix.
In practical applications, the first weight information and the second weight information corresponding to the first residual information and the second residual information may be calculated, where the first residual information and the second residual information may represent deviations between the second map data and the first map data after being converted according to a corresponding positioning matrix, such as rotational translation, and the like, and when the deviations are larger, the accuracy of obtaining the positioning information according to the positioning matrix conversion is lower, so when calculating the weights, the residual values are inversely proportional to the weights, that is, when the residual values are larger, the corresponding weights are smaller.
After the first weight information and the second weight information are determined, the first positioning matrix and the second positioning matrix can be subjected to weighted fusion calculation through the first weight information and the second weight information, the first positioning matrix is updated according to a new positioning matrix value obtained through weighted fusion calculation, and the updated first positioning matrix is the positioning matrix required by the positioning correction.
By performing weighted fusion according to the residual error value, the drift error caused by the error in the previous positioning correction can be reduced, and the robustness is improved by the positioning information obtained by calculation.
For example: second residual error information in the latest positioning matching process according to the track mapAnd the historical (last update) residual value Ck-1(first residual information) and calculating the corresponding weight(second weight information) and w (C)k-1) (first weight information), wherein the weight and the residual are inversely proportional.
The second location matrix may then be usedCorresponding angle of rotationAnd amount of translationAnd history positioning matrix Tk-1(first positioning matrix) corresponding to Δ Pk-1And Δ θk-1Weighted average is performed to obtain the current rotation angle delta PkAnd amount of translation Δ θkAs shown in the following equation:
thus, the final positioning matrix T of the current position can be obtainedk(updated first positioning matrix), as shown in the following equation:
and step 106, carrying out positioning correction by adopting the updated first positioning matrix.
After the first positioning matrix is updated, the updated first positioning matrix can be adopted to perform positioning correction on the latest track point or semantic landmark information in the second map data.
In an example, the real-time track information may be acquired from a track map constructed in real time, and the track pose information of the current position in the track map may be determined, so that a first pose matrix may be determined, and then a second pose matrix converted into a scene map may be determined using the updated first pose matrix and the pose matrix, so that the corrected pose information may be obtained according to the second pose matrix.
For example: the track pose of the current position in the track map is PtMap,iAnd thetatMap,iFrom this, the current pose matrix T can be derivedtMap,iBy means of the latest updated positioning matrix Tk(the updated first positioning matrix) to calculate a pose matrix T converted to the scene mapvMap,iAnd its corresponding position PvMap,iAnd attitude θvMap,i. As shown in the following equation.
TvMap,i=Tk·TtMap,i
In an embodiment of the present invention, the method further includes:
and updating the first residual error information according to the second residual error information.
After the second residual error information is obtained, the second residual error information in the latest positioning matching process of the track map can be obtainedAnd the historical (last update) residual value Ck-1(first residual information) and calculating the current residual value Ck(updated first residual information), as shown in the following equation:
the embodiment of the invention obtains the preset first map data, obtains the second map data collected in real time, obtains the first positioning matrix and the first residual error information aiming at the first positioning matrix, generates the second positioning matrix and the second residual error information aiming at the second positioning matrix according to the first map data and the second map data, updates the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix, adopts the updated first positioning matrix to carry out positioning correction, realizes that the positioning correction is carried out by adopting the updated first matrix, on one hand, the positioning correction speed can be improved, on the other hand, the updated first positioning matrix fully considers the first residual error information and the second residual error information, the error influence can be reduced, and the robustness of the positioning information is improved.
Referring to fig. 2, a flowchart illustrating steps of another data processing method according to an embodiment of the present invention is shown, which may specifically include the following steps:
in an example, the target track point may be track information from initial track information of the start point of the collected data to first track information, or may be track information from second track information of the position of the vehicle in the previous correction in the track map to the first track information.
After the first positioning matrix and the first residual information are obtained, the first track information of the current position can be determined in the second map data collected in real time, so that the target track information including the first track information can be obtained.
Step 205, when the target trajectory information acquired in real time meets a first preset condition, generating a second positioning matrix and second residual error information for the second positioning matrix according to the first map data, the second map data, and the first positioning matrix;
in an example, the first preset condition is that a track distance in the real-time acquired target track information is greater than a first preset distance; and/or the first preset condition is that the number of semantic elements in the real-time acquired target track information is larger than a first preset number.
