Laser radar distance ambiguity elimination method

Abstract

An embodiment of the present invention provides a method for eliminating distance ambiguity of a laser radar, the method comprising: obtaining point cloud data to be processed, determining a pulse repetition frequency corresponding to the point cloud data; determining a pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency; and determining a mapping sequence between a plurality of echo information points and pulse emission intervals. According to the mapping sequence and the pulse emission time sequence, the first mapping relationship is adjusted to obtain target point cloud data for distance detection. Through the mapping sequence, the pulse emission interval in which the plurality of echo information points are located is determined, and in combination with the pulse emission time sequence, the actual pulse emission time corresponding to the plurality of echo information points is determined to establish a correct mapping relationship between the echo pulse and the pulse emission time, thereby correcting the erroneous position of the distance-ambiguous point cloud data to eliminate the distance ambiguity problem existing in the point cloud data.

Description

Technical Field

[0001] The present invention relates to the field of laser radar technology, and in particular to a laser radar distance ambiguity elimination method. Background Art

[0002] The LiDAR (Light Detection and Ranging) system is an advanced optical remote sensing technology that measures the distance and other parameters of the target by emitting pulsed lasers to the scanning object, and then calculates the system orientation information provided by the Position and Orientation (POS) system to achieve direct measurement of the coordinates of the ground object.

[0003] Since the first commercial airborne LiDAR appeared in the 1990s, the hardware system of airborne LiDAR has developed rapidly, which has led to a continuous increase in the pulse repetition frequency of the laser, which inevitably brings about the problem of range ambiguity. Range ambiguity refers to the laser scanner incorrectly recording the distance to the target, which will cause a certain number of points in the point cloud data to appear in the wrong position visible to the naked eye, perhaps lurking below the ground or suspended in the air. The range ambiguity problem caused by the LiDAR system will reduce the quality of the point cloud data, making the point cloud limited in usable range or completely unusable. Summary of the invention

[0004] An embodiment of the present invention provides a laser radar distance ambiguity elimination method, which can eliminate the distance ambiguity problem existing in the laser radar system.

[0005] In a first aspect, an embodiment of the present invention provides a method for eliminating laser radar distance ambiguity, the method comprising:

[0006] Acquire point cloud data to be processed, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments;

[0007] Determining a pulse repetition frequency corresponding to the point cloud data;

[0008] Determining a pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency;

[0009] If the plurality of echo information points are non-isolated data, determining a mapping sequence of the plurality of echo information points and pulse emission intervals, wherein the pulse emission interval is determined based on the pulse repetition frequency;

[0010] According to the mapping sequence and the pulse emission time sequence, the first mapping relationship is adjusted to obtain target point cloud data for realizing distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

[0011] In a second aspect, an embodiment of the present invention provides a laser radar distance ambiguity elimination device, the device comprising:

[0012] An acquisition module, used for acquiring point cloud data to be processed, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments;

[0013] A first determination module, used to determine the pulse repetition frequency corresponding to the point cloud data;

[0014] A second determination module is used to determine a pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency;

[0015] A third determination module, configured to determine a mapping sequence of the plurality of echo information points and pulse emission intervals if the plurality of echo information points are non-isolated data, wherein the pulse emission interval is determined based on the pulse repetition frequency;

[0016] An adjustment module is used to adjust the first mapping relationship according to the mapping sequence and the pulse emission time sequence to obtain target point cloud data for realizing distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

[0017] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor; wherein the memory stores executable code, and when the executable code is executed by the processor, the processor can at least implement the distance ambiguity elimination method as described in the first aspect.

[0018] In a fourth aspect, an embodiment of the present invention provides a non-temporary machine-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor can at least implement the distance ambiguity elimination method as described in the first aspect.

[0019] The technical solution provided by the embodiment of the present invention first obtains the point cloud data to be processed, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments, then determines the pulse repetition frequency corresponding to the point cloud data, and determines the pulse emission moment sequence corresponding to the point cloud data according to the pulse repetition frequency. If the plurality of echo information points are non-isolated data, then determines the mapping sequence of the plurality of echo information points and the pulse emission interval, and the pulse emission interval is determined based on the pulse repetition frequency. Finally, according to the mapping sequence and the pulse emission moment sequence, the first mapping relationship is adjusted to obtain the target point cloud data for realizing distance detection, wherein the target point cloud data includes a second mapping relationship between a plurality of echo information points and a plurality of actual emission moments, and the first mapping relationship is different from the second mapping relationship.

[0020] In the above scheme, through the mapping sequence of multiple echo information points and pulse emission intervals, the pulse emission intervals in which the multiple echo information points are located can be determined respectively, and then combined with the complete pulse emission time sequence, the actual pulse emission time corresponding to the multiple echo information points can be determined, and then the first mapping relationship between the multiple echo information points in the point cloud data and the multiple original emission time is adjusted to establish a correct mapping relationship between the echo pulse and the pulse emission time, so as to correct the erroneous position of the point cloud data with distance ambiguity, so as to obtain the target point cloud data for distance detection, and distance measurement based on the target point cloud data can eliminate the distance ambiguity problem in the point cloud data. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 A flowchart of a laser radar distance ambiguity elimination method provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a flow chart of determining a pulse emission time sequence corresponding to point cloud data according to a pulse repetition frequency provided in an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of a process for determining a mapping sequence between multiple echo information points and pulse emission intervals provided by an embodiment of the present invention;

[0025] Figure 4A schematic diagram of a process for determining a first sub-mapping sequence between a plurality of echo information points in a first scan line and a pulse emission interval provided by an embodiment of the present invention;

[0026] Figure 5 A schematic diagram of a flow chart of determining a second sub-mapping sequence between a plurality of echo information points outside a first scanning line and a pulse emission interval provided by an embodiment of the present invention;

[0027] Figure 6 A schematic diagram of a process for adjusting the first mapping relationship to obtain target point cloud data for distance detection provided by an embodiment of the present invention;

[0028] Figure 7 A schematic diagram of a flow chart of another distance ambiguity elimination method provided by an embodiment of the present invention;

[0029] Figure 8 A schematic diagram of the structure of a laser radar distance blur device provided by an embodiment of the present invention;

[0030] Fig. 9 For Figure 8 A schematic diagram of the structure of an electronic device corresponding to the laser radar range blurring device provided in the illustrated embodiment. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0033] Definition of terms:

[0034] Pulse repetition frequency: The number of pulses emitted by a laser per unit time.

[0035] Distance ambiguity: In laser pulse radar, when the distance between the target object and the laser radar system is greater than the maximum distance corresponding to the pulse repetition period, the target echo does not fall within this period, making the measured target distance not the real distance.

[0036] In order to facilitate those skilled in the art to understand the technical solutions provided by the embodiments of the present invention, the following is a brief description of the relevant technologies: the laser radar system integrates multiple units such as a laser scanner, an inertial measurement unit, a global positioning system, and a high-speed storage device. Among them, the laser scanner instructs the laser pulse transmitter to emit a laser pulse at an inherent angle. After the laser pulse contacts the target surface, it is reflected and received by the laser pulse receiver. The distance between the transmitting target and the laser scanner is measured by the difference between the laser pulse emission time and the laser pulse reception time, so as to achieve spatial sampling of the measurement target, and the sampling results are stored in the data under its own coordinate system.

[0037] With the rapid development of airborne lidar hardware systems, the pulse repetition frequency of the laser is continuously improved to achieve higher density spatial sampling. At the same time, the reflected laser pulse will fall into the pulse emission interval of the next laser emission, causing the pulse emission interval to which the laser emission moment of the recorded echo pulse belongs to deviate, which inevitably brings about the problem of distance ambiguity. In order to eliminate the distance ambiguity problem and ensure the accuracy of target distance measurement, the following implementation methods are provided in the prior art: by using a staggered emission mode to expand the distance difference of the distance ambiguity point cloud. Or by adding a micro-sequence tag to each laser pulse, the correct emission pulse emission interval is directly matched when the pulse echo is received. Or through global DEM data, additional coordinate information of the scanning area is obtained as data for comparison to determine the correct emission pulse emission interval corresponding to the pulse echo.

