WO2021102980A1 - ***误差标定方法、装置和计算机可读存储介质 - Google Patents

***误差标定方法、装置和计算机可读存储介质 Download PDF

Info

Publication number
WO2021102980A1
WO2021102980A1 PCT/CN2019/122145 CN2019122145W WO2021102980A1 WO 2021102980 A1 WO2021102980 A1 WO 2021102980A1 CN 2019122145 W CN2019122145 W CN 2019122145W WO 2021102980 A1 WO2021102980 A1 WO 2021102980A1
Authority
WO
WIPO (PCT)
Prior art keywords
error
scanning device
direction vector
light emitted
scanning
Prior art date
Application number
PCT/CN2019/122145
Other languages
English (en)
French (fr)
Inventor
吴特思
陈涵
许友
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to PCT/CN2019/122145 priority Critical patent/WO2021102980A1/zh
Publication of WO2021102980A1 publication Critical patent/WO2021102980A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/48Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
    • G01S7/497Means for monitoring or calibrating

Definitions

  • This application generally relates to the field of laser detection technology, and more specifically relates to a system error calibration method, device, and computer-readable storage medium.
  • the optical system of the current lidar is mostly composed of prisms and plane mirrors, which form a specific scanning mode through refraction, reflection and device rotation.
  • prisms and mirrors may be stationary or rotate around a specific axis.
  • a system error calibration method includes: acquiring at least two sets of system input parameters of the scanning device when the scanning device scans at least two reference points, the reference points being the scanning view A reference point with a known direction vector in the field, and the system input parameters include the direction vector of the light emitted by the laser light source of the scanning device and the axis rotation parameters of one or more components in the scanning device; The direction vectors of at least two reference points, the at least two sets of system input parameters, and a preset set of error terms determine the systematic error of the scanning device.
  • a system error calibration method comprising: obtaining a first direction vector from a scanning device, the first direction vector being calculated by the scanning device; the first direction vector Is the direction vector in the scanning field of view of the light emitted by the scanning device when the scanning device scans to the reference point in the scanning field of view; the second direction vector is obtained from the measuring instrument, and the second direction vector is the measurement The direction vector of the reference point in the scanning field of view measured by the instrument; and the system error compensation value in the target area corresponding to the reference point is determined according to the first direction vector and the second direction vector.
  • an error calibration device of a scanning device includes a memory and a processor, the memory stores a computer program run by the processor, and the processor is used for Execute when the calculator program is running: Obtain at least two sets of system input parameters of the scanning device when the scanning device scans at least two reference points, where the reference point is a reference point with a known direction vector in the scanning field of view, and the system
  • the input parameters include the direction vector of the light emitted by the laser light source of the scanning device and the axis rotation parameters of one or more components in the scanning device; according to the direction vectors of the at least two reference points, the at least two reference points
  • a set of system input parameters and a preset set of error items determine the system error of the scanning device.
  • an error calibration device of a scanning device includes a memory and a processor, and a computer program run by the processor is stored in the memory. Executed when the processor is running: obtain a first direction vector from a scanning device, the first direction vector is calculated by the scanning device; the first direction vector is scanned by the scanning device into the scanning field of view The reference point is the direction vector of the light emitted by the scanning device in the scanning field of view; the second direction vector is obtained from the measuring instrument, and the second direction vector is the reference point measured by the measuring instrument in the scanning view.
  • the direction vector in the field; the system error compensation value in the target area corresponding to the reference point is determined according to the first direction vector and the second direction vector.
  • a computer-readable storage medium having a computer program stored on the computer-readable storage medium, and when the computer program is running on a processor, the processor executes the above-mentioned error calibration method .
  • the system error calibration method, device, and computer-readable storage medium determine the scan based on the direction vectors of the at least two reference points, the at least two sets of system input parameters, and a preset set of error terms.
  • the system error of the device or the system error compensation value in the target area corresponding to the reference point is determined according to the first direction vector and the second direction vector, so as to obtain a universal, high-efficiency, fully automatic, and high-precision system. , It can support the error calibration program with the advantages of mass production and simple calibration environment.
  • Fig. 1 shows a schematic diagram of a lidar structure according to an embodiment of the present application.
  • Fig. 2 shows a schematic diagram of the optical path of the double prism structure lidar in the first case according to an embodiment of the present application.
  • Fig. 3 shows a schematic diagram of the optical path of the double-prism structure laser radar in the second case according to the embodiment of the present application.
  • FIG. 4 shows a schematic diagram of the laser radar error of the double prism structure according to an embodiment of the present application.
  • Fig. 5 shows a schematic flowchart of a first error calibration method according to an embodiment of the present application.
  • Fig. 6 shows a schematic diagram of FOV gridding according to an embodiment of the present application.
  • Fig. 7 shows a schematic flowchart of a second error calibration method according to an embodiment of the present application.
  • Fig. 8 shows a framework diagram of an application scenario according to an embodiment of the present application.
  • Fig. 9 shows a schematic diagram of an error calibration device provided according to another aspect of the present application.
  • Radar is an active sensing sensor. Its basic principle is that the radar actively emits light pulse signals. The signal encounters objects in the propagation path and produces reflected echoes. Part of the echo signals are captured by the radar's receiving module, which is transmitted and received by the radar. The time difference between the signals calculates the depth of the measured object from the radar. According to the radar's transmitting direction, the angle information of the measured object relative to the radar is obtained. Based on the depth and angle information, the three-dimensional space position of the measured object relative to the radar can be calculated.
  • the prism and reflector are important components of the internal optical system of the mechanical lidar.
  • the prism and the reflector may be stationary or rotate around a specific axis of rotation, which are formed by the combination of reflection and refraction under the dynamic change of the reflector and the prism Specific scan mode.
  • the lidar structure shown in FIG. 1 As shown in FIG. 1, after the laser is excited, it first passes through an axis-rotating prism, then passes through a stationary mirror and finally exits. It may include two prisms that rotate around an axis, or replace the stationary mirror with a stationary prism or a rotating prism, for example, it may also be replaced with a MID-40 model lidar.
  • the laser light passes through two prisms rotating around the axis in sequence after being excited, and finally exits.
  • the types, arrangement sequence, and number of lenses in Figures 1 and 2 are only examples. In practical applications, the detection device may include other types and numbers of lenses.
  • the detection device in this application includes but is not limited to radar, and the radar in this application includes but is not limited to: lidar, electromagnetic wave radar, millimeter wave radar, or ultrasonic radar.
  • FIG. 2 shows a schematic diagram of an optical path in the second case according to an embodiment of the present application.
  • the direction vector of the light emitted by the laser light source of the scanning device after passing through the prism 1 and prism 2 the direction of the light emitted by the scanning device in the scanning field of view
  • the vector is consistent with the calculated value of the scanning device, that is, the A-space point is the point actually emitted by the scanning device, and it is also the solution point of the scanning device under ideal conditions, and the two are the same point.
  • the direction vector of the light emitted by the laser light source of the scanning device is estimated using the design nominal value.
  • the design nominal value refers to the value of the direction vector of the light rays emitted by the scanning device when the optical component is assumed to be in the design posture.
  • the A space point in Figure 3 is the actual exit point of the scanning device, and the A'space point is the exit point calculated by the scanning device, which corresponds to the design nominal value.
  • the A space point and the A'space point are not the same point. There are errors.
  • the reason for the difference between the A space point and the A'space point is that there is an installation error between the prism 1 and the prism 2.
  • the ideal installation position of the prism 1 is that the right side is perpendicular to the optical axis.
  • the ideal installation position of prism 2 is the left side perpendicular to the optical axis, as shown by the dotted line of prism 2.
  • prism 1 and prism 2 are installed as solid lines The position is shown, which leads to the difference between the A space point and the A'space point actually emitted.
  • FIG. 3 only shows the situation where there is an installation error of the prism 1 and the prism 2 in the optical axis direction
  • FIG. 4 shows a schematic diagram of the laser radar error of the double prism structure according to an embodiment of the present application.
  • the solid line in the figure shows the actual pose of the prism
  • the dotted line in the figure shows
  • the actual normal vector of the 11# surface is like
  • the actual normal vector of face 12# is As shown
  • the long vector on the left shows the actual vector of the prism
  • the actual angle deviation of the rotation axis is shown.
  • the possible errors include the following error items: the rotation axis normal axis deviation ⁇ y1 , ⁇ z1 , the prism/plane mirror initial normal deviation ⁇ y1 , ⁇ z1 , the axis rotation angle measurement deviation ⁇ 1.
  • the components in the Y-axis and Z-axis directions, the directions of the Y-axis and Z-axis are as shown in Figure 4.
  • the initial normal deviation of the prism/plane mirror ⁇ y1 , ⁇ z1 is and / or The components of the long vector on the left in the Y-axis and Z-axis directions, similarly, the second component: prism 2, in the direction perpendicular to the paper, including 21# surface and 22# surface, the solid line in the figure shows the prism
  • the solid line in the figure shows the prism
  • the actual pose of the prism the dotted line in the figure shows the design pose of the prism
  • the actual normal direction of face 21# is as As shown
  • the actual normal vector of surface 22# is As shown
  • the long vector on the left shows the actual vector of the prism
  • the actual angle deviation of the rotation axis is shown.
  • the possible errors include the following error items: the rotation axis normal axis deviation ⁇ y2 , ⁇ z2 , the prism/plane mirror initial normal deviation ⁇ y2 , ⁇ z2 , the axis rotation angle measurement deviation ⁇ 2.
  • the rotation axis direction of the deviation ⁇ y2, ⁇ z2 is The components in the Y-axis and Z-axis directions, the directions of the Y-axis and Z-axis are as shown in Figure 4.
  • the initial normal deviation of the prism/plane mirror ⁇ y2 , ⁇ z2 is and / or The components of the long vector on the left in the Y and Z directions.
  • the optical system consists of a prism and/or a plane mirror, which forms a specific scanning mode through refraction, reflection and device rotation.
  • This type of optical system is relatively complicated, involves many parameters and has a high degree of coupling. It is quite difficult to solve the error calibration problem of this type of system.
  • This application provides a calibration scheme from two perspectives to obtain an efficient, fully automatic, and simple calibration environment. Effect.
  • the frame diagram includes a scanning device, a measuring instrument and an error calibration device.
  • the scanning device is exemplarily a double-prism structured radar, and the detection signal emitted by it is a laser.
  • Its lidar can collect its own data, including point cloud data, the direction vector of the laser light source at each point when it is emitted, the rotation angle of each component inside the lidar, and so on.
  • the scanning device sends the obtained data to the error calibration device, wherein the data of the scanning device can be selected to be sent to the error calibration device in real time, or it can be selected to send all the data to the error calibration device after the scanning is completed.
  • the scanning device can also be replaced with electromagnetic wave radar, millimeter wave radar or ultrasonic radar. When it is replaced with another radar, the corresponding detection signal also changes.
  • the measuring instrument is located in the scanning field of view of the scanning device, which can accurately measure the direction vector of the marker (or marking point, reference point, etc.) in the space in the scanning field of view.
  • the measuring instrument can calculate the direction vector of the marker (or marker point) in the space in the scanning space based on the space coordinate system of the lidar.
  • the error calibration device can receive the data of the scanning device and the measuring instrument, and process the received data to perform error calibration.
  • the specific error calibration process will be described in detail in conjunction with the following drawings.
  • the error calibration method 100 may include the following steps:
  • step S110 acquire at least two sets of system input parameters of the scanning device when the scanning device scans at least two reference points, where the reference points are reference points with known direction vectors in the scanning field of view, and the system input parameters include The direction vector of the light emitted by the laser light source of the scanning device and the axis rotation parameters of one or more components in the scanning device.
  • step S120 the system error of the scanning device is determined according to the direction vectors of the at least two reference points, the at least two sets of system input parameters, and a preset set of error items.
  • a scanning device may be used to transmit a light pulse signal to the object to be measured, and receive a reflected pulse signal corresponding to the light pulse signal.
  • the detection device includes, but is not limited to, laser radar, electromagnetic wave radar, millimeter wave radar, or ultrasonic radar.
  • At least two sets of system input parameters of the scanning device when the scanning device scans at least two reference points are acquired, and the reference points are reference points with known direction vectors in the scanning field of view, so
  • the system input parameters include the direction vector of the light emitted by the laser light source of the scanning device and the axis rotation parameters of one or more components in the scanning device.
  • the solution of the A'space point can be obtained by the following model calculations. First, establish the model, And estimate the exit angle under ideal conditions: the laser light follows the reflection law when passing through the reflection plane:
  • Is the exit unit vector Is the incident unit vector
  • a prism needs to consider two refractions, that is, there are two refractive surfaces.
  • is always zero.
  • ⁇ , ⁇ , ⁇ are vector deviations
  • ⁇ , n are numerical deviations.
  • the vector deviation can be decomposed into two directions of rotation around the y axis and around the z axis ⁇ y , ⁇ z , there are:
  • Figure 4 is a schematic diagram of error terms
  • Table 1 is a summary of the error terms.
  • the measuring instrument in step 110, in the scanning field of view, accurately detects the position of the scanning laser.
  • the direction vector of the reference point is known (For example, the direction vector of the reference point is pre-measured by the measuring instrument).
  • the scanning device scans to a reference point, acquire at least two sets of system input parameters of the scanning device when the scanning device scans to at least two reference points.
