CN115309630A - Method, device and equipment for generating automatic driving simulation data and storage medium - Google Patents
Method, device and equipment for generating automatic driving simulation data and storage medium Download PDFInfo
- Publication number
- CN115309630A CN115309630A CN202210751550.1A CN202210751550A CN115309630A CN 115309630 A CN115309630 A CN 115309630A CN 202210751550 A CN202210751550 A CN 202210751550A CN 115309630 A CN115309630 A CN 115309630A
- Authority
- CN
- China
- Prior art keywords
- data
- target
- actual
- simulated
- measurement unit
- 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
- 238000004088 simulation Methods 0.000 title claims abstract description 119
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000005259 measurement Methods 0.000 claims abstract description 107
- 230000001133 acceleration Effects 0.000 claims abstract description 55
- 238000004422 calculation algorithm Methods 0.000 claims description 52
- 238000012795 verification Methods 0.000 claims description 36
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 6
- 238000009434 installation Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3684—Test management for test design, e.g. generating new test cases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3696—Methods or tools to render software testable
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of automatic driving, and discloses a method, a device, equipment and a storage medium for generating automatic driving simulation data, which are used for reducing the generation difficulty of the simulation data. The generation method of the automatic driving simulation data comprises the following steps: acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, wherein the actual observation data comprises actual observation acceleration and actual observation rotation angular velocity; carrying out calibration parameter estimation through actual laser radar odometer data and actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit; and generating simulated laser radar odometer data through the external reference and the pose of the target.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a method, a device, equipment and a storage medium for generating automatic driving simulation data.
Background
With the development of the automatic driving technology, the accuracy of the automatic driving algorithm is higher and higher, and the verification of the accuracy of the automatic driving algorithm usually needs simulation data to support.
The existing simulation data generation technology is usually realized through a simulator, but the simulator is high in cost and difficult to simulate regular simulation data, so that the simulation data generation difficulty is high, and the threshold is high.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for generating automatic driving simulation data, which are used for reducing the generation difficulty of the simulation data.
The invention provides a method for generating automatic driving simulation data in a first aspect, which comprises the following steps:
acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, wherein the actual observation data comprises actual observation acceleration and actual observation rotational angular velocity;
carrying out calibration parameter estimation through the actual laser radar odometer data and the actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit;
and generating simulated laser radar odometer data through the external target parameters and the pose of the target.
Optionally, the target calibration parameters further include: after the target bias data of the inertial measurement unit generates simulated lidar odometry data through the target external parameters and the target pose, the method for generating the automatic driving simulation data further comprises the following steps:
and performing error removal processing on the actual observation data through the target bias data to obtain simulated observation data of the inertial measurement unit.
Optionally, the target offset data includes a target acceleration offset and a target rotation angular velocity offset, and the obtaining the simulated observation data of the inertial measurement unit by performing error removal processing on the actual observation data through the target offset data includes:
calculating the difference between the actual observed acceleration and the target acceleration bias to obtain the simulated observed acceleration of the inertial measurement unit;
calculating a difference value between the actual observation rotation angular velocity and the target rotation angular velocity offset to obtain a simulation observation rotation angular velocity of the inertia measurement unit;
and combining the simulated observation acceleration and the simulated observation rotation angular velocity to obtain simulated observation data of the inertial measurement unit.
Optionally, after the actual observation data is subjected to error removal processing through the target bias data to obtain the simulated observation data of the inertial measurement unit, the method for generating the automatic driving simulation data further includes:
and integrating the simulated observation acceleration and the simulated observation rotation angular velocity to obtain the simulated observation pose of the inertial measurement unit.
Optionally, the estimating calibration parameters through the actual lidar odometry data and the actual observation data to obtain target calibration parameters includes:
calculating calibration parameters of the actual laser radar odometer data, the actual observation acceleration and the actual observation rotation angular velocity through a preset target calibration algorithm to obtain target calibration parameters;
after generating the simulated lidar odometry data through the target external participation and the target pose, the method for generating the automatic driving simulation data further comprises the following steps:
and verifying the target calibration algorithm through the simulated laser radar odometer data and the actual laser radar odometer data to obtain a verification result, wherein the verification result is used for indicating the accuracy of the target calibration algorithm.
Optionally, the verifying the target calibration algorithm by using the simulated lidar odometry data and the actual lidar odometry data to obtain a verification result includes:
error calculation is carried out on the simulated laser radar odometer data and the actual laser radar odometer data to obtain a target error value;
and carrying out accuracy judgment on the target calibration algorithm through the target error value to obtain a verification result.