After the target track information is obtained, when the target track information meets a first preset condition, a second positioning matrix and second residual error information for the second positioning matrix may be generated according to the first map data, the second map data, and the first positioning matrix.
In an embodiment of the present invention, when the track information collected in real time meets a first preset condition, generating a second positioning matrix according to the first map data, the second map data, and the first positioning matrix includes:
converting the second map data by using the first positioning matrix; generating an intermediate positioning matrix according to the converted second map data and the first map data; and combining the intermediate positioning matrix and the first positioning matrix to obtain a second positioning matrix.
In one example, further comprising:
and generating second residual error information aiming at the middle positioning matrix according to the converted second map data and the first map data.
After the second positioning matrix is generated, since there is a possibility that the first map data and the second map data may deviate and may not completely overlap each other, second residual information for the second positioning matrix may also be generated.
In practical application, when the real-time track information meets a first preset condition, track data with fixed distance or fixed semantic landmark quantity including first track information can be determined in second map data, a first positioning matrix can be collected to convert the second map data, the converted track data is matched with the semantic landmark information in the first map data, an intermediate positioning matrix can be generated according to the track data corresponding to the matched semantic landmark information and the first map data, and then the intermediate positioning matrix and the first positioning matrix can be combined to obtain the second positioning matrix.
For example: front window matching: selecting a short-distance sliding window on the current real-time track map according to a certain track distance or semantic number, and triggering front-end matching when the window meets a sliding condition (a first preset condition).
Front-end matching is mainly achieved by using the previously calculated positioning transformation matrix Tk-1(first positioning matrix) in the latest short-distance map window of the track map (second map data) to be constructed in real timeAnd converting all semantic landmark information into data in a coordinate system of a constructed scene map, and finding out semantic landmark information matched with the semantic landmark information in the first map data in a window selected by the track map by utilizing the position similarity of the semantic landmark information. The conversion matrix delta T of the front end can be obtained according to the matched semantic landmark informationF,k(intermediate matrix) and post-transform matching residual CF,k. (second residual information) calculating the current estimated positioning matrix(second positioning matrix), and the current predicted residual(second residual information).
When the matching positioning is carried out without depending on the prior positioning information, the calculation amount is large, and the consumed time is long, so that the positioning correction of the map is too slow, the user requirements cannot be met, the calculation amount of the front-end matching is small, and the correction speed can be improved.
in an example, the updating the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix includes:
determining first weight information corresponding to the first residual error information and second weight information corresponding to the second residual error information; and updating the first positioning matrix according to the first weight information, the second weight information and the second positioning matrix.
In practical applications, the first weight information and the second weight information corresponding to the first residual information and the second residual information may be calculated, where the first residual information and the second residual information may represent deviations between the second map data and the first map data after being converted according to a corresponding positioning matrix, such as rotational translation, and the like, and when the deviations are larger, the accuracy of obtaining the positioning information according to the positioning matrix conversion is lower, so when calculating the weights, the residual values are inversely proportional to the weights, that is, when the residual values are larger, the corresponding weights are smaller.
After the first weight information and the second weight information are determined, the first positioning matrix and the second positioning matrix can be subjected to weighted fusion calculation through the first weight information and the second weight information, the first positioning matrix is updated according to a new positioning matrix value obtained through weighted fusion calculation, and the updated first positioning matrix is the positioning matrix required by the positioning correction.
By performing weighted fusion according to the residual error value, the drift error caused by the error in the previous positioning correction can be reduced, and the robustness is improved by the positioning information obtained by calculation.
For example: second residual error information in the latest positioning matching process according to the track mapAnd the historical (last update) residual value Ck-1(first residual information) and calculating the corresponding weight(second weight information) and w (C)k-1) (first weight information), wherein the weight and the residual are inversely proportional.
The second location matrix may then be usedCorresponding angle of rotationAnd amount of translationAnd history positioning matrix Tk-1(first positioning matrix) corresponding to Δ Pk-1And Δ θk-1Weighted average is performed to obtain the current rotation angle delta PkTranslation amount Δ θkAs shown in the following equation:
thus, the final positioning matrix T of the current position can be obtainedk(updated first positioning matrix), as shown in the following equation:
and step 207, adopting the updated first positioning matrix to perform positioning correction.