  • the system input parameters include the direction vector of the light emitted by the laser light source of the scanning device and
  • the axis rotation parameters of one or more components in the scanning device the light emitted by the laser light source of the scanning device is used as the incident light of the optical component, and its direction vector represents the incident direction vector V in , one or more of the scanning devices
  • the axis rotation parameters of the plurality of components are, for example, ⁇ 1 , ⁇ 2 ... ⁇ n .
  • step 120 the system error of the scanning device is determined according to the direction vectors of the at least two reference points, the at least two sets of system input parameters, and a preset set of error items.
  • step 110 at least the direction vectors of two reference points with known truth values are obtained In each In the equation of And ⁇ 1 , ⁇ 2 ?? ⁇ n are known, and multiple The equations are listed as a system of equations. In theory, the number of equations in the equation system is greater than or equal to the number of unknown terms, and the value of the error term ⁇ can be solved.
  • the various error items summarized in Table 1 in the scanning device can be calibrated.
  • the error value of each component can be obtained separately, and therefore, it has at least an accurate and efficient calibration effect.
  • step 110 said determining the system error of the scanning device according to the direction vectors of the at least two reference points, the at least two sets of system input parameters and a preset set of error terms, The method includes: substituting the direction vectors of the at least two reference points, the at least two sets of system input parameters, and the set of error terms into a preset error model to calculate the system error of the scanning device.
  • the error is calculated according to the preset error model, and the error value can be obtained concisely and effectively, making the calibration environment simple.
  • the preset error model is a functional relationship equation based on the laws of optics. among them, It is obtained based on the reflection law and Snell's law.
  • the functional relationship equation based on the optical law can obtain accurate error values, and ensure the validity and accuracy of the error calibration method and device. Moreover, the above functional relationship equation can be applied to scanning devices containing various optical components. For scanning devices containing different components, only the functional relationship equations need to be replaced according to the optical law to obtain the error value of the scanning device. Therefore, the general Strong adaptability.
  • the solution of the functional relationship equation adopts a nonlinear fitting method.
  • ⁇ 1 , ⁇ 2 & ⁇ n are known, and multiple The equations are listed as a set of equations.
  • the solution of the function relationship f is a nonlinear least squares problem, and it is difficult to solve it directly. Therefore, a set of optimized parameters is obtained by selecting a nonlinear fitting method.
  • the non-linear least squares algorithm that can be used is the Levenbert-Marquardt algorithm (referred to as the LM algorithm).
  • the LM algorithm uses the gradient to find the maximum value of the function. The specific steps will not be repeated here.
  • a set of error parameters obtained by the LM algorithm is used as the final calibration result.
  • the method of solving the equations can adopt the nonlinear fitting method, of course, other methods can also be adopted. There are many methods to solve the nonlinear least squares problem, which can be selected according to the actual situation. This application does not deal with the method of solving the functional relation equation. limited.
  • the functional relationship equation can be optimized through the method of nonlinear fitting, so as to ensure that the error value obtained by solving the functional relationship equation is closer to the true value, and the calibration error is more accurate and precise.
  • each error item in the error item set has a corresponding sensitivity, and the sensitivity is used to characterize the degree of influence of the error item on the accuracy of the measurement result of the scanning device.
  • the set of error terms includes a first error term and a second error term, and if the first error term changes the first threshold and the second error term does not change, the scan
  • the accuracy of the direction vector of the light emitted by the device in the scanning field of view is greater than when the first error term does not change and the second error term changes the first threshold value, the light emitted by the scanning device is in the scanning field of view If the accuracy of the direction vector is, the sensitivity of the first error term is higher than the sensitivity of the second error term.
  • Sensitivity is defined as the degree of influence on the accuracy of the result when a single parameter changes in the error term set ⁇ ′ ⁇ .
  • the sensitivity of the k-th error term ⁇ k in the set is In the actual calculation, the calculation can be simplified, and a single parameter can be increased or decreased by 1 unit, and the deviation value of the calculation result can be used as the sensitivity of the parameter.
  • each error term in the error term set ⁇ can be obtained.
  • the error item in the error item set is an error item whose sensitivity is greater than or equal to a preset threshold.
  • the error parameters can be rationally screened according to the actual situation of the system (working error control, etc.) and accuracy requirements, to filter out insensitive parameters, and retain sensitive parameters.
  • the standard for filtering or retaining the error term is to set a preset threshold. When the sensitivity of the error term is less than the preset threshold, it is considered that the change in the direction vector of the light emitted by the system caused by the error value is small. Therefore, filtering it out will reduce the number of items to be solved, thereby reducing the difficulty of the solution, simplifying the calibration environment, and improving the efficiency of the solution.
  • it further includes: when any two error items in the set of error items have a proportional relationship to the degree of influence on the accuracy of the direction vector of the light emitted by the scanning device in the scanning field of view When, combine the any two error terms into one error term and add it to the set of error terms.
  • the coupling between two error parameters can be obtained by the following method: (1) Increase or decrease the error term parameter 1 by one unit, keep the remaining error term parameters unchanged, traverse all error term combinations, and calculate all error term combinations Lower angle offset; (2) Repeat step (1) for parameter 2; (3) Calculate the correlation coefficient of the angle offset value obtained in (1) and (2) as the coupling degree between the two parameters. For the parameter items with higher coupling degree, they are "integrated" into an error item.
  • the number of error term parameter sets will have a relatively obvious drop, written as ⁇ .
  • the scanning device Based on laser light error model Taking the error term set ⁇ as an unknown number (assuming that once the whole machine is assembled, the value in ⁇ does not change), the scanning device obtains Substituting into the model as internal parameters, an equation can be obtained for each group of error terms. Collect enough sets of equations in the calibration scene to solve the set of error terms.
  • the number of items to be solved can be reduced, thereby reducing the difficulty of the solution, simplifying the calibration environment, and improving the efficiency of the solution.
  • the method further includes: sending the system error of the scanning device to the scanning device, and the system error is used by the scanning device to calculate that the light emitted by the scanning device is within the scanning field of view.
  • Direction vector is used by the scanning device to calculate that the light emitted by the scanning device is within the scanning field of view.
  • the scanning device obtains the value of the error term ⁇ calculated by the error calibration device, combined with Value, bring it into That is, the direction vector of the light rays actually emitted by the scanning device can be obtained.
  • the lidar can collect its own data, including point cloud data.
  • the scanning device sends the obtained data to the error calibration device, wherein the data of the scanning device can be selected to be sent to the error calibration device in real time, or it can be selected to send all the data to the error calibration device after the scanning is completed.
  • the point in the point cloud data collected by the lidar has a corresponding relationship with the direction vector of the reference point measured by the measuring instrument.
  • the error calibration device can determine the respective reference points corresponding to each set of system input parameters, and then combine the Vin and angle combinations in each set of input parameters and the corresponding reference points.
  • the error term set can be solved by substituting the direction vector into multiple sets of equations.
  • the error calibration device is exemplarily a PC, or other hardware devices including a memory and a processor.
  • the direction vector of the light emitted by the scanning device in the scanning field of view is the direction of the light emitted by the laser light source of the scanning device after passing through the optical system of the scanning device vector.
  • the axis rotation parameter includes an axis rotation angle.
  • the component includes a mirror and/or a prism.
  • the component as shown in Figure 1 includes a mirror and a prism.
  • the reflector is also called a plane mirror.
  • the component in Figure 2 includes two prisms.
  • FIG. 1 and FIG. 2 are only exemplary displays. In other embodiments, the component may only include mirrors, and the present application does not limit the number of mirrors and/or prisms contained therein.
  • the error items in the set of error items include axis normal vector deviation, mirror and/or prism initial normal vector deviation, laser light source emission deviation, axis rotation angle measurement deviation or prism refraction One or more of the rate deviations.
  • system input parameters of different groups include at least one axis rotation parameter of a different component.
  • axis rotation parameters ⁇ For different groups of system input parameters, it contains at least one of the axis rotation parameters ⁇ of different components, so as to realize the solution of the axis rotation parameters.
  • the reference point is a marker located in the scanning field of view of the scanning device.
  • the selection of the reference point can be preset.
  • a marker is preset to identify the location of the reference point, and when the scanning device scans the marker, the location is recognized as the reference point position.
  • the direction vector is a direction vector in the spatial coordinate system of the scanning device.
  • the aforementioned direction vectors are all direction vectors in the spatial coordinate system of the scanning device. That is, the measurement of the direction vector is calibrated in the spatial coordinate system where the scanning device is located.
  • the direction vector of the light emitted by the laser light source of the scanning device is the direction vector of the light emitted by the laser light source of the scanning device before passing through the optical system of the scanning device.
  • the direction vector of the light emitted by the laser light source of the scanning device is the direction vector of the light emitted by the laser light source of the scanning device before passing through the optical system of the scanning device.
  • the position of the laser as shown in Fig. 2 and Fig. 3 is the direction vector before the laser light emitted by the laser light source passes through the optical system: prism 1 and prism 2.
  • the scanning device includes any one or more of laser radar, millimeter wave radar, or ultrasonic radar. This application does not limit the type of radar, as long as it can realize the detection function, it falls into the scope of the scanning device of this application.
  • the error calibration method described above or the device adopting the above method to calibrate the error of the scanning device the error value of each component can be obtained separately. Therefore, it is at least universal, efficient, fully automatic, and highly accurate, and can support Mass production, simple calibration effect in calibration environment.
  • the error calibration method 200 may include the following steps:
  • a first direction vector is obtained from the scanning device, the first direction vector is calculated by the scanning device; the first direction vector is the first direction vector when the scanning device scans to a reference point in the scanning field of view The direction vector of the light emitted by the scanning device in the scanning field of view.
  • a second direction vector is obtained from a measuring instrument, where the second direction vector is a direction vector of the reference point measured by the measuring instrument in the scanning field of view.
  • step S230 a system error compensation value in the target area corresponding to the reference point is determined according to the first direction vector and the second direction vector.
  • the second error calibration method can also be called the compensation method.
  • the specific error transmission is regarded as a black box, and only the actual exit direction under different conditions is compared with the standard model (no error is introduced). Item case) the difference between the two exit directions calculated, and these errors are compensated through a certain mechanism to reduce the error of the final result.
  • the specific values of each error item within the system can be ignored, and the system can be compensated only according to the changes before and after the error transmission. Therefore, this method is suitable for various complex systems, and has strong universality. Efficient, fully automatic, high precision, can support mass production, simple calibration environment and other advantages.
  • step S210 Obtain a first direction vector from a scanning device, where the first direction vector is calculated by the scanning device;
  • the reference point in the field is the direction vector of the light emitted by the scanning device in the scanning field of view.
  • Step S220 Obtain a second direction vector from a measuring instrument, where the second direction vector is a direction vector of the reference point measured by the measuring instrument in the scanning field of view.
  • the direction vector of the reference point is obtained by the measuring instrument, and the direction vector of the reference point is taken as the second direction vector.
  • Step S230 Determine a system error compensation value in the target area corresponding to the reference point according to the first direction vector and the second direction vector.
  • the system error compensation value in the target area corresponding to the reference point is determined according to the relationship between the first direction vector and the second direction vector.
  • the first direction vector is the ideal value
  • the second direction vector is the actual output value. Through the relationship between the two, the error value of the system can be determined, and then the compensation value can be obtained.
  • the target area in step S230 is obtained by dividing the scanning field of view.
  • the target area is obtained by dividing the scanning field of view, for example, one reference point corresponds to one target area, or four reference points are used as the four vertices of the target area.
  • the target area is defined by the azimuth angle and the zenith angle of the light emitted by the scanning device compared to the scanning device.
  • step S230 includes using the difference between the first direction vector and the second direction vector as the system error compensation value in the target area corresponding to the reference point.
  • the compensation value is directly taken as Among them, based on the standard model f, the marker direction vector of the specific angle combination of the j-th optical component in the region i is calculated
  • the nominal direction vector of the marker in the measured area i in the radar coordinate system The nominal direction vector refers to the actual direction vector of the light emitted by the scanning device obtained by the measuring instrument, or is called the true direction vector.
  • the area i has a corresponding relationship with the specific angle combination of the j-th optical component.
  • the scanning device includes one or more components, the permutation and combination of the angles of the components in the one or more components constitute an angle combination, and the angle is a rotation angle around an axis.
  • the method further includes: generating a correspondence relationship between at least one angle combination cluster and the system error compensation value in the at least one target area, and each angle combination cluster includes one or more angle combinations, wherein The at least one angle combination cluster includes a first angle combination cluster, the at least one target area includes a first target area, and the first angle combination cluster corresponds to a system error compensation value in the first target area.