Optionally, the generating simulated lidar odometry data by the target external parameter and the target pose includes:
and converting the target pose by the target external parameters to obtain simulated laser radar odometer data.
A second aspect of the present invention provides an automatic driving simulation data generation apparatus, including:
the system comprises an acquisition module, a display module and a control module, wherein the acquisition module is used for acquiring actual laser radar odometer data generated in the automatic driving process and actual observation data of an inertia measurement unit, and the actual observation data comprises actual observation acceleration and actual observation rotation angular velocity;
the estimation module is used for estimating calibration parameters through the actual laser radar odometer data and the actual observation data to obtain target calibration parameters, and the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit;
and the generating module is used for generating simulated laser radar odometer data through the target external parameters and the target pose.
Optionally, the target calibration parameters further include: the device for generating the automatic driving simulation data further comprises:
and the error removing module is used for removing errors of the actual observation data through the target bias data to obtain the simulated observation data of the inertial measurement unit.
Optionally, the target offset data includes a target acceleration offset and a target rotation angular velocity offset, and the error removal module is specifically configured to:
calculating the difference between the actual observed acceleration and the target acceleration bias to obtain the simulated observed acceleration of the inertial measurement unit;
calculating a difference value between the actual observation rotation angular velocity and the target rotation angular velocity offset to obtain a simulation observation rotation angular velocity of the inertia measurement unit;
and combining the simulated observation acceleration and the simulated observation rotation angular velocity to obtain simulated observation data of the inertial measurement unit.
Optionally, the device for generating automatic driving simulation data further includes:
and the integration module is used for integrating the simulation observation acceleration and the simulation observation rotation angular velocity to obtain the simulation observation pose of the inertia measurement unit.
Optionally, the estimation module is specifically configured to:
calculating calibration parameters of the actual laser radar odometer data, the actual observation acceleration and the actual observation rotation angular velocity through a preset target calibration algorithm to obtain target calibration parameters;
the generation device of the automatic driving simulation data further includes:
and the verification module is used for verifying the target calibration algorithm through the simulated laser radar odometer data and the actual laser radar odometer data to obtain a verification result, and the verification result is used for indicating the accuracy of the target calibration algorithm.
Optionally, the verification module is specifically configured to:
error calculation is carried out on the simulated laser radar odometer data and the actual laser radar odometer data to obtain a target error value;
and carrying out accuracy judgment on the target calibration algorithm through the target error value to obtain a verification result.
Optionally, the generating module is specifically configured to:
and converting the laser radar coordinate system to the target pose through the target external parameters to obtain simulated laser radar odometer data.
A third aspect of the present invention provides an automatic driving simulation data generation device, including: a memory and at least one processor, the memory having a computer program stored therein; the at least one processor calls the computer program in the memory to cause the generation device of the automated driving simulation data to execute the above-described generation method of the automated driving simulation data.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein a computer program which, when run on a computer, causes the computer to execute the above-described method of generating automatic driving simulation data.
In the technical scheme provided by the invention, actual laser radar odometer data and actual observation data of an inertia measurement unit, which are generated in the automatic driving process, are obtained, wherein the actual observation data comprise actual observation acceleration and actual observation rotational angular velocity; performing calibration parameter estimation through the actual laser radar odometer data and the actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit; and generating simulated laser radar odometer data through the target external parameters and the target pose. In the embodiment of the invention, the calibration parameters are generated through the laser radar odometer data and the observation data of the inertia measurement unit generated in the actual automatic driving process, and the calibration parameters are used as intermediate parameters and can be used for generating simulation automatic driving data such as simulation laser radar odometer data and the like, so that the simulation automatic driving data does not depend on a simulator, and the generation difficulty and the generation cost of the simulation data are greatly reduced.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a method for generating automatic driving simulation data according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of another embodiment of a method for generating automatic driving simulation data according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an embodiment of an apparatus for generating autopilot simulation data in an embodiment of the present invention;
fig. 4 is a schematic diagram of another embodiment of the automatic driving simulation data generation device according to the embodiment of the invention;
fig. 5 is a schematic diagram of an embodiment of an automatic driving simulation data generation device in the embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method, a device, equipment and a storage medium for generating automatic driving simulation data, which are used for reducing the generation difficulty of the simulation data.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be implemented in other sequences than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It is to be understood that the executing subject of the present invention may be a device for generating automatic driving simulation data, and may also be a terminal or a server, which is not limited herein. The embodiment of the present invention is described by taking a terminal as an execution subject.