In an example, the real-time track information may be acquired from a track map constructed in real time, the track pose information of the current position in the track map may be determined, and the planning may be performed, so that a first pose matrix may be determined, and then a second pose matrix converted into a scene map may be determined using the updated first pose matrix and the pose matrix, so that the corrected pose information may be obtained according to the second pose matrix.
For example: the track pose of the current position in the track map is PtMap,iAnd thetatMap,iFrom this, the current pose matrix T can be derivedtMap,iBy means of the latest updated positioning matrix Tk(the updated first positioning matrix) to calculate a pose matrix T converted to the scene mapvMap,iAnd its corresponding position PvMap,iAnd attitude θvMap,i. As shown in the following equation.
TvMap,i=Tk·TtMap,i
In an embodiment of the present invention, the method further includes:
and updating the first residual error information according to the second residual error information.
After the second residual error information is obtained, the second residual error information in the latest positioning matching process of the track map can be obtainedAnd the historical (last update) residual value Ck-1(first residual information) and calculating the current residual value Ck(updated first residual information), as shown in the following equation:
in the embodiment of the invention, by acquiring preset first map data, acquiring real-time acquired second map data, acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix, acquiring real-time acquired target track information, when the real-time acquired target track information meets a first preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data, the second map data and the first positioning matrix, updating the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix, and performing positioning correction by adopting the updated first positioning matrix, the positioning correction by adopting the updated first positioning matrix is realized, compared with the correction only by depending on the second positioning matrix, on one hand, the positioning correction speed can be improved, and on the other hand, the updated first positioning matrix fully considers the first residual error information and the second residual error information, so that the error influence can be reduced, and the robustness of the positioning information is improved.
Referring to fig. 3, a flowchart illustrating steps of another data processing method according to an embodiment of the present invention is shown, which may specifically include the following steps:
303, acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix;
in an example, the target track point may be track information from initial track information of the start point of the collected data to the first track information, or may be track information from second track information of the position of the vehicle at the previous correction to the first track information.
After the first positioning matrix and the first residual information are obtained, the first track information of the current position can be determined in the second map data collected in real time, so that the target track information including the first track information can be obtained.
305, when the real-time track information meets a second preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data;
in an example, the second preset condition is that the track distance in the real-time track information is greater than a second preset distance; wherein the second preset distance is greater than the first preset distance; the second preset condition is that the number of semantic road signs in the real-time collected target track information is larger than a second preset number; wherein the second preset number is greater than the first preset number.
After the target track information is obtained, when the target track information meets a second preset condition, the back-end matching can be triggered, so that a second positioning matrix and second residual error information aiming at the second positioning matrix can be generated according to the first map data, the second map data and the first positioning matrix.
In an implementation of the present invention, when the target trajectory information collected in real time meets a second preset condition, generating a second positioning matrix according to the first map data and the second map data includes:
converting the second map data according to the first map data; and generating a second positioning matrix according to the converted second map data and the first map data.
When the real-time trajectory information meets the first preset condition, trajectory data with a fixed distance or a fixed semantic quantity including the first trajectory information can be determined in the second map data, and the probability of successful key semantic matching between the trajectory data and the first map data is higher, so that matched key semantic landmark information can be determined between the trajectory data and the first map data, for example: and semantic landmark information such as an entrance, a slope exit point, a slope entry point and a road node is converted into second map data through key semantic landmark information, so that the converted track data and the first map data can be matched with semantic landmark information, and a second matrix can be generated through the matched semantic landmark information.
Rear window matching: and selecting a long-distance sliding window on the current real-time track map according to a certain track distance or the number of semantic road signs, and triggering rear-end matching when the window meets a sliding condition (a second preset condition).
In back-end matching, the matching of several key semantic road sign information (such as entrance, slope point, road node) can be determined in a track map and a scene map through the correlation among all semantics, after the temporary conversion is carried out according to the matching result, the latest long-distance window is selected, the matched semantic road sign information in the window is determined through the position similarity of the semantic road sign information, and the conversion matrix delta T is determined according to the matched semantic road sign informationB,kAnd post-transform matched residual CB,k. I.e. the current estimated positioning matrix(second positioning matrix), and the current predicted residual(second residual information).
