WO2020113357A1 - 目标检测方法和装置、航迹管理方法和装置以及无人机 - Google Patents

目标检测方法和装置、航迹管理方法和装置以及无人机 Download PDF

Info

Publication number
WO2020113357A1
WO2020113357A1 PCT/CN2018/118861 CN2018118861W WO2020113357A1 WO 2020113357 A1 WO2020113357 A1 WO 2020113357A1 CN 2018118861 W CN2018118861 W CN 2018118861W WO 2020113357 A1 WO2020113357 A1 WO 2020113357A1
Authority
WO
WIPO (PCT)
Prior art keywords
current
target
coordinate information
coordinate
detected
Prior art date
Application number
PCT/CN2018/118861
Other languages
English (en)
French (fr)
Inventor
王俊喜
林灿龙
王春明
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/CN2018/118861 priority Critical patent/WO2020113357A1/zh
Priority to CN201880069284.6A priority patent/CN111279215A/zh
Publication of WO2020113357A1 publication Critical patent/WO2020113357A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • G01S13/60—Velocity or trajectory determination systems; Sense-of-movement determination systems wherein the transmitter and receiver are mounted on the moving object, e.g. for determining ground speed, drift angle, ground track
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/277—Analysis of motion involving stochastic approaches, e.g. using Kalman filters