For convenience of understanding, a detailed flow of an embodiment of the present invention is described below, and referring to fig. 1, an embodiment of a method for generating automatic driving simulation data according to an embodiment of the present invention includes:
101. acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, wherein the actual observation data comprises actual observation acceleration and actual observation rotation angular velocity;
it should be noted that, the actual lidar odometer data and the actual observation data observed by the Inertial Measurement Unit (IMU) are generated during the actual automatic driving process of the autonomous vehicle equipped with the autonomous driving system, wherein the lidar odometer data is used to indicate pose information of the lidar, and may be obtained by registering point clouds scanned by the lidar at different times, or may be obtained by a solid-state lidar odometer, or may be obtained by any other method capable of obtaining the pose of the lidar, and the specific details are not limited herein.
In this embodiment, the inertial measurement unit is a sensor for measuring acceleration and rotation angular velocity, and therefore, the actual observation data of the inertial measurement unit includes the actual observation acceleration and the actual observation rotation angular velocity, and in one embodiment, in addition to the actual observation data of the inertial measurement unit, the actual observation data of any other sensor, such as the actual observation data of a camera, a millimeter wave radar, an ultrasonic wave, and the like, may be obtained for performing calibration parameter estimation with the actual lidar odometer data to obtain target calibration parameters, and then generating simulated lidar odometer data. For example, the actual observation data of the inertial measurement unit may be replaced by the actual observation data of the camera, and then the actual observation data includes the original image captured by the camera and the attribute parameters captured by the original image, such as the capturing time, the image size, and the camera parameters, which are used for the subsequent calibration parameter estimation.
In this embodiment, the actual lidar odometry data generated during the automatic driving process and the actual observation data of the inertial measurement unit, which are acquired by the automatic driving terminal, have time synchronization, that is, the actual lidar odometry data and the actual observation data are data collected at the same time or within the same time period, for example, if the actual lidar odometry data is data generated during the actual automatic driving process within the time period of 00-9.
102. Estimating calibration parameters through actual laser radar odometer data and actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit;
in this embodiment, the target calibration parameters are used as intermediate data generated by the automatic driving simulation data, and the actual calibration parameters, that is, the target calibration parameters, can be generated by performing calibration parameter estimation through actual lidar odometer data and actual observation data generated in the actual automatic driving process, wherein the target external parameters in the target calibration parameters can be used as external parameter true values for generation and conversion of the simulation data. After the target calibration parameters are obtained, automatic driving data of different scenes can be generated in a simulation mode, even the automatic driving data of regular motion which cannot be generated in the actual automatic driving process can be used for verifying the accuracy of an automatic driving algorithm or system under different scenes, and therefore the overall safety of automatic driving is improved.
In one implementation mode, the automatic driving terminal carries out calibration parameter estimation on actual laser radar odometer data and actual observation data through a preset calibration algorithm to obtain target external parameters between the inertia measurement unit and the laser radar, and then converts the laser radar pose corresponding to the actual laser radar odometer data into a coordinate system of an inertia measurement unit through the target external parameters to obtain the target pose of the inertia measurement unit, so that target calibration parameters including the target external parameters between the inertia measurement unit and the laser radar and the target pose of the inertia measurement unit are obtained. According to the embodiment, quick pose conversion can be performed based on external parameters among different sensors, so that the generation efficiency of simulation data is improved, and the difficulty is reduced.
In an embodiment, the target calibration parameters can be generated by any algorithm capable of estimating the calibration parameters, the calibration algorithms corresponding to different sensors are different, and the actual observation data can be replaced by the actual observation data observed by any other sensor except the inertial measurement unit, so that the target calibration parameters can include target external parameters between any other sensor and the laser radar except the inertial measurement unit and target poses of any other sensor.
103. And generating simulated laser radar odometer data through the external reference and the pose of the target.
In one embodiment, after the target external parameter between the inertial measurement unit and the laser radar and the target pose of the inertial measurement unit are involved, the automatic driving terminal integrates the target pose to obtain the simulated observation pose of the inertial measurement unit, and then the simulated observation pose of the inertial measurement unit is converted into a laser radar coordinate system through the target external parameter to obtain simulated laser radar odometer data. The method and the device can generate simulation data close to actual laser radar odometer data for verification of an automatic driving algorithm or system, so that the generation of the automatic driving simulation data does not depend on a simulator, and the generation difficulty is greatly reduced.
In one embodiment, after the automatic driving terminal obtains the target external parameter and the target pose, the automatic driving terminal can generate simulated lidar odometer data for actual observation data of the inertial measurement unit at different moments or different time periods, specifically, the automatic driving terminal obtains a first pose of the inertial measurement unit, wherein the first pose is used for indicating the pose information of the inertial measurement unit observed at the acquisition moment or the acquisition time period different from the actual lidar odometer data, and the first pose is converted into a lidar coordinate system through the target external parameter to obtain the simulated lidar odometer data. For example, the actual lidar odometer data acquisition period is 9. In one possible implementation, in order to acquire simulation data with a larger data volume, the automatic driving terminal can acquire the observation time length of the first pose as much as possible to generate simulation laser radar odometer data with a longer time length, such as that the observation time length of the first pose can be 00-23.