  • the one or more angle combinations included in the first angle combination cluster are the angle combinations of one or more components in the scanning device when the light rays emitted by the scanning device fall into the target area. Compensation is done by traversing variable combinations.
  • the exit angle error is approximately equal: the exit direction falls within the same subdivision area, and the angle combination is similar.
  • the angle combination is approximately defined as follows: the two angle combinations are regarded as two equal-length vectors a, b, if It is considered that the combination of the two angles is similar.
  • T is the set threshold, which can be adjusted according to the actual situation.
  • the angle combination can be divided into n clusters, set as (m 1 , m 2 ..., m n ), and each cluster corresponds to the same angle error compensation value And the same subdivision area i.
  • One or more markers can be set in each subdivided area i to perform the calculations in the above steps 210-230 to obtain the system error compensation value corresponding to the subdivided area i.
  • the component includes a mirror and/or a prism.
  • the mirror is also called a flat mirror.
  • the system error compensation value is used by the scanning device according to the direction vector of the light emitted by the laser light source of the scanning device, and the angle combination of one or more components in the scanning device.
  • the system error compensation value in the area corresponding to the angle combination of the one or more components determines the actual direction vector of the light emitted by the scanning device.
  • the steps in the previous embodiment include: the scanning device obtains the second angle combination cluster where the angle combination of the one or more components is located, and obtains the second angle combination The system error compensation value in the area corresponding to the cluster,
  • the scanning device according to the direction vector of the light emitted by the laser light source of the scanning device, the second angle combination cluster where the angle combination of the one or more components is located, and the system in the area corresponding to the second angle combination cluster
  • the error compensation value determines the actual direction vector of the light rays emitted by the scanning device.
  • the direction vector of the light emitted by the laser light source of the scanning device is V in
  • the cluster corresponding to the angle combination of one or more components is the k-th cluster
  • the area corresponding to the k-th cluster is area n
  • the area n has an error
  • the compensation value c n is calculated based on the standard model f to obtain the marker direction vector V out of the k-th cluster in the region n, and the scanning device can determine the actual direction vector of the light emitted by the scanning device according to c n and V out.
  • the second angle combination cluster may also be the first angle combination cluster corresponding to the target area, or it may not be the first angle combination cluster.
  • the actual direction vector of the light emitted by the scanning device is determined by the scanning device according to the direction vector of the light emitted by the scanning device and the area corresponding to the second angle combination cluster.
  • the system error compensation value is added together, and the direction vector of the light emitted by the scanning device is calculated by the scanning device according to the direction vector of the light emitted by the laser light source.
  • the variable in the actual operation of the lidar is the angle of rotation of each prism or plane mirror around the axis. Therefore, it can be compensated by traversing variable combinations.
  • equal-interval sampling traversal is not a reasonable traversal method, because this does not guarantee that the exit direction corresponding to the selected variable combination is uniformly distributed in the field of view.
  • the errors in the error term set are controlled by a certain degree of work difference, which is a relatively small value.
  • the combined effect of all error terms will not cause a significant deviation of the laser exit angle (e.g.>1 degree) in each area of the imaging.
  • we subdivide the field of view into a large number of areas for example, according to the azimuth angle of 1 degree * zenith angle of 1 degree to divide the area, then the azimuth angle is between 0 and 1 degree and the zenith angle is between 0 and 1 degree.
  • the direction vector is a subdivision area).
  • the exit angle error of the angle combination that meets the following conditions is approximately equal: the exit direction falls within the same subdivision area and the angle combination is similar.
  • the angle combination is approximately defined as follows: the two angle combinations are regarded as two equal-length vectors a, b, if It is considered that the combination of the two angles is similar. Among them, T is the set threshold, which can be adjusted according to the actual situation.
  • the angle combination can be divided into n clusters, set as (m 1 , m 2 ..., m n ), and each cluster corresponds to the same angle error compensation value And the same subdivision area i.
  • the angle error compensation value can be obtained by the following method:
  • the compensation method is as follows:
  • f is the standard calculation function when no internal parameters are introduced
  • I the compensation value of the k-th cluster.
  • step S230 includes determining the difference between the first direction vector and the second direction vector, and the N reference points correspond to a target area, and according to the N reference points The error compensation value of the point obtains the error compensation value of the target area, and N is a positive integer greater than or equal to 2.
  • the target area is a rectangular grid.
  • the target area corresponding to the four reference points is followed by the error compensation value of the four reference points to obtain the error compensation value of the target area.
  • FOV field of view
  • Figure 6 is an example of gridding, where the reference points are reference markers, and every four reference markers form a grid. (For example, the area enclosed by marker a, b, c, d forms a rectangular grid, and so on).
  • the obtaining the error compensation value of the target area according to the error compensation values of the N reference points includes averaging the error compensation values of the N reference points as the The error compensation value of the target area.
  • this mapping relationship can be obtained by enclosing the grid marker's real value-solving value relationship.
  • the calculated value corresponds to the value of the direction vector of the emitted light calculated by the scanning device when the optical component is in the design pose.
  • the true value refers to the actual direction vector of the light emitted by the scanning device measured by the measuring instrument, or called the true direction vector.
  • the calculation of the average value is relatively simple, so the efficiency of the error calibration method can be guaranteed.
  • the averaging the N differences as the error compensation value of the target area includes taking an arithmetic average or a root mean square average of the error compensation values of the N reference points value.
  • a is an actual value of a reference point
  • a' is a calculated value of a reference point
  • a-a' is an error value of a reference point.
  • b, c, and d are the actual values of the reference point
  • mapping relationship F(i) can be any kind of mapping relationship with abcd as a variable.
  • obtaining the error compensation value of the target area according to the error compensation values of the N reference points includes a functional relationship fitted according to the error compensation values of the N reference points As the error compensation value of the target area. Among them, each target area has an error compensation value.
  • the functional relationship fitted according to the error compensation values of the N reference points is used as the compensation value, including obtaining the said compensation value by interpolation based on the error compensation values of the N reference points
  • the functional relationship is used as the error compensation value of the target area. There is an approximate mapping relationship between the measured value and the true value in the same grid.
  • the method further includes: generating a correspondence relationship between at least one area and at least one system error compensation value, wherein the at least one area includes the target area.
  • the system error compensation value is used for the scanning device to determine the actual direction vector of the light emitted by the scanning device according to the direction vector of the light emitted by the scanning device in the scanning field of view. After obtaining the error compensation value, according to the application scenario shown in FIG. 8, the actual direction vector of the light emitted by the scanning device can be obtained.
  • the compensation method is as follows:
  • the scanning device obtains the direction vector of the light emitted by the scanning device in the scanning field of view
  • the scanning device obtains the error value corresponding to the area in the error compensation compensation device corresponding to the direction vector obtained in the previous step;
  • the actual direction vector of the light emitted by the scanning device is obtained.
  • the scanning device obtains the system error compensation value corresponding to the first area, so as to obtain the actual direction vector of the light emitted by the scanning device, and the first area It is the area where the direction vector of the light emitted by the scanning device falls in the scanning field of view. It can be seen that there is a corresponding relationship between the area and the system error compensation value.
  • the direction vector of the light emitted by the scanning device in the scanning field of view is calculated according to the direction vector of the light emitted by the laser light source of the scanning device.
  • the direction vector of the light emitted by the scanning device in the scanning field of view is calculated based on the ideal model.
  • the ideal output value calculated based on the ideal model is used as the basis of error compensation to obtain the direction vector of the actual light emitted by the scanning device.
  • the direction vector is a direction vector in the spatial coordinate system of the scanning device.
  • the aforementioned direction vectors are all direction vectors in the spatial coordinate system of the scanning device. That is, the measurement of the direction vector is calibrated in the spatial coordinate system where the scanning device is located.
  • the direction vector of the light emitted by the laser light source of the scanning device is the direction vector of the light emitted by the laser light source of the scanning device before passing through the optical system of the scanning device.
  • the direction vector of the light emitted by the laser light source of the scanning device is the direction vector of the light emitted by the laser light source of the scanning device before passing through the optical system of the scanning device.
  • the position of the laser shown in Fig. 2 and Fig. 3 is the direction vector before the laser light emitted by the laser light source passes through the optical system: prism 1 and prism 2.
  • the direction vector of the light emitted by the scanning device in the scanning field of view is the direction of the light emitted by the laser light source of the scanning device after passing through the optical system of the scanning device vector.
  • the direction vector of the light emitted by the scanning device in the scanning field of view is the direction vector of the light emitted by the laser light source of the scanning device after passing through the optical system of the scanning device.
  • the position of the laser shown in Fig. 2 and Fig. 3 is the directional vector of the laser light emitted by the laser light source after passing through the optical system: prism 1 and prism 2.
  • the scanning device includes any one or more of laser radar, millimeter wave radar, or ultrasonic radar.
  • This application does not limit the type of radar, as long as it can realize the detection function, it falls into the scope of the scanning device of this application.
  • a scanning device may be used to transmit a light pulse signal to the object to be measured, and receive a reflected pulse signal corresponding to the light pulse signal.
  • the detection device includes, but is not limited to, laser radar, electromagnetic wave radar, millimeter wave radar, or ultrasonic radar.
  • the following exemplary description is given:
  • the area needs to be subdivided as much as possible.
  • a large number of subdivision areas will cause a great traversal interval, which will greatly affect the calibration efficiency.
  • the second compensation scheme grids the FOV during calibration, and the intersection of the grids is the reference marker.
  • Figure 6 is an example of gridding, where the reference points are reference markers, and every four reference markers form a grid. (For example, the area enclosed by marker a, b, c, d forms a rectangular grid, and so on).
  • the calibration process can be seen in Figure 6: After obtaining the mapping relationships in all grids, the error calibration device sends the mapping relationships between the true values and the calculated values of all grids (target areas) to the scanning device, and the scanning device is in the actual In the process of use, the corresponding compensation value is found from the mapping relationship, and specifically, the scanning device can obtain the compensated emission direction in the following manner:
  • the system error calibration method, device, and computer-readable storage medium are based on the direction vectors of the at least two reference points, the at least two sets of system input parameters, and preset error terms.
  • the above exemplarily describes the system error calibration method according to the embodiment of the present application.
  • the error calibration device according to the embodiment of the present application can be used to implement the error calibration method according to the embodiment of the present application described above.
  • the main structure and function of the error calibration device are described below, and some specific details that have been described above are omitted.
  • the error calibration device 1000 provided according to another aspect of the present application will be described below with reference to FIG. 9.
  • the error calibration device 1000 according to the embodiment of the present application can be used to implement the error calibration method 100, 200 according to the embodiment of the present application described above.
  • the error calibration device 1000 For the sake of brevity, only the main structure and function of the error calibration device 1000 are described below, and some specific details that have been described above are omitted.
  • the error calibration device 1000 may include a processor 1010 and a memory 1020, wherein the memory 1020 stores a computer program run by the processor 1010, and the processor 1010 is used to run a calculator program When performing the aforementioned error calibration method.
  • the error calibration device may include, but is not limited to, a PC, or other hardware devices including a memory and a processor.
  • a computer-readable storage medium having a computer program stored on the computer-readable storage medium, and the computer program executes the error calibration method according to the embodiment of the present application when the computer program is running.
  • the computer-readable storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk Read only memory (CD-ROM), USB memory, or any combination of the above storage media.
  • the computer-readable storage medium may be any combination of one or more computer-readable storage media.
  • the disclosed device and method may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or It can be integrated into another device, or some features can be ignored or not implemented.
  • the various component embodiments of the present application may be implemented by hardware, or by software modules running on one or more processors, or by a combination of them.
  • a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some modules according to the embodiments of the present application.
  • This application can also be implemented as a device program (for example, a computer program and a computer program product) for executing part or all of the methods described herein.
  • Such a program for implementing the present application may be stored on a computer-readable storage medium, or may have the form of one or more signals.
  • Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Optical Radar Systems And Details Thereof (AREA)