[0038] The above two implementation methods both rely on the LiDAR hardware system. If the LiDAR system is adaptively modified or special hardware modules are added, the point cloud data collected by different hardware systems will not be universal, and the cost of the LiDAR system will also increase. The third implementation method not only relies on additional prior information, but also requires DEM data to have sufficient accuracy. If the resolution of the DEM data is too low and the elevation details of the grid are missing, the reliability of the collision between the virtual echo and the prior digital model will be reduced.

[0039] In order to solve the technical problem of distance ambiguity caused by the increase of pulse repetition frequency, this embodiment provides a distance ambiguity elimination method. This technical solution determines the pulse emission interval where multiple echo information points are located through a mapping sequence of multiple echo information points and pulse emission intervals, and determines the actual pulse emission time corresponding to multiple echo information points in combination with the pulse emission time sequence, so as to establish a correct mapping relationship between the echo pulse and the pulse emission time, thereby correcting the wrong position of the distance ambiguous point cloud data to eliminate the distance ambiguity problem existing in the point cloud data. That is, the technical solution provided by the embodiment of the present invention no longer relies on the hardware system and prior information, and can directly achieve ambiguity elimination based on the acquired point cloud data.

[0040] Figure 1 A flowchart of a laser radar distance ambiguity elimination method provided by an embodiment of the present invention, such as Figure 1 As shown, the execution subject of the method may be a distance ambiguity elimination device. Specifically, the distance elimination method includes the following steps:

[0041] Step 101: Acquire point cloud data to be processed, where the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments.

[0042] Step 102: Determine the pulse repetition frequency corresponding to the point cloud data.

[0043] Step 103: Determine the pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency.

[0044] Step 104: If the multiple echo information points are non-isolated data, a mapping sequence between the multiple echo information points and the pulse emission interval is determined, and the pulse emission interval is determined based on the pulse repetition frequency.

[0045] Step 105: Adjust the first mapping relationship according to the mapping sequence and the pulse emission time sequence to obtain target point cloud data for distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

[0046] The laser radar system realizes spatial sampling of the measurement target by emitting and receiving laser pulses, and stores the sampling results in the data of the laser scanner's own coordinate system (SOCS for short). In the embodiment of the present invention, the point cloud data under the laser scanner is used as the processing object, the point cloud data is processed, and the coordinate correction of the echo information point can be completed under SOCS after the processing is completed.

[0047] The point cloud data is a massive point set used to represent the surface characteristics of the target, and can reflect the real situation of the surface with high accuracy. In the embodiment of the present invention, the point cloud data includes various information corresponding to the multiple echo information points, which may specifically include: a first mapping relationship between the multiple echo information points and the multiple original emission times, a scanning angle corresponding to each of the multiple echo information points, a measurement distance corresponding to each of the multiple echo information points, and the like.

[0048] The echo information point is used to represent the information corresponding to the echo pulse. The echo pulse is generated by the emission pulse after it contacts the surface of the target object. Assuming that the emission time interval between the two emission pulses is short, the first echo pulse signal is generated by the first emission pulse being emitted by the target object, but the echo pulse falls within the pulse emission time interval of the second emission pulse. The laser radar system will assume that this echo pulse signal is generated by the reflection of the second emission pulse, so that the emission time corresponding to the echo information point collected by the laser radar system is wrong, resulting in a distance ambiguity problem. Based on this, the technical solution provided by the embodiment of the present invention is mainly to determine the correct emission time corresponding to each echo information point in the point cloud data, adjust the first mapping relationship between multiple echo information points in the point cloud data and the original pulse emission time, so as to correct the wrong pulse emission time and obtain the target point cloud data for distance detection.

[0049] Specifically, firstly, the point cloud data to be processed is obtained, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original emission times. It should be noted here that if the laser pulse repetition frequency is large, so that the echo information point falls within the subsequent laser pulse emission time interval, the mapping relationship between the echo information point and the original emission time recorded by the system will be wrong. Then, the pulse repetition frequency corresponding to the point cloud data is determined.

[0050] There is no limitation on the method of determining the pulse repetition frequency. The laser pulse repetition frequency corresponding to the point cloud data can be determined by directly obtaining the laser pulse repetition frequency set by the laser radar system. The pulse repetition frequency can be set according to the actual application needs. When measuring the distance of the target, a laser pulse repetition frequency will be set in advance. The distance ambiguity elimination device can directly obtain the set laser pulse repetition frequency, thereby determining the pulse repetition frequency corresponding to the point cloud data. The pulse repetition frequency corresponding to the point cloud data can also be determined according to the number of laser pulses emitted by the laser transmitter per unit time.

[0051] After determining the pulse repetition frequency corresponding to the point cloud data, the pulse emission time sequence corresponding to the point cloud data can be determined according to the pulse repetition frequency. Next, the data types of multiple echo information points are detected. The noise energy of the echo information points with distance ambiguity that may be caused by isolated data when they appear in the wrong position is smaller. By judging the data types corresponding to multiple echo information points, the influence of isolated data on the processing effect of point cloud data can be avoided. Optionally, the specific implementation method of detecting the data type of multiple echo information points may include: determining the spatial continuity corresponding to each echo information point in the point cloud data, and determining whether each echo information point is non-isolated data according to the spatial continuity.

[0052] The spatial continuity degree corresponding to the echo information point can be compared with a first preset value. If the spatial continuity degree corresponding to the echo information point is greater than or equal to the first preset value, the echo information point is determined to be non-isolated data; or, if the spatial continuity degree corresponding to the echo information point is less than the first preset value, the echo information point is determined to be isolated data. The first preset value here can be set according to actual conditions, for example, the first preset value can be set to 0.8.

[0053] Since most of the target objects are continuous, the echo information points with high spatial continuity are determined as objects, and the echo information points with low spatial continuity are determined as cables. In the embodiment of the present invention, different methods are used to process the objects and cables respectively to achieve accurate measurement of the target distance. If multiple echo information points are non-isolated data, a mapping sequence of multiple echo information points and pulse emission intervals is determined, wherein the pulse emission interval is determined based on the pulse repetition frequency. There may be deviations in the original pulse emission time corresponding to the echo information points in the collected point cloud data. The deviation in the first mapping relationship between the collected multiple echo information points and the multiple original emission times is mainly due to the fact that the echo information points may fall within the subsequent pulse emission interval, resulting in a deviation in the pulse emission interval to which the original pulse emission time recorded by the echo information points belongs. Therefore, the correct pulse emission time corresponding to each echo information point can be determined by determining the pulse emission interval in which each echo information point is located.

[0054] Specifically, by determining the pulse emission interval in which each echo information point is located, the correct pulse emission time corresponding to each echo information point can be determined as follows: based on the mapping sequence of multiple echo information points and pulse emission intervals and the pulse emission time sequence, the correct pulse emission time corresponding to each echo information point can be determined.