Definitions

  • the invention relates to the technical field of detection, in particular to a target detection method, a target detection device, a track management method, a track management device and a drone.
  • Radar is mainly used to detect the target.
  • the detection result of the radar for the target is not accurate, and problems such as missed detection, target superposition, target expansion, and noise false alarm may occur.
  • the radar is not fixed, but it is set in the mounting platform.
  • the mounting platform is mobile.
  • the radar moves with the mounting platform. In this case, it will bring about the detection effect of the radar. The greater the impact, the more inaccurate the test results.
  • Embodiments of the present invention provide a target detection method, a target detection device, a track management method, a track management device, and a drone, to solve technical problems in related technologies.
  • a target detection method which is applicable to a radar and the radar is installed in a platform, and the method includes:
  • a method for trajectory management including the method described in the above embodiment, and further including:
  • a target detection device which is suitable for a radar, and the radar is provided in a platform, the device includes a processor, and the processor is used to
  • a track management device including a processor, the processor configured to determine the current correction coordinate information of the target determined by the target detection device according to the foregoing embodiment Describe the trajectory of the target;
  • an unmanned aerial vehicle including the target detection device and/or the track management device according to any one of the preceding claims.
  • the radar since the radar is installed in the mounting platform, and the mounting platform is movable, this causes the radar in the mounting platform to move relative to the detection target.
  • the current motion information of the mounted platform is considered, so the current predicted coordinate information of the target obtained by the prediction is more accurate, and then estimated based on the more accurate target's current predicted coordinate information and current detected coordinate information
  • the current corrected coordinate information of the obtained target is also more accurate. Therefore, it is helpful to improve the accuracy of detecting the target, so as to accurately determine the position and trajectory of the target.
  • FIG. 1 is a schematic flowchart of a target detection method according to an embodiment of the present disclosure.
  • FIG. 2 is a schematic flowchart illustrating acquiring current detected coordinate information of a target and motion information of the mounted platform according to an embodiment of the present disclosure.
  • FIG. 3 is a schematic diagram illustrating the positional relationship between the radar and the mounting platform according to an embodiment of the present disclosure.
  • FIG. 4 is another schematic flowchart of acquiring detected coordinate information of a target and motion information of the mounted platform according to an embodiment of the present disclosure.
  • FIG. 5 is a schematic flowchart illustrating determining the current predicted coordinate information of the target according to the coordinate information detected at the previous moment of the target and the current motion information of the mounted platform according to an embodiment of the present disclosure.
  • FIG. 6 is a schematic flowchart illustrating an estimation based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information of the target according to an embodiment of the present disclosure.
  • FIG. 7 is another schematic flowchart illustrating that the current predicted coordinate information and the current detected coordinate information are estimated according to an embodiment of the present disclosure to obtain current corrected coordinate information of the target.
  • FIG. 8 is a schematic flowchart illustrating determining the current detected coordinate information in at least one coordinate detected at a current time by a preset association algorithm according to an embodiment of the present disclosure.
  • FIG. 9 is another schematic flowchart of determining the current detected coordinate information in at least one coordinate detected at a current time by a preset association algorithm according to an embodiment of the present disclosure.
  • FIG. 10 is a schematic diagram illustrating determining current detection coordinate information according to an embodiment of the present disclosure.
  • FIG. 11 is a schematic flowchart illustrating a filter determination method according to the motion model of the platform and/or the type of the earth coordinate system according to an embodiment of the present disclosure.
  • FIG. 12 is a schematic flowchart of a track management method according to an embodiment of the present disclosure.
  • FIG. 13 is a schematic flowchart of another track management method according to an embodiment of the present disclosure.
  • FIG. 14 is a schematic flowchart of still another track management method according to an embodiment of the present disclosure.
  • FIG. 15 is a schematic flowchart of determining the reliability of the detected trajectories of multiple targets according to an embodiment of the present disclosure.
  • Embodiments of the present disclosure provide a target detection method, which is suitable for target detection devices, such as radar, image acquisition equipment, unmanned aerial vehicles, etc., which can detect the target to be detected.
  • target detection devices such as radar, image acquisition equipment, unmanned aerial vehicles, etc.
  • the target detection device is mobile, or the target detection device itself does not move, but can be mounted on a movable platform and can detect stationary targets. According to the target detection value and its own motion information, the motion information of the target relative to the target detection device is estimated.
  • the target detection device is mobile, or the target detection device itself does not move, but can be mounted on a movable platform and can detect the moving target. According to the target detection value and its own motion information, the motion information of the target relative to the target detection device is estimated.
  • the target detection device can detect at least one target, and estimate the motion information of the target relative to the target detection device according to the target detection value.
  • FIG. 1 is a schematic flowchart of a target detection method according to an embodiment of the present disclosure. The following mainly uses radar as a target detection device for example.
  • the target detection method described in this embodiment can be applied to a radar, which is installed in a carrying platform, which can be a road vehicle, such as a vehicle, or an air vehicle, such as an unmanned aerial vehicle, or Water vehicles, such as ships, are not limited in this disclosure.
  • the platform can be mobile, and the radar can move with the platform.
  • the radar can improve the measurement accuracy and enhance the adaptability to complex environments by implementing the target detection method of the embodiments of the present disclosure.
  • the target detection method may include the following steps:
  • Step S1 acquiring the detected coordinate information of the target and the motion information of the mounted platform
  • the detection information of the target can be obtained, where K is a non-negative integer.
  • K is a non-negative integer.
  • Previous moment involved in the following embodiments may refer to moment K
  • current may refer to the current moment K+1, which is between the previous moment K and the latter moment K+2 between.
  • the detected coordinate information of the target may include the distance and azimuth of the target.
  • the distance of the target may include the horizontal distance of the target to the radar
  • the azimuth of the target may include the angle of the target in the radar coordinate system, such as the yaw angle of the mounted platform, and the detection information of the target may be obtained by radar detection.
  • the detection information of the target is obtained by radar, and the radar can collect the echo signal of the target, and the signal detection signal of the echo signal can be used to obtain the detection coordinate information of the target.
  • the radar can be rotated, and the radar can collect the detected coordinate information of the target by rotating a preset angle range.
  • the preset angle range may be set as needed, for example, the front of the radar is 0°, and the preset angle range may be -90° to +90°.
  • the motion information of the platform can include yaw (heading angle) direction information, speed information, etc.
  • the motion information of the platform can be obtained by sensors mounted on the carrier platform, for example, the motion information of the platform is passed through a GPS sensor (not shown) Get out).
  • Step S2 Determine the current predicted coordinate information of the target based on the detected coordinate information of the target at the previous moment and the current motion information of the mounted platform;
  • the current predicted coordinate information of the target may be determined based on the detected coordinate information of the target at the previous moment and the current motion information of the mounted platform based on the motion model of the mounted platform.
  • the motion model may be a uniform velocity model or a uniform acceleration model.
  • a uniform velocity model is used as an example for description.
  • the current motion information of the platform can include speed information, and the target can be stationary or moving. This embodiment will be described by taking an example where the target is stationary.
  • Step S3 Estimate according to the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information of the target.
  • the radar since the radar is installed in the mounting platform, and the mounting platform is movable, this causes the radar in the mounting platform to move relative to the detection target.
  • the current motion information of the mounted platform is taken into account, so the current predicted coordinate information of the target obtained by the prediction is more accurate, and then estimated based on the current predicted coordinate information and current detected coordinate information of the accurate target
  • the current corrected coordinate information of the obtained target is also more accurate. Therefore, it is helpful to improve the accuracy of detecting the target, so as to accurately determine the position and trajectory of the target.
  • FIG. 2 is a schematic flowchart of acquiring detected coordinate information of a target and motion information of the mounted platform according to an embodiment of the present disclosure.
  • the detected coordinate information of the acquired target and the motion information of the mounted platform include:
  • Step S11 Detect the target by radar to determine the first coordinate of the target in the coordinate system of the radar;
  • Step S12 Determine the current detected coordinate information corresponding to the first target in the geodetic coordinate system of the mounted platform according to the positional relationship between the radar and the mounted platform.
  • FIG. 3 is a schematic diagram showing the positional relationship between the radar and the mounted platform according to the embodiment of the present disclosure.
  • the coordinate system of the radar may be a polar coordinate system
  • the distance between the target and the radar is R1
  • the deflection angle with respect to the Y axis is ⁇
  • the center of the Y axis is connected to the radar The lines coincide
  • the first coordinate of the target in the radar's coordinate system is:
  • the first coordinate of the target in the radar coordinate system needs to be converted into the geodetic coordinate system of the mounted platform.
  • the positive direction of the horizontal axis is east (E), and the vertical axis
  • the positive direction is north (N).
  • the first coordinates of the target in the radar's coordinate system can be converted into the geodetic coordinate system of the mounted platform to obtain the current detection coordinate information corresponding to the target in the geodetic coordinate system of the mounted platform for subsequent consideration
  • the target is detected with platform motion information.
  • FIG. 4 is another schematic flowchart of acquiring detected coordinate information of a target and motion information of the mounted platform according to an embodiment of the present disclosure.
  • the detected coordinate information of the acquired target and the motion information of the mounted platform include:
  • Step S13 Compensate the position deviation according to the positional relationship between the radar and the mounted platform to determine the detection coordinate information.
  • the radar is not set in the center of the platform, then R2 ⁇ 0.
  • the position deviation needs to be compensated according to the positional relationship between the radar and the platform, that is, the first target is determined
  • the corresponding current detection coordinate information in the geodetic coordinate system on which the platform is mounted must first be compensated based on R2. For example, when the coordinate system is converted, the installation error is compensated to ensure the accuracy of subsequent calculations.
  • FIG. 5 shows a method for detecting the coordinate information and the current motion information of the mounted platform based on the previous time of the target to determine the target’s
  • the determining the current predicted coordinate information of the target based on the detected coordinate information of the target at the previous moment and the current motion information of the mounted platform includes:
  • step S21 the motion model of the mounted platform is determined, and the current predicted coordinate information of the target is determined according to the detected motion information of the motion model and the target at the previous moment and the current motion information of the mounted platform.
  • the mounting platform is movable, and the radar moves along with the mounting platform on the mounting platform, the displacement of the mounting platform varies from the previous moment to the current based on the different motion modes, therefore, In order to predict the current predicted coordinate information of the target, the prediction can be made based on the motion model of the mounted platform and the detected coordinate information of the target at the previous moment.
  • the motion model of the platform can be a uniform acceleration model, a uniform velocity model, etc.
  • the state variable of the target K at the previous moment can be taken:
  • x is the abscissa of the target at the previous time K
  • y is the ordinate of the target at the previous time K
  • t is the time interval between the current time K+1 and the previous time K.
  • the obtained target has the abscissa of the predicted coordinates at the current moment as The ordinate is
  • FIG. 6 shows an estimation based on the current predicted coordinate information and the current detected coordinate information according to an embodiment of the present disclosure to obtain the current corrected coordinates of the target Schematic flow chart of information.
  • the estimating based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information of the target includes:
  • Step S31 a filter is determined according to the motion model of the mounted platform and/or the type of the earth coordinate system where the mounted platform is located;
  • Step S32 through the filter, estimate based on the current predicted coordinate information and the current detected coordinate information to obtain the current modified coordinate information, where the current modified coordinate information is used to determine the predicted coordinate at the next moment information.
  • the process of estimating based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information of the target may be implemented by a filter, and different filters have different Filter model.
  • the filter can be determined according to the motion model of the mounted platform and/or the type of the earth coordinate system where the mounted platform is located, so that the current The predicted coordinate information and the current detected coordinate information are estimated.
  • the filter may be a linear filter; for example, if the type of the earth coordinate system is a rectangular coordinate system, then the filter may also be a linear filter.
  • the linear filter may be an ⁇ - ⁇ filter, a Kalman filter, or the like.
  • the filter may be a nonlinear filter; for example, if the type of the earth coordinate system is a polar coordinate system, then the filter may also be a nonlinear filter.
  • the nonlinear filter may be an extended Kalman filter (EKF) and a lossless Kalman filter (UKF).
  • FIG. 7 is another example according to an embodiment of the present disclosure, which estimates based on the current predicted coordinate information and the current detected coordinate information to obtain the current correction of the target Schematic flow chart of coordinate information.
  • the estimating based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information of the target further includes:
  • Step S33 Determine the current detected coordinate information in the at least one second coordinate detected at the current moment through a preset association algorithm
  • the passing through the filter and estimating based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information includes:
  • Step S321 Calculate the current modified coordinate information according to the first weight and the current predicted coordinate information, and the second weight and the current detected coordinate information.
  • At least one second coordinate can be obtained, for example, the target can be approximated as a point, then the detected second coordinate is one, in this case, the second coordinate Determine the current detection coordinate information, that is, use the second coordinate as the current detection coordinate.
  • the detected second coordinates are multiple.
  • the target is regarded as a point, that is, multiple second based on the target If one of the coordinates is to be calculated, then it is necessary to determine the most reasonable second coordinate among the multiple second coordinates.
  • a second coordinate can be determined among the multiple coordinates through a preset association algorithm. That is, the determined second coordinate is used as the current detection coordinate.
  • the current predicted coordinate information can be weighted by the first weight value, and the detected coordinate information can be weighted by the second weight value.
  • the first weight value is ⁇
  • the second weight value is ⁇
  • the current time is K+1
  • the current correction coordinate information is calculated as follows:
  • Z(K+1) is the current detected coordinate information
  • K) is the current predicted coordinate information
  • FIG. 8 is a schematic flowchart illustrating determining the current detected coordinate information in at least one coordinate detected at a current time by a preset association algorithm according to an embodiment of the present disclosure.
  • the determining the current detected coordinate information in the at least one coordinate detected at the current moment by a preset association algorithm includes:
  • Step S331 Calculate the distance from the at least one second coordinate to the predicted coordinate corresponding to the current predicted coordinate information
  • Step S332 Determine the current detected coordinate information according to the coordinate with the smallest distance from the predicted coordinate in the at least one second coordinate.
  • the coordinate with the smallest distance to the predicted coordinate is the second coordinate, so the second coordinate can be used as the current detection coordinate information.
  • the distance to the predicted coordinates can be calculated for each second coordinate.
  • FIG. 9 is another schematic flowchart of determining the current detected coordinate information in at least one coordinate detected at a current time by a preset association algorithm according to an embodiment of the present disclosure .
  • the determining the current detected coordinate information in the at least one coordinate detected at the current time by a preset association algorithm includes:
  • Step S333 Determine at least one associated coordinate located in the preset area (including the edge located in the preset area) in at least one second coordinate;
  • Step S334 calculating the distance from the at least one associated coordinate to the predicted coordinate corresponding to the current predicted coordinate information
  • Step S335 Determine the current detected coordinate information according to the coordinate with the smallest distance to the predicted coordinate among the at least one associated coordinate.
  • the associated coordinates located in the preset area are determined in the detected second coordinates, wherein, the preset area may be an area containing predicted coordinates, for example, an area centered on the predicted coordinates, or an area located near the predicted coordinates, but not including the predicted coordinates.
  • coordinates that are not located in the preset area can be removed from the at least one second coordinate, so as to reduce the number of related coordinates that need to be calculated to the distance to the predicted coordinate in the future, thereby reducing the overall calculation amount.
  • FIG. 10 is a schematic diagram illustrating determining current detection coordinate information according to an embodiment of the present disclosure.
  • the predicted coordinates are A(x 0 , y 0 ), and the three detected second coordinates are B(x 1 , y 1 ), C( x 2 , y 2 ), D(x 3 , y 3 ), the preset area is a circular area with the predicted coordinate A as the center and the radius as DIS.
  • step S333 it can be determined that point D is outside the preset area, so that point D is eliminated, and points B and C are retained as associated coordinates, and then the distance to point A is calculated for points B and C, respectively:
  • the preset area is a circular area with the predicted coordinates as the center of the circle and the first preset distance DIS as the radius;
  • the first weight is equal to the ratio of the distance from the detection coordinate corresponding to the current detection coordinate information to the predicted coordinate and the first preset distance
  • the second weight is equal to 1 and the first Weight difference
  • ⁇ represents the weight of the item X(K+1
  • the higher the reliability of the current predicted coordinate information, the The higher the weight should be, by setting ⁇ e 01 /DIS, it can be ensured that the first weight ⁇ is positively correlated with e 01 , that is, the closer the current detected coordinate information is to the current predicted coordinate information, indicating that the predicted coordinates are approximately closer to the detected coordinates , That is, the more accurate the prediction, that is, the higher the reliability of the current predicted coordinate information, then the larger ⁇ .
  • the first preset distance DIS can be set according to parameters such as the accuracy of the radar, the moving speed of the mounted platform, and the distance between the radar and the target. For example, the higher the accuracy of the radar, the smaller the DIS; the greater the speed of the platform, the greater the DIS; the farther the distance between the radar and the target, the greater the DIS.
  • FIG. 11 is a schematic flowchart of determining a filter according to the motion model of the mounted platform and/or the type of the earth coordinate system according to an embodiment of the present disclosure.