Abstract

一种***误差标定方法、装置和计算机可读存储介质,包括:获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数;根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。从而获得具有普适性强,高效,全自动,精度高,可支持量产,标定环境简单等优点的误差标定方案。

Description

***误差标定方法、装置和计算机可读存储介质 技术领域
本申请总体上涉及激光探测技术领域,更具体地涉及一种***误差标定方法、装置和计算机可读存储介质。
背景技术
机械式激光雷达激光出射后,普遍需要经过一个较为复杂的光学***,当前激光雷达的光学***多由棱镜、平面镜组成,通过折射、反射与器件旋转的方式形成特定的扫描模式,激光雷达工作期间,以满足特定扫描需求,棱镜和反射镜可能静止或绕特定转轴旋转。
然而,棱镜和反射镜在安装时存在不可避免的工差,继而使得出光角度的解算值偏离实际值形成角度误差,最终造成成像失真。此类光学***较为复杂,涉及参数多且耦合度高,目前缺乏一套标定方法针对性解决此类内参标定问题。这一问题若无法得到解决,将成为产品量产效能,精度性能指标的瓶颈。
发明内容
本申请实施例提供一种***误差的标定方案,该标定方案具有普适性强,高效,全自动,精度高,可支持量产,标定环境简单等优点。下面简要描述本申请提出的***误差标定方法,更多细节将在后续结合附图在具体实施方式中加以描述。
根据本申请一方面,提供了一种***误差标定方法,所述方法包括:获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数;根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
根据本申请另一方面,提供了一种***误差标定方法,所述方法包括:从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量;从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量;根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。
根据本申请再一方面,提供了一种扫描装置的误差标定装置,所述误差标定装置包括存储器和处理器,所述存储器上存储有由所述处理器运行的计算机程序,所述处理器用于运行计算器程序时执行:获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数;根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
根据本申请又一方面,提供了一种扫描装置的误差标定装置,所述误差标定装置包括存储器和处理器,所述存储器上存储有由所述处理器运行的计算机程序,所述计算机程序在被所述处理器运行时执行:从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量;从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量;根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。
根据本申请又一方面,提供了一种计算机可读存储介质,所述计算机可读存储介质上存储有计算机程序,所述计算机程序在处理器上运行时所述处理器执行上述的误差标定方法。
根据本申请实施例的***误差标定方法、装置和计算机可读存储介质在根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差或者根据所述第一方向向量以 及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,从而获得具有普适性强,高效,全自动,精度高,可支持量产,标定环境简单等优点的误差标定方案。
附图说明
图1示出根据本申请实施例的激光雷达结构示意图。
图2示出根据本申请实施例的双棱镜结构激光雷达在第一情况下的光路示意性图。
图3示出根据本申请实施例的双棱镜结构激光雷达在第二情况下的光路示意图。
图4示出根据本申请实施例的双棱镜结构激光雷达误差的示意图。
图5示出根据本申请实施例的第一误差标定方法的示意性流程图。
图6示出根据本申请实施例的FOV网格化的示意图。
图7示出根据本申请实施例的第二误差标定方法的示意流程图。
图8示出根据本申请实施例的应用场景框架图。
图9示出根据本申请另一方面提供的误差标定装置示意图。
具体实施方式
为了使得本申请的目的、技术方案和优点更为明显,下面将参照附图详细描述根据本申请的示例实施例。显然,所描述的实施例仅仅是本申请的一部分实施例,而不是本申请的全部实施例,应理解,本申请不受这里描述的示例实施例的限制。
在下文的描述中,给出了大量具体的细节以便提供对本申请更为彻底的理解。然而,对于本领域技术人员而言显而易见的是,本申请可以无需一个或多个这些细节而得以实施。在其他的例子中,为了避免与本申请发生混淆,对于本领域公知的一些技术特征未进行描述。
应当理解的是,本申请能够以不同形式实施,而不应当解释为局限于这里提出的实施例。相反地,提供这些实施例将使公开彻底和完全,并且将本申请的范围完全地传递给本领域技术人员。
在此使用的术语的目的仅在于描述具体实施例并且不作为本申请的 限制。在此使用时,单数形式的“一”、“一个”和“所述/该”也意图包括复数形式,除非上下文清楚指出另外的方式。还应明白术语“组成”和/或“包括”,当在该说明书中使用时,确定所述特征、整数、步骤、操作、元件和/或部件的存在,但不排除一个或更多其它的特征、整数、步骤、操作、元件、部件和/或组的存在或添加。在此使用时,术语“和/或”包括相关所列项目的任何及所有组合。
为了彻底理解本申请,将在下列的描述中提出详细的步骤以及详细的结构,以便阐释本申请提出的技术方案。除了本申请详细描述的实施例外,本申请还可以具有其他实施方式。
雷达是一种主动式感知传感器,其基本原理为雷达主动发射光脉冲信号,信号在传播途径中遇到物体产生反射回波,部分回波信号由雷达的接收模块所捕获,由发射信号和接收信号之间的时间差计算被测物距离雷达的深度,根据雷达的发射方向,获得被测物相对雷达的角度信息,基于深度和角度信息可以计算被测物相对于雷达的三维空间位置。
而棱镜与反射镜是机械式激光雷达内部光学***的重要组件,机械式激光雷达工作时,棱镜和反射镜可能静止或绕特定转轴旋转,通过反射镜及棱镜动态变化下的反射、折射组合形成特定的扫描模式。
示例性地,可以参考图1所示的激光雷达结构示意图:如图1所示,激光激发后先经过一个绕轴旋转的棱镜,然后经过一个静止的反射镜并最终出射。其中可以包括两个绕轴旋转的棱镜,或者将静止的反射镜替换为静止的棱镜或旋转的棱镜,示例性地,还可以替换为MID-40型号激光雷达。
示例性地,可以参考图2所示的光路原理示意图:如图2所示,激光激发后依次经过两个绕轴旋转的棱镜并最终出射。图1和图2中镜片的类型、排列顺序以及数量仅作为示例,实际应用中,探测装置可以包括其他类型、数量的镜片。
其中,本申请中的探测装置包括但不限于雷达,且本申请中的雷达包括但不限于:激光雷达、电磁波雷达、毫米波雷达或者超声波雷达。
下面参照图2描述根据本申请实施例的第一情况下的光路示意性图。参照图3示出根据本申请实施例的第二情况下的光路示意图。
如图2所示,若光学组件实际位姿与设计位姿相符,则扫描装置的激 光光源发射的光线的方向向量经过棱镜1和棱镜2后,扫描装置出射的光线在扫描视场内的方向向量与扫描装置的解算值相符,也就是说,A空间点为扫描装置实际出射的点,同时也是理想情况下扫描装置的解算点,两者为同一点。
如图3所示,若光学组件实际位姿与设计位姿存在误差,则使用设计名义值估算的扫描装置的激光光源发射的光线的方向向量经过棱镜1和棱镜2后,扫描装置出射的光线在扫描视场内的方向向量与解算值不符,从而引入角度误差。所述设计名义值是指假设光学组件处于设计位姿时,扫描装置出射的光线的方向向量的值。具体地,图3中的A空间点为扫描装置实际出射的点,A’空间点为扫描装置解算出的出射点,其对应于设计名义值,A空间点与A’空间点并非同一点,存在着误差。造成A空间点和A’空间点存在差异的原因在于,棱镜1和棱镜2存在着安装误差,示例性地,如图3所示,棱镜1的理想安装位置为右侧面垂直于光轴设置,如棱镜1的虚线位置所示,棱镜2的理想安装位置为左侧面垂直光轴设置,如棱镜2的虚线位置所示,而实际安装过程中,将棱镜1和棱镜2安装为实线位置所示,从而导致了实际出射的A空间点与A’空间点存在差异。
图3仅示出了棱镜1和棱镜2在光轴方向存在安装误差的情况,图4示出根据本申请实施例的双棱镜结构激光雷达误差的示意图。如图4所示,对于第一组件:棱镜1,在垂直于纸面的方向,包括11#面和12#面,图中实线示出了棱镜的实际位姿,图中虚线位置示出了棱镜的设计位姿,11#面实际的法向向量如
Figure PCTCN2019122145-appb-000001
所示,12#面实际的法向向量如
Figure PCTCN2019122145-appb-000002
所示,
Figure PCTCN2019122145-appb-000003
左侧的长向量示出了棱镜的实际向量,
Figure PCTCN2019122145-appb-000004
示出了实际的旋转轴角度偏差,可能存在的误差包括以下误差项:转轴法向轴偏差α y1z1、棱镜/平面镜初始法线偏差β y1z1、绕轴旋转角度测量偏差ε 1、入射光偏差λ 12,折射率偏差n,其中旋转轴方向偏差α y1z1
Figure PCTCN2019122145-appb-000005
在Y轴和Z轴方向的分量,Y轴和Z轴的方向如图4中所所示,棱镜/平面镜初始法线偏差β y1z1
Figure PCTCN2019122145-appb-000006
和/或
Figure PCTCN2019122145-appb-000007
左侧的长向量在Y轴和Z轴方向的分量,类似的,第二组件:棱镜2,在垂直于纸面的方向,包括21#面和22#面,图中实线示出了棱镜的实际位姿,图中虚线位置示出了棱镜的设计位姿,21#面实际的法向方向如
Figure PCTCN2019122145-appb-000008
所示,22#面实际的法向向量如
Figure PCTCN2019122145-appb-000009
所示,
Figure PCTCN2019122145-appb-000010
左侧的长向量示出了棱镜的实际向量,
Figure PCTCN2019122145-appb-000011
示出了实际的旋转轴角度偏差,可能存在的误差包括以下误差项:转轴法向轴偏差α y2z2、棱镜/平面镜初始法线偏差β y2z2、绕轴旋转角度测量偏差ε 2、入射光偏差λ 12,折射率偏差n。其中旋转轴方向偏差α y2z2
Figure PCTCN2019122145-appb-000012
在Y轴和Z轴方向的分量,Y轴和Z轴的方向如图4中所所示,棱镜/平面镜初始法线偏差β y2z2
Figure PCTCN2019122145-appb-000013
和/或
Figure PCTCN2019122145-appb-000014
左侧的长向量在Y轴和Z轴方向的分量。
将上述误差项归纳如下表所示。
Figure PCTCN2019122145-appb-000015
表1双棱镜激光雷达误差项
根据前述分析,以双棱镜激光雷达为例,其光学***由棱镜和/或平面镜,通过折射、反射与器件旋转的方式形成特定的扫描模式。此类光学***较为复杂,涉及参数多且耦合度高,解决此类***的误差标定问题具有相当的难度,本申请从两个角度来给出标定方案,以获得高效、全自动、标定环境简单的效果。
在介绍本申请的标定方案之前,参照附图8来描述实现本申请技术方案的应用场景框架图。
如图8所示,框架图中包括扫描装置、测量仪器和误差标定装置。
其中,扫描装置示例性地为双棱镜结构雷达,其发射的探测信号为激光。其激光雷达能够采集自身的数据,包括点云数据,每个点在发射时的激光光源的方向向量、激光雷达内部各个组件的旋转角度等等。扫描装置将得到的数据发送给误差标定装置,其中,扫描装置的数据可以选择为实时发送给误差标定装置,也可以选择扫描结束后,将所有的数据一起发送 给误差标定装置。
示例性地,所述扫描装置还可以替换为电磁波雷达、毫米波雷达或者超声波雷达,当替换为其他雷达时,相应的探测信号也发生改变。
其中,测量仪器位于扫描装置的扫描视场中,其能够精确测量空间中的标记物(或标记点、参考点等)在扫描视场中的方向向量。其中,测量仪器可以基于激光雷达的空间坐标系来计算空间中的标记物(或标记点)在扫描空间内的方向向量。
其中,误差标定装置能够接收到扫描装置以及测量仪器的数据,并对接收的数据进行处理,从而进行误差标定。具体的误差标定过程将结合下面的附图进行详细说明。
下面参照图5描述本申请实施例的第一误差标定方法100的示意性流程图。
如图5所示,误差标定方法100可以包括如下步骤:
在步骤S110,获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数。
在步骤S120,根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
在本申请的实施例中,可以采用扫描装置向被测物发射光脉冲信号,并接收所述光脉冲信号对应的反射脉冲信号。其中,所述探测装置包括但不限于激光雷达、电磁波雷达、毫米波雷达或者超声波雷达等。
在本申请的一个实施例中,获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数。
为了理解上述步骤,我们首先来描述一下扫描视场内方向向量的计算过程,具体地,附图3中,A’空间点的解算示例性地可以由以下模型计算获得,首先,建立模型,并进行理想情况下出射角度估算:激光光线经过反射平面时遵循反射定律:
Figure PCTCN2019122145-appb-000016
其中
Figure PCTCN2019122145-appb-000017
为出射光线单位向量,
Figure PCTCN2019122145-appb-000018
为入射光线单位向量,
Figure PCTCN2019122145-appb-000019
为镜面单位法向量。
激光光线经过棱镜平面时遵循折射斯涅耳定律(snell’s law):
Figure PCTCN2019122145-appb-000020
其中
Figure PCTCN2019122145-appb-000021
为出射单位向量,
Figure PCTCN2019122145-appb-000022
为入射单位向量,r为入射介质反射率与出射介质反射率的比值,即:r=n in/n out
Figure PCTCN2019122145-appb-000023
为棱镜平面法向量。通常一个棱镜需要考虑两次折射,即存在两个折射面。
当棱镜或反射镜绕轴旋转时,上述公式中的平面法向量
Figure PCTCN2019122145-appb-000024
为平面初始法向量
Figure PCTCN2019122145-appb-000025
转轴向量
Figure PCTCN2019122145-appb-000026
及当前绕转轴旋转角度θ的函数,关系式为:
Figure PCTCN2019122145-appb-000027
其中,当棱镜或反射镜静止时,可以认为是上述函数的特殊情况,即θ恒为零。
给定任意多棱镜及平面镜组合(绕轴旋转或静止)的光学***,基于上述公式,都能得到函数关系
Figure PCTCN2019122145-appb-000028
其中
Figure PCTCN2019122145-appb-000029
为出射光线单位向量,
Figure PCTCN2019122145-appb-000030
为入射光线单位向量,(θ 12...,θ n)为每个器件当前绕轴旋转的角度(若器件m始终保持静止,θ m始终取0)。需要说明的是,上述公式中的
Figure PCTCN2019122145-appb-000031
即为扫描装置计算得到的激光经过激光雷达的光学***后出射光线的方向向量。
可以看到,上述函数关系是基于设计值得到的,通过上述模型可以计算得到A’空间点的解算值。
然后,引入光学***误差项,实际情况中,无论是原料还是安装过程都会引入工差。这些工差使得由函数关系f(·)计算得到的结果偏离实际值。可能的误差项包括但不限于如下参数:
1.转轴法向量偏差α
2.棱镜/平面镜初始法向量偏差β
3.激光光源出射偏差(入射光偏差)λ
4.绕轴旋转角度测量偏差ε
5.棱镜折射率偏差n
其中α,β,λ为向量偏差,ε,n为数值偏差。向量偏差可以分解为绕y轴 和绕z轴两个方向的旋转Δ yz,有:
Figure PCTCN2019122145-appb-000032
其中