Because the back-end matching does not depend on the first positioning matrix for matching positioning, the influence of drift errors is not easy to be caused.
in an example, the updating the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix includes:
determining first weight information corresponding to the first residual error information and second weight information corresponding to the second residual error information; and updating the first positioning matrix according to the first weight information, the second weight information and the second positioning matrix.
In practical applications, the first weight information and the second weight information corresponding to the first residual information and the second residual information may be calculated, where the first residual information and the second residual information may represent deviations between the second map data and the first map data after being converted according to a corresponding positioning matrix, such as rotational translation, and the like, and when the deviations are larger, the accuracy of obtaining the positioning information according to the positioning matrix conversion is lower, so when calculating the weights, the residual values are inversely proportional to the weights, that is, when the residual values are larger, the corresponding weights are smaller.
After the first weight information and the second weight information are determined, the first positioning matrix and the second positioning matrix can be subjected to weighted fusion calculation through the first weight information and the second weight information, the first positioning matrix is updated according to a new positioning matrix value obtained through weighted fusion calculation, and the updated first positioning matrix is the positioning matrix required by the positioning correction.
By performing weighted fusion according to the residual error value, the drift error caused by the error in the previous positioning correction can be reduced, and the robustness is improved by the positioning information obtained by calculation.
For example: maximum according to the track mapSecond residual error information in new positioning matching processAnd the historical (last update) residual value Ck-1(first residual information) and calculating the corresponding weight(second weight information) and w (C)k-1) (first weight information), wherein the weight value is inversely proportional to the residual value.
The second location matrix may then be usedCorresponding angle of rotationAnd amount of translationAnd history positioning matrix Tk-1(first positioning matrix) corresponding to Δ Pk-1And Δ θk-1Weighted average is performed to obtain the current rotation angle delta PkTranslation amount Δ θkAs shown in the following equation:
thus, the final positioning matrix T of the current position can be obtainedk(updated first positioning matrix), as shown in the following equation:
and 307, positioning and correcting by using the updated first positioning matrix.
In an example, the real-time track information may be acquired from a track map constructed in real time, the track pose information of the current position in the track map may be determined, and the planning may be performed, so that a first pose matrix may be determined, and then a second pose matrix converted into a scene map may be determined using the updated first pose matrix and the pose matrix, so that the corrected pose information may be obtained according to the second pose matrix.
For example: the track pose of the current position in the track map is PtMap,iAnd thetatMap,iFrom this, the current pose matrix T can be derivedtMap,iBy means of the latest updated positioning matrix Tk(the updated first positioning matrix) to calculate a pose matrix T converted to the scene mapvMap,iAnd its corresponding position PvMap,iAnd attitude θvMap,i. As shown in the following equation.
TvMap,i=Tk·TtMap,i
In an embodiment of the present invention, the method further includes:
and updating the first residual error information according to the second residual error information.
After the second residual error information is obtained, the second residual error information in the latest positioning matching process of the track map can be obtainedAnd the historical (last update) residual value Ck-1(first residual information) and calculating the current residual value Ck(updated first residual information), as shown in the following equation:
the embodiment of the invention obtains the second map data collected in real time by obtaining the preset first map data, obtains the first positioning matrix and the first residual error information aiming at the first positioning matrix, obtains the target track information collected in real time, generates the second positioning matrix and the second residual error information aiming at the second positioning matrix according to the first map data, the second map data and the first positioning matrix when the target track information collected in real time meets the second preset condition, updates the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix, adopts the updated first positioning matrix to carry out positioning correction, realizes the positioning correction by adopting the updated first matrix, and fully considers the first residual error information and the second residual error information in the updated first positioning matrix, the error influence can be reduced, and the robustness of the positioning information is improved.
The invention is illustrated below with reference to fig. 4:
a. and carrying out map matching, positioning and initialization on the online track map and the constructed scene map.
And carrying out initial matching on the first map data to obtain an initial first positioning matrix and first residual error information aiming at the initial first positioning matrix.