  • the determining filter according to the motion model of the mounted platform and/or the type of the earth coordinate system includes:
  • Step S311 when the motion model of the mounted platform is a linear model, and/or the type of the earth coordinate system is a rectangular coordinate system, a linear filter is determined;
  • Step S312 when the motion model of the mounted platform is a nonlinear model, and/or the type of the earth coordinate system is a polar coordinate system, a nonlinear filter is determined.
  • the linear filter facilitates the operation of the data in the linear model and the rectangular coordinate system
  • the nonlinear filter facilitates the operation of the data in the nonlinear model and the polar coordinate system
  • the linear filter includes at least one of the following:
  • the non-linear filter includes at least one of the following:
  • the motion information of the mounted platform includes at least one of the following: position and speed.
  • the state frame at time K can indicate the position of the platform, with Respectively represent the speed of the mounted platform along the horizontal and vertical axes in the geodetic coordinate system.
  • the motion information of the platform mounted when the uniform speed model is used as an example.
  • the motion model is a uniform acceleration model
  • the motion information of the platform mounted may also include acceleration.
  • the process of determining the current corrected coordinate information in the above embodiments may be performed for one target or multiple targets.
  • the target detection method described in the above embodiment can be applied to radar or other devices, such as the image acquisition device mentioned above.
  • it can be constructed based on the image acquisition device A coordinate system, in which the image acquisition device can determine the position of the target relative to itself based on its own posture information when acquiring the image and the depth information of the target in the image, thereby serving as the detection coordinate information of the target.
  • Track management mainly refers to track start, track maintenance, and track end.
  • This embodiment discloses a trajectory management method, which can be used for the above-mentioned trajectory management. Specifically, it can manage the free point of the target, the reliable trajectory, and the destruction of the trajectory.
  • the track management method described in this embodiment includes the target detection method described in any of the above embodiments as shown in FIG. 12, and the track management method further includes:
  • Step S1' determining the trajectory of the target according to the current modified coordinate information of the target
  • the radar detects the target at a certain moment, which may be caused by interference, so that the detected traces include not only the target traces, but also clutter traces.
  • the radar can determine the trajectory of the target based on the current correction coordinate information of the target determined in the foregoing embodiment, and further based on the current correction coordinate information, for example, the current correction determined from N times from 0 to N
  • the trajectory obtained by connecting the coordinate information in chronological order is used as the trajectory of the target, where the target may be one or more.
  • Step S2' determining the credibility of the detected trajectories of multiple targets
  • the trajectory of multiple targets can be managed by introducing credibility for the trajectories of multiple targets.
  • the trajectory of each target can be determined for its reliability, because the target's trajectory is determined according to the current modified coordinate information, and the current modified coordinate information is determined according to the current predicted coordinate information and the current detected coordinate information According to the estimation, the current predicted coordinate information used is predicted, so it is not necessarily accurate, so the current corrected coordinate information estimated based on this is not necessarily accurate, and then based on the current corrected coordinate information The determined trajectory is not necessarily accurate, so the accuracy of the trajectory can be expressed by the credibility, where the manner of determining the credibility is exemplified in the subsequent embodiments.
  • Step S3' delete the track whose credibility is lower than the preset credibility.
  • the radar can manage the trajectories of multiple targets, the accuracy of the trajectories with lower credibility (eg, lower than the preset credibility) is poor, that is, the real If the trajectories are far apart, then there is no need to continue to monitor them. Therefore, trajectories with a credibility lower than the preset credibility can be deleted, so that only the trajectories with higher credibility are retained, which is helpful to reduce the radar's The load makes the radar detect only the targets corresponding to the trajectories with higher reliability.
  • lower credibility eg, lower than the preset credibility
  • the execution frequency of the steps in the track management method shown in this embodiment that is, the frequency of updating the trajectory of the target
  • the execution frequency of the steps in the target detection method of the foregoing embodiment that is, determining the current target
  • the frequency of the correction coordinate information may be the same or different.
  • the trajectory of the target is updated according to the newly determined current correction coordinate information.
  • the trajectory of the target may be updated based on the current correction coordinate information determined last time. If the execution frequency of the steps in the track management method is greater than the execution frequency of the steps in the target detection method, that is, after determining the current correction coordinate information once, the trajectory of the target will be updated multiple times. Modified coordinate information of the nearest neighbor time determined before the current trajectory update.
  • FIG. 13 is a schematic flowchart of another track management method according to an embodiment of the present disclosure. As shown in FIG. 13, the track management method further includes:
  • Step S4' calculating the trajectory distance of each trajectory to the origin of the geodetic coordinate system at the current moment
  • Step S5' sort each track according to the track distance.
  • the radar is installed on the platform, it is moving relative to the target, so the target is also moving relative to the radar, and for the trajectory of a target, the distance from the target to the radar will occur at different times.
  • Change that is, the point corresponding to the target on the trajectory, the trajectory distance to the origin of the geodetic coordinate system will change, so that when multiple targets are detected, in the trajectory corresponding to the multiple targets, the radar is closest to the radar at different times.
  • the trajectory is different, and generally speaking, the closer the trajectory to the radar trajectory distance, the more likely the trajectory to collide with the mounted platform, that is, the higher the degree of threat, so the trajectory can be sorted according to the trajectory distance.
  • the order can be identified by number. For example, the smaller the track distance, the higher the order, and the smaller the number, so that the radar can display the track with higher threat level forward, which is convenient for the user to respond in time.
  • the deleted track its number can be re-assigned to other tracks, so as to be sorted.
  • FIG. 14 is a schematic flowchart of still another track management method according to an embodiment of the present disclosure. As shown in FIG. 14, the track management method further includes:
  • step S6' the trajectory is output according to the sorting result.
  • the user can determine the order of each trajectory, where the output method can be output through the screen display, or the logo of the trajectory can be played in order by audio.
  • the credibility is inversely related to the number of times the target is determined to be the current predicted coordinate information, and the number of times the distance from the detected coordinate corresponding to the current predicted coordinate information of the calculated target to the predicted coordinate corresponding to the current predicted coordinate information is positive
  • Correlation is inversely related to the distance between the detected coordinates of the target and the predicted coordinates.
  • the trajectory is determined based on the current modified coordinate information at multiple times, and the current modified coordinate information at each time is based on the current predicted coordinate information and the current detection
  • the coordinate information is estimated, and the current detected coordinate information is based on the detected coordinates corresponding to the current predicted coordinate information of the target (such as the second coordinate in the above embodiment) to the predicted coordinates corresponding to the current predicted coordinate information
  • the distance is determined, and the weight value needs to be set based on the distance from the detected coordinate to the predicted coordinate (for example, e 01 above).
  • the specific determination method may be performed according to the embodiment shown in FIG. 8 or FIG. 9.
  • the process of determining the current correction coordinate information is essentially a prediction process, due to the amount of prediction used in the prediction process (for example, in the embodiment shown in FIG. 5 with ) May change, so the probability of a larger deviation is higher, which will result in less accurate determination of the current corrected coordinate information.
  • the process of determining the current detection coordinate information is essentially a detection process, which includes the above process of calculating the distance from the detection coordinate to the predicted coordinate, and the detection coordinate is actually detected, so the probability of a large deviation is small and will This makes it more accurate to determine the current corrected coordinate information.
  • the distance between the detection coordinate and the prediction coordinate represents the difference between the detection result and the prediction result.
  • the greater the distance between the detection coordinate and the prediction coordinate the greater the difference between the detection result and the prediction result, indicating that the prediction process is less accurate and the current correction coordinates are determined The less accurate the information.
  • the more times the current predicted coordinate information is determined the less accurate the current corrected coordinate information is. That is, the lower the accuracy of the determined trajectory, the more times the distance between the detection coordinates corresponding to the current predicted coordinate information of the target and the predicted coordinates corresponding to the current predicted coordinate information is calculated, the more accurate the current corrected coordinate information is determined, that is, The higher the accuracy of the determined trajectory, the more inaccurate the predicted trajectory. The greater the distance between the detected coordinates of the target and the predicted coordinates, the more inaccurate the determined current modified coordinate information, that is, the lower the accuracy of the determined trajectory .
  • the credibility is set to be inversely related to the number of times the target is determined to be the current predicted coordinate information, and the number of times the distance from the detected coordinate corresponding to the current predicted coordinate information of the calculated target to the predicted coordinate corresponding to the current predicted coordinate information is positive
  • Correlation which is inversely related to the distance between the detected coordinates of the target and the predicted coordinates, can ensure the accuracy of the credibility calculation in order to accurately determine whether the trajectory of the target is credible.
  • FIG. 15 is a schematic flowchart of determining the reliability of the detected trajectories of multiple targets according to an embodiment of the present disclosure.
  • the reliability of determining the trajectories of the detected multiple targets includes:
  • Step S21' determining whether the newly detected target belongs to the recorded track according to the current modified coordinate information of the newly detected target
  • Step S22' if it does not belong to the recorded track, initialize the track and credibility of the newly detected target;
  • Step S23' each time the current predicted coordinate information of the newly detected target is predicted, the first preset confidence level is subtracted from the initialized confidence level;
  • Step S24' each time the distance between the detected coordinates of the newly detected target and the predicted coordinates is calculated, a second preset confidence level is added from the initialized confidence level, wherein the second preset confidence level It is inversely related to the distance from the detected coordinates of the newly detected target to the predicted coordinates.
  • the credibility is inversely related to the number of times the target is determined to be the current predicted coordinate information
  • the number of times that the distance between the detected coordinate corresponding to the current predicted coordinate information of the target and the predicted coordinate corresponding to the current predicted coordinate information is positively correlated , which is inversely related to the distance from the detected coordinates of the target to the predicted coordinates, so each time the current predicted coordinate information of the newly detected target is predicted, the first preset confidence level can be subtracted from the initial confidence level, and each new calculation The distance between the detection coordinates of the detected target and the predicted coordinates, a second preset confidence level is added from the initial confidence level, and the added second preset confidence level and the detection of the newly detected target The distance from the coordinate to the predicted coordinate is inversely related.
  • the initial reliability is 30 points
  • 0.5 points are subtracted from the initialized reliability
  • the detection coordinates of the newly detected target are calculated each time The distance to the predicted coordinates, add 10--(e 01 -0.3)*10/(3-0.3) points from the initial confidence level, and finally you can delete the confidence level below the preset according to the calculation result of the confidence level Tracks with credibility (for example, 20 points), and sort and output the remaining tracks.
  • the execution frequency of the steps in the track management method is 15 Hz and the execution frequency of the steps in the target detection method is 100 Hz, then each time the trajectory is updated, the current predicted coordinate information of the newly detected target 6 times needs to be predicted, thereby reducing Go 3 points.
  • a flag can be set for the current revised coordinate information, and whenever the trajectory needs to be updated, it can be determined whether the current revised coordinate information has a flag, and if there is a flag, based on the current revised coordinate information Predict the current predicted coordinate information of the newly detected target, and calculate the distance from the detected coordinate of the newly detected target to the predicted coordinate, and then delete the mark. If there is no mark, it means that the newly detected target has been calculated for the current modified coordinate information. The distance from the detected coordinate to the predicted coordinate, then there is no need to predict the current predicted coordinate information of the newly detected target based on the current modified coordinate information. According to this, it can be guaranteed whether the distance between the detected coordinates of the newly detected target and the predicted coordinates has been calculated for the current corrected coordinate information obtained each time, and then the points are accurately added.
  • the distance between the detected coordinates of the newly detected target and the predicted coordinates of the second preset confidence level is inversely correlated within the first preset range, and is less than the lower limit value of the first preset range
  • the ground range is equal to the first preset value, and the range is greater than the upper limit of the first preset range to the second preset value, where the first preset value is greater than the second preset value.
  • the distance between the second preset confidence level and the detected coordinates of the newly detected target to the predicted coordinates may be inversely related only within the first preset range, for example, at e 01 ⁇ (0.3 m, 3m) Within this range, the second preset confidence level is equal to 10-(e 01 -0.3)*10/(3-0.3), and within the range of e 01 ⁇ 0.3m, the second preset confidence level can be equal to 10. In the range of e 01 ⁇ 0.3 m, the second preset confidence level may be equal to 0.
  • the first preset range can be set as needed.
  • the present disclosure also proposes embodiments of the target detection device and track management device.
  • An embodiment of the present disclosure proposes a target detection device, which is suitable for a radar.
  • the radar is installed in a platform, and the device includes a processor.
  • the processor is used to:
  • the processor is configured to detect the target by radar to determine the first coordinate of the target in the coordinate system of the radar;
  • the detection coordinate information corresponding to the first target in the geodetic coordinate system of the mounted platform is determined according to the positional relationship between the radar and the mounted platform.
  • the processor is configured to compensate a position deviation according to the positional relationship between the radar and the mounted platform to determine the detected coordinate information.
  • the processor is used to determine a motion model of the mounted platform, and determine based on the detected motion information of the motion model and the target at the previous moment and the current motion information of the mounted platform The current predicted coordinate information of the target.
  • the processor is used to determine a filter according to the motion model of the mounted platform and/or the type of the earth coordinate system where the mounted platform is located;
  • an estimation is performed based on the current predicted coordinate information and the current detected coordinate information to obtain the current modified coordinate information, where the current modified coordinate information is used to determine the predicted coordinate information at the next moment.
  • the processor is configured to determine the current detected coordinate information in at least one second coordinate detected at a current moment through a preset association algorithm
  • the passing through the filter and estimating based on the current predicted coordinate information and the current detected coordinate information to obtain the current corrected coordinate information includes:
  • the current corrected coordinate information is calculated according to the first weight value and the current predicted coordinate information, and the second weight value and the current detected coordinate information.
  • the processor is configured to calculate the distance from the at least one second coordinate to the predicted coordinate corresponding to the current predicted coordinate information
  • the current detected coordinate information is determined according to the coordinate with the smallest distance from the predicted coordinate in the at least one second coordinate.
  • the processor is configured to determine at least one associated coordinate located in the preset area in at least one second coordinate;
  • the current detected coordinate information is determined according to the coordinate with the smallest distance to the predicted coordinate among the at least one associated coordinate.
  • the preset area is a circular area with the predicted coordinates as the center of the circle and the first preset distance as the radius;
  • the first weight is equal to the ratio of the distance from the detection coordinate corresponding to the current detection coordinate information to the predicted coordinate and the first preset distance
  • the second weight is equal to 1 and the first Weight difference
  • the processor is used to determine a linear filter when the motion model of the mounted platform is a linear model and/or the type of the earth coordinate system is a rectangular coordinate system;
  • a nonlinear filter is determined.
  • the linear filter includes at least one of the following:
  • the non-linear filter includes at least one of the following:
  • the motion information of the mounted platform includes at least one of the following: position and speed.
  • the target is one or more.
  • An embodiment of the present disclosure proposes a track management device including a processor.
  • the processor determines, based on the current correction coordinate information of the target determined by the target detection device according to any of the foregoing embodiments, Trajectory
  • the processor is further configured to calculate a trajectory distance from each trajectory to the origin of the geodetic coordinate system at the current moment;
  • the processor is further configured to output the trajectory according to the sorted result.
  • the credibility is inversely related to the number of times the target is determined to be the current predicted coordinate information, and the distance between the detected coordinate corresponding to the current predicted coordinate information of the calculated target and the predicted coordinate corresponding to the current predicted coordinate information
  • the frequency is positively correlated, which is inversely related to the distance from the detected coordinate of the target to the predicted coordinate.
  • the processor is configured to determine whether the newly detected target belongs to a recorded trajectory according to the current modified coordinate information of the newly detected target;
  • the first preset confidence level is subtracted from the initial confidence level
  • a second preset confidence level is added to the initialized confidence level, wherein the second preset confidence level and the new The distance between the detected coordinates of the detected target and the predicted coordinates is inversely correlated.
  • the distance between the detected coordinates of the newly detected target and the predicted coordinates of the second preset confidence level is inversely correlated within the first preset range, below the first preset range
  • the range within the limit is equal to the first preset value
  • the range within the range greater than the upper limit of the first preset range is equal to the second preset value, wherein the first preset value is greater than the second preset value .
  • An embodiment of the present disclosure proposes an unmanned aerial vehicle, including the target detection device and/or the track management device described in any of the above embodiments.
  • the system, device, module or unit explained in the above embodiments may be implemented by a computer chip or entity, or by a product with a certain function.
  • the functions are divided into various units and described separately.
  • the functions of each unit may be implemented in one or more software and/or hardware.
  • the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware.
  • the present invention may take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