Figure PCTCN2019122145-appb-000033
为理想值,
Figure PCTCN2019122145-appb-000034
为考虑向量偏差后的实际值。
根据实际***引入误差项集合{Δ},可以得到新的函数关系:
Figure PCTCN2019122145-appb-000035
以双棱镜结构的激光雷达***为例,图4为误差项示意图,表1为误差项汇总情况。
在本申请的一个实施例中,在步骤110中,在扫描视场内,测量仪器精确探测扫描激光的位置,在扫描视场内有一个或多个参考点,参考点的方向向量是已知的(例如,参考点的方向向量是由测量仪器预先测量的)。当扫描装置扫描到参考点时,获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数,扫描装置的激光光源发射的光线作为光学组件的入射光线,其方向向量即代表了入射方向向量V in,扫描装置中的一个或多个组件的绕轴旋转参数例如是θ 1,θ 2……θ n
在步骤120中,根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
根据前面的分析,在步骤110中,至少获得了两个真值已知的参考点的方向向量
Figure PCTCN2019122145-appb-000036
在每一个
Figure PCTCN2019122145-appb-000037
的方程式中,
Figure PCTCN2019122145-appb-000038
和θ 1,θ 2……θ n是已知的,将获得的多个
Figure PCTCN2019122145-appb-000039
等式列为方程组,理论上,方程组中包含方程的数目大于或者等于未知项的个数即可求解出其中误差项{Δ}的值。
如此,扫描装置中表1中汇总的各种误差项均可以被标定。通过这种方法或采用上述方法的装置对扫描装置的误差进行标定,能够分别获得各个组件的误差值,因此,至少具有精确地、高效地标定效果。
在本申请的一个实施例中,步骤110:所述根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差,包括:将所述至少两个参考点的方向向量、所述至 少两组***输入参数以及所述误差项集合代入预设误差模型计算所述扫描装置的***误差。
基于前面的分析,
Figure PCTCN2019122145-appb-000040
为预设的误差模型,根据该误差模型,在每一个
Figure PCTCN2019122145-appb-000041
的方程式中,
Figure PCTCN2019122145-appb-000042
和θ 1,θ 2……θ n是已知的,将获得的多个
Figure PCTCN2019122145-appb-000043
等式列为方程组,理论上,方程组中包含方程的数目大于或者等于未知项的个数即可求解出其中误差项{Δ}的值。
根据预设误差模型计算误差,能够简洁有效地获得误差值,使得标定环境简单。
在本申请的一个实施例中,所述预设误差模型为基于光学定律的函数关系方程。其中,
Figure PCTCN2019122145-appb-000044
为基于反射定律以及斯涅耳定律获得。
基于光学定律的函数关系方程能够获得精确的误差值,保证误差标定方法和装置的有效性和精确性。而且上述函数关系方程能应用于包含各种光学组件的扫描装置,对于包含不同组件的扫描装置,仅需根据光学定律将函数关系方程进行替换,即可获得扫描装置的误差值,因此,其普适性较强。
在本申请的一个实施例中,所述函数关系方程的求解采用非线性拟合方法。在每一个
Figure PCTCN2019122145-appb-000045
的方程式中,
Figure PCTCN2019122145-appb-000046
和θ 1,θ 2……θ n是已知的,将获得的多个
Figure PCTCN2019122145-appb-000047
等式列为方程组,函数关系f的求解是一个非线性最小二乘问题,直接求解较为困难,因此选择非线性拟合的方式求得一组优化参数。示例性地,可以采用非线性最小二乘算法是Levenbert-Marquardt算法(简称LM算法)。LM算法利用梯度来求得函数最值,具体步骤在此不进行赘述,由LM算法求得的一组误差参数作为最后的标定结果。该方程组的求解方法可采用非线性拟合方法,当然也可以采用其他方法,解决非线性最小二乘问题的方法很多,可以根据实际情况进行选择,本申请对于函数关系方程的求解方法不做限定。
通过非线性拟合的方法函数关系方程能够获得优化的解,从而保证通过求解函数关系方程得到的误差值更接近于真实值,标定的误差更准确、精度高。
在本申请的一个实施例中,所述误差项集合中的每个误差项具有对应 的敏感度,所述敏感度用于表征误差项对所述扫描装置的测量结果准确度的影响程度。
根据前面的分析,在误差标定过程中,当光学***较为复杂时,其中包含的组件较多,由于每个组件会引入多个误差项,因此,当组件较多时,误差项参数过多。由于误差项参数均为未知数,因此待求解项的增多将造成(1)求解难度增高;(2)参考点方向向量收集难度增大导致的标定环境复杂化;(3)求解效率降低等问题。
前文根据光学定律对双棱镜光学建立了理论模型,并通过引入误差项集合对上述理论模型进行拟合求解,从而获得各个误差项的数值,这种获得误差项数值的方法可称为建模拟合法,因此,根据本申请提出的建模拟合法将先进行简单的误差项敏感度分析。识别每个误差项的敏感度将有助于降低求解难度,简化标定环境,提升求解效率。
在本申请的一个实施例中,所述误差项集合中包括第一误差项和第二误差项,若所述第一误差项变动第一阈值且所述第二误差项不变动时所述扫描装置出射的光线在扫描视场内的方向向量的准确度大于所述第一误差项不变动且所述第二误差项变动所述第一阈值时所述扫描装置出射的光线在扫描视场内的方向向量的准确度,则所述第一误差项的敏感度高于所述第二误差项的敏感度。
敏感度定义为误差项集合{Δ′}内单一参数变动时对结果准确度的影响程度。集合内第k个误差项Δ k的敏感度为
Figure PCTCN2019122145-appb-000048
实际计算时可以简化计算,令单一参数增加或减少1个单位,以计算结果偏差值作为该参数敏感度。
对于不同的误差项,其具有不同的敏感度,因此不同的误差项的敏感能够相互比较,根据误差项敏感度的绝对值以及相对大小,可以得到误差项集合{Δ}内各个误差项。,为后续对误差项的处理提供了基础,并且将有助于降低求解难度,简化标定环境,提升求解效率。
在本申请的一个实施例中,所述误差项集合中的误差项为敏感度大于等于预设阈值的误差项。
在得到所有误差项参数的敏感度后,可以依据***实际情况(工差控制等)及精度要求对误差参数进行合理筛选,滤除不敏感参数,保留敏感 参数。其中,对于误差项进行滤除或保留的标准为设置一个预设阈值,当误差项的敏感度小于预设阈值时,认为其误差值所导致的***出射的光线的方向向量的变化较小,因此,将其滤除,滤除后将减少待求解项的数目,进而降低求解难度,简化标定环境,提升求解效率。
在本申请的一个实施例中,还包括:当所述误差项集合中的任意两个误差项对所述扫描装置出射的光线在扫描视场内的方向向量的准确度的影响程度具有正比例关系时,将所述任意两个误差项合并成一个误差项添加到所述误差项集合中。
进行参数选择时另一个需要考虑的问题是参数耦合。复杂***中,可能会出现某一项误差项参数变化对最终结果的影响能在一定程度上被另一误差项参数的变化所替代。若这种替换性很强,表明两个参数间存在很强的相关关系,则可以考虑将两个参数进行“整合”,用单一变量来进行表示,减少待求解项的数目,进而降低求解难度,简化标定环境,提升求解效率。
例如:两个误差参数间的耦合性可以由以下方法获得:(1)令误差项参数1增加或减少一个单位,保持其余误差项参数不变,遍历所有误差项组合,计算所有误差项组合情况下角度偏移;(2)对参数2重复步骤(1);(3)计算(1)、(2)中所得角度偏移值的相关系数,作为两个参数间的耦合度。对于耦合度较高的参数项,将其“整合”为一个误差项。
具体地,经过前述的滤除和“整合”等合理分析筛选后,误差项参数集数量会有一个较为明显的下降,写作{Δ}。基于激光光线误差模型
Figure PCTCN2019122145-appb-000049
将误差项集合{Δ}作为未知数(假设整机一旦装配完成,{Δ}中的值不再改变),扫描装置获得的
Figure PCTCN2019122145-appb-000050
作为内参代入模型,每一组误差项可以得到一个等式。在标定场景中收集测得足够多组等式即可对误差项集合进行求解。
经过上述的过程后,可减少待求解项的数目,进而降低求解难度,简化标定环境,提升求解效率。
在本申请的一个实施例中,还包括:将所述扫描装置的***误差发送给所述扫描装置,所述***误差用于所述扫描装置计算所述扫描装置出射的光线在扫描视场内的方向向量。
示例性地如图8所示,扫描装置获得误差标定装置计算出的误差项的 值{Δ},结合
Figure PCTCN2019122145-appb-000051
的值,将其带入
Figure PCTCN2019122145-appb-000052
即可以获得扫描装置实际出射的光线的方向向量。
示例性地如图8所示,激光雷达能够采集自身的数据,包括点云数据,激光雷达扫到空间中每个点时,激光雷达发射的激光光源的方向向量、激光雷达内部各个组件的旋转角度等等。扫描装置将得到的数据发送给误差标定装置,其中,扫描装置的数据可以选择为实时发送给误差标定装置,也可以选择扫描结束后,将所有的数据一起发送给误差标定装置。
其中,激光雷达采集的点云数据中的点与测量仪器测得的参考点的方向向量具有对应关系。误差标定装置在获得激光雷达发送的多组***输入参数后,可以确定每组***输入参数各自对应的参考点,进而将每组输入参数中的Vin、角度组合以及与之对应的参考点的已知方向向量代入多组等式即可对误差项集合进行求解。
其中误差标定装置示例性地为PC,或其他包含存储器和处理器的硬件设备。
在本申请的一个实施例中,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光线经过了所述扫描装置的光学***后出射的光线的方向向量。
在本申请的一个实施例中,所述绕轴旋转参数包括绕轴旋转角度。
如表1中所示的ε 1、ε 2
在本申请的一个实施例中,所述组件包括反射镜和/或棱镜。
如图1中的组件包含了反射镜和棱镜,其中,反射镜又称平面镜,如图2中的组件包含了两个棱镜。图1和图2仅为示例性展示,在其他的实施方式中,组件可以仅包含反射镜,且本申请对其包含的反射镜和/或棱镜的个数不做限定。
在本申请的一个实施例中,所述误差项集合中的误差项包括转轴法向量偏差、反射镜和/或棱镜的初始法向量偏差、激光光源出射偏差、绕轴旋转角度测量偏差或棱镜折射率偏差中的一个或多个。
图表1中所列举的误差项。然而,本申请对于上述误差项不做限定,其还可以包括任何组件由于误差所引入的误差项。
在本申请的一个实施例中,不同组的***输入参数中至少包括一个不 同的组件的绕轴旋转参数。
对于不同组的***输入参数,其至少包含一个不同的组件的绕轴旋转参数ε,从而实现对绕轴旋转参数的求解。
在本申请的一个实施例中,所述参考点为位于所述扫描装置的扫描视场内的标记物。
参考点的选择可以是预先设定的,例如预先设定标记物,标识出参考点所在位置,当扫描装置扫到该标记物时,即认定其所处位置为参考点位置。
在本申请的一个实施例中,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。
前述的方向向量均为扫描装置的空间坐标系下的方向向量。即方向向量的测量是以扫描装置所处的空间坐标系来标定的。
在本申请的一个实施例中,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。如图2和图3所示的激光的位置,其为激光光源发射的激光经过光学***:棱镜1和棱镜2之前的方向向量。
在本申请的一个实施例中,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。对于雷达的种类,本申请不做限定,只要能实现探测的功能,均落入本申请的扫描装置的范围。
根据前文所述的误差标定方法或采用上述方法的装置对扫描装置的误差进行标定,能够分别获得各个组件的误差值,因此,至少具有普适性强,高效,全自动,精度高,可支持量产,标定环境简单的标定效果。
下面将结合附图6和附图7从另一个角度来描述本申请所涉及的误差标定方法。
下面参照图7描述本申请实施例的第二误差标定方法200的示意性流程图。
如图7所示,误差标定方法200可以包括如下步骤:
在步骤S210,从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量。
在步骤S220,从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量。
在步骤S230,根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。
第二误差标定方法,也可以称为补偿法,在使用补偿法误差标定的过程中,将误差传导的具体情况视作黑盒,仅关注不同情况下实际出射方向与按照标准模型(不引入误差项的情况)计算所得的出射方向两者间的差异,并通过一定的机制将这些误差予以补偿,减小最终结果的误差。
采用上述第二误差标定方法可以忽略掉***内部各个误差项的具体数值,仅根据误差传导前后的变化,对***进行补偿,所以该方法适用于各种复杂***,普适性强,而且还具有高效,全自动,精度高,可支持量产,标定环境简单等优点。
在本申请的一个实施例中,步骤S210:从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量。当扫描装置扫描到扫描视场内的参考点时,计算此时扫描装置的理想出射光线的方向向量,此时计算得到的方向向量为第一方向向量。
步骤S220:从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量。参考点的方向向量由测量仪器获得,其中,并将参考点的方向向量作为第二方向向量。
步骤S230:根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。根据第一方向向量和第二方向向量的关系确定参考点所对应的目标区域内的***误差补偿值。第一方向向量为理想值,第二方向向量为实际输出的值,通过二者之间的关系,可以确定***的误差值,进而得到补偿值。
在本申请的一个实施例中,步骤S230中的目标区域是通过划分扫描视场得到的。目标区域通过对扫描视场划分得到,例如,一个参考点对应 一个目标区域,或者四个参考点作为目标区域的四个顶点。
以下将结合具体的实施例说明补偿法所包含的第一补偿方案:
在本申请的一个实施例中,所述目标区域由所述扫描装置发射的光线相较于所述扫描装置的方位角和天顶角限定。通过上述方法可以将视场细分为大量区域(例如按照方位角1度*天顶角1度做区域划分,则方位角在0~1度且天顶角在0~1度间的方向向量为一个细分区域)。
在本申请的一个实施例中,步骤S230中包括将所述第一方向向量与所述第二方向向量的差值作为所述参考点对应的目标区域内的***误差补偿值。补偿值直接取为
Figure PCTCN2019122145-appb-000053
其中,基于标准模型f计算得到第j个光学组件的特定角度组合在区域i的marker方向向量
Figure PCTCN2019122145-appb-000054
测得区域i的marker在雷达坐标系下的名义方向向量
Figure PCTCN2019122145-appb-000055
所述名义方向向量是指测量仪所获得的扫描装置出射的光线的实际方向向量,或称为真实方向向量。区域i与第j个光学组件的特定角度组合具有对应关系。
在本申请的一个实施例中,所述扫描装置包括一个或多个组件,所述一个或多个组件中各组件的角度的排列组合构成角度组合,所述角度为绕轴旋转角度。
在本申请的一个实施例中,还包括:生成至少一个角度组合集群与至少一个目标区域内的***误差补偿值的对应关系,每个角度组合集群中包括一个或多个角度组合,其中,所述至少一个角度组合集群中包括第一角度组合集群,所述至少一个目标区域中包括第一目标区域,所述第一角度组合集群对应所述第一目标区域内的***误差补偿值,所述第一角度组合集群中包含的一个或多个角度组合是所述扫描装置出射的光线落入所述目标区域时,所述扫描装置中的一个或多个组件的角度组合。通过遍历变量组合的方式来进行补偿。对各个组件的角度组合进行分类,可以认为满足以下条件的角度组合出射角度误差近似相等:出射方向落在同一细分区域内、角度组合近似。角度组合近似定义如下:将两个角度组合视作两个等长向量a,b,若
Figure PCTCN2019122145-appb-000056
则认为两个角度组合近似。其中T为设定的阈值,可视实际情况调整。
由上述分组规则可以将角度组合分为n个集群,设为(m 1,m 2...,m n),每个集群对应同一个角度误差补偿值