[0055] After determining the correct pulse emission time corresponding to each echo information point, the first mapping relationship needs to be adjusted so that the user can achieve accurate distance measurement based on the adjusted target point cloud data. The specific implementation method can be: according to the mapping sequence and the pulse emission time sequence, the first mapping relationship is adjusted to obtain the target point cloud data for distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

[0056] The embodiment of the present invention can respectively determine the pulse emission intervals where the multiple echo information points are located through the mapping sequence of multiple echo information points and pulse emission intervals, and then determine the actual pulse emission moments corresponding to the multiple echo information points in combination with the complete pulse emission time sequence, and then adjust the first mapping relationship between the multiple echo information points in the point cloud data and the multiple original emission moments to establish the correct mapping relationship between the echo pulse and the pulse emission moment, so as to correct the erroneous position of the point cloud data with distance ambiguity, so as to obtain the target point cloud data for distance detection, and the distance measurement based on the target point cloud data can eliminate the distance ambiguity problem in the point cloud data. That is, the essence of solving the distance ambiguity is to determine the correct pulse emission moment corresponding to each echo information point. This implementation process does not need to rely on a redesigned hardware system, does not need to obtain a priori terrain data, and can eliminate the distance ambiguity directly based on the collected point cloud data.

[0057] Figure 2 A flow chart of determining a pulse emission time sequence corresponding to point cloud data according to a pulse repetition frequency is provided in an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 2 As shown, this embodiment provides an implementation method for determining a pulse emission time sequence corresponding to point cloud data, which may specifically include:

[0058] Step 201: Determine the original pulse emission time sequence based on the point cloud data.

[0059] Step 202: Determine the pulse time interval according to the pulse repetition frequency.

[0060] Step 203: Determine the pulse emission time sequence corresponding to the point cloud data according to the original pulse emission time sequence and the pulse time interval.

[0061] In the point cloud data to be processed, each echo information point corresponds to an original pulse emission time. The original pulse emission time corresponding to each echo information point in the point cloud data is extracted, and the original pulse emission time sequence corresponding to the point cloud data can be determined based on each original pulse emission time.

[0062] Since the storage of the LiDAR system is limited, data is generally stored selectively. For example, the limited carrying capacity of the UAV carrier makes it impossible for the UAV LiDAR system to carry a large-scale high-speed storage device, resulting in the abandonment of data storage. If no echo pulse is received within a pulse emission interval, the emission time of the pulse will not be recorded in the point cloud data, resulting in the loss of time tags. For echo information point data without distance ambiguity, this storage strategy will not affect the data, but for echo information points with distance ambiguity, it cannot be ruled out that the actual pulse emission time corresponding to the target echo information point with distance ambiguity is not recorded.

[0063] In order to avoid the problem of unrecorded pulse emission time in point cloud data, after determining the original pulse emission time sequence corresponding to the point cloud data, the pulse time interval is determined according to the pulse repetition frequency, and then the pulse emission time sequence corresponding to the point cloud data is determined according to the original pulse emission time sequence and the pulse time interval. The pulse emission time sequence determined at this time is a complete pulse emission time, which can avoid the existence of unrecorded pulse emission time in the original pulse emission time sequence, so that each echo information point with distance ambiguity has the possibility of being corrected, making the processed data more accurate, thereby improving the effect of data processing.

[0064] In an optional embodiment, a pulse emission time sequence corresponding to the point cloud data is determined based on the original pulse emission time sequence and the pulse time interval. The specific implementation method may include: determining the time difference between adjacent original pulse emission times; if the time difference is a multiple of the pulse time interval, determining that there are unrecorded pulse emission times in the original pulse emission time sequence, and determining the pulse emission time sequence corresponding to the point cloud data based on the original pulse emission time sequence and the unrecorded pulse emission time; if the time difference is the pulse time interval, determining the original pulse emission time sequence as the pulse emission time sequence corresponding to the point cloud data.

[0065] After obtaining the original pulse emission time corresponding to each echo information point, the original pulse emission time difference between adjacent echo information points is calculated. If the calculated time difference is a multiple of the pulse time interval, it is determined that there is an unrecorded pulse emission time in the original pulse emission time sequence, and the unrecorded pulse emission time needs to be added to the original pulse emission time sequence to obtain the pulse emission time sequence corresponding to the point cloud data. The multiple of the pulse time interval here can be an integer multiple of the pulse time interval or a non-integer multiple of the pulse time interval, that is, the sum of multiple time intervals. In an optional embodiment, a special mark can be used to mark the unrecorded pulse emission time to ensure the strict correctness of the distance calculation.

[0066] If the time difference is the pulse time interval, the original pulse emission time sequence is determined as the pulse emission time sequence corresponding to the point cloud data. When the calculated time difference is exactly the pulse time interval, it means that there is no unrecorded pulse emission time between the two adjacent original pulse emission times used to calculate the time difference. In addition, the time difference may be 0, which means that the two adjacent echo information points are in the same pulse time interval. In this case, there is no unrecorded pulse emission time, and the original pulse emission time sequence is directly determined as the pulse emission time sequence corresponding to the point cloud data.

[0067] In an embodiment of the present invention, the pulse repetition frequency is used to determine the pulse time interval, and the pulse emission time sequence corresponding to the point cloud data is determined based on the original pulse emission time sequence and the pulse time interval. This can avoid the situation where the laser radar system loses the pulse emission time due to the storage strategy, thereby ensuring the accuracy of the processing results.

[0068] Figure 3 A flowchart of determining a mapping sequence of multiple echo information points and pulse transmission intervals provided by an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 3 As shown, in this embodiment, the point cloud data includes a scanning angle, and provides an implementation method for determining a mapping sequence of multiple echo information points and pulse emission intervals, which may specifically include:

[0069] Step 301 : segment the point cloud data according to the scanning angle in the point cloud data, and determine the scanning lines corresponding to each echo information point in the point cloud data, wherein the scanning lines include the first scanning line.

[0070] Step 302: Determine a target data point in the first scan line corresponding to a scan angle, where the scan angle is 0° or close to 0°.

[0071] Step 303: Determine an initial mapping sequence between target data points and pulse emission intervals according to the pulse repetition frequency corresponding to the point cloud data.

[0072] Step 304: Using a dynamic programming algorithm, backtracking is performed starting from the initial mapping sequence to determine a first sub-mapping sequence between a plurality of echo information points in the first scanning line and the pulse emission interval.

[0073] Step 305: Determine a second sub-mapping sequence between a plurality of echo information points outside the first scanning line and the pulse emission interval according to the first sub-mapping sequence and the dynamic programming algorithm.

[0074] Step 306: Determine a mapping sequence of multiple echo information points and pulse transmission intervals based on the first sub-mapping sequence and the second sub-mapping sequence.

[0075] When determining the mapping sequence between each echo information point and the pulse emission interval in the point cloud data, in order to process each echo information point more quickly, each echo information point on the scan line can be processed according to the scan line. Specifically, the point cloud data is segmented according to the scan angle in the point cloud data to determine the scan line corresponding to each echo information point in the point cloud data, and the scan line includes the first scan line.

[0076] In the embodiment of the present invention, it is no longer necessary to use the prior information global DEM data, and each echo information point in the first scan line is used as a ground object reference. Therefore, it is necessary to ensure that the first sub-mapping sequence between each echo information point in the first scan line and the pulse emission interval is accurate. In order to ensure that the first sub-mapping sequence in the first scan line is accurate, the first scan line is processed separately, and the target data point corresponding to the scanning angle in the first scan line is first determined, where the scanning angle is 0° or close to 0°, and then the initial mapping sequence between the target data point and the pulse emission interval is determined according to the pulse repetition frequency corresponding to the point cloud data, and then the dynamic programming algorithm is used to backtrack with the initial mapping sequence as the starting point to determine the first sub-mapping sequence between multiple echo information points in the first scan line and the pulse emission interval.

[0077] In an optional embodiment, determining the initial mapping sequence between the target data point and the pulse emission interval based on the pulse repetition frequency corresponding to the point cloud data may specifically include: obtaining the flight altitude of the lidar system above the ground and the pulse repetition frequency of the lidar system, calculating the correct pulse emission interval where the target data point is located, and determining the initial mapping sequence between the target data point and the pulse emission interval based on the calculated correct pulse emission interval.