一种目标检测方法，包括：获取目标的检测坐标信息以及所述搭载平台的运动信息（S1）；根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息（S2）；根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息（S3）。所述方法有利于提高对于目标进行检测的准确率，以便准确地确定目标的位置和轨迹等信息。

Description

目标检测方法和装置、航迹管理方法和装置以及无人机 技术领域
本发明涉及检测技术领域，尤其涉及目标检测方法、目标检测装置、航迹管理方法、航迹管理装置和无人机。
背景技术
雷达主要用于对目标进行检测，然而由于噪声的存在，导致雷达对于目标的检测结果并不准确，可能出现目标漏检、目标叠加、目标扩展、噪点虚警等问题。
而且目前很多场景下，雷达并不是固定不动的，而是设置在搭载平台中的，搭载平台是移动的，雷达随着搭载平台移动，在这种情况下，会对雷达的检测效果带来更大的影响，导致检测结果更加不准确。
发明内容
本发明实施例提供一种目标检测方法、目标检测装置、航迹管理方法、航迹管理装置和无人机，以解决相关技术中的技术问题。
根据本公开实施例的第一方面，提出一种目标检测方法，适用于雷达，所述雷达设置在搭载平台中，所述方法包括：
获取目标的检测坐标信息以及所述搭载平台的运动信息；
根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
根据本公开实施例的第二方面，提出一种航迹管理方法，包括上述实施例所述的方法，还包括：
根据所述目标的当前修正坐标信息确定所述目标的轨迹；
确定所检测的多个目标的轨迹的可信度；
删除可信度低于预设可信度的轨迹。
根据本公开实施例的第三方面，提出一种目标检测装置，适用于雷达，所述雷达设置在搭载平台中，所述装置包括处理器，所述处理器用于，
获取目标的检测坐标信息以及所述搭载平台的运动信息；
根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
根据本公开实施例的第四方面，提出一种航迹管理装置，包括处理器，所述处理器用于，根据上述实施例所述的目标检测装置确定的所述目标的当前修正坐标信息确定所述目标的轨迹；
确定所检测的多个目标的轨迹的可信度；
删除可信度低于预设可信度的轨迹。
根据本公开实施例的第五方面，提出一种无人飞行器，包括上述任一项权利要求所述的目标检测装置和/或航迹管理装置。
根据本公开的实施例，由于雷达设置在搭载平台中，而搭载平台是可以移动的，这会导致搭载平台中的雷达相对于检测目标移动。在这种情况下，可以获取目标的前一时刻的检测坐标信息、当前检测坐标信息和搭载平台的运动信息，然后根据目标的前一时刻的检测坐标信息以及搭载平台的当前运动信息，确定目标的当前预测坐标信息，进而根据当前预测坐标信息与当前检测坐标信息进行估算，以获得目标的当前修正坐标信息。
由于确定目标的当前预测坐标信息时，考虑到了搭载平台的当前运动信息，因此进行预测得到的目标的当前预测坐标信息更准确，进而根据更准确的目标的当前预测坐标信和当前检测坐标信息进行估算得到的目标的当前修正坐标信息也更为准确。从而有利于提高对于目标进行检测的准确率，以便 准确地确定目标的位置和轨迹等信息。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是根据本公开的实施例示出的一种目标检测方法的示意流程图。
图2是根据本公开的实施例示出的一种获取目标的当前检测坐标信息以及所述搭载平台的运动信息的示意流程图。
图3是根据本公开的实施例示出的雷达和搭载平台位置关系的示意图。
图4是根据本公开的实施例示出的另一种获取目标的检测坐标信息以及所述搭载平台的运动信息的示意流程图。
图5是根据本公开的实施例示出的一种根据所述目标的前一时刻检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息的示意流程图。
图6是根据本公开的实施例示出的一种根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息的示意流程图。
图7是根据本公开的实施例示出的另一种根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息的示意流程图。
图8是根据本公开的实施例示出的一种通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息的示意流程图。
图9是根据本公开的实施例示出的另一种通过预设关联算法在当前时刻 检测的至少一个坐标中确定所述当前检测坐标信息的示意流程图。
图10是根据本公开的实施例示出的一种确定当前检测坐标信息的示意图。
图11是根据本公开的实施例示出的一种根据所述搭载平台的运动模型和/或所述大地坐标系的类型确定滤波器的示意流程图。
图12是根据本公开的实施例示出的一种航迹管理方法的示意流程图。
图13是根据本公开的实施例示出的另一种航迹管理方法的示意流程图。
图14是根据本公开的实施例示出的又一种航迹管理方法的示意流程图。
图15是根据本公开的实施例示出的一种确定所检测的多个目标的轨迹的可信度的示意流程图。
具体实施方式
本公开实施例提供一种目标检测方法，适用于目标检测装置，例如，雷达、图像采集设备、无人飞行器等，可以对待检测目标进行检测。
在一个实施例中，该目标检测装置是移动的，或者该目标检测装置本身不移动，而能搭载在可移动平台上，能够对静止的目标的进行检测。并根据目标检测值、以及自身的运动信息，对目标相对于目标检测装置的运动信息进行估算。
在一个实施例中，该目标检测装置是移动的，或者该目标检测装置本身不移动，而能搭载在可移动平台上，能够对移动的目标进行检测。并根据目标检测值、以及自身的运动信息，对目标的相对于目标检测装置的运动信息进行估算。
在一个实施例中，该目标检测装置能够对至少一个目标进行检测，并根据目标检测值对目标相对于目标检测装置的运动信息进行估算。
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行 清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。另外，在不冲突的情况下，下述的实施例及实施例中的特征可以相互组合。图1是根据本公开的实施例示出的一种目标检测方法的示意流程图。以下主要以雷达作为目标检测装置进行示例性说明。本实施例所述目标检测方法可以适用于雷达，所述雷达设置在搭载平台中，所述搭载平台可以是路面交通工具，例如车辆，也可以是空中交通工具，例如无人飞行器，还可以是水上交通工具，例如船舶，对此，本公开不作限制。搭载平台可以是移动的，该雷达可以随着搭载平台移动。该雷达可以通过实施本公开实施例的目标检测方法，提升测量精度，增强对复杂环境的适应能力。
如图1所示，所述目标检测方法可以包括以下步骤：
步骤S1，获取目标的检测坐标信息以及所述搭载平台的运动信息；
需要说明的是，在每一时刻，例如，K、K+1、K+2时刻，都可以获取目标的检测信息，其中，K为非负整数。下述实施例中涉及的“前一时刻”一词，可以是指K时刻，“当前”一词，可以是指当前时刻K+1，其处于前一时刻K和后一时刻K+2之间。
其中，目标的检测坐标信息可以包括目标的距离、方位角。具体的，目标的距离可以包括目标到雷达的水平距离，目标的方位角可以包括目标在雷达的坐标系下的角度，例如搭载平台的偏航角度，目标的检测信息可以由雷达检测获得。
其中，通过雷达获取目标的检测信息，所述雷达可以采集目标的回波信号，通过对回波信号进行信号处理，可以获得目标的检测坐标信息。
在一个实施例中，雷达是可以旋转的，雷达可以通过旋转预设角度范围来采集目标的检测坐标信息。所述预设角度范围可以根据需要进行设置，例如，以雷达的正前方为0°，预设角度范围可以是-90°到+90°。
搭载平台的运动信息可以包括yaw(航向角)方向信息、速度信息等， 搭载平台的运动信息可以由搭载在载体平台上的传感器获得，例如，搭载平台的运动信息通过GPS传感器(图中未示出)获得。
步骤S2，根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
具体的，可以基于搭载平台的运动模型，根据目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息。其中，运动模型可以是匀速模型、匀加速模型。本实施例以匀速模型为例进行说明。搭载平台的当前运动信息可以包括速度信息，目标可以是静止的或移动的。本实施例以目标是静止的为例进行说明。
步骤S3，根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
在一个实施例中，由于雷达设置在搭载平台中，而搭载平台是可以移动的，这会导致搭载平台中的雷达相对于检测目标移动。在这种情况下，可以获取目标的前一时刻的检测坐标信息、当前检测坐标信息和搭载平台的运动信息，然后根据目标的前一时刻的检测坐标信息以及搭载平台的当前运动信息，确定目标的当前预测坐标信息，进而根据当前预测坐标信息与当前检测坐标信息进行估算，以获得目标的当前修正坐标信息。
由于确定目标的当前预测坐标信息时，考虑到了搭载平台的当前运动信息，因此进行预测得到的目标的当前预测坐标信息更准确，进而根据跟准确的目标的当前预测坐标信和当前检测坐标信息进行估算得到的目标的当前修正坐标信息也更为准确。从而有利于提高对于目标进行检测的准确率，以便准确地确定目标的位置和轨迹等信息。
在图1所示实施例的基础上，图2是根据本公开的实施例示出的一种获取目标的检测坐标信息以及所述搭载平台的运动信息的示意流程图。如图2所示，所述获取目标的检测坐标信息以及所述搭载平台的运动信息，包括：
步骤S11，通过雷达检测目标，以确定所述目标在所述雷达的坐标系中的所述第一坐标；
步骤S12，根据所述雷达与所述搭载平台的位置关系，以确定所述第一目标在所述搭载平台的大地坐标系中对应的所述当前检测坐标信息。
在图1或图2所示实施例的基础上，图3是根据本公开的实施例示出的雷达和搭载平台位置关系的示意图。
在一个实施例中，如图3所示，雷达的坐标系可以是极坐标系，目标与雷达的距离为R1，相对于Y轴偏转角度为θ，Y轴与搭载平台的中心到雷达的连线重合，那么目标在雷达的坐标系中的第一坐标为：
x＝R1*sinθ，y＝R1*cosθ。
而为了考虑搭载平台的运动状态，需要将目标在雷达的坐标系中的第一坐标转换到搭载平台的大地坐标系中，在大地坐标系中，横轴正方向为东(E)，纵轴正方向为北(N)。其中，大地坐标系的圆心与搭载平台的中心重合，雷达与搭载平台的中心之间距离为R2，并且相对于N方向偏转角度为φ，而在一般情况下，可以将雷达设置在搭载平台的中心，那么R2＝0，则第一目标在搭载平台的大地坐标系中对应的当前检测坐标信息为：
x＝R1*sin(φ+θ)，y＝R1*cos(φ+θ)。
据此，可以将目标在雷达的坐标系中的所述第一坐标，转换到搭载平台的大地坐标系中，得到目标在搭载平台的大地坐标系中对应的当前检测坐标信息，以便后续在考虑搭载平台运动信息的情况下对目标进行检测。
在图1所示实施例的基础上，图4是根据本公开的实施例示出的另一种获取目标的检测坐标信息以及所述搭载平台的运动信息的示意流程图。如图4所示，所述获取目标的检测坐标信息以及所述搭载平台的运动信息，包括：
步骤S13，根据所述雷达与所述搭载平台的位置关系补偿位置偏差，以确定检测坐标信息。
在一个实施例中，若雷达并不设置在搭载平台的中心，那么R2≠0，在这种情况下，就需要根据雷达与搭载平台的位置关系补偿位置偏差，也即在确定第一目标在搭载平台的大地坐标系中对应的当前检测坐标信息时，首先需要基于R2进行补偿，例如，在进行坐标系转换的时候，补偿安装误差，以 便保证后续计算的准确性。
在图1所示实施例的基础上，图5是根据本公开的实施例示出的一种根据所述目标的前一时刻检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息的示意流程图。如图5所示，所述根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息，包括：
步骤S21，确定所述搭载平台的运动模型，并根据所述运动模型以及所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息。
在一个实施例中，由于搭载平台是可以运动的，而雷达在搭载平台上随着搭载平台一同运动，搭载平台基于运动模式的不同，从前一时刻到当前所产生的位移也有所不同，因此，为了预测目标的当前预测坐标信息，可以基于搭载平台的运动模型和目标的前一时刻的检测坐标信息进行预测。
其中，搭载平台的运动模型可以是匀加速模型、匀速模型等，以下主要以匀速模型为例进行说明。
可以取目标在前一时刻K的状态变量：
Figure PCTCN2018118861-appb-000001
其中，x为目标在前一时刻K的横坐标，y为目标在前一时刻K的纵坐标，
Figure PCTCN2018118861-appb-000002
为目标在前一时刻K沿横轴的速度，
Figure PCTCN2018118861-appb-000003