In one embodiment, contrary to the previous embodiment, after obtaining the external target parameter and the position and pose of the target, the automatic driving terminal may further perform simulation data generation of the simulated inertial measurement unit on actual lidar odometer data at different times or time periods, specifically, after step 103, the automatic driving terminal acquires target lidar odometer data, wherein the target lidar odometer data is used for indicating lidar odometer data at a collecting time or a collecting time period different from actual observation data of the inertial measurement unit, and then converts the target lidar odometer data into a coordinate system of the inertial measurement unit through the external target parameter to obtain a target simulated position and pose of the inertial measurement unit, and the target simulation is used for comparing with the actual position and pose to verify the accuracy of the actual position and pose. For example, the acquisition period of the actual observation data of the inertial measurement unit is 9. In this embodiment, after obtaining the target simulation pose of the inertia measurement unit, the autopilot terminal compares the target simulation pose with the actual pose at the same time or within the same time period to obtain a target error value, and ranks the accuracy of the actual pose by the target error value to verify whether the actual pose of the inertia measurement unit is accurate, thereby providing support data for verification of an autopilot algorithm or system and enabling optimization of the autopilot algorithm or system to have directivity.
In an embodiment, after obtaining the target external parameter and the target pose, the automatic driving terminal may further perform simulation generation for different automatic driving algorithms or automatic driving systems, specifically, after step 103, the automatic driving system obtains first lidar odometer data, where the first lidar odometer data is used to indicate a lidar odometer data generated by an automatic driving algorithm or system different from the automatic driving algorithm or system generated by the actual lidar odometer data, and if the actual lidar odometer data is generated by the automatic driving system a, the first lidar odometer data is generated by the automatic driving system B, and the automatic driving terminal converts the first lidar odometer data to an inertial measurement unit coordinate system through the target external parameter to obtain a first simulation pose of the inertial measurement unit, where the first simulation pose is used to compare with an actual pose observed by the inertial measurement unit to verify the accuracy of the actual pose, and further verify the accuracy of the actual pose generation algorithm or system. For example, the actual pose is generated by the autopilot system a, and then the accuracy of the autopilot system a can be verified by comparing the first simulation pose with the actual pose. The method and the system can verify the precision of different automatic driving algorithms or systems through simulation data generated by the different automatic driving algorithms or systems so as to ensure that the automatic driving algorithms or systems can meet the precision requirement and ensure the safety of automatic driving.
In one embodiment, after the autopilot terminal obtains the target external parameters and the target pose, any autopilot data that may be generated by the target external parameters may be simulated, including but not limited to: the inertial measurement unit and the laser radar respectively correspond to speed, acceleration, rotation angular velocity, track, installation parameters, sensor parameters and the like, wherein the installation parameters comprise parameters such as installation height and installation position, and the sensor parameters comprise parameters such as sensor precision and internal parameters. The same applies when the inertial measurement unit is replaced by any other sensor, and the specific description is not limited herein.
In one embodiment, after the simulated lidar odometry data are generated, the automatic driving terminal compares the simulated lidar odometry data with actual lidar odometry data to obtain a target deviation value, and then performs precision grading on an automatic driving algorithm or system through the target deviation value to obtain target precision grade information. The present embodiment can verify the accuracy of an autonomous driving algorithm or system through simulated autonomous driving data, thereby improving safety of autonomous driving.
In the embodiment of the invention, the calibration parameters are generated through the laser radar odometer data and the observation data of the inertia measurement unit generated in the actual automatic driving process, and the calibration parameters are used as intermediate parameters and can be used for generating simulation automatic driving data such as simulation laser radar odometer data and the like, so that the simulation automatic driving data does not depend on a simulator, and the generation difficulty and the generation cost of the simulation data are greatly reduced.
Referring to fig. 2, another embodiment of the method for generating automatic driving simulation data according to the embodiment of the present invention includes:
201. acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, wherein the actual observation data comprises actual observation acceleration and actual observation rotation angular velocity;
the execution process of step 201 is similar to the execution process of step 101, and detailed description thereof is omitted here.