[0078] Specifically, assuming that the laser radar system is at a flight height of H above the ground, the pulse repetition frequency of the laser radar system is F, and the propagation speed of the laser pulse is c, obtain the product of the flight height and the pulse repetition frequency H·F, obtain the product of the flight height and the pulse repetition frequency and the inverse of the propagation speed H·F / c, and round up the value of 2 times the product to determine the correct pulse time interval where the target data point is located, that is, the correct pulse time interval where the target data point is located is

[0079] Since the original pulse emission time corresponding to the multiple echo information points in the acquired point cloud data is not the actual pulse emission time corresponding to the echo information point, the echo information point may be located in the subsequent pulse emission time interval. In order to accurately determine the pulse emission time interval where each echo information point is located, a dynamic programming algorithm can be used to obtain the optimal solution for the pulse emission time interval where each echo information point is located, so that the pulse emission time interval determined is more accurate. Therefore, after determining the initial mapping sequence between the target data point and the pulse emission interval, the dynamic programming algorithm can be used to backtrack from the initial mapping sequence as the starting point to determine the first sub-mapping sequence between the multiple echo information points in the first scan line and the pulse emission interval.

[0080] The energy of the laser pulse signal decays as the propagation distance in the medium increases. The maximum pulse emission interval in the laser radar system is set to max, that is, the laser radar can receive echo information that is at most beyond max-1 pulse emission intervals and less than max+1 pulse emission intervals. In an embodiment of the present invention, the pulse emission interval in which each echo information point is located is expanded to the point cloud space under 1-max pulse emission intervals, and the correct pulse emission interval in which each echo information point is located is determined. Since all echo information points, except for the first echo information point as the starting point of the path calculation, meet the minimum noise energy of the correct path in different pulse emission intervals under the 1-max pulse emission interval. Then the problem of solving the echo information point with ambiguous distance can be abstracted as a combinatorial problem of selecting pulse emission intervals.

[0081] The dynamic programming algorithm can be used to solve the selection and combination problem of the pulse emission intervals in which each echo information point is located. The dynamic programming algorithm is applied to each scan line, and each echo information point is processed separately according to the order recorded in the point cloud data. Since the lowest noise energy of each echo information point in 1-max different pulse emission intervals will be used multiple times when processing the next echo information point of the echo information point, the selection path of the lowest noise energy of any echo information point in 1-max different pulse emission intervals must be determined by the minimum value of the selection path of the lowest noise energy of the previous echo information point in each different pulse emission interval and the sum of the noise energies between the two echo information points. Based on this, the dynamic programming algorithm is used to backtrack with the initial mapping sequence as the starting point to determine the multiple echo information points in the first scan line.

[0082] The first sub-mapping sequence between the echo information points and the pulse transmission interval.

[0083] Next, according to the first sub-mapping sequence and the dynamic programming algorithm, a second sub-mapping sequence between multiple echo information points outside the first scan line and the pulse emission interval is determined. When determining each scan line outside the first scan line, in addition to the previous echo information point, each echo information point in the previous scan line is used as a reference to determine the second sub-mapping sequence between multiple echo information points outside the first scan line and the pulse emission interval. Finally, based on the first sub-mapping sequence and the second sub-mapping sequence, a mapping sequence between multiple echo information points and the pulse emission interval is determined. It should be noted here that, assuming that the sub-mapping sequence corresponding to the second scan line needs to be determined at this time, it needs to be based on the first scan line as a terrain reference, and after determining the sub-mapping sequence corresponding to the second scan line, the second scan line needs to be updated to the latest terrain reference so that the third scan line can determine its corresponding sub-mapping sequence based on the latest terrain reference.

[0084] In an embodiment of the present invention, the point cloud data is segmented and processed according to the scanning angle in the point cloud data, the scanning line corresponding to each echo information point in the point cloud data is determined, and the echo information points on each scanning line are processed respectively. First, a first sub-mapping sequence between multiple echo information points in the first scanning line and the pulse emission interval is determined, and then a second sub-mapping sequence between multiple echo information points outside the first scanning line and the pulse emission interval is determined based on the first sub-mapping sequence and the dynamic programming algorithm. Finally, based on the first sub-mapping sequence and the second sub-mapping sequence, a mapping sequence between multiple echo information points and the pulse emission interval is determined. The mapping sequence obtained in this way is more accurate and more in line with the actual situation, and also provides a basis for the subsequent accurate elimination of distance ambiguity.

[0085] Figure 4 A schematic diagram of a flow chart of determining a first sub-mapping sequence between a plurality of echo information points in a first scanning line and a pulse emission interval provided in an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 4 As shown, in this embodiment, the point cloud data includes a measured distance, and a method for implementing a first sub-mapping sequence between a plurality of echo information points in the first scanning line and the pulse emission interval by backtracking using a dynamic programming algorithm as a starting point is provided, which may specifically include:

[0086] Step 401: Determine the current pulse emission time corresponding to each of the plurality of echo information points in the first scanning line in the pulse emission time sequence.

[0087] Step 402: Acquire multiple historical pulse emission times before the current pulse emission time.

[0088] Step 403: Determine two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to the multiple echo information points, the current pulse emission time, and the multiple historical pulse emission times.

[0089] Step 404: Determine two-dimensional noise energy information corresponding to a plurality of echo information points according to the two-dimensional distance information.

[0090] Step 405: Determine two-dimensional pointer information corresponding to a plurality of echo information points according to the two-dimensional noise energy information.

[0091] Step 406: Detect whether the multiple historical pulse emission times include unrecorded pulse emission times.

[0092] Step 407: if the unrecorded pulse emission time is not included, backtracking is performed based on the two-dimensional pointer information and the two-dimensional noise energy information, starting from the initial mapping sequence, to determine the first sub-mapping sequence between the plurality of echo information points in the first scan line and the pulse emission interval.

[0093] Step 408: If it includes unrecorded pulse emission moments, the column information corresponding to the echo information points in the two-dimensional distance information, the column information corresponding to the echo information points in the two-dimensional noise energy information, and the column information corresponding to the echo information points in the two-dimensional pointer information are all set to the second preset value.

[0094] When using a dynamic programming algorithm to determine a first sub-mapping sequence between a plurality of echo information points and pulse emission intervals in a first scan line, it is first necessary to determine the two-dimensional distance information, two-dimensional noise energy information, and two-dimensional pointer information corresponding to the plurality of echo information points used by the dynamic programming algorithm for backtracking. The two-dimensional distance information includes the distance information of each echo information point in the first scan line in each pulse emission interval, the two-dimensional noise energy information includes the noise energy of each echo information point in the first scan line in each pulse emission interval, and the two-dimensional pointer information includes the minimum noise energy of each echo information point in the first scan line in each pulse emission interval inherited from the pulse emission interval corresponding to the noise energy of the previous echo information point.

[0095] Before determining the two-dimensional distance information, two-dimensional noise energy information, and two-dimensional pointer information corresponding to the multiple echo information points, the double cursor method is first used to determine the current pulse emission time corresponding to the multiple echo information points in the first scanning line in the pulse emission time sequence, and obtain multiple historical pulse emission times before the current pulse emission time. Then, based on the measured distances corresponding to the multiple echo information points, the current pulse emission time, and the historical pulse emission time, the two-dimensional distance information corresponding to the multiple echo information points is determined.

[0096] Specifically, in an optional embodiment, the two-dimensional distance information corresponding to the multiple echo information points can be determined by: determining the distance information corresponding to each of the multiple echo information points in each pulse emission interval according to the measured distances corresponding to each of the multiple echo information points, the current pulse emission time, and the multiple historical pulse emission times. Determine the two-dimensional distance information corresponding to the multiple echo information points according to the distance information corresponding to each of the multiple echo information points.