In practical application, a parking lot entrance can be used as a starting point, data acquisition is started, initial matching can be carried out on the acquired data and first map data, semantic landmark information capable of being matched is obtained, cloud computing is carried out according to positioning coordinates of the semantic landmark information, and therefore a first positioning matrix and first residual error information after conversion according to the first positioning matrix can be obtained.
b. During the real-time update of the track map, determine whether to trigger front-end matching and/or back-end matching? And c, when the front end matching and/or the back end matching are triggered, entering step e, and when neither the front end matching nor the back end matching is triggered.
c. And when the target track information meets a first preset condition, triggering front-end matching, and performing front-end map matching to obtain a second positioning matrix and second residual error information. (step 204-step 205)
And when the target track information meets a second preset condition, triggering rear-end matching, and performing rear-end map matching to obtain a second positioning matrix and second residual error information. (step 304-305.)
When no front-end or back-end match is triggered,
d. updating a positioning matrix: and after the front-end map matching or the rear-end map matching is carried out, updating the latest first positioning matrix by the second positioning matrix, the second residual error information and the first residual error information. (e.g., step 206, step 306)
e. Positioning update in the scene map: and adopting the latest first positioning matrix to perform positioning update to obtain a pose matrix under a scene map coordinate system. (step 207, step 3087)
f. Pose updating in the scene map: and obtaining the position and the posture of the current position of the vehicle according to the posture matrix.
When both front-end matching and back-end matching are triggered, the first positioning matrix may be updated with the second positioning matrix obtained by the front-end matching, and then the updated first positioning matrix may be updated with the second positioning matrix obtained by the back-end matching.
Or, when both front-end matching and back-end matching are triggered, the first positioning matrix may be updated by the second positioning matrix obtained by back-end matching, and then the updated first positioning matrix may be updated by the second positioning matrix obtained by front-end matching and the updated first positioning matrix may be matched by the back-end.
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. 5, a schematic structural diagram of a game speech generation apparatus according to an embodiment of the present invention is shown, which may specifically include the following modules:
a first obtaining module 501, configured to obtain preset first map data;
a second obtaining module 502, configured to obtain second map data collected in real time;
a third obtaining module 503, configured to obtain a first positioning matrix and first residual information for the first positioning matrix;
a generating module 504, configured to generate a second positioning matrix and second residual information for the second positioning matrix according to the first map data and the second map data;
a first positioning matrix updating module 505, configured to update the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix;
and a positioning correction module 506, configured to perform positioning correction by using the updated first positioning matrix.
In an embodiment of the present invention, the apparatus may further include:
and the first residual error information updating module is used for updating the first residual error information according to the second residual error information.
In an embodiment of the present invention, the apparatus may further include:
and the initial matching module is used for performing initial matching on the first map data to obtain an initial first positioning matrix and first residual error information aiming at the initial first positioning matrix.
In an embodiment of the present invention, the generating module 504 may include:
the target track information acquisition submodule is used for acquiring target track information acquired in real time;
the first preset condition submodule is used for generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data, the second map data and the first positioning matrix when the target track information acquired in real time meets a first preset condition;
and the second preset condition submodule is used for generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data when the real-time track information meets a second preset condition.
In an embodiment of the present invention, the first preset condition is that a track distance in the real-time acquired target track information is greater than a first preset distance, and the second preset condition is that the track distance in the real-time track information is greater than a second preset distance; wherein the second preset distance is greater than the first preset distance; and/or the first preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a first preset number, and the second preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a second preset number; wherein the second preset number is greater than the first preset number.
In an embodiment of the present invention, the first preset condition submodule may include:
a first conversion unit, configured to convert the second map data by using the first positioning matrix;
the intermediate positioning matrix generating unit is used for generating an intermediate positioning matrix according to the converted second map data and the first map data;
and the first generating unit is used for combining the intermediate positioning matrix and the first positioning matrix to obtain a second positioning matrix.
In an embodiment of the present invention, the second preset condition submodule may include:
a second conversion unit configured to convert the second map data according to the first map data;
and the second generating unit is used for generating a second positioning matrix according to the converted second map data and the first map data.
According to the embodiment of the invention, the preset first map data is obtained, the second map data collected in real time is obtained, the first positioning matrix and the first residual error information aiming at the first positioning matrix are obtained, the second positioning matrix and the second residual error information aiming at the second positioning matrix are generated according to the first map data and the second map data, the first positioning matrix is updated according to the first residual error information, the second residual error information and the second positioning matrix, and the updated first positioning matrix is adopted for positioning correction, so that the positioning correction is realized by adopting the updated first matrix, the first residual error information and the second residual error information are fully considered by the updated first positioning matrix, the error influence can be reduced, and the robustness of the positioning information is improved.