Figure PCTCN2019122145-appb-000057
及同一个细分区域i。每个细分区域 i内可以设置一个或多个标记物进行上述步骤210-230的计算得到该细分区域i对应的***误差补偿值。
在本申请的一个实施例中,所述组件包括反射镜和/或棱镜。其中,反射镜又称平面镜。
在本申请的一个实施例中,所述***误差补偿值用于所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述扫描装置中的一个或多个组件的角度组合和所述一个或多个组件的角度组合对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量。
在本申请的一个实施例中,前一个实施例中的步骤中包括:所述扫描装置获取所述一个或多个组件的角度组合所在的第二角度组合集群,并获取所述第二角度组合集群对应的区域内的***误差补偿值,
所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述一个或多个组件的角度组合所在的第二角度组合集群和所述第二角度组合集群对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量。
例如,扫描装置的激光光源发射的光线的方向向量为V in,一个或多个组件的角度组合对应的集群为第k个集群,第k个集群对应的区域为区域n,且区域n具有误差补偿值c n,基于标准模型f计算得到第k个集群在区域n的marker方向向量V out,扫描装置根据c n和V out可以确定所述扫描装置出射的光线的实际方向向量。
其中,第二角度组合集群也可以是目标区域对应的第一角度组合集群,也可以不是第一角度组合集群。
在本申请的一个实施例中,所述扫描装置出射的光线的实际方向向量是由所述扫描装置根据所述扫描装置出射的光线的方向向量与所述第二角度组合集群对应的区域内的***误差补偿值相加得到的,所述扫描装置出射的光线的方向向量是由所述扫描装置根据所述激光光源发射的光线的方向向量计算得到的。
为了全面说明第一补偿方案,给出如下示例性说明:激光雷达实际运行中的变量为各棱镜或平面镜绕轴旋转的角度。因此可以通过遍历变量组合的方式来进行补偿。但需要注意的是,由于映射函数的非线性,等间隔 抽样遍历并不是合理的遍历方式,因为这样并不能保证选取的变量组合对应的出射方向在视场内均匀分布。
针对上述问题,我们先引入一个假设:误差项集合中的误差都受到一定程度上的工差管控,为相对小值。所有误差项的综合作用在成像各区域都不会使得激光出射角度存在明显偏差(e.g.>1度)。在此基础上,我们将视场细分为大量区域(例如按照方位角1度*天顶角1度做区域划分,则方位角在0~1度且天顶角在0~1度间的方向向量为一个细分区域)。
基于上述假设和区域划分,可以认为满足以下条件的角度组合出射角度误差近似相等:出射方向落在同一细分区域内、角度组合近似。
角度组合近似定义如下:将两个角度组合视作两个等长向量a,b,若
Figure PCTCN2019122145-appb-000058
则认为两个角度组合近似。其中T为设定的阈值,可视实际情况调整。
由上述分组规则可以将角度组合分为n个集群,设为(m 1,m 2...,m n),每个集群对应同一个角度误差补偿值
Figure PCTCN2019122145-appb-000059
及同一个细分区域i。针对第j个集群,角度误差补偿值可以由下述方法得到:
1.在每个细分区域布置一个marker
2.基于标准模型f计算得到第j个集群在区域i的marker方向向量
Figure PCTCN2019122145-appb-000060
3.测得区域i的marker在雷达坐标系下的名义方向向量
Figure PCTCN2019122145-appb-000061
4.补偿值
Figure PCTCN2019122145-appb-000062
扫描装置在计算出射光线的方向向量时补偿方法如下:
1.根据当前角度组合找到其对应的集群(假设为第k个集群)
2.根据下述公式计算出射方向:
Figure PCTCN2019122145-appb-000063
其中
Figure PCTCN2019122145-appb-000064
表示经过补偿的出射方向最终值,f为不引入内参情况时的标准计算函数,
Figure PCTCN2019122145-appb-000065
为第k个集群的补偿值。
以下将结合具体的实施例说明补偿法所包含的第二补偿方案:
在本申请的一个实施例中,步骤S230中,包括确定所述第一方向向量与所述第二方向向量的差值,N个所述参考点对应一个目标区域,并根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,N为大于等于2的正整数。
在本申请的一个实施例中,所述目标区域为矩形网格。
例如,如图6所示,将四个参考点所对应的目标区域,跟据四个参考点的误差补偿值得到目标区域的误差补偿值。本方案标定时将视角(或称为视场角)(field of view,FOV)网格化,网格的交点为参考marker。图6为网格化的一种实例,其中参考点为参考marker,每四个参考marker围成一个网格。(如marker a,b,c,d围成的区域形成一个矩形网格,依次类推)。
在本申请的一个实施例中,所述根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括对所述N个参考点的误差补偿值取平均值作为所述目标区域的误差补偿值。同一个网格内解算值与真实值存在一个近似的映射关系,示例性地,此映射关系能通过围成该网格marker的真实值-解算值关系得到。其中,解算值对应于光学组件处于设计位姿时,扫描装置计算出的出射的光线的方向向量的值。真实值是指测量仪器测量的扫描装置出射的光线的实际方向向量,或称为真实方向向量。
平均值的计算相对简单,因此能够保证误差标定方法的高效性。
在本申请的一个实施例中,所述对N个差值取平均值作为所述目标区域的误差补偿值,包括对所述N个参考点的误差补偿值取算数平均值或均方根平均值。
例如,区域i对应的误差补偿值的映射关系示例性地表示为F(i)=((a-a')+(b-b')+(c-c')+(d-d'))/4,或均方根平均值。其中,a为参考点之一实际值,a'为参考点之一解算值,a-a'为参考点之一误差值。类似的,b,c,d均为参考点的实际值,b',c',d'均为参考点的解算值。假设x为abcd参考点所对应区域内的任意一点的解算值,则根据上述映射关系F(i)可以确定x对应的实际值x'。F(i)可以是任意一种以abcd为变量的映射关系。
在本申请的一个实施例中,所述根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括根据N个所述参考点的误差补偿值拟合出的函数关系作为所述目标区域的误差补偿值。其中,每个目标区域具有一个误差补偿值。
在本申请的一个实施例中,所述根据N个所述参考点的误差补偿值拟合出的函数关系作为所述补偿值,包括根据N个所述参考点的误差补偿值插值得到所述函数关系作为所述目标区域的误差补偿值。同一个网格内测 量值与真实值存在一个近似的映射关系。
在本申请的一个实施例中,还包括:生成至少一个区域与至少一个***误差补偿值的对应关系,其中,所述至少一个区域中包括所述目标区域。
在本申请的一个实施例中,所述***误差补偿值用于所述扫描装置根据所述扫描装置出射的光线在扫描视场内的方向向量确定所述扫描装置出射的光线的实际方向向量。在得到误差补偿值后,根据附图8所示的应用场景,可以得到扫描装置出射的光线的实际方向向量。
扫描装置在计算出射光线的方向向量时补偿方法如下:
扫描装置获取扫描装置出射的光线在扫描视场内的方向向量,
扫描装置获取误差补偿补偿装置中的对应前一步骤中获得方向向量的区域所对应的误差值;
根据扫描装置出射的光线在扫描视场内的方向向量,以及所获得的误差值,得到扫描装置出射光线的实际方向向量。在本申请的一个实施例中,在前一个实施例中包括所述扫描装置获取第一区域对应的***误差补偿值,从而获得所述扫描装置出射的光线的实际方向向量,所述第一区域是所述扫描装置出射的光线在扫描视场内的方向向量落入的区域。由此可见,区域与***误差补偿值之间具有对应关系。
在本申请的一个实施例中,所述扫描装置出射的光线在扫描视场内的方向向量是根据所述扫描装置的激光光源发射的光线的方向向量计算得到的。扫描装置出射的光线在扫描视场内的方向向量是基于理想模型计算出射方向,基于理想模型计算得到的理想输出值,作为误差补偿的基础,能够获得扫描装置实际出射的光线的方向向量。
在本申请的一个实施例中,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。前述的方向向量均为扫描装置的空间坐标系下的方向向量。即方向向量的测量是以扫描装置所处的空间坐标系来标定的。
在本申请的一个实施例中,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。如图2和图 3所示的激光的位置,其为激光光源发射的激光经过光学***:棱镜1和棱镜2之前的方向向量。
在本申请的一个实施例中,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光经过了所述扫描装置的光学***后出射的光线的方向向量。
扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光经过了所述扫描装置的光学***后出射的光线的方向向量。如图2和图3所示的激光的位置,其为激光光源发射的激光经过了光学***:棱镜1和棱镜2之后的方向向量。
在本申请的一个实施例中,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。
对于雷达的种类,本申请不做限定,只要能实现探测的功能,均落入本申请的扫描装置的范围。
在本申请的实施例中,可以采用扫描装置向被测物发射光脉冲信号,并接收所述光脉冲信号对应的反射脉冲信号。其中,所述探测装置包括但不限于激光雷达、电磁波雷达、毫米波雷达或者超声波雷达等。
为了全面说明第二补偿方案,给出如下示例性说明:第一补偿方案中,为保证补偿精度,需要将区域尽可能细分。而大量的细分区域会造成极大的遍历区间从而极大的影响标定效率。同时在大量的细小区域内均匀布置marker在实操上存在难度。
为了进一步改进第一补偿方案,第二补偿方案在标定时将FOV网格化,网格的交点为参考marker。图6为网格化的一种实例,其中参考点为参考marker,每四个参考marker围成一个网格。(如marker a,b,c,d围成的区域形成一个矩形网格,依次类推)。
在此方案中,我们同样需要引入一个假设:同一个网格内解算值与真实值存在一个近似的映射关系,且此映射关系能通过围成该网格marker的真实值-解算值关系得到。
基于上述假设,标定流程可参见图6:得到所有网格中的映射关系后,误差标定装置发送所有网格(目标区域)的真实值与解算值的映射关系给扫描装置,扫描装置在实际使用过程中从映射关系中找对应的补偿值,并且 具体地,扫描装置可以由下述方式得到补偿后的出射方向:
(1)基于理想模型计算出射方向
Figure PCTCN2019122145-appb-000066
(2)根据出射方向找到对应的网格,假设为网格i
(3)补偿后的出射方向为
Figure PCTCN2019122145-appb-000067
基于上面的描述,根据本申请实施例的***误差标定方法、装置和计算机可读存储介质在根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差或者根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,从而获得具有普适性强,高效,全自动,精度高,可支持量产,标定环境简单等优点的误差标定方案。
以上示例性地描述了根据本申请实施例的***误差标定方法。根据本申请实施例的误差标定装置可以用于实施上文中描述的根据本申请实施例的误差标定方法。为了简洁,下文中仅对误差标定装置的主要结构和功能进行描述,而省略上文中已经描述的部分具体细节。
下面结合图9描述根据本申请另一方面提供的误差标定装置1000。根据本申请实施例的误差标定装置1000可以用于实施上文中描述的根据本申请实施例的误差标定方法100,200。为了简洁,下文中仅对误差标定装置1000的主要结构和功能进行描述,而省略上文中已经描述的部分具体细节。
如图9所示,误差标定装置1000可以包括处理器1010、存储器1020,其中,所述存储器1020上存储有由所述处理器1010运行的计算机程序,所述处理器1010用于运行计算器程序时执行前述误差标定方法。示例性地,所述误差标定装置可以包括但不限于PC,或其他包含存储器和处理器的硬件设备。
根据本申请的再一方面,还提供了一种计算机可读存储介质,所述计算机可读存储介质上存储有计算机程序,所述计算机程序在运行时执行根据本申请实施例的误差标定方法。所述计算机可读存储介质例如可以包括智能电话的存储卡、平板电脑的存储部件、个人计算机的硬盘、只读存储器(ROM)、可擦除可编程只读存储器(EPROM)、便携式紧致盘只读存储器(CD-ROM)、USB存储器、或者上述存储介质的任意组合。所述计算机可 读存储介质可以是一个或多个计算机可读存储介质的任意组合。
尽管这里已经参考附图描述了示例实施例,应理解上述示例实施例仅仅是示例性的,并且不意图将本申请的范围限制于此。本领域普通技术人员可以在其中进行各种改变和修改,而不偏离本申请的范围和精神。所有这些改变和修改意在被包括在所附权利要求所要求的本申请的范围之内。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中,应该理解到,所揭露的设备和方法,可以通过其它的方式实现。例如,以上所描述的设备实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个设备,或一些特征可以忽略,或不执行。
在此处所提供的说明书中,说明了大量具体细节。然而,能够理解,本申请的实施例可以在没有这些具体细节的情况下实践。在一些实例中,并未详细示出公知的方法、结构和技术,以便不模糊对本说明书的理解。
类似地,应当理解,为了精简本申请并帮助理解各个发明方面中的一个或多个,在对本申请的示例性实施例的描述中,本申请的各个特征有时被一起分组到单个实施例、图、或者对其的描述中。然而,并不应将该本申请的方法解释成反映如下意图:即所要求保护的本申请要求比在权利要求中所明确记载的特征更多的特征。更确切地说,如相应的权利要求书所反映的那样,其发明点在于可以用少于某个公开的单个实施例的所有特征的特征来解决相应的技术问题。因此,遵循具体实施方式的权利要求书由此明确地并入该具体实施方式,其中权利要求本身都作为本申请的单独实施例。
本领域的技术人员可以理解,除了特征之间相互排斥之外,可以采用任何组合对本说明书(包括伴随的权利要求、摘要和附图)中公开的所有特征以及如此公开的任何方法或者设备的所有过程或单元进行组合。除非 另外明确陈述,本说明书(包括伴随的权利要求、摘要和附图)中公开的特征可以由提供相同、等同或相似目的的替代特征来代替。
此外,本领域的技术人员能够理解,尽管在此所述的一些实施例包括其它实施例中所包括的某些特征而不是其它特征,但是不同实施例的特征的组合意味着处于本申请的范围之内并且形成不同的实施例。例如,在权利要求书中,所要求保护的实施例的任意之一都可以以任意的组合方式来使用。
本申请的各个部件实施例可以以硬件实现,或者以在一个或者多个处理器上运行的软件模块实现,或者以它们的组合实现。本领域的技术人员应当理解,可以在实践中使用微处理器或者数字信号处理器(DSP)来实现根据本申请实施例的一些模块的一些或者全部功能。本申请还可以实现为用于执行这里所描述的方法的一部分或者全部的装置程序(例如,计算机程序和计算机程序产品)。这样的实现本申请的程序可以存储在计算机可读存储介质上,或者可以具有一个或者多个信号的形式。这样的信号可以从因特网网站上下载得到,或者在载体信号上提供,或者以任何其他形式提供。
应该注意的是上述实施例对本申请进行说明而不是对本申请进行限制,并且本领域技术人员在不脱离所附权利要求的范围的情况下可设计出替换实施例。在权利要求中,不应将位于括号之间的任何参考符号构造成对权利要求的限制。本申请可以借助于包括有若干不同元件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中,这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。
以上所述,仅为本申请的具体实施方式或对具体实施方式的说明,本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。本申请的保护范围应以权利要求的保护范围为准。