[0097] The measured distance, current pulse emission time and multiple historical pulse emission time corresponding to each echo information point are obtained, and then the distance information corresponding to each of the multiple echo information points is determined in each pulse emission interval according to the obtained measured distance, current pulse emission time and multiple historical pulse emission time. Specifically, the pulse emission time in the interval corresponding to the multiple echo information points in the pulse emission interval is obtained, and the time difference between the current pulse emission time and the pulse emission time in the interval is obtained. According to the time difference and the laser propagation speed, the spatial distance corresponding to each of the multiple echo information points is determined. The distance sum value between the measured distance and the spatial distance corresponding to the echo information point is obtained. The distance sum value is determined as the distance information corresponding to each of the multiple echo information points in each pulse emission interval.

[0098] Next, two-dimensional distance information corresponding to the multiple echo information points is determined based on the distance information corresponding to each of the multiple echo information points. Next, two-dimensional noise energy information corresponding to the multiple echo information points is determined based on the two-dimensional distance information.

[0099] In an optional embodiment, the specific implementation method of determining the two-dimensional noise energy information corresponding to the multiple echo information points according to the two-dimensional distance information can be: obtaining the data relationship between the multiple echo information points and the target data point. If the multiple echo information points are data obtained before the target data point, then according to the two-dimensional distance information, starting from the first echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset ascending order. If the echo information point is data obtained after the target echo information point, then according to the two-dimensional distance information, starting from the last echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset descending order.

[0100] When determining the first mapping relationship between each echo information point and the pulse emission interval, backtracking is performed with the target data point as the starting point. Then, when determining the noise energy value of each echo information point on the first scanning line in each pulse emission interval, it is also necessary to use the target data point as the boundary to determine the noise energy value corresponding to the echo information points before and after the target data point. Therefore, when determining the noise energy corresponding to each echo information point in each pulse emission interval, first obtain the data relationship between multiple echo information points and the target data point. If the multiple echo information points are data obtained before the target data point, then according to the two-dimensional distance information, starting with the first echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset increasing order. If the echo information point is data obtained after the target echo information point, then according to the two-dimensional distance information, starting with the last echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset decreasing order.

[0101] Specifically, according to the two-dimensional distance information, starting from the first echo information point, determining the two-dimensional noise energy information corresponding to each of the multiple echo information points in a preset ascending order may include: setting the column information corresponding to the first echo information point in the two-dimensional noise energy information to an initial state. Starting from the first echo information point, in an ascending order, sequentially obtaining the distance values ​​corresponding to the echo information points at each position in the two-dimensional distance information and the previous distance value corresponding to the previous echo information point. Obtaining the previous noise energy value corresponding to the previous echo information point. According to the distance values ​​corresponding to each echo information point, the previous distance value and the previous noise energy value, determining the noise energy values ​​of the multiple echo information points at each position. According to the noise energy values ​​of the multiple echo information points at each position, determining the two-dimensional noise energy information corresponding to the multiple echo information points.

[0102] If multiple echo information points are data obtained before the target data point, each echo information point is determined in order starting from the first echo information point. That is, the noise energy values ​​of multiple echo information points at each position are determined according to the distance value, the previous distance value and the previous noise energy value corresponding to each echo information point. The noise energy value of the echo information point at each position is determined based on the lowest noise energy value of the previous echo information point at each position. In this way, when the path is traced back, it can be ensured that the lowest noise energy of any echo information point in any pulse emission interval is calculated by the lowest noise energy value in each pulse emission interval of the previous echo information point, so that the pulse emission interval where each echo information point is located is more accurate.

[0103] Specifically, according to the two-dimensional distance information, starting from the last echo information point, determining the two-dimensional noise energy information corresponding to each of the multiple echo information points in a preset descending order may include: setting the column information corresponding to the last echo information point in the two-dimensional noise energy information to an initial state. Starting from the last echo information point, in a descending order, sequentially obtaining the distance values ​​corresponding to the echo information points at each position in the two-dimensional distance information and the next distance value corresponding to the next echo information point. Obtaining the next noise energy value corresponding to the next echo information point. According to the distance values, the next distance value and the next noise energy value corresponding to each echo information point, determining the noise energy values ​​of the multiple echo information points at each position. According to the noise energy values ​​of the multiple echo information points at each position, determining the two-dimensional noise energy information corresponding to the multiple echo information points.

[0104] The noise energy values ​​of the multiple echo information points at various positions (in the pulse emission interval) determined based on the above method are all determined based on the next echo information point, and the noise energy value of the echo information point at various positions is determined based on the lowest noise energy value of the next echo information point at various positions. In this way, when the path is traced back from the target data point in a preset increasing order, it can be ensured that the lowest noise energy of any echo information point in any pulse emission interval is calculated by the lowest noise energy value in each pulse emission interval of the next echo information point, so that the pulse emission interval where each echo information point is located is more accurately determined.

[0105] Then, according to the two-dimensional noise energy information, the two-dimensional pointer information corresponding to the multiple echo information points is determined. Specifically, according to the noise energy value of the position of each echo information point in the two-dimensional noise energy information, the pointer value corresponding to each echo information point is determined, and according to the pointer value corresponding to each echo information point, the two-dimensional pointer information corresponding to each echo information point is determined. Among them, since the noise energy value is determined in sequence from the first echo information point and the last echo information point respectively, the determination of the pointer value is also determined in sequence from the first echo information point and the last echo information point, and the pointer value of the echo information point at each position records the noise energy value of this echo information point at this position inherited from the position of the previous echo information point or the next echo information point.

[0106] After determining the two-dimensional noise energy information and two-dimensional pointer information corresponding to each echo information point on the first scan line, it is detected whether the multiple historical pulse emission moments include the unrecorded pulse emission moment. In the process of processing, it is impossible to determine whether the actual pulse emission moment corresponding to the echo information point with ambiguous distance is the unrecorded pulse emission moment. In order to achieve better processing effect, the unrecorded pulse emission moment with special marks encountered in the process of processing is not processed and is set to a determined value. Specifically, if the unrecorded pulse emission moment is not included in the process of processing, the first sub-mapping sequence between the multiple echo information points in the first scan line and the pulse emission interval is determined based on the two-dimensional pointer information and the two-dimensional noise energy information and backtracked from the initial mapping sequence as the starting point. If the unrecorded pulse emission moment is included, the column information corresponding to the echo information point in the two-dimensional distance information, the column information corresponding to the echo information point in the two-dimensional noise energy information, and the column information corresponding to the echo information point in the two-dimensional pointer information are directly set to the second preset value, wherein the second preset value can be set according to the design requirements or actual application requirements, for example, the second preset value is set to -1. Then, during the processing, the columns corresponding to the unrecorded pulse emission moments are skipped, and based on the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed starting from the initial mapping sequence to determine the first sub-mapping sequence between the multiple echo information points in the first scan line and the pulse emission interval.

[0107] In an optional embodiment, based on the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed from the initial mapping sequence as the starting point to determine the first sub-mapping sequence between the multiple echo information points and the pulse emission interval in the first scan line, which may specifically include: obtaining the data relationship between the multiple echo information points and the target data point. If the multiple echo information points are data obtained before the target data point, backtracking is performed in descending order from the initial mapping sequence according to the two-dimensional pointer information and the two-dimensional noise energy information to determine the mapping sequence corresponding to the multiple echo information points. If the multiple echo information points are data obtained after the target data point, backtracking is performed in ascending order from the initial mapping sequence according to the two-dimensional pointer information and the two-dimensional noise energy information to determine the mapping sequence corresponding to the multiple data points.