为目标在前一时刻K沿纵轴的速度。
目标在前一时刻K对当前时刻K+1的当前预测坐标信息：
X(K+1)＝φ*X(K)；
Figure PCTCN2018118861-appb-000004
其中，t为当前时刻K+1与前一时刻K的时间间隔。
那么得到的目标在当前时刻预测坐标中横坐标为
Figure PCTCN2018118861-appb-000005
纵坐标为
Figure PCTCN2018118861-appb-000006
在图5所示实施例的基础上，图6是根据本公开的实施例示出的一种根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息的示意流程图。如图6所示，所述根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息，包括：
步骤S31，根据所述搭载平台的运动模型和/或所述搭载平台所在的大地坐标系的类型确定滤波器；
步骤S32，通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息，其中，当前修正坐标信息用于确定下一时刻的预测坐标信息。
在一个实施例中，根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息的过程，可以通过滤波器实现，而不同的滤波器具备不同的滤波模型，为了使得当前预测坐标信息与前检测坐标信息与滤波器相适应，可以根据搭载平台的运动模型和/或搭载平台所在的大地坐标系的类型确定滤波器，以便后续通过滤波器对当前预测坐标信息与当前检测坐标信息进行估算。
例如搭载平台的运动模型为线性模型，那么滤波器可以是线性滤波器；例如所述大地坐标系的类型为直角坐标系，那么滤波器也可以是线性滤波器。其中，线性滤波器可以是α-β滤波器，卡尔曼滤波器等。
相应地，例如搭载平台的运动模型为非线性模型，那么滤波器可以是非线性滤波器；例如所述大地坐标系的类型为极坐标系，那么滤波器也可以是非线性滤波器。其中，非线性滤波器可以是扩展卡尔曼滤波器(EKF)，无损卡尔曼滤波器(UKF)。
在图6所示实施例的基础上，图7是根据本公开的实施例示出的另一种根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息的示意流程图。如图7所示，所述根据所述当前预 测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息，还包括：
步骤S33，通过预设关联算法在当前时刻检测的至少一个第二坐标中确定所述当前检测坐标信息；
所述通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息包括：
步骤S321，根据第一权值和所述当前预测坐标信息，以及第二权值和所述当前检测坐标信息计算所述当前修正坐标信息。
在一个实施例中，针对目标进行检测，可以得到至少一个第二坐标，例如目标可以近似为一个点，那么检测得到的第二坐标是一个，在这种情况下，可以根据这一个第二坐标确定当前检测坐标信息，也即将该第二坐标作为当前检测坐标。
而若目标形状较为复杂，例如是具有较多树枝的树木，那么检测得到的第二坐标是多个，然而一般进行计算时，是将目标视作一个点，也即基于目标的多个第二坐标中的一个坐标进行计算，那么就需要在多个第二坐标中确定一个最为合理的第二坐标，基于本实施例，可以通过预设关联算法来在多个坐标中确定一个第二坐标，也即将所确定的第二坐标作为当前检测坐标。
进而通过滤波器进行计算时，可以通过第一权值对当前预测坐标信息进行加权，通过第二权值对检测坐标信息进行加权，以α-β滤波器为例，第一权值为α，第二权值为β，当前时刻为K+1，那么当前修正坐标信息计算方式如下：
Figure PCTCN2018118861-appb-000007
其中，Z(K+1)是当前检测坐标信息，X(K+1|K)是当前预测坐标信息，当K＝1时，那么在首次计算X(K+1|K)时，X(K+1|K)＝X(K+1)＝φ*X(K)，而当K大于1时，也即非首次计算X(K+1|K)时，X(K+1|K)＝X(K+1)＝φ*X(K)，其中的X(K)用
Figure PCTCN2018118861-appb-000008
来替换，在这种情况下
Figure PCTCN2018118861-appb-000009
在图7所示实施例的基础上，图8是根据本公开的实施例示出的一种通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息的示意流程图。如图8所示，所述通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息包括：
步骤S331，计算所述至少一个第二坐标到所述当前预测坐标信息对应的预测坐标的距离；
步骤S332，根据所述至少一个第二坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
在一个实施例中，在对目标检测只得到一个第二坐标时，到预测坐标的距离最小的坐标就是这个第二坐标，因此可以将这个第二坐标作为当前检测坐标信息。
在对目标检测得到多个第二坐标时，可以针对每个第二坐标计算其到预测坐标的距离，距离越小，说明检测的坐标与预测的坐标越近，更能体现目标真实位置对应的坐标，因此将其中最小的距离对应的第二坐标作为当前检测坐标信息。
在图7所示实施例的基础上，图9是根据本公开的实施例示出的另一种通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息的示意流程图。如图9所示，所述通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息包括：
步骤S333，在至少一个第二坐标中确定位于预设区域内(包括位于预设区域的边沿)的至少一个关联坐标；
步骤S334，计算所述至少一个关联坐标到所述当前预测坐标信息对应的预测坐标的距离；
步骤S335，根据所述至少一个关联坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
在一个实施例中，与图8所示的实施例不同的，在计算所检测的坐标到预测坐标的距离之前，先在所检测到的第二坐标中确定位于预设区域内的关 联坐标，其中，预设区域可以是包含预测坐标的区域，例如以预测坐标为中心的区域，也可以是位于预测坐标附近，但是不包含预测坐标的区域。
基于预设区域可以从至少一个第二坐标中剔除没有位于预设区域内的坐标，以便减少后续需要计算到预测坐标的距离的关联坐标的数量，从而降低总体的计算量。
在图9所示实施例的基础上，图10是根据本公开的实施例示出的一种确定当前检测坐标信息的示意图。
如图10所示，以雷达和搭载平台的中心重合为例，预测坐标为A(x 0,y 0)，检测到的三个第二坐标分别为B(x 1,y 1)，C(x 2,y 2)，D(x 3,y 3)，预设区域为以预测坐标A为圆心，半径为DIS的圆形区域。
首先根据步骤S333，可以确定D点位于预设区域外，从而将D点剔除，保留B点和C点作为关联坐标，进而针对B点和C点分别计算到A点的距离，其中：
Figure PCTCN2018118861-appb-000010
Figure PCTCN2018118861-appb-000011
根据计算结果可以确定e 01小于e 02，从而将e 01对应的点B的坐标作为当前检测坐标信息。
可选地，所述预设区域为以所述预测坐标为圆心，以第一预设距离DIS为半径的圆形区域；
其中，所述第一权值等于所述当前检测坐标信息对应的检测坐标到所述预测坐标的距离与所述第一预设距离的比值，所述第二权值等于1与所述第一权值之差。
在一个实施例中，基于图9所示的实施例，第一权值α＝e 01/DIS，第二权值β＝1-α，其中，e 01的最大值为DIS，据此设置第一权值和第二权值，由于DIS是固定值，可以使得第一权值α与e 01正相关，根据当前修正坐标信息的计算公式：
Figure PCTCN2018118861-appb-000012
据此可知，α表示X(K+1|K)这一项的权重，而X(K+1|K)这一项是当前预测坐标信息，当前预测坐标信息的可信度越高，其权重应该越高，通过设置α＝e 01/DIS，可以保证第一权值α与e 01正相关，也即当前检测坐标信息越接近当前预测坐标信息，说明预测的坐标与检测的坐标约接近，也就说明预测的越准，即当前预测坐标信息的可信度越高，那么α就越大。
需要说明的是，第一预设距离DIS，可以根据雷达的精度、搭载平台的运动速度、雷达与目标之间的距离等参数进行设置。例如雷达的精度越高，DIS越小；搭载平台的速度越大，DIS越大；雷达与目标之间的距离越远，DIS越大。
在图6所示实施例的基础上，图11是根据本公开的实施例示出的一种根据所述搭载平台的运动模型和/或所述大地坐标系的类型确定滤波器的示意流程图。如图11所示，所述根据所述搭载平台的运动模型和/或所述大地坐标系的类型确定滤波器包括：
步骤S311，在所述搭载平台的运动模型为线性模型，和/或所述大地坐标系的类型为直角坐标系时，确定线性滤波器；
步骤S312，在所述搭载平台的运动模型为非线性模型，和/或所述大地坐标系的类型为极坐标系时，确定非线性滤波器。
在一个实施例中，由于线性滤波器便于对线性模型和直角坐标系内的数据进行运算，而非线性滤波器便于对非线性模型和极坐标系内的数据进行运算，因此根据本实施例确定滤波器，便后续通过滤波器对当前预测坐标信息与当前检测坐标信息进行估算。
可选地，所述线性滤波器包括以下至少之一：
α-β滤波器，卡尔曼滤波器。
可选地，所述非线性滤波器包括以下至少之一：
扩展卡尔曼滤波器，无损卡尔曼滤波器。
可选地，所述搭载平台的运动信息包括如下至少一种：位置、速度。
仍以图5所示的实施例为例，K时刻的状态边框，
Figure PCTCN2018118861-appb-000013
其中x和y可以表示搭载平台的位置，
Figure PCTCN2018118861-appb-000014
和
Figure PCTCN2018118861-appb-000015
分别表示搭载平台沿在大地坐标系中沿横轴和纵轴的速度。
当然，这是以匀速模型为例时搭载平台的运动信息，在运动模型为匀加速模型时，搭载平台的运动信息还可以包括加速度。
可选地，所述目标为一个或多个。
在一个实施例中，上述实施例中确定当前修正坐标信息的过程，可以针对一个目标进行，也可以针对多个目标进行。
需要说明的是，上述实施例中所述的目标检测方法，可以应用于雷达，也可以应用于其他设备，例如上文提到的图像采集设备，在这种情况下，可以基于图像采集设备构建坐标系，其中，图像采集设备可以基于采集图像时自身的姿态信息，以及图像中目标的深度信息，确定目标相对于自己的位置，从而作为目标的检测坐标信息。
图12是根据本公开的实施例示出的一种航迹管理方法的示意流程图。航迹管理主要是指航迹起始、航迹维持、航迹终结。本实施例公开一种航迹管理方法，可以用于上述航迹管理，具体的，可以对目标的自由点、可靠航迹以及航迹的销毁进行管理。本实施所述的航迹管理方法包括上述任一实施例所述的目标检测方法如图12所示，航迹管理方法还包括：
步骤S1’，根据所述目标的当前修正坐标信息确定所述目标的轨迹；
雷达在某个时刻对目标进行检测，可能因受到干扰，导致所检测的点迹不仅包括目标点迹，还包括杂波点迹。在本实施例中，雷达能够根据上述实施例确定的目标的当前修正坐标信息，并进一步基于当前修正坐标信息确定所述目标的轨迹，例如可以将从0至N这N个时刻确定的当前修正坐标信息按照时间顺序相连得到的轨迹，作为目标的轨迹，其中所述目标可以为一个或多个。
步骤S2’，确定所检测的多个目标的轨迹的可信度；
由于雷达可以同时对多个目标进行航迹管理，通过为多个目标的轨迹引入可信度，以对多个目标的轨迹进行管理。在一个实施例中，对每个目标的轨迹可以确定其可信度，由于目标的轨迹是根据当前修正坐标信息确定的，而当前修正坐标信息确定又是根据当前预测坐标信息与当前检测坐标信息进行估算得到的，由于所采用的当前预测坐标信息是预测得到的，所以并不一定是准确的，因此据此估算得到的当前修正坐标信息也并不一定是准确的，进而基于当前修正坐标信息确定的轨迹也也并不一定是准确的，因此可以通过可信度来表达轨迹的准确性，其中，可信度的确定方式在后续实施例进行示例性说明。
步骤S3’，删除可信度低于预设可信度的轨迹。
在一个实施例中，由于雷达可以对多个目标的轨迹进行管理，而可信度较低(例如低于预设可信度)的轨迹其准确性是较差的，也即与目标的真实运动轨迹相差较远，那么就无需对其进行继续监测，因此对于可信度低于预设可信度的轨迹可以删除，从而仅保留可信度较高的轨迹，据此有利于降低雷达的负荷，使得雷达仅对于可信度较高的轨迹对应的目标进行检测。
需要说明的是，本实施例所示的航迹管理方法中步骤的执行频率，即对目标进行更新轨迹的频率，与前述实施例中目标检测方法中步骤的执行频率，即对目标进行确定当前修正坐标信息的频率，可以是相同的，也可以是不同的。
在两者是相同的情况下，针对目标每确定一次当前修正坐标信息，就根据新确定的当前修正坐标信息更新一次目标的轨迹。