Claims (85)

  1. 一种***误差标定方法,其特征在于,包括:
    获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数;
    根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
  2. 根据权利要求1所述的误差标定方法,其特征在于,所述根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差,包括:
    将所述至少两个参考点的方向向量、所述至少两组***输入参数以及所述误差项集合代入预设误差模型计算所述扫描装置的***误差。
  3. 根据权利要求2所述的误差标定方法,其特征在于,所述预设误差模型为基于光学定律的函数关系方程。
  4. 根据权利要求3所述的误差标定方法,其特征在于,所述函数关系方程的求解采用非线性拟合方法。
  5. 根据权利要求1-4任一项所述的误差标定方法,其特征在于,所述误差项集合中的每个误差项具有对应的敏感度,所述敏感度用于表征误差项对所述扫描装置的测量结果准确度的影响程度。
  6. 根据权利要求5所述的误差标定方法,其特征在于,所述误差项集合中包括第一误差项和第二误差项,若所述第一误差项变动第一阈值且所述第二误差项不变动时所述扫描装置出射的光线在扫描视场内的方向向量的准确度大于所述第一误差项不变动且所述第二误差项变动所述第一阈值时所述扫描装置出射的光线在扫描视场内的方向向量的准确度,则所述第一误差项的敏感度高于所述第二误差项的敏感度。
  7. 根据权利要求5或6所述的误差标定方法,其特征在于,所述误差项集合中的误差项为敏感度大于等于预设阈值的误差项。
  8. 根据权利要求1-7任一项所述的误差标定方法,其特征在于,还包括:当所述误差项集合中的任意两个误差项对所述扫描装置出射的光线在 扫描视场内的方向向量的准确度的影响程度具有正比例关系时,将所述任意两个误差项合并成一个误差项添加到所述误差项集合中。
  9. 根据权利要求1-8任一项所述的误差标定方法,其特征在于,还包括:
    将所述扫描装置的***误差发送给所述扫描装置,所述***误差用于所述扫描装置计算所述扫描装置出射的光线在扫描视场内的方向向量。
  10. 根据权利要求9所述的误差标定方法,其特征在于,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光线经过了所述扫描装置的光学***后出射的光线的方向向量。
  11. 根据权利要求1-10任一项所述的误差标定方法,其特征在于,所述绕轴旋转参数包括绕轴旋转角度。
  12. 根据权利要求1-11任一项所述的误差标定方法,其特征在于,所述组件包括反射镜和/或棱镜。
  13. 根据权利要求1-12任一项所述的误差标定方法,其特征在于,所述误差项集合中的误差项包括转轴法向量偏差、反射镜和/或棱镜的初始法向量偏差、激光光源出射偏差、绕轴旋转角度测量偏差或棱镜折射率偏差中的一个或多个。
  14. 根据权利要求1-13任一项所述的误差标定方法,其特征在于,不同组的***输入参数中至少包括一个不同的组件的绕轴旋转参数。
  15. 根据权利要求1-14任一项所述的误差标定方法,其特征在于,所述参考点为位于所述扫描装置的扫描视场内的标记物。
  16. 根据权利要求1-15任一项所述的误差标定方法,其特征在于,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。
  17. 根据权利要求1-16任一项所述的误差标定方法,其特征在于,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
  18. 根据权利要求1-17任一项所述的误差标定方法,其特征在于,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。
  19. 一种***误差标定方法,其特征在于,包括:
    从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计 算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量;
    从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量;
    根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。
  20. 根据权利要求19所述的误差标定方法,其特征在于,所述目标区域是通过划分扫描视场得到的。
  21. 根据权利要求19或20所述的误差标定方法,其特征在于,所述根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,
    包括将所述第一方向向量与所述第二方向向量的差值作为所述参考点对应的目标区域内的***误差补偿值。
  22. 根据权利要求21所述的误差标定方法,其特征在于,所述目标区域由所述扫描装置发射的光线相较于所述扫描装置的方位角和天顶角限定。
  23. 根据权利要求19-22任一项所述的误差标定方法,其特征在于,所述扫描装置包括一个或多个组件,所述一个或多个组件中各组件的角度的排列组合构成角度组合,所述角度为绕轴旋转角度。
  24. 根据权利要求23所述的误差标定方法,其特征在于,还包括:
    生成至少一个角度组合集群与至少一个目标区域内的***误差补偿值的对应关系,每个角度组合集群中包括一个或多个角度组合,其中,所述至少一个角度组合集群中包括第一角度组合集群,所述至少一个目标区域中包括第一目标区域,所述第一角度组合集群对应所述第一目标区域内的***误差补偿值,所述第一角度组合集群中包含的一个或多个角度组合是所述扫描装置出射的光线落入所述目标区域时,所述扫描装置中的一个或多个组件的角度组合。
  25. 根据权利要求23或24所述的误差标定方法,其特征在于,所述组件包括反射镜和/或棱镜。
  26. 根据所述权利要求24或25所述的误差标定方法,其特征在于, 所述***误差补偿值用于所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述扫描装置中的一个或多个组件的角度组合和所述一个或多个组件的角度组合对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量。
  27. 根据所述权利要求26所述的误差标定方法,其特征在于,所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述扫描装置中的一个或多个组件的角度组合和所述一个或多个组件的角度组合对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量,包括:
    所述扫描装置获取所述一个或多个组件的角度组合所在的第二角度组合集群,并获取所述第二角度组合集群对应的区域内的***误差补偿值,
    所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述一个或多个组件的角度组合所在的第二角度组合集群和所述第二角度组合集群对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量。
  28. 根据所述权利要求27所述的误差标定方法,其特征在于,所述扫描装置出射的光线的实际方向向量是由所述扫描装置根据所述扫描装置出射的光线的方向向量与所述第二角度组合集群对应的区域内的***误差补偿值相加得到的,所述扫描装置出射的光线的方向向量是由所述扫描装置根据所述激光光源发射的光线的方向向量计算得到的。
  29. 根据权利要求19所述的误差标定方法,其特征在于,所述根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,
    包括确定所述第一方向向量与所述第二方向向量的差值,N个所述参考点对应一个目标区域,并根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,N为大于等于2的正整数。
  30. 根据权利要求29所述的误差标定方法,其特征在于,所述根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括对所述N个参考点的误差补偿值取平均值作为所述目标区域的误差补偿值。
  31. 根据权利要求29所述的误差标定方法,其特征在于,所述对N 个差值取平均值作为所述目标区域的误差补偿值,包括对所述N个参考点的误差补偿值取算数平均值或均方根平均值。
  32. 根据权利要求29所述的误差标定方法,其特征在于,所述根据N个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括根据N个所述参考点的误差补偿值拟合出的函数关系作为所述目标区域的误差补偿值。
  33. 根据权利要求29-32任一项所述的误差标定方法,其特征在于,所述根据N个所述参考点的误差补偿值拟合出的函数关系作为所述补偿值,包括根据N个所述参考点的误差补偿值插值得到所述函数关系作为所述目标区域的误差补偿值。
  34. 根据权利要求29-33任一项所述的误差标定方法,其特征在于,所述目标区域为矩形网格。
  35. 根据权利要求29-34任一项所述的误差标定方法,其特征在于,还包括:
    生成至少一个区域与至少一个***误差补偿值的对应关系,其中,所述至少一个区域中包括所述目标区域。
  36. 根据权利要求35所述的误差标定方法,其特征在于,所述***误差补偿值用于所述扫描装置根据所述扫描装置出射的光线在扫描视场内的方向向量确定所述扫描装置出射的光线的实际方向向量。
  37. 根据权利要求36所述的误差标定方法,其特征在于,所述扫描装置根据所述扫描装置出射的光线在扫描视场内的方向向量确定所述扫描装置出射的光线的实际方向向量,包括:
    所述扫描装置获取第一区域对应的***误差补偿值,从而获得所述扫描装置出射的光线的实际方向向量,所述第一区域是所述扫描装置出射的光线在扫描视场内的方向向量落入的区域。
  38. 根据所述权利要求36或37所述的误差标定方法,其特征在于,
    所述扫描装置出射的光线在扫描视场内的方向向量是根据所述扫描装置的激光光源发射的光线的方向向量计算得到的。
  39. 根据权利要求19-38任一项所述的误差标定方法,其特征在于,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。
  40. 根据权利要求19-39任一项所述的误差标定方法,其特征在于,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
  41. 根据权利要求19-40任一项所述的误差标定方法,其特征在于,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光经过了所述扫描装置的光学***后出射的光线的方向向量。
  42. 根据权利要求19-41任一项所述的误差标定方法,其特征在于,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。
  43. 一种扫描装置的误差标定装置,其特征在于,所述误差标定装置包括存储器和处理器,所述存储器上存储有由所述处理器运行的计算机程序,所述处理器用于运行计算器程序时执行:
    获取扫描装置扫描到至少两个参考点时所述扫描装置的至少两组***输入参数,所述参考点为扫描视场内方向向量已知的参考点,所述***输入参数包括所述扫描装置的激光光源发射的光线的方向向量以及所述扫描装置中的一个或多个组件的绕轴旋转参数;
    根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差。
  44. 根据权利要求43所述的误差标定装置,其特征在于,
    所述处理器用于根据所述至少两个参考点的方向向量、所述至少两组***输入参数以及预设的误差项集合确定所述扫描装置的***误差,包括:
    将所述至少两个参考点的方向向量、所述至少两组***输入参数以及所述误差项集合代入预设误差模型计算所述扫描装置的***误差。
  45. 根据权利要求44所述的误差标定装置,其特征在于,所述预设误差模型为基于光学定律的函数关系方程。
  46. 根据权利要求45所述的误差标定装置,其特征在于,所述函数关系方程的求解采用非线性拟合方法。
  47. 根据权利要求43-46任一项所述的误差标定装置,其特征在于,所述误差项集合中的每个误差项具有对应的敏感度,所述敏感度用于表征误差项对所述扫描装置的测量结果准确度的影响程度。
  48. 根据权利要求47所述的误差标定装置,其特征在于,所述误差 项集合中包括第一误差项和第二误差项,若所述第一误差项变动第一阈值且所述第二误差项不变动时所述扫描装置出射的光线在扫描视场内的方向向量的准确度大于所述第一误差项不变动且所述第二误差项变动所述第一阈值时所述扫描装置出射的光线在扫描视场内的方向向量的准确度,则所述第一误差项的敏感度高于所述第二误差项的敏感度。
  49. 根据权利要求43-48任一项所述的误差标定装置,其特征在于,所述误差项集合中的误差项为敏感度大于等于预设阈值的误差项。
  50. 根据权利要求43-49任一项所述的误差标定装置,其特征在于,所述处理器还用于:当所述误差项集合中的任意两个误差项对所述扫描装置出射的光线在扫描视场内的方向向量的准确度的影响程度具有正比例关系时,将所述任意两个误差项合并成一个误差项添加到所述误差项集合中。
  51. 根据权利要求43-50任一项所述的误差标定装置,其特征在于,所述处理器还用于:
    将所述扫描装置的***误差发送给所述扫描装置,所述***误差用于所述扫描装置计算所述扫描装置出射的光线在扫描视场内的方向向量。
  52. 根据权利要求43-51任一项所述的误差标定装置,其特征在于,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光线经过了所述扫描装置的光学***后出射的光线的方向向量。
  53. 根据权利要求43-52所述的误差标定装置,其特征在于,所述绕轴旋转参数包括绕轴旋转角度。
  54. 根据权利要求43-53任一项所述的误差标定装置,其特征在于,所述组件包括反射镜和/或棱镜。
  55. 根据权利要求43-54任一项所述的误差标定装置,其特征在于,所述误差项集合中的误差项包括转轴法向量偏差、反射镜和/或棱镜的初始法向量偏差、激光光源出射偏差、绕轴旋转角度测量偏差或棱镜折射率偏差中的一个或多个。
  56. 根据权利要求55所述的误差标定装置,其特征在于,不同组的***输入参数中至少包括一个不同的组件的绕轴旋转参数。
  57. 根据权利要求43-56任一项所述的误差标定装置,其特征在于,所述参考点为位于所述扫描装置的扫描视场内的标记物。
  58. 根据权利要求43-57任一项所述的误差标定装置,其特征在于,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。
  59. 根据权利要求43-58任一项所述的误差标定装置,其特征在于,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
  60. 根据权利要求43-59任一项所述的误差标定装置,其特征在于,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。
  61. 一种扫描装置的误差标定装置,其特征在于,所述误差标定装置包括存储器和处理器,所述存储器上存储有由所述处理器运行的计算机程序,所述计算机程序在被所述处理器运行时执行:
    从扫描装置获取第一方向向量,所述第一方向向量是所述扫描装置计算的;所述第一方向向量是所述扫描装置扫描到扫描视场内的参考点时所述扫描装置出射的光线在扫描视场内的方向向量;
    从测量仪器获取第二方向向量,所述第二方向向量是所述测量仪器测量的所述参考点在所述扫描视场内的方向向量;
    根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值。
  62. 根据权利要求61所述的误差标定装置,其特征在于,所述目标区域是通过划分扫描视场得到的。
  63. 根据权利要求61或62所述的误差标定装置,其特征在于,所述处理器用于根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,
    包括将所述第一方向向量与所述第二方向向量的差值作为所述参考点对应的目标区域内的***误差补偿值。
  64. 根据权利要求63所述的误差标定装置,其特征在于,所述目标区域由所述扫描装置发射的激光相较于所述扫描装置的方位角和天顶角限定。
  65. 根据权利要求61-64任一项所述的误差标定装置,其特征在于,所述扫描装置包括一个或多个组件,所述一个或多个组件中各组件的角度的排列组合构成角度组合,所述角度为绕轴旋转角度。
  66. 根据权利要求65所述的误差标定装置,其特征在于,所述处理器还用于:
    生成至少一个角度组合集群与至少一个目标区域内的***误差补偿值的对应关系,每个角度组合集群中包括一个或多个角度组合,其中,所述至少一个角度组合集群中包括第一角度组合集群,所述至少一个目标区域中包括第一目标区域,所述第一角度组合集群对应所述第一目标区域内的***误差补偿值,所述第一角度组合集群中包含的一个或多个角度组合是所述扫描装置出射的光线落入所述目标区域时,所述扫描装置中的一个或多个组件的角度组合。
  67. 根据权利要求65或66所述的误差标定装置,其特征在于,所述组件包括反射镜和/或棱镜。
  68. 根据所述权利要求66或67所述的误差标定装置,其特征在于,所述***误差补偿值用于所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述扫描装置中的一个或多个组件的角度组合和所述一个或多个组件的角度组合对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量。
  69. 根据所述权利要求68所述的误差标定装置,其特征在于,
    所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述扫描装置中的一个或多个组件的角度组合和所述一个或多个组件的角度组合对应的区域内的***误差补偿值确定所述扫描装置出射的光线的实际方向向量,包括:
    所述扫描装置获取所述一个或多个组件的角度组合所在的第一角度集群,并获取所述第一角度集群对应的区域内的***误差补偿值,
    所述扫描装置根据所述扫描装置的激光光源发射的光线的方向向量、所述一个或多个组件的角度组合所在的第一角度集群和所述第一角度集群对应的区域内的***误差补偿值确定从而获得所述扫描装置出射的光线的实际方向向量。
  70. 根据所述权利要求69所述的误差标定装置,其特征在于,所述扫描装置出射的光线的实际方向向量是由所述扫描装置根据所述扫描装置出射的光线的方向向量与所述第一角度集群对应的区域内的***误差补偿 值相加得到的,所述扫描装置出射的光线的方向向量是由所述扫描装置根据所述激光光源发射的光线的方向向量计算得到的。
  71. 根据权利要求61所述的误差标定装置,其特征在于,所述处理器用于根据所述第一方向向量以及所述第二方向向量确定所述参考点对应的目标区域内的***误差补偿值,
    包括确定所述第一方向向量与所述第二方向向量的差值,四个所述参考点对应一个目标区域,并根据四个所述参考点的误差补偿值得到所述目标区域的误差补偿值,所述误差补偿值为误差补偿数值或误差补偿关系。
  72. 根据权利要求71所述的误差标定装置,其特征在于,所述处理器用于根据四个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括对四个参考点的误差补偿值取平均值作为所述目标区域的误差补偿值。
  73. 根据权利要求71所述的误差标定装置,其特征在于,所述对四个差值取平均值作为所述目标区域的误差补偿值,包括对四个参考点的误差补偿值取算数平均值或均方根平均值。
  74. 根据权利要求71所述的误差标定装置,其特征在于,所述处理器用于根据四个所述参考点的误差补偿值得到所述目标区域的误差补偿值,包括根据四个所述参考点的误差补偿值拟合出的函数关系作为所述目标区域的误差补偿值。
  75. 根据权利要求71-74任一项所述的误差标定装置,其特征在于,所述所处理器用于根据四个所述参考点的误差补偿值拟合出的函数关系作为所述补偿值,包括根据四个所述参考点的误差补偿值插值得到所述函数关系作为所述目标区域的误差补偿值。
  76. 根据权利要求71-75任一项所述的误差标定装置,其特征在于,所述目标区域为矩形网格。
  77. 根据所述权利要求71-76任一项所述的误差标定装置,其特征在于,所述处理器用于生成至少一个区域与至少一个***误差补偿值的对应关系,其中,所述至少一个区域中包括所述目标区域。
  78. 根据所述权利要求77所述的误差标定装置,其特征在于,所述***误差补偿值用于所述扫描装置根据所述扫描装置出射的光线在扫描视场内的方向向量确定所述扫描装置出射的光线的实际方向向量。
  79. 根据权利要求78所述的误差标定装置,所述扫描装置根据所述扫描装置出射的光线在扫描视场内的方向向量确定所述扫描装置出射的光线的实际方向向量,包括:
    所述扫描装置获取第一区域对应的***误差补偿值,从而获得所述扫描装置出射的光线的实际方向向量,所述第一区域是所述扫描装置出射的光线在扫描视场内的方向向量落入的区域。
  80. 根据所述权利要求78或79所述的误差标定装置,其特征在于,
    所述扫描装置出射的光线在扫描视场内的方向向量是根据所述扫描装置的激光光源发射的光线的方向向量计算得到的。
  81. 根据权利要求61-80任一项所述的误差标定装置,其特征在于,所述方向向量为在所述扫描装置的空间坐标系下的方向向量。
  82. 根据权利要求61-81任一项所述的误差标定装置,其特征在于,所述扫描装置的激光光源发射的光线的方向向量为所述扫描装置的激光光源发射的光线经过所述扫描装置的光学***前的方向向量。
  83. 根据权利要求61-82所述的误差标定装置,其特征在于,所述扫描装置出射的光线在扫描视场内的方向向量为所述扫描装置的激光光源发射的光线经过了所述扫描装置的光学***后出射的光线的方向向量。
  84. 根据权利要求61-83任一项所述的误差标定装置,其特征在于,所述扫描装置包括激光雷达、毫米波雷达或超声波雷达中任意一种或多种。
  85. 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质上存储有计算机程序,所述计算机程序在处理器上运行时所述处理器执行如权利要求1-42中的任一项所述的误差标定方法。
PCT/CN2019/122145 2019-11-29 2019-11-29 ***误差标定方法、装置和计算机可读存储介质 WO2021102980A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/122145 WO2021102980A1 (zh) 2019-11-29 2019-11-29 ***误差标定方法、装置和计算机可读存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/122145 WO2021102980A1 (zh) 2019-11-29 2019-11-29 ***误差标定方法、装置和计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2021102980A1 true WO2021102980A1 (zh) 2021-06-03