[0108] Since the two-dimensional noise energy information of the data obtained before the target data point starts from the first echo information point, and the noise energy corresponding to each echo information point is determined in ascending order, when the optimal path is backtracked, it is necessary to start from the initial mapping sequence and backtrack in descending order to determine the mapping sequence corresponding to multiple echo information points. Similarly, since the two-dimensional noise energy information of the data obtained after the target data point starts from the last echo information point, and the noise energy corresponding to each echo information point is determined in descending order, when the optimal path is backtracked, it is necessary to start from the initial mapping sequence and backtrack in ascending order to determine the mapping sequence corresponding to multiple echo information points. The pulse emission interval corresponding to each echo information point in the mapping sequence corresponding to multiple echo information points determined based on the above method is the optimal solution, that is, the pulse emission interval where each echo information point is located is more accurate.

[0109] In the embodiment of the present invention, two-dimensional distance information corresponding to multiple echo information points is determined by the measured distances corresponding to the multiple echo information points, the current pulse emission time, and the multiple historical pulse emission times. According to the two-dimensional distance information, the two-dimensional noise energy information corresponding to the multiple echo information points is determined. According to the two-dimensional noise energy information, the two-dimensional pointer information corresponding to the multiple echo information points is determined. And according to the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed with the initial mapping sequence as the starting point to determine the first sub-mapping sequence between the multiple echo information points in the first scan line and the pulse emission interval. The first sub-mapping sequence determined in this way is more in line with the actual situation and more accurate, thereby improving the processing effect of the target point cloud data.

[0110] Figure 5 A schematic diagram of a flow chart of determining a second sub-mapping sequence between a plurality of echo information points outside the first scanning line and a pulse emission interval provided in an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 5 As shown, in this embodiment, the point cloud data includes the measured distance, and an implementation method for determining the second sub-mapping sequence between the plurality of echo information points outside the first scanning line and the pulse emission interval according to the first sub-mapping sequence and the dynamic programming algorithm is provided, which may specifically include:

[0111] Step 501: Determine the current pulse emission time corresponding to each of the plurality of echo information points outside the first scanning line in the pulse emission time sequence.

[0112] Step 502: Acquire multiple historical pulse emission times before the current pulse emission time.

[0113] Step 503: Determine the two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to the multiple echo information points, the current pulse emission time, and the multiple historical pulse emission times.

[0114] Step 504: Determine two-dimensional noise energy information corresponding to a plurality of echo information points according to the two-dimensional distance information and the first sub-mapping sequence.

[0115] Step 505: Determine two-dimensional pointer information corresponding to a plurality of echo information points according to the two-dimensional noise energy information.

[0116] Step 506: Detect whether the multiple historical pulse emission moments include unrecorded pulse emission.

[0117] Step 507: If the unrecorded pulse emission time is not included, determine the target mapping sequence between the last echo information point and the pulse emission interval, and based on the two-dimensional pointer information and the two-dimensional noise energy information, backtrack from the target mapping sequence as the starting point to determine the second sub-mapping sequence between the multiple echo information points outside the first scanning line and the pulse emission interval.

[0118] Step 508: If it includes unrecorded pulse emission moments, the column information corresponding to the echo information points in the two-dimensional distance information, the column information corresponding to the echo information points in the two-dimensional noise energy information, and the column information corresponding to the echo information points in the two-dimensional pointer information are all set to the second preset value.

[0119] When determining the second mapping sequence between multiple echo information points outside the first scanning line and the pulse emission interval, in addition to determining the information corresponding to the next echo information point in sequence based on the information corresponding to the previous echo information point, it is also necessary to refer to the information of each echo information point on the previous scanning line, so that the determined second mapping sequence is more accurate.

[0120] Specifically, firstly, the current pulse emission time corresponding to each of the multiple echo information points outside the first scan line is determined in the pulse emission time sequence. And multiple historical pulse emission times located before the current pulse emission time are obtained. Then, according to the measured distances corresponding to each of the multiple echo information points, the current pulse emission time and the multiple historical pulse emission times, the two-dimensional distance information corresponding to the multiple echo information points is determined. Among them, the determination method of the two-dimensional distance information corresponding to the multiple echo information points outside the first scan line can be consistent with the determination method of the multiple echo information points in the first scan line. The implementation process of determining the two-dimensional distance information corresponding to the multiple echo information points can refer to the relevant description of steps 4031-step 4032 in the aforementioned embodiment, which will not be repeated here.

[0121] After determining the two-dimensional distance information corresponding to the plurality of echo information points, the two-dimensional noise energy information corresponding to the plurality of echo information points is determined according to the two-dimensional distance information and the first sub-mapping sequence. In an optional embodiment, determining the two-dimensional noise energy information corresponding to the plurality of echo information points according to the two-dimensional distance information and the first sub-mapping sequence may specifically include: determining the two-dimensional noise energy information corresponding to the plurality of echo information points in ascending order starting from the first echo information point according to the two-dimensional distance table and the mapping sequence of the plurality of echo information points and the pulse emission interval in the first scan line. It should be noted here that: if the point cloud data is divided into a plurality of scan lines, when determining the second scan line, the two-dimensional noise energy information corresponding to the plurality of echo information points is determined based on the two-dimensional distance information and the first sub-mapping sequence, and after determining the two-dimensional noise energy information corresponding to the second scan line, when determining the third scan line, it is necessary to determine the sub-mapping sequence and the two-dimensional distance information corresponding to the second scan line. That is, the two-dimensional noise energy information corresponding to each echo information point on each scan line is determined based on the two-dimensional distance information and the mapping sequence corresponding to the previous scan line, and the two-dimensional noise energy information corresponding to the plurality of echo information points is determined.

[0122] Among them, the method for determining the two-dimensional pointer information corresponding to the multiple echo information points outside the first scan line is consistent with the method for determining the multiple echo information points corresponding to the data obtained before the target in the first scan line. According to the noise energy value of the position of each echo information point in the two-dimensional noise energy information, the pointer value corresponding to each echo information point is determined, and according to the pointer value corresponding to each echo information point, the two-dimensional pointer information corresponding to each echo information point is determined. The pointer value of the echo information point at each position (in the pulse transmission interval) records the noise energy value of this echo information point at this position inherited from the position of the previous echo information point. The implementation process of determining the two-dimensional pointer information corresponding to multiple echo information points can refer to the relevant description of the steps in the aforementioned embodiment, which will not be repeated here.

[0123] Next, it is detected whether the multiple historical pulse emission moments include the unrecorded pulse emission moments, so that the influence of the unrecorded pulse emission moments on the processing effect can be reduced. When encountering the specially marked pulse emission moment, the column information corresponding to the echo information point in the two-dimensional distance information, the column information corresponding to the echo information point in the two-dimensional noise energy information, and the column information corresponding to the echo information point in the two-dimensional pointer information are directly set to the second preset value. If the unrecorded pulse emission moment is not included, the target mapping sequence between the last echo information point and the pulse emission interval is determined, and according to the two-dimensional pointer information and the two-dimensional noise energy information, the target mapping sequence is used as the starting point for backtracking to determine the second sub-mapping sequence between the multiple echo information points outside the first scanning line and the pulse emission interval. The mapping relationship between the echo information points and the pulse emission interval in the second sub-mapping sequence determined in this way is the optimal solution, which can more accurately determine the correct pulse emission interval where each echo information point is located.

[0124] In the embodiment of the present invention, the two-dimensional distance information corresponding to the multiple echo information points is determined by the measured distances, the current pulse emission time and the multiple historical pulse emission time corresponding to the multiple echo information points, and the noise energy information corresponding to the multiple echo information points is determined according to the two-dimensional distance information and the first sub-mapping sequence, and the two-dimensional pointer information corresponding to the multiple echo information points is determined according to the noise energy information. In addition, the target mapping sequence between the last echo information point and the pulse time interval is determined, and the second sub-mapping sequence between the multiple echo information points outside the first scanning line and the pulse emission interval is determined by backtracking based on the two-dimensional pointer information and the two-dimensional noise energy information, so that the second sub-mapping sequence determined in this way is more in line with the actual situation and more accurate, thereby improving the processing effect of the target point cloud data.