An embodiment of the present invention also provides a vehicle, which may include a processor, a memory, and a computer program stored on the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method of data processing as above.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the above data processing method.
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.
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 preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be 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 and apparatus for data processing provided above are described in detail, and a specific example is applied herein to illustrate the principles and embodiments of the present invention, and the above description of the embodiment is only used to help understand the method and core ideas 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 (10)
1. A method of data processing, the method comprising:
acquiring preset first map data;
acquiring second map data acquired in real time;
acquiring a first positioning matrix and first residual error information aiming at the first positioning matrix;
generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data;
updating the first positioning matrix according to the first residual error information, the second residual error information and the second positioning matrix;
and performing positioning correction by adopting the updated first positioning matrix.
2. The method of claim 1, further comprising:
and updating the first residual error information according to the second residual error information.
3. The method of claim 2, wherein generating a second positioning matrix and second residual information for the second positioning matrix from the first map data and the second map data comprises:
acquiring target track information acquired in real time;
when the target track information collected in real time meets a first preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data, the second map data and the first positioning matrix;
and when the real-time track information meets a second preset condition, generating a second positioning matrix and second residual error information aiming at the second positioning matrix according to the first map data and the second map data.
4. The method according to claim 3, wherein when the real-time trajectory information satisfies a first preset condition, generating a second positioning matrix according to the first map data, the second map data, and the first positioning matrix includes:
converting the second map data by using the first positioning matrix;
generating an intermediate positioning matrix according to the converted second map data and the first map data;
and combining the intermediate positioning matrix and the first positioning matrix to obtain a second positioning matrix.
5. The method according to claim 3 or 4, wherein when the target trajectory information collected in real time meets a second preset condition, generating a second positioning matrix according to the first map data and the second map data comprises:
converting the second map data according to the first map data;
and generating a second positioning matrix according to the converted second map data and the first map data.
6. The method according to claim 3, wherein the first preset condition is that a track distance in the real-time acquired target track information is greater than a first preset distance, and the second preset condition is that the track distance in the real-time track information is greater than a second preset distance;
wherein the second preset distance is greater than the first preset distance;
and/or the first preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a first preset number, and the second preset condition is that the number of semantic elements in the real-time acquired target track information is greater than a second preset number;
wherein the second preset number is greater than the first preset number.
7. The method of claim 1, further comprising, prior to the obtaining the latest first positioning matrix and the first residual information for the first positioning matrix:
and carrying out initial matching on the first map data to obtain an initial first positioning matrix and first residual error information aiming at the initial first positioning matrix.
8. An apparatus for data processing, the apparatus comprising:
the first acquisition module is used for acquiring preset first map data;
the second acquisition module is used for acquiring second map data acquired in real time;
a third obtaining module, configured to obtain a first positioning matrix and first residual information for the first positioning matrix;
a generating module, configured to generate a second positioning matrix and second residual information for the second positioning matrix according to the first map data and the second map data;
a first positioning matrix updating module, configured to update the first positioning matrix according to the first residual information, the second residual information, and the second positioning matrix;
and the positioning correction module is used for performing positioning correction by adopting the updated first positioning matrix.