202. Estimating calibration parameters through actual laser radar odometer data and actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit; the target calibration parameters further comprise: target bias data for the inertial measurement unit; the target offset data includes a target acceleration offset and a target rotational angular velocity offset;
specifically, step 202 includes: calculating calibration parameters of actual laser radar odometer data, actual observation acceleration and actual observation rotation angular velocity through a preset target calibration algorithm to obtain target calibration parameters; after step 203, the method further comprises: and verifying the target calibration algorithm through the simulated laser radar odometer data and the actual laser radar odometer data to obtain a verification result, wherein the verification result is used for indicating the accuracy of the target calibration algorithm.
In the embodiment, the automatic driving terminal calculates calibration parameters of actual laser radar odometer data, actual observation acceleration and actual observation rotation angular velocity through a preset target calibration algorithm, and after target calibration parameters are obtained, the automatic driving terminal carries out precision verification on the target calibration algorithm through generated simulated laser radar odometer data and actual laser radar odometer data in a reverse mode to obtain a verification result, wherein the verification result is used for indicating the accuracy of the target calibration algorithm. Specifically, the automatic driving terminal calculates deviation values of the simulated laser radar odometer data and the actual laser radar odometer data to obtain a target deviation value, and then performs precision verification on a target calibration algorithm through the target deviation value to obtain a verification result. According to the method and the system, the accuracy of the actual automatic driving data can be reversely deduced through the simulation automatic driving data, so that the difficulty of automatic driving data verification is reduced, and the efficiency is improved.
Further, the target calibration algorithm is verified through the simulated laser radar odometer data and the actual laser radar odometer data, and a verification result is obtained, wherein the verification result comprises the following steps: carrying out error calculation on the simulated laser radar odometry data and the actual laser radar odometry data to obtain a target error value; and carrying out accuracy judgment on the target calibration algorithm through the target error value to obtain a verification result.
In the embodiment, the automatic driving terminal obtains a target error value by calculating an error between the simulated laser radar odometer data and the actual laser radar odometer data, compares the target error value with a preset error value interval to perform accuracy grading on the target calibration algorithm to obtain a verification result, and the verification result is used for indicating accuracy grade information of the target calibration algorithm.
203. Generating simulated laser radar odometer data through target external parameters and target pose;
specifically, step 203 includes: and converting the target pose by the target external parameters to obtain simulated laser radar odometer data.
In one implementation mode, after the target external parameters and the target pose are obtained, the automatic driving terminal projects the target pose to a laser radar coordinate system through the target external parameters to obtain simulated laser radar odometer data of the laser radar coordinate system, so that the effect and the purpose of automatic driving data simulation are achieved, the difficulty of simulation data generation is reduced, the efficiency of simulation data generation is improved, and the efficiency of automatic driving algorithm or system verification is improved.
204. And performing error removal processing on the actual observation data through the target bias data to obtain the simulated observation data of the inertial measurement unit.
In the embodiment, the actual observation data observed by the inertia measurement unit usually contains offset and noise, and when the performance analysis of the automatic driving algorithm under some ideal states is performed, simulation data which does not contain offset and noise is often needed, therefore, the automatic driving terminal performs the de-offset and de-noise processing on the actual observation data through the target offset data to obtain the simulation observation data of the inertia measurement unit, and the simulation observation data does not contain offset and noise.
Specifically, step 204 includes: calculating the difference between the actual observed acceleration and the target acceleration bias to obtain the simulated observed acceleration of the inertia measurement unit; calculating the difference between the actual observation rotation angular velocity and the target rotation angular velocity offset to obtain the simulated observation rotation angular velocity of the inertia measurement unit; and combining the simulated observation acceleration and the simulated observation rotation angular velocity to obtain the simulated observation data of the inertia measurement unit.
In the embodiment, after the target acceleration offset and the target rotation angular velocity offset are obtained, the automatic driving terminal subtracts the target acceleration offset from the actual observation acceleration to obtain the simulated observation acceleration of the inertia measurement unit, and then subtracts the target rotation angular velocity offset from the actual observation rotation angular velocity to obtain the simulated observation rotation angular velocity of the inertia measurement unit, so that the simulated observation data including the simulated observation acceleration and the simulated observation rotation angular velocity is obtained. According to the embodiment, the actual observation data of the inertia measurement unit can be subjected to bias removal processing to obtain the non-biased simulated observation data, so that the generation difficulty of the simulation data is reduced, and the efficiency of an automatic driving algorithm or system verification is improved.
Further, after step 204, the method further includes: and integrating the simulated observation acceleration and the simulated observation rotation angular velocity to obtain the simulated observation pose of the inertial measurement unit.