[0125] Figure 6 A schematic diagram of a process for adjusting the first mapping relationship to obtain target point cloud data for distance detection provided by an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 6 As shown, this embodiment provides a method for adjusting the first mapping relationship according to the mapping sequence and the pulse emission time sequence to obtain the target point cloud data for distance detection, which may specifically include:

[0126] Step 601: Determine actual pulse emission times corresponding to each of a plurality of echo information points according to a mapping sequence and a pulse emission time sequence.

[0127] Step 602: Update the first mapping relationship based on multiple echo information and the actual emission time corresponding to each echo information to obtain target point cloud data for distance detection.

[0128] The mapping sequence includes the pulse emission interval where each echo information point is located. Based on the pulse emission interval where each echo information point is located and the pulse emission time sequence, the actual pulse emission time corresponding to each echo information point is determined, and then the first mapping relationship is updated based on the actual pulse emission time to obtain the second mapping relationship, and the previous echo information point data is updated to obtain the target point cloud data. In this way, when the target distance is measured based on the target point cloud data, not only the distance ambiguity problem in the point cloud data can be eliminated, but also a more accurate distance measurement can be obtained.

[0129] Figure 7 A flow chart of another distance ambiguity elimination method provided by an embodiment of the present invention; based on the above embodiment, continue to refer to the attached Figure 7 As shown, in order to increase the applicability of the embodiment of the present invention, the method may further include:

[0130] Step 701: If the multiple echo information points are isolated data, determine the virtual echo positions corresponding to the multiple echo information points.

[0131] Step 702: Determine the ground object target points corresponding to each of the multiple echo information points.

[0132] Step 703: Adjust the first mapping relationship according to the ground object target point and the virtual echo position to obtain target point cloud data for distance detection.

[0133] In practical applications, the target object to be measured may be a cable. If the isolated data is not processed directly during the processing, the final data may have a large deviation. Therefore, in addition to performing distance ambiguity elimination processing on the non-isolated data, the isolated data is also processed separately in the embodiment of the present invention. When processing the isolated data, it is first necessary to determine the virtual echo position corresponding to each echo information point.

[0134] Specifically, if the plurality of echo information points are isolated data, then the virtual echo positions corresponding to the plurality of echo information points are determined. Determining the virtual echo positions corresponding to the plurality of echo information points includes:

[0135] Step 7011: Obtain the measurement distances corresponding to each of the multiple echo information points.

[0136] Step 7012: Obtain the current pulse emission time corresponding to each of the multiple echo information points.

[0137] Step 7013, obtaining the pulse emission time in the interval corresponding to multiple echo information points in the pulse emission interval.

[0138] Step 7014, obtain the time difference between the current pulse emission time and the pulse emission time within the interval.

[0139] Step 7015: Determine the spatial distances corresponding to the multiple echo information points based on the time difference and the laser propagation speed.

[0140] Step 7016: Obtain the distance and value between the measured distance and the spatial distance corresponding to the echo information point.

[0141] Step 7017: determine the distance and value as the virtual echo position corresponding to each of the multiple echo information points in each pulse transmission interval.

[0142] After determining the virtual echo positions corresponding to the multiple echo information points, the ground object target points corresponding to the multiple echo information points are then determined. The specific implementation method can be: obtain multiple adjacent echo information points adjacent to the multiple data points, the multiple adjacent echo information points are non-isolated data, and the multiple adjacent echo information points are echo information points obtained after the distance ambiguity elimination processing operation has been performed in the above process. Among the multiple adjacent echo information points, determine the target adjacent echo information point corresponding to the maximum measurement distance, and determine the target adjacent echo information points corresponding to the multiple echo information points as the ground object target points corresponding to the multiple echo information points. The determined ground object target point is used as a reference point, and by comparing the virtual echo position corresponding to the echo information point with the position of the ground object target point, it is determined whether the echo information point is a cable or a noise data point recorded incorrectly, so as to determine the specific position corresponding to the cable.

[0143] Specifically, adjusting the first mapping relationship according to the ground object target point and the virtual echo position to obtain the target point cloud data for distance detection may include: obtaining a tolerance for distance detection operation, obtaining a difference between the measured distance of the ground object target point and the virtual echo position, and adjusting the first mapping relationship according to the tolerance and the difference to obtain the target point cloud data for distance detection. The tolerance includes a downward tolerance and an upward tolerance, which may be determined according to a downward tolerance threshold and an upward tolerance threshold.

[0144] In the process of adjusting the first mapping relationship according to the tolerance and the difference to obtain the target point cloud data for distance detection, if the difference is greater than the upward tolerance, the echo information point is likely to have distance ambiguity and is in an unreasonable spatial position, and it is necessary to continue to find the correct actual pulse emission time, and update the first mapping relationship based on the actual pulse emission time; if the difference is greater than or equal to the downward tolerance and less than or equal to the upward tolerance, the echo information point is likely to have no distance ambiguity and is in a reasonable spatial position, that is, the virtual echo position corresponding to the echo information point is likely to be the correct position, and then continue to search forward to determine the correct actual pulse emission time after comparison, and update the first mapping relationship based on the actual pulse emission time; wherein the downward tolerance includes a sign, which is a negative value; if the difference is less than the downward tolerance, the echo information point has submerged below the surface, this situation cannot really exist, and the virtual echo position corresponding to this echo information point is wrong, so it is not processed again to find the next marked isolated data for processing.

[0145] In an embodiment of the present invention, a virtual echo position corresponding to each echo information point is determined, and then an echo information point farthest from the laser radar system is searched in the vicinity of the virtual echo position as a ground object target point. Finally, the position of the ground object target point is compared with the virtual echo position to determine the correct position of each echo information point, that is, the correct pulse emission interval of each echo information point is determined, and based on the pulse emission interval and the pulse emission sequence, the actual pulse emission time corresponding to each echo information point is determined, and the first mapping relationship is adjusted to obtain target point cloud data for distance detection. This can not only eliminate distance ambiguity, but also achieve accurate distance measurement based on the target point cloud data.

[0146] In specific application, this application embodiment provides a method of eliminating the laser radar distance ambiguity by determining the pulse emission interval in which each echo information point is located to determine the actual pulse emission time corresponding to each echo information point. Specifically, the method may include the following steps:

[0147] 1. Obtain the point cloud data to be processed.

[0148] The point cloud data is data in SOCS coordinates. The embodiment of the present invention directly eliminates distance ambiguity in SOCS coordinates to obtain target data points that are ultimately used to achieve target distance measurement.

[0149] 2. Preprocess the point cloud data.

[0150] Read SOCS data line by line. When a scan angle jump is detected, store the previous multiple lines of data as a scan line in the data structure. Divide the echo information points in the SOCS data into scan lines according to the scan angle, and determine the scan line corresponding to each echo information point.

[0151] 3. Process each echo information point on the first scan line according to the set method.

[0152] First, the pulse emission time sequence of the first scan line is obtained, and the pulse repetition frequency is used to calculate the pulse time interval. If the time difference between adjacent pulse emission times is a multiple of the pulse time interval, the unrecorded pulse emission time is filled with a marker to ensure the strict accuracy of the distance calculation. Then, according to the set point cloud filter, all the echo information points of the first scan line are divided into: echo information points with high spatial continuity (ground features) and echo information points with low spatial continuity (cables).

[0153] Specifically, all the data on the first scanning line are traversed, and according to the set point cloud filtering, the spatial continuity of each laser pulse signal corresponding to the echo position is calculated, and the data is divided into echo information points with high spatial continuity and echo information points with low spatial continuity.

[0154] The specific implementation process of calculating the spatial continuity of the echo position corresponding to each laser pulse signal may include:

[0155] For a scan line containing N echoes; obtain the minimum line spacing C of the overhead transmission line l 、Minimum distance of overhead transmission lines C g ; Get the flight altitude H above the ground and the field of view angle A fov .