9. A vehicle comprising a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing a method of data processing according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out a method of data processing according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011005875.2A CN112130567A (en) | 2020-09-22 | 2020-09-22 | Data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011005875.2A CN112130567A (en) | 2020-09-22 | 2020-09-22 | Data processing method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112130567A true CN112130567A (en) | 2020-12-25 |
Family
ID=73842579
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011005875.2A Pending CN112130567A (en) | 2020-09-22 | 2020-09-22 | Data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112130567A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107727104A (en) * | 2017-08-16 | 2018-02-23 | 北京极智嘉科技有限公司 | Positioning and map building air navigation aid, apparatus and system while with reference to mark |
CN110000786A (en) * | 2019-04-12 | 2019-07-12 | 珠海市一微半导体有限公司 | A kind of historical map or atlas of view-based access control model robot utilizes method |
CN110288710A (en) * | 2019-06-26 | 2019-09-27 | Oppo广东移动通信有限公司 | A kind of processing method of three-dimensional map, processing unit and terminal device |
CN110704562A (en) * | 2019-09-27 | 2020-01-17 | Oppo广东移动通信有限公司 | Map fusion method and device, equipment and storage medium |
CN111091602A (en) * | 2018-10-24 | 2020-05-01 | 舜宇光学(浙江)研究院有限公司 | SLAM back end optimization method, optimization device and electronic equipment |
-
2020
- 2020-09-22 CN CN202011005875.2A patent/CN112130567A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107727104A (en) * | 2017-08-16 | 2018-02-23 | 北京极智嘉科技有限公司 | Positioning and map building air navigation aid, apparatus and system while with reference to mark |
CN111091602A (en) * | 2018-10-24 | 2020-05-01 | 舜宇光学(浙江)研究院有限公司 | SLAM back end optimization method, optimization device and electronic equipment |
CN110000786A (en) * | 2019-04-12 | 2019-07-12 | 珠海市一微半导体有限公司 | A kind of historical map or atlas of view-based access control model robot utilizes method |
CN110288710A (en) * | 2019-06-26 | 2019-09-27 | Oppo广东移动通信有限公司 | A kind of processing method of three-dimensional map, processing unit and terminal device |
CN110704562A (en) * | 2019-09-27 | 2020-01-17 | Oppo广东移动通信有限公司 | Map fusion method and device, equipment and storage medium |
Non-Patent Citations (1)
Title |
---|
邢伯阳: "ArUco-SLAM:基于ArUco二维码阵列的单目实时建图定位系统", 北京理工大学学报, vol. 40, no. 4, pages 427 - 433 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11320836B2 (en) | Algorithm and infrastructure for robust and efficient vehicle localization | |
CN110388924B (en) | System and method for radar-based vehicle positioning in connection with automatic navigation | |
US10380890B2 (en) | Autonomous vehicle localization based on walsh kernel projection technique | |
CN108820042B (en) | Automatic driving method and device | |
US11188089B2 (en) | Localization for autonomous vehicles using gaussian mixture models | |
CN110533943A (en) | Determine that the lane of automatic driving vehicle changes the method for track | |
US10054945B2 (en) | Method for determining command delays of autonomous vehicles | |
CN110345955A (en) | Perception and planning cooperation frame for automatic Pilot | |
CN109491375A (en) | The path planning based on Driving Scene for automatic driving vehicle | |
CN110239562A (en) | The real-time perception adjustment based on surrounding vehicles behavior of automatic driving vehicle is adjusted with driving | |
CN108981730A (en) | For generating the method and system of reference path for operation automatic driving vehicle | |
CN110096053A (en) | Driving locus generation method, system and machine readable media for automatic driving vehicle | |
CN110389580A (en) | Method for planning the drift correction in the path of automatic driving vehicle | |
CN108891417A (en) | For operating the method and data processing system of automatic driving vehicle | |
JP6776707B2 (en) | Own vehicle position estimation device | |
US11158065B2 (en) | Localization of a mobile unit by means of a multi hypothesis kalman filter method | |
CN109345015B (en) | Method and device for selecting route | |
CN110096054A (en) | For using multiple threads to generate the method and system of the reference line for automatic driving vehicle | |
US10066953B2 (en) | Destination estimating system and destination estimating method | |
JP2019191653A (en) | Map information update system and map information update program | |
CN110389582A (en) | Utilize multiple clue tracking objects | |
US20220017106A1 (en) | Moving object control device, moving object control learning device, and moving object control method | |
CN112130567A (en) | Data processing method and device | |
CN114639263B (en) | Vehicle parking position identification method and device | |
JP2019060669A (en) | Lane change support system and lane change support program |
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 | ||
CB02 | Change of applicant information |
Address after: Room 46, room 406, No. 1, Yichuang street, Zhongxin knowledge city, Huangpu District, Guangzhou, Guangdong 510725 Applicant after: Guangzhou Xiaopeng Automatic Driving Technology Co.,Ltd. Address before: Room 46, room 406, No.1, Yichuang street, Zhongxin knowledge city, Huangpu District, Guangzhou City, Guangdong Province Applicant before: Guangzhou Xiaopeng Automatic Driving Technology Co.,Ltd. |
|
CB02 | Change of applicant information |