In the embodiment, the automatic driving terminal can also integrate the simulated observation acceleration and the simulated observation rotation angular velocity to obtain a simulated observation pose, the simulated observation pose can be used for verifying the accuracy of the actual observation pose, can also be used for converting the pose of a laser radar, and can also be used for verifying and calibrating an automatic driving algorithm or system, so that the generation difficulty of the simulated data is reduced, and the verification efficiency of the automatic driving algorithm or system is improved.
In the embodiment of the invention, the calibration parameters are generated through the laser radar odometer data and the observation data of the inertia measurement unit generated in the actual automatic driving process, and the calibration parameters are used as intermediate parameters and can be used for generating simulation automatic driving data such as simulation laser radar odometer data and simulation observation data of the inertia measurement unit, so that the simulation automatic driving data does not depend on a simulator, and the generation difficulty and the generation cost of the simulation data are greatly reduced.
With reference to fig. 3, the method for generating the automatic driving simulation data according to the embodiment of the present invention is described above, and a device for generating the automatic driving simulation data according to the embodiment of the present invention is described below, where an embodiment of the device for generating the automatic driving simulation data according to the embodiment of the present invention includes:
an obtaining module 301, configured to obtain actual lidar odometer data generated in an automatic driving process and actual observation data of an inertial measurement unit, where the actual observation data includes an actual observation acceleration and an actual observation rotational angular velocity;
an estimation module 302, configured to perform calibration parameter estimation according to the actual lidar odometer data and the actual observation data to obtain target calibration parameters, where the target calibration parameters include a target external parameter between the inertial measurement unit and the lidar and a target pose of the inertial measurement unit;
and the generating module 303 is configured to generate simulated lidar odometry data by the external reference of the target and the pose of the target.
In the embodiment of the invention, the calibration parameters are generated through the laser radar odometer data and the observation data of the inertia measurement unit generated in the actual automatic driving process, and the calibration parameters are used as intermediate parameters and can be used for generating simulation automatic driving data such as simulation laser radar odometer data and the like, so that the simulation automatic driving data does not depend on a simulator, and the generation difficulty and the generation cost of the simulation data are greatly reduced.
Referring to fig. 4, another embodiment of the device for generating automatic driving simulation data according to the embodiment of the present invention includes:
an obtaining module 301, configured to obtain actual lidar odometer data generated in an automatic driving process and actual observation data of an inertial measurement unit, where the actual observation data includes an actual observation acceleration and an actual observation rotational angular velocity;
an estimation module 302, configured to perform calibration parameter estimation on the actual lidar odometer data and the actual observation data to obtain target calibration parameters, where the target calibration parameters include target external parameters between the inertial measurement unit and the lidar and a target pose of the inertial measurement unit;
and the generating module 303 is configured to generate simulated lidar odometry data by the external reference of the target and the pose of the target.
Optionally, the target calibration parameters further include: target bias data of the inertial measurement unit, the generation device of the automatic driving simulation data further includes:
an error removing module 304, configured to perform error removing processing on the actual observation data through the target bias data to obtain simulated observation data of the inertial measurement unit.
Optionally, the target offset data includes a target acceleration offset and a target rotation angular velocity offset, and the error removing module 304 is specifically configured to:
calculating the difference between the actual observed acceleration and the target acceleration bias to obtain the simulated observed acceleration of the inertial measurement unit;
calculating a difference value between the actual observation rotation angular velocity and the target rotation angular velocity offset to obtain a simulated observation rotation angular velocity of the inertial measurement unit;
and combining the simulated observation acceleration and the simulated observation rotation angular velocity to obtain simulated observation data of the inertial measurement unit.
Optionally, the device for generating automatic driving simulation data further includes:
and an integrating module 305, configured to integrate the simulated observation acceleration and the simulated observation rotational angular velocity to obtain a simulated observation pose of the inertial measurement unit.
Optionally, the estimating module 302 is specifically configured to:
calculating calibration parameters of the actual laser radar odometer data, the actual observation acceleration and the actual observation rotation angular velocity through a preset target calibration algorithm to obtain target calibration parameters;
the generation device of the automatic driving simulation data further includes:
a verification module 306, configured to verify the target calibration algorithm according to the simulated lidar odometry data and the actual lidar odometry data to obtain a verification result, where the verification result is used to indicate accuracy of the target calibration algorithm.
Optionally, the verification module 306 is specifically configured to:
error calculation is carried out on the simulated laser radar odometer data and the actual laser radar odometer data to obtain a target error value;
and carrying out accuracy judgment on the target calibration algorithm through the target error value to obtain a verification result.
Optionally, the generating module 303 is specifically configured to:
and converting the laser radar coordinate system to the target pose through the target external parameters to obtain simulated laser radar odometer data.