Family

ID=76129057

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/122145 WO2021102980A1 (zh) 2019-11-29 2019-11-29 ***误差标定方法、装置和计算机可读存储介质

Country Status (1)

Country Link
WO (1) WO2021102980A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140094970A1 (en) * 2012-10-02 2014-04-03 Fanuc Corporation Translation/rotation error compensation amount creating device
CN107290734A (zh) * 2017-08-22 2017-10-24 北京航空航天大学 一种基于自制地基激光雷达垂直度误差的点云误差校正方法
CN107290735A (zh) * 2017-08-22 2017-10-24 北京航空航天大学 一种基于自制地基激光雷达铅垂度误差的点云误差校正方法
CN107560563A (zh) * 2017-07-28 2018-01-09 华南理工大学 一种线激光三维测量装置标定及误差补偿方法
CN107703499A (zh) * 2017-08-22 2018-02-16 北京航空航天大学 一种基于自制地基激光雷达对准误差的点云误差校正方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140094970A1 (en) * 2012-10-02 2014-04-03 Fanuc Corporation Translation/rotation error compensation amount creating device
CN107560563A (zh) * 2017-07-28 2018-01-09 华南理工大学 一种线激光三维测量装置标定及误差补偿方法
CN107290734A (zh) * 2017-08-22 2017-10-24 北京航空航天大学 一种基于自制地基激光雷达垂直度误差的点云误差校正方法
CN107290735A (zh) * 2017-08-22 2017-10-24 北京航空航天大学 一种基于自制地基激光雷达铅垂度误差的点云误差校正方法
CN107703499A (zh) * 2017-08-22 2018-02-16 北京航空航天大学 一种基于自制地基激光雷达对准误差的点云误差校正方法

Similar Documents

Publication Publication Date Title
Wujanz et al. An intensity-based stochastic model for terrestrial laser scanners
Gudmundsson et al. Environmental effects on measurement uncertainties of time-of-flight cameras
Kurillo et al. Evaluating the accuracy of the azure kinect and kinect v2
Janßen et al. Decreasing the uncertainty of the target center estimation at terrestrial laser scanning by choosing the best algorithm and by improving the target design
CN107917701A (zh) 基于主动式双目立体视觉的测量方法及rgbd相机***
Tan et al. Investigation of TLS intensity data and distance measurement errors from target specular reflections
Medić et al. Towards system calibration of panoramic laser scanners from a single station
CN111372005B (zh) Tof摄像模组自动曝光补偿方法及其***
CN111913169B (zh) 激光雷达内参、点云数据的修正方法、设备及存储介质
CN112014829B (zh) 一种激光雷达扫描仪的性能指标测试方法与装置
CN105678076A (zh) 点云测量数据质量评估优化的方法及装置
Kermarrec et al. On the sensitivity of the parameters of the intensity-based stochastic model for terrestrial laser scanner. Case study: B-spline approximation
Guidi Metrological characterization of 3D imaging devices
CN116008964A (zh) 激光雷达及其反射率标定方法和反射率测量方法
Shahbazi et al. Range camera self-calibration based on integrated bundle adjustment via joint setup with a 2D digital camera
Bräuer-Burchardt et al. Underwater 3D measurements with advanced camera modelling
Jaafar et al. Terrestrial laser scanner error quantification for the purpose of monitoring
Kot et al. Using virtual scanning to find optimal configuration of a 3D scanner turntable for scanning of mechanical parts
Li et al. Lab-built terrestrial laser scanner self-calibration using mounting angle error correction
Curto et al. An experimental assessment of depth estimation in transparent and translucent scenes for Intel RealSense D415, SR305 and L515
CN108776338B (zh) 信号源空间传感方法、装置及主动式传感***
Ozendi et al. A generic point error model for TLS derived point clouds
WO2021102980A1 (zh) ***误差标定方法、装置和计算机可读存储介质
CN116299319B (zh) 多激光雷达的同步扫描及点云数据处理方法和雷达***
CN116381726A (zh) 基于数据自身的无人机激光点云精度自评估方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19954383

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19954383

Country of ref document: EP

Kind code of ref document: A1