在两者是不同的情况下，若航迹管理方法中步骤的执行频率，小于目标检测方法中步骤的执行频率，也即在确定多次当前修正坐标信息后，才会更新一次目标的轨迹，在这种情况下，可以基于最后一次确定的当前修正坐标信息，来更新目标的轨迹。若航迹管理方法中步骤的执行频率，大于目标检测方法中步骤的执行频率，也即在确定一次当前修正坐标信息，会更新多次 目标的轨迹，在这种情况下，每次更新轨迹可以基于当前轨迹更新之前所确定的最近邻时刻的修正坐标信息。
以下主要在航迹管理方法中步骤的执行频率(例如15Hz)，小于目标检测方法中步骤的执行频率(例如100Hz)的情况下，对本公开的实施例进行示例性说明。
在图12所示实施例的基础上，图13是根据本公开的实施例示出的另一种航迹管理方法的示意流程图。如图13所示，所述航迹管理方法还包括：
步骤S4’，计算每条轨迹在当前时刻到所述大地坐标系的原点的轨迹距离；
步骤S5’，根据所述轨迹距离对每条轨迹进行排序。
在一个实施例中，由于雷达设置在搭载平台上，相对于目标是运动的，从而目标相对于雷达也是运动的，而对于某个目标的轨迹而言，在不同时刻目标到雷达的距离会发生变化，也即目标在轨迹上对应的点，到大地坐标系的原点的轨迹距离会发生变化，从而在对于多个目标进行检测时，在多个目标对应的轨迹中，在不同时刻到雷达最近的轨迹有所不同，而一般而言，轨迹到雷达的轨迹距离越近，说明该轨迹越可能与搭载平台发生碰撞，也即威胁程度越高，因此，可以根据轨迹距离对轨迹进行排序。
其中，可以通过编号标识排序，例如越小的轨迹距离对应的轨迹，其排序越靠前，编号越小，从而使得雷达可以将威胁程度较高的轨迹靠前显示，便于用户及时作出应对操作。另外，对于被删除的轨迹，其编号可以重新被赋予其他轨迹，以便进行排序。
在图13所示实施例的基础上，图14是根据本公开的实施例示出的又一种航迹管理方法的示意流程图。如图14所示，所述航迹管理方法还包括：
步骤S6’，根据排序的结果输出所述轨迹。
在一个实施例中，通过根据排序的结果输出轨迹，可以使得用户能够确定每条轨迹的排序，其中，输出的方式可以通过屏幕显示输出，也可以通过音频按照排序播放轨迹的标识。
可选地，所述可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关。
在一个实施例中，针对一个目标的轨迹而言，由于该轨迹是基于多个时刻的当前修正坐标信息确定的，而每个时刻的当前修正坐标信息，又是根据当前预测坐标信息与当前检测坐标信息进行估算得到的，其中的当前检测坐标信息，则是基于目标的当前预测坐标信息对应的检测坐标(例如上述实施例中的第二坐标)到所述当前预测坐标信息对应的预测坐标的距离来确定的，并且需要基于检测坐标到预测坐标的距离(例如上述e 01)来设置权值，具体确定方式可以按照图8或图9所示的实施例进行。
其中，确定当前修正坐标信息的过程，本质上是一个预测的过程，由于预测的过程中用于预测的量(例如图5所示实施例中的
Figure PCTCN2018118861-appb-000016
和
Figure PCTCN2018118861-appb-000017
)可能会发生变化，因此出现较大偏差的概率较高，会导致确定当前修正坐标信息较为不准。
确定当前检测坐标信息的过程，本质上是一个检测的过程，其中包括上述计算检测坐标到预测坐标的距离的过程，而检测坐标是实际检测得到的，因此出现较大偏差的概率较小，会使得确定当前修正坐标信息较为准确。
检测坐标到预测坐标的距离，代表检测结果与预测结果的差异，检测坐标到预测坐标的距离越大，那么检测结果与预测结果的差异越大，说明预测过程就越不准确，确定当前修正坐标信息就越不准。
因此，由于由于该轨迹是基于多个时刻的当前修正坐标信息确定的，在确定目标的轨迹的过程中，确定当前预测坐标信息的次数越多，确定的当前修正坐标信息就越不准，也即确定的轨迹准确性越低，计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数越多，确定的当前修正坐标信息就越准，也即确定的轨迹准确性越高，那么所预测的轨迹就越不准，目标的检测坐标到预测坐标的距离越大，确定 的当前修正坐标信息就越不准，也即确定的轨迹准确性越低。
从而按照本实施例设置可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关，可以保证可信度计算的准确性，以便准确地确定目标的轨迹是否可信。
在图12所示实施例的基础上，图15是根据本公开的实施例示出的一种确定所检测的多个目标的轨迹的可信度的示意流程图。如图15所示，所述确定所检测的多个目标的轨迹的可信度包括：
步骤S21’，根据新检测的目标的当前修正坐标信息确定所述新检测的目标是否属于已记录的轨迹；
步骤S22’，若不属于已记录的轨迹，初始化所述新检测的目标的轨迹和可信度；
步骤S23’，每预测一次所述新检测的目标的当前预测坐标信息，从初始化的可信度中减去第一预设可信度；
步骤S24’，每计算一次所述新检测的目标的检测坐标到预测坐标的距离，从初始化的可信度中加上第二预设可信度，其中，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离反相关。
在一个实施例中，由于可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关，因此每预测一次新检测的目标的当前预测坐标信息，可以从初始化的可信度中减去第一预设可信度，每计算一次新检测的目标的检测坐标到预测坐标的距离，从初始化的可信度中加上第二预设可信度，并且所加上的第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离反相关。
例如初始化的可信度为30分，那么每预测一次所述新检测的目标的当前预测坐标信息，从初始化的可信度中减去0.5分，每计算一次所述新检测的目 标的检测坐标到预测坐标的距离，从初始化的可信度中加上10-(e 01-0.3)*10/(3-0.3)分，最终可以根据可信度的计算结果删除可信度低于预设可信度(例如20分)的轨迹，并对剩余的轨迹排序和输出。
例如在航迹管理方法中步骤的执行频率为15Hz，目标检测方法中步骤的执行频率为100Hz的情况下，那么每更新一次轨迹，需要预测6次新检测的目标的当前预测坐标信息，从而减去3分。
另外，当得到当前修正坐标信息时，可以为当前修正坐标信息设置标识(flag)，而每当需要更新轨迹时，可以确定当前修正坐标信息是否存在标识，若存在标识，则基于当前修正坐标信息预测新检测的目标的当前预测坐标信息，并计算一次新检测的目标的检测坐标到预测坐标的距离，然后删除该标识，若不存在标识，说明针对当前修正坐标信息已经计算过新检测的目标的检测坐标到预测坐标的距离，那么就无需基于当前修正坐标信息预测新检测的目标的当前预测坐标信息。据此，可以保证针对每次得到的当前修正坐标信息，是否已计算一次新检测的目标的检测坐标到预测坐标的距离，进而准确地进行加分。
可选地，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离，在第一预设范围内反相关，在小于第一预设范围的下限值地范围内等于第一预设值，在大于第一预设范围的上限值的范围内等于第二预设值，其中，所述第一预设值大于所述第二预设值。
在一个实施例中，第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离，可以仅在第一预设范围内反相关，例如在e 01∈(0.3m,3m)这个范围内，第二预设可信度等于10-(e 01-0.3)*10/(3-0.3)，在e 01≤0.3m的范围内，第二预设可信度可以等于10，在e 01≥0.3m的范围内，第二预设可信度可以等于0。其中，第一预设范围可以根据需要进行设定。
与前述目标检测方法和航迹管理方法的实施例相对应地，本公开还提出了目标检测装置和航迹管理装置的实施例。
本公开的实施例提出一种目标检测装置，适用于雷达，所述雷达设置在 搭载平台中，所述装置包括处理器，所述处理器用于，
获取目标的检测坐标信息以及所述搭载平台的运动信息；
根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
在一个实施例中，所述处理器用于，通过雷达检测目标，以确定所述目标在雷达的坐标系中的所述第一坐标；
根据所述雷达与所述搭载平台的位置关系，以确定所述第一目标在所述搭载平台的大地坐标系中对应的所述检测坐标信息。
在一个实施例中，所述处理器用于，根据所述雷达与所述搭载平台的位置关系补偿位置偏差，以确定所述检测坐标信息。
在一个实施例中，所述处理器用于，确定所述搭载平台的运动模型，并根据所述运动模型以及所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息。
在一个实施例中，所述处理器用于，根据所述搭载平台的运动模型和/或所述搭载平台所在的大地坐标系的类型确定滤波器；
通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息，其中，当前修正坐标信息用于确定下一时刻的预测坐标信息。
在一个实施例中，所述处理器用于，通过预设关联算法在当前时刻检测的至少一个第二坐标中确定所述当前检测坐标信息；
所述通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息包括：
根据第一权值和所述当前预测坐标信息，以及第二权值和所述当前检测坐标信息计算所述当前修正坐标信息。
在一个实施例中，所述处理器用于，计算所述至少一个第二坐标到所述 当前预测坐标信息对应的预测坐标的距离；
根据所述至少一个第二坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
在一个实施例中，所述处理器用于，在至少一个第二坐标中确定位于预设区域内的至少一个关联坐标；
计算所述至少一个关联坐标到所述当前预测坐标信息对应的预测坐标的距离；
根据所述至少一个关联坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
在一个实施例中，所述预设区域为以所述预测坐标为圆心，以第一预设距离为半径的圆形区域；
其中，所述第一权值等于所述当前检测坐标信息对应的检测坐标到所述预测坐标的距离与所述第一预设距离的比值，所述第二权值等于1与所述第一权值之差。
在一个实施例中，所述处理器用于，在所述搭载平台的运动模型为线性模型，和/或所述大地坐标系的类型为直角坐标系时，确定线性滤波器；
在所述搭载平台的运动模型为非线性模型，和/或所述大地坐标系的类型为极坐标系时，确定非线性滤波器。
在一个实施例中，所述线性滤波器包括以下至少之一：
α-β滤波器，卡尔曼滤波器。
在一个实施例中，所述非线性滤波器包括以下至少之一：
扩展卡尔曼滤波器，无损卡尔曼滤波器。
在一个实施例中，所述搭载平台的运动信息包括如下至少一种：位置、速度。
在一个实施例中，所述目标为一个或多个。
本公开的实施例提出一种航迹管理装置，包括处理器，所述处理器由于，根据上述任一实施例所述的目标检测装置确定的所述目标的当前修正坐标信 息确定所述目标的轨迹；
确定所检测的多个目标的轨迹的可信度；
删除可信度低于预设可信度的轨迹。
在一个实施例中，所述处理器还用于，计算每条轨迹在当前时刻到所述大地坐标系的原点的轨迹距离；
根据所述轨迹距离对每条轨迹进行排序。
在一个实施例中，所述处理器还用于，根据排序的结果输出所述轨迹。
在一个实施例中，所述可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关。
在一个实施例中，所述处理器用于，根据新检测的目标的当前修正坐标信息确定所述新检测的目标是否属于已记录的轨迹；
若不属于已记录的轨迹，初始化所述新检测的目标的轨迹和可信度；
每预测一次所述新检测的目标的当前预测坐标信息，从初始化的可信度中减去第一预设可信度；
每计算一次所述新检测的目标的检测坐标到预测坐标的距离，从初始化的可信度中加上第二预设可信度，其中，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离反相关。
在一个实施例中，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离，在第一预设范围内反相关，在小于第一预设范围的下限值地范围内等于第一预设值，在大于第一预设范围的上限值的范围内等于第二预设值，其中，所述第一预设值大于所述第二预设值。
本公开的实施例提出一种无人飞行器，包括上述任一实施例所述的目标检测装置和/或航迹管理装置。
上述实施例阐明的系统、装置、模块或单元，具体可以由计算机芯片或 实体实现，或者由具有某种功能的产品来实现。为了描述的方便，描述以上装置时以功能分为各种单元分别描述。当然，在实施本申请时可以把各单元的功能在同一个或多个软件和/或硬件中实现。本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上所述仅为本申请的实施例而已，并不用于限制本申请。对于本领域技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本申请的权利要求范围之内。