In the embodiment of the invention, the calibration parameters are generated through the laser radar odometer data and the observation data of the inertia measurement unit generated in the actual automatic driving process, and the calibration parameters are used as intermediate parameters and can be used for generating simulation automatic driving data such as simulation laser radar odometer data and simulation observation data of the inertia measurement unit, so that the simulation automatic driving data does not depend on a simulator, and the generation difficulty and the generation cost of the simulation data are greatly reduced.
Fig. 3 and 4 describe the generation device of the automated driving simulation data in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the generation device of the automated driving simulation data in the embodiment of the present invention is described in detail from the perspective of the hardware processing.
Fig. 5 is a schematic structural diagram of an automatic driving simulation data generation device according to an embodiment of the present invention, where the automatic driving simulation data generation device 500 may generate relatively large differences due to different configurations or performances, and may include one or more processors (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing applications 533 or data 532. Memory 520 and storage media 530 may be, among other things, transient or persistent storage. The program stored on the storage medium 530 may include one or more modules (not shown), each of which may include a series of computer program operations in the generation device 500 for the automated driving simulation data. Still further, the processor 510 may be configured to communicate with the storage medium 530 to execute a series of computer program operations in the storage medium 530 on the generation device 500 of the automated driving simulation data.
The autopilot simulation data generation apparatus 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input-output interfaces 560, and/or one or more operating systems 531, such as Windows Server, mac OS X, unix, linux, freeBSD, and the like. Those skilled in the art will appreciate that the configuration of the generation device of autopilot simulation data shown in fig. 5 does not constitute a limitation of the generation device of autopilot simulation data and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
The present invention also provides a computer device, which includes a memory and a processor, wherein the memory stores a computer readable computer program, and when the computer readable computer program is executed by the processor, the processor executes the steps of the method for generating automatic driving simulation data in the above embodiments.
The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium, and which may also be a volatile computer-readable storage medium, having stored therein a computer program, which, when run on a computer, causes the computer to execute the steps of the method for generating autopilot simulation data.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes several computer programs to enable a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a portable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of generating automated driving simulation data, the method comprising:
acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, wherein the actual observation data comprises actual observation acceleration and actual observation rotational angular velocity;
carrying out calibration parameter estimation through the actual laser radar odometer data and the actual observation data to obtain target calibration parameters, wherein the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit;
and generating simulated laser radar odometer data through the target external parameters and the target pose.
2. The method of generating autopilot simulation data according to claim 1 wherein the target calibration parameters further include: after the target bias data of the inertial measurement unit generates simulated lidar odometry data through the target external parameters and the target pose, the method for generating the automatic driving simulation data further comprises the following steps:
and performing error removal processing on the actual observation data through the target bias data to obtain simulated observation data of the inertial measurement unit.
3. The method for generating the automatic driving simulation data according to claim 2, wherein the target offset data includes a target acceleration offset and a target rotational angular velocity offset, and the de-error processing of the actual observation data by the target offset data to obtain the simulated observation data of the inertial measurement unit includes:
calculating the difference between the actual observed acceleration and the target acceleration bias to obtain the simulated observed acceleration of the inertial measurement unit;
calculating a difference value between the actual observation rotation angular velocity and the target rotation angular velocity offset to obtain a simulated observation rotation angular velocity of the inertial measurement unit;
and combining the simulated observation acceleration and the simulated observation rotation angular velocity to obtain simulated observation data of the inertial measurement unit.
4. The method for generating automatic driving simulation data according to claim 3, wherein after the actual observation data is subjected to the error removal processing by the target bias data to obtain the simulated observation data of the inertia measurement unit, the method for generating automatic driving simulation data further comprises:
and integrating the simulated observation acceleration and the simulated observation rotation angular velocity to obtain the simulated observation pose of the inertial measurement unit.
5. The method for generating automatic driving simulation data according to claim 1, wherein the estimating calibration parameters through the actual lidar odometry data and the actual observation data to obtain target calibration parameters comprises:
calculating calibration parameters of the actual laser radar odometer data, the actual observation acceleration and the actual observation rotation angular velocity through a preset target calibration algorithm to obtain target calibration parameters;
after generating the simulated lidar odometry data through the target external participation and the target pose, the method for generating the automatic driving simulation data further comprises the following steps:
and verifying the target calibration algorithm through the simulated laser radar odometry data and the actual laser radar odometry data to obtain a verification result, wherein the verification result is used for indicating the accuracy of the target calibration algorithm.
6. The method for generating automatic driving simulation data according to claim 5, wherein the verifying the target calibration algorithm by the simulated lidar odometry data and the actual lidar odometry data to obtain a verification result comprises:
error calculation is carried out on the simulated laser radar odometer data and the actual laser radar odometer data to obtain a target error value;
and carrying out accuracy judgment on the target calibration algorithm through the target error value to obtain a verification result.
7. The method of generating autopilot simulation data according to claim 1 wherein said generating simulated lidar odometry data by the target external participation and the target pose comprises:
and converting the target pose by the target external parameters to obtain simulated laser radar odometer data.
8. An automatic driving simulation data generation device, characterized by comprising:
the system comprises an acquisition module, a display module and a control module, wherein the acquisition module is used for acquiring actual laser radar odometer data and actual observation data of an inertia measurement unit, which are generated in the automatic driving process, and the actual observation data comprise actual observation acceleration and actual observation rotational angular velocity;
the estimation module is used for carrying out calibration parameter estimation through the actual laser radar odometer data and the actual observation data to obtain target calibration parameters, and the target calibration parameters comprise target external parameters between the inertial measurement unit and the laser radar and target poses of the inertial measurement unit;
and the generating module is used for generating simulated laser radar odometer data through the target external parameters and the target pose.
9. An automatic driving simulation data generation device, characterized by comprising: a memory and at least one processor, the memory having stored therein a computer program;
the at least one processor invokes the computer program in the memory to cause the generation apparatus of autopilot simulation data to perform the generation method of autopilot simulation data as set forth in any one of claims 1-7.
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 generating autopilot simulation data according to one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210751550.1A CN115309630A (en) | 2022-06-28 | 2022-06-28 | Method, device and equipment for generating automatic driving simulation data and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210751550.1A CN115309630A (en) | 2022-06-28 | 2022-06-28 | Method, device and equipment for generating automatic driving simulation data and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115309630A true CN115309630A (en) | 2022-11-08 |
Family
ID=83854701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210751550.1A Pending CN115309630A (en) | 2022-06-28 | 2022-06-28 | Method, device and equipment for generating automatic driving simulation data and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115309630A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672049A (en) * | 2024-01-31 | 2024-03-08 | 深圳风向标教育资源股份有限公司 | Intelligent networking automobile integrated sensor teaching experiment device |
-
2022
- 2022-06-28 CN CN202210751550.1A patent/CN115309630A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672049A (en) * | 2024-01-31 | 2024-03-08 | 深圳风向标教育资源股份有限公司 | Intelligent networking automobile integrated sensor teaching experiment device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2959315B1 (en) | Generation of 3d models of an environment | |
| Prescott et al. | Line-based correction of radial lens distortion | |
| WO2020253260A1 (en) | Time synchronization processing method, electronic apparatus, and storage medium | |
| CN112285676B (en) | Laser radar and IMU external parameter calibration method and device | |
| CN110470333B (en) | Calibration method and device of sensor parameters, storage medium and electronic device | |
| CN110501036A (en) | The calibration inspection method and device of sensor parameters | |
| EP2922023A1 (en) | Three-dimensional object recognition device and three-dimensional object recognition method | |
| CN109781163B (en) | Calibration parameter validity checking method, device, equipment and storage medium | |
| KR101890612B1 (en) | Method and apparatus for detecting object using adaptive roi and classifier | |
| CN115309630A (en) | Method, device and equipment for generating automatic driving simulation data and storage medium | |
| CN111998870B (en) | Calibration method and device of camera inertial navigation system | |
| CN115728753A (en) | External parameter calibration method and device for laser radar and integrated navigation and intelligent vehicle | |
| CN117392241B (en) | Sensor calibration method and device in automatic driving and electronic equipment | |
| CN113137973A (en) | Image semantic feature point truth value determining method and device | |
| CN114821497A (en) | Method, device and equipment for determining position of target object and storage medium | |
| Irmisch et al. | Simulation framework for a visual-inertial navigation system | |
| CN111812668A (en) | Winding inspection device, positioning method thereof and storage medium | |
| CN115082290A (en) | Projection method, device and equipment of laser radar point cloud and storage medium | |
| CN113256728A (en) | IMU equipment parameter calibration method and device, storage medium and electronic device | |
| CN117367458A (en) | Map accuracy verification method and device, terminal equipment and medium | |
| CN114879168A (en) | Laser radar and IMU calibration method and system | |
| CN114034303A (en) | Moving target object positioning method and device based on Kalman filtering | |
| CN111982108A (en) | Mobile robot positioning method, device, equipment and storage medium | |
| CN117935096B (en) | Structural vibration monitoring and correcting method, device and storage medium based on unmanned aerial vehicle | |
| CN115265581B (en) | Calibration parameter determining method of laser radar and inertial measurement unit and related equipment |
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 |