Claims (41)

  1. 一种目标检测方法，其特征在于，适用于雷达，所述雷达设置在搭载平台中，所述方法包括：
    获取目标的检测坐标信息以及所述搭载平台的运动信息；
    根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
    根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
  2. 根据权利要求1所述的方法，其特征在于，所述获取目标的检测坐标信息以及所述搭载平台的运动信息，包括：
    通过雷达检测目标，以确定所述目标在雷达的坐标系中的所述第一坐标；
    根据所述雷达与所述搭载平台的位置关系，以确定所述第一目标在所述搭载平台的大地坐标系中对应的所述检测坐标信息。
  3. 根据权利要求1所述的目标检测方法，其特征在于，所述获取目标的检测坐标信息以及所述搭载平台的运动信息，包括：
    根据所述雷达与所述搭载平台的位置关系补偿位置偏差，以确定所述检测坐标信息。
  4. 根据权利要求1所述的目标检测方法，其特征在于，所述根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息，包括：
    确定所述搭载平台的运动模型，并根据所述运动模型以及所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息。
  5. 根据权利要求4所述的目标检测方法，其特征在于，所述根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息，包括：
    根据所述搭载平台的运动模型和/或所述搭载平台所在的大地坐标系的类 型确定滤波器；
    通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息，其中，当前修正坐标信息用于确定下一时刻的预测坐标信息。
  6. 根据权利要求5所述的方法，其特征在于，所述根据所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息，还包括：
    通过预设关联算法在当前时刻检测的至少一个第二坐标中确定所述当前检测坐标信息；
    所述通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息包括：
    根据第一权值和所述当前预测坐标信息，以及第二权值和所述当前检测坐标信息计算所述当前修正坐标信息。
  7. 根据权利要求6所述的方法，其特征在于，所述通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息包括：
    计算所述至少一个第二坐标到所述当前预测坐标信息对应的预测坐标的距离；
    根据所述至少一个第二坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
  8. 根据权利要求6所述的方法，其特征在于，所述通过预设关联算法在当前时刻检测的至少一个坐标中确定所述当前检测坐标信息包括：
    在至少一个第二坐标中确定位于预设区域内的至少一个关联坐标；
    计算所述至少一个关联坐标到所述当前预测坐标信息对应的预测坐标的距离；
    根据所述至少一个关联坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
  9. 根据权利要求8所述的方法，其特征在于，所述预设区域为以所述预 测坐标为圆心，以第一预设距离为半径的圆形区域；
    其中，所述第一权值等于所述当前检测坐标信息对应的检测坐标到所述预测坐标的距离与所述第一预设距离的比值，所述第二权值等于1与所述第一权值之差。
  10. 根据权利要求5所述的方法，其特征在于，所述根据所述搭载平台的运动模型和/或所述大地坐标系的类型确定滤波器包括：
    在所述搭载平台的运动模型为线性模型，和/或所述大地坐标系的类型为直角坐标系时，确定线性滤波器；
    在所述搭载平台的运动模型为非线性模型，和/或所述大地坐标系的类型为极坐标系时，确定非线性滤波器。
  11. 根据权利要求10所述的方法，其特征在于，所述线性滤波器包括以下至少之一：
    α-β滤波器，卡尔曼滤波器。
  12. 根据权利要求10所述的方法，其特征在于，所述非线性滤波器包括以下至少之一：
    扩展卡尔曼滤波器，无损卡尔曼滤波器。
  13. 根据权利要求1至12中任一项所述的方法，其特征在于，所述搭载平台的运动信息包括如下至少一种：位置、速度。
  14. 根据权利要求1至12中任一项所述的方法，其特征在于，所述目标为一个或多个。
  15. 一种航迹管理方法，其特征在于，包括权利要求1-14任一项所述的方法，还包括：
    根据所述目标的当前修正坐标信息确定所述目标的轨迹；
    确定所检测的多个目标的轨迹的可信度；
    删除可信度低于预设可信度的轨迹。
  16. 根据权利要求15所述的方法，其特征在于，还包括：
    计算每条轨迹在当前时刻到所述大地坐标系的原点的轨迹距离；
    根据所述轨迹距离对每条轨迹进行排序。
  17. 根据权利要求16所述的方法，其特征在于，还包括：
    根据排序的结果输出所述轨迹。
  18. 根据权利要求15所述的方法，其特征在于，所述可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关。
  19. 根据权利要求18所述的方法，其特征在于，所述确定所检测的多个目标的轨迹的可信度包括：
    根据新检测的目标的当前修正坐标信息确定所述新检测的目标是否属于已记录的轨迹；
    若不属于已记录的轨迹，初始化所述新检测的目标的轨迹和可信度；
    每预测一次所述新检测的目标的当前预测坐标信息，从初始化的可信度中减去第一预设可信度；
    每计算一次所述新检测的目标的检测坐标到预测坐标的距离，从初始化的可信度中加上第二预设可信度，其中，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离反相关。
  20. 根据权利要求19所述的方法，其特征在于，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离，在第一预设范围内反相关，在小于第一预设范围的下限值地范围内等于第一预设值，在大于第一预设范围的上限值的范围内等于第二预设值，其中，所述第一预设值大于所述第二预设值。
  21. 一种目标检测装置，其特征在于，适用于雷达，所述雷达设置在搭载平台中，所述装置包括处理器，所述处理器用于，
    获取目标的检测坐标信息以及所述搭载平台的运动信息；
    根据所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息；
    根据所述当前预测坐标信息与当前检测坐标信息进行估算，以获得所述目标的当前修正坐标信息。
  22. 根据权利要求21所述的装置，其特征在于，所述处理器用于，通过雷达检测目标，以确定所述目标在雷达的坐标系中的所述第一坐标；
    根据所述雷达与所述搭载平台的位置关系，以确定所述第一目标在所述搭载平台的大地坐标系中对应的所述检测坐标信息。
  23. 根据权利要求21所述的目标检测装置，其特征在于，所述处理器用于，根据所述雷达与所述搭载平台的位置关系补偿位置偏差，以确定所述检测坐标信息。
  24. 根据权利要求21所述的目标检测装置，其特征在于，所述处理器用于，确定所述搭载平台的运动模型，并根据所述运动模型以及所述目标的前一时刻的检测坐标信息以及所述搭载平台的当前运动信息，确定所述目标的当前预测坐标信息。
  25. 根据权利要求24所述的目标检测装置，其特征在于，所述处理器用于，
    根据所述搭载平台的运动模型和/或所述搭载平台所在的大地坐标系的类型确定滤波器；
    通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息，其中，当前修正坐标信息用于确定下一时刻的预测坐标信息。
  26. 根据权利要求25所述的装置，其特征在于，所述处理器用于，通过预设关联算法在当前时刻检测的至少一个第二坐标中确定所述当前检测坐标信息；
    所述通过所述滤波器，基于所述当前预测坐标信息与所述当前检测坐标信息进行估算，以获得所述当前修正坐标信息包括：
    根据第一权值和所述当前预测坐标信息，以及第二权值和所述当前检测坐标信息计算所述当前修正坐标信息。
  27. 根据权利要求26所述的装置，其特征在于，所述处理器用于，计算所述至少一个第二坐标到所述当前预测坐标信息对应的预测坐标的距离；
    根据所述至少一个第二坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
  28. 根据权利要求26所述的装置，其特征在于，所述处理器用于，在至少一个第二坐标中确定位于预设区域内的至少一个关联坐标；
    计算所述至少一个关联坐标到所述当前预测坐标信息对应的预测坐标的距离；
    根据所述至少一个关联坐标中到所述预测坐标的距离最小的坐标确定所述当前检测坐标信息。
  29. 根据权利要求28所述的装置，其特征在于，所述预设区域为以所述预测坐标为圆心，以第一预设距离为半径的圆形区域；
    其中，所述第一权值等于所述当前检测坐标信息对应的检测坐标到所述预测坐标的距离与所述第一预设距离的比值，所述第二权值等于1与所述第一权值之差。
  30. 根据权利要求25所述的装置，其特征在于，所述处理器用于，在所述搭载平台的运动模型为线性模型，和/或所述大地坐标系的类型为直角坐标系时，确定线性滤波器；
    在所述搭载平台的运动模型为非线性模型，和/或所述大地坐标系的类型为极坐标系时，确定非线性滤波器。
  31. 根据权利要求30所述的装置，其特征在于，所述线性滤波器包括以下至少之一：
    α-β滤波器，卡尔曼滤波器。
  32. 根据权利要求30所述的装置，其特征在于，所述非线性滤波器包括以下至少之一：
    扩展卡尔曼滤波器，无损卡尔曼滤波器。
  33. 根据权利要求21至32中任一项所述的装置，其特征在于，所述搭 载平台的运动信息包括如下至少一种：位置、速度。
  34. 根据权利要求21至32中任一项所述的装置，其特征在于，所述目标为一个或多个。
  35. 一种航迹管理装置，其特征在于，包括处理器，所述处理器由于，根据权利要求21至34中任一项所述的目标检测装置确定的所述目标的当前修正坐标信息确定所述目标的轨迹；
    确定所检测的多个目标的轨迹的可信度；
    删除可信度低于预设可信度的轨迹。
  36. 根据权利要求35所述的装置，其特征在于，所述处理器还用于，计算每条轨迹在当前时刻到所述大地坐标系的原点的轨迹距离；
    根据所述轨迹距离对每条轨迹进行排序。
  37. 根据权利要求36所述的装置，其特征在于，所述处理器还用于，根据排序的结果输出所述轨迹。
  38. 根据权利要求35所述的装置，其特征在于，所述可信度与目标被确定当前预测坐标信息的次数反相关，与计算目标的当前预测坐标信息对应的检测坐标到所述当前预测坐标信息对应的预测坐标的距离的次数正相关，与目标的检测坐标到预测坐标的距离反相关。
  39. 根据权利要求38所述的装置，其特征在于，所述处理器用于，根据新检测的目标的当前修正坐标信息确定所述新检测的目标是否属于已记录的轨迹；
    若不属于已记录的轨迹，初始化所述新检测的目标的轨迹和可信度；
    每预测一次所述新检测的目标的当前预测坐标信息，从初始化的可信度中减去第一预设可信度；
    每计算一次所述新检测的目标的检测坐标到预测坐标的距离，从初始化的可信度中加上第二预设可信度，其中，所述第二预设可信度与所述新检测的目标的检测坐标到预测坐标的距离反相关。
  40. 根据权利要求39所述的装置，其特征在于，所述第二预设可信度与 所述新检测的目标的检测坐标到预测坐标的距离，在第一预设范围内反相关，在小于第一预设范围的下限值地范围内等于第一预设值，在大于第一预设范围的上限值的范围内等于第二预设值，其中，所述第一预设值大于所述第二预设值。
  41. 一种无人飞行器，其特征在于，包括上述任一项权利要求所述的目标检测装置和/或航迹管理装置。
PCT/CN2018/118861 2018-12-03 2018-12-03 目标检测方法和装置、航迹管理方法和装置以及无人机 WO2020113357A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/118861 WO2020113357A1 (zh) 2018-12-03 2018-12-03 目标检测方法和装置、航迹管理方法和装置以及无人机
CN201880069284.6A CN111279215A (zh) 2018-12-03 2018-12-03 目标检测方法和装置、航迹管理方法和装置以及无人机

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/118861 WO2020113357A1 (zh) 2018-12-03 2018-12-03 目标检测方法和装置、航迹管理方法和装置以及无人机

Publications (1)

Publication Number Publication Date
WO2020113357A1 true WO2020113357A1 (zh) 2020-06-11

Family

ID=70974043

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/118861 WO2020113357A1 (zh) 2018-12-03 2018-12-03 目标检测方法和装置、航迹管理方法和装置以及无人机

Country Status (2)

Country Link
CN (1) CN111279215A (zh)
WO (1) WO2020113357A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113766426B (zh) * 2020-06-29 2024-04-12 京东城市(北京)数字科技有限公司 车辆行驶风险的预警方法、预警装置、电子设备及存储介质
CN111986522B (zh) * 2020-07-29 2022-03-22 广州市新航科技有限公司 基于ads-b信号的机载设备定位方法、机载设备及其存储介质
CN112179375B (zh) * 2020-08-20 2022-08-09 中国科学院深圳先进技术研究院 安全保护装置的控制方法及相关装置
CN112799421B (zh) * 2021-03-30 2021-07-02 北京三快在线科技有限公司 无人机的控制方法、装置、存储介质和电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101609149A (zh) * 2009-07-13 2009-12-23 北京航空航天大学 一种提高机载激光雷达定姿精度的方法
JP6041547B2 (ja) * 2012-06-08 2016-12-07 三菱電機株式会社 追尾装置
CN106814753A (zh) * 2017-03-20 2017-06-09 成都通甲优博科技有限责任公司 一种目标位置矫正方法、装置及系统
CN108898624A (zh) * 2018-06-12 2018-11-27 浙江大华技术股份有限公司 一种运动物体跟踪的方法、装置、电子设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101609149A (zh) * 2009-07-13 2009-12-23 北京航空航天大学 一种提高机载激光雷达定姿精度的方法
JP6041547B2 (ja) * 2012-06-08 2016-12-07 三菱電機株式会社 追尾装置
CN106814753A (zh) * 2017-03-20 2017-06-09 成都通甲优博科技有限责任公司 一种目标位置矫正方法、装置及系统
CN108898624A (zh) * 2018-06-12 2018-11-27 浙江大华技术股份有限公司 一种运动物体跟踪的方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN111279215A (zh) 2020-06-12

Similar Documents

Publication Publication Date Title
WO2020113357A1 (zh) 目标检测方法和装置、航迹管理方法和装置以及无人机
CA2614977C (en) Automatic past error corrections for location and inventory tracking
US9990732B2 (en) Entity recognition system
US8838374B2 (en) Automatic correction of past position errors for location and inventory tracking
Fakharian et al. Adaptive Kalman filtering based navigation: An IMU/GPS integration approach
US7489255B2 (en) Self-position identification apparatus and self-position identification method
US20150241560A1 (en) Apparatus and method for providing traffic control service
CN109781099A (zh) 一种自适应ukf算法的导航方法及系统
CN109212521A (zh) 一种基于前视相机与毫米波雷达融合的目标跟踪方法
US20180154901A1 (en) Method and system for localizing a vehicle
US7417583B2 (en) Methods and apparatus for providing target altitude estimation in a two dimensional radar system
CN109564285A (zh) 用于检测在移动单元的交通环境下的地面标志的方法和系统
EP2923175A1 (en) Improving kalman filtering with indirect noise measurements
CN113419258B (zh) 一种定位异常检测方法及其相关设备
US10866307B2 (en) Method for analyzing error and existence probability of multi-sensor fusion of obstacle detection
CN106197415A (zh) 终端对可穿戴设备的跟踪方法、装置
Forno et al. Techniques for improving localization applications running on low-cost IoT devices
CA2958759A1 (en) Enhanced positioning method for moving target in mine shaft based on witness nodes under internet of things architecture
JP5604893B2 (ja) 目標物管理装置および目標物管理方法
CN114861725A (zh) 一种目标感知跟踪的后处理方法、装置、设备及介质
CN114281832A (zh) 基于定位结果的高精地图数据更新方法、装置和电子设备
GB2555779A (en) Acquisition and/or tracking of remote object
US10830906B2 (en) Method of adaptive weighting adjustment positioning
CN114078084A (zh) 一种基于坐标映射实现目标定位及跟踪的方法
JP6783681B2 (ja) 演算装置、演算方法およびプログラム

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: 18942636

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: 18942636

Country of ref document: EP

Kind code of ref document: A1