[0156] Step 1: Declare and initialize a decision array J of length N; use the following formula to calculate the upper limit of the number of neighborhood points M n ;

[0157]

[0158] Step 2: Traverse the point cloud data P; for the echo information point P in P i ,sequence Record The echo storage order is P i Points before and after; variable C n Record Sequence Zhong and P i The number of points determined as spatial neighbors is initialized to 0;

[0159] Among them, P i =(x i ,yi , z i ),

[0160] for Calculate P j With P i The distance D ij ; if D ij <C l , C n Increment by 1;

[0161] D ij =(x i -x j ) 2 +(y i -y j ) 2 +(z i -z j ) 2

[0162] Echo information point P i After all neighborhood points are counted; if C n / M n ≥0.8,P i The echo information point judged as having high spatial continuity is judged as the object J in array J. i Set to 1; if C n / M n <0.8,P i The echo information point judged as having low spatial continuity is judged as object J in array J. i Set to 0.

[0163] Then, for echo information points with high spatial continuity, the MTA interval to which the echo information points with a scanning angle close to zero degrees in the first scan line belong is calculated based on the flight altitude of the drone above the ground given by the user and the pulse repetition frequency. Use bidirectional dynamic programming from both ends to the middle of the scan line to trace back from the position of the echo information point with a scanning angle close to zero degrees with the correct MTA interval calculated above as the starting point to obtain a mapping sequence. Use the mapping sequence in combination with the pulse emission time sequence to calculate the correct position of each echo information point with a high degree of spatial continuity in the first scan line, and use the array of this correct position as a terrain reference array. Among them, the MTA interval refers to the pulse emission interval. In the embodiments of the present invention, for the sake of ease of description, it is uniformly referred to as the MTA interval. The mapping sequence contains an array of the MTA interval to which each echo information point belongs.

[0164] Specifically, the echo information points with high spatial continuity are traversed, and according to the set dynamic programming algorithm, the flight altitude above the ground given by the user, the pulse repetition frequency of the lidar system, the index of the laser pulse signal with a scanning angle close to zero degrees, and the pulse emission time corresponding to the echo information points with high spatial continuity are solved to obtain the echo pulse signal with correct ranging, and the echo pulse position with high spatial continuity that eliminates the distance ambiguity of the first scanning line is recorded as a terrain reference.

[0165] Specifically, assuming that for the first scan line containing N echoes and with a maximum MTA interval of M, the index of the laser pulse signal with a scanning angle close to zero is I, the pulse emission time sequence is T, and the point cloud data is P; the flight height above the ground surface H and the pulse repetition frequency F of the laser radar system are obtained. The implementation method of determining the pulse emission time corresponding to the echo information point with a high degree of spatial continuity may include:

[0166] Step 1: Declare and initialize the distance table R with N columns and M rows; the noise energy table E with N columns and M rows; the pointer table P with N columns and M rows; the mapping array A with a length of N; and the terrain reference array G with a length of N.

[0167] Step 2: Use the double cursor method to convert the object P of the point cloud data P i The moment S of the pulse emission time series T t Match, that is, P i =(x i ,y i , z i , S t ). Read S in T in sequence t ,S t-1 ,S t-2 ,,S t-M+1 There are M moments in total. If the moment contains the missing mark, the M elements in the i-th column of the distance table R are set to "-1" as a mark; if the M moments do not contain special marks, the i-th column of the table R is filled according to the following formula:

[0168]

[0169] Step 3: For all objects P i∈[0I) i , the position of P0 in different MTA intervals is the starting point of the dynamic programming process, and the first column of Table E and P is filled in using the following formula: E 0j =0,P 0j =j.

[0170] Next, process them in the order of increasing i, set a natural number g to record the distance between the previous unmarked column and the current column; for columns whose elements in the distance table R are all "-1", set the corresponding columns in tables E and P to "-1", and g increases by 1; for unmarked column i in R, use the formula to fill in tables E and P:

[0171] ΔE ijk =(R ij -R (i-g)k ) 2 +E (i-g)k

[0172] E ij =min(ΔE ij0 ΔE ij1 ΔE ijM )

[0173] P ij = k where ΔE ijk =E ij

[0174] Element E in the noise energy table E ij Record the lowest noise energy when passing through this position, and P ij Record E ij The noise energy value inherited from any position of the previous unmarked echo; after the calculation of the non-"-1" column is completed, jumpStep is reset to "1".

[0175] Step 4: For all objects P i∈[I,N) i , P N-1 The location under different MTA zones is the starting point of the dynamic planning process. The last column of Table E and Table P is filled with the following formula: (N-1)j =0,P (N-1)j =j.

[0176] Next, start from i=N-1 and process in descending order of i; set a natural number g to record the distance between the previous unmarked column and the current column; for columns whose elements in the distance table R are all "-1", set the corresponding columns in tables E and P to "-1", and g is reduced by one; for the unmarked column i in R, use the formula to fill in tables E and P:

[0177] ΔE ijk =(R ij -R (i+g)k ) 2 +E (i+g)k

[0178] E ij =min(ΔE ij0 ΔE ij1 ΔEijM )

[0179] P ij = k where ΔE ijk =E ij

[0180] Element E in the noise energy table E ij Record the lowest noise energy when passing through this position, and P ij Record E ij The noise energy value inherited from any position of the previous unmarked echo; after the calculation of the non-"-1" column is completed, jumpStep is reset to "1".

[0181] Step 5: Using the given flight altitude H above the ground and the pulse repetition frequency F of the lidar system, calculate the correct MTA interval Z to which the laser pulse with a scan angle close to zero belongs:

[0182] Step 6: After the filling of tables E and P is completed, the path is backtracked to obtain the correct mapping between the echo and the emission time, and filled into the mapping array A; from the I-1 column of the noise energy table E, the nearest non-"-1" column is searched forward, and the same column and Zth row in the pointer table P are used as the starting point of the backtracking. According to the direction of the pointer, the correct mapping relationship is continuously read forward and filled into the corresponding position of the mapping table A, and the first column of the table P is read and stopped; from the I column of the noise energy table E, the nearest non-"-1" column is searched backward, and the same column and Zth row in the pointer table P are used as the starting point of the backtracking. According to the direction of the pointer, the correct mapping relationship is continuously read backward and filled into the corresponding position of the mapping table A, and the last column of the table P is read and stopped; for the point cloud data with time S t Object P i , use the following formula group to calculate the correct position of the echo:

[0183]

[0184] Correcting the laser pulse signal using the calculated Cartesian coordinates and the correct pulse emission time;

[0185]

[0186] Fill the corrected data into the terrain reference array G, and record the length N of the terrain reference array G. l = N. Finally, for the echo information points with low spatial continuity, the terrain reference array is used as the terrain reference. For each echo information point with high spatial dispersion, the maximum value of the scanner distance is found in the field of its index (array subscript) as the ground object anchor point of the echo, and the potential correct virtual echo position of the echo is compared with the ground object anchor point position to calculate the correct position of the echo with high spatial dispersion.

[0187] Specifically, the echo information points with low spatial continuity in the data are traversed, and according to the set terrain prediction algorithm, the POS data subset, the upward tolerance threshold, the downward tolerance threshold, and the terrain reference are used to solve the pulse emission time of the laser pulse signal with low spatial continuity to obtain the echo pulse signal with correct ranging.

[0188] For a scan line containing N echoes and having a maximum MTA interval of M, the specific implementation process of obtaining the number of neighboring points q, the upward tolerance threshold uH, the downward tolerance threshold dH, and the terrain reference array G of the corresponding scan line and obtaining the echo information point with correct ranging may include:

[0189] Step 1: Use the double cursor method to convert the object P of the point cloud data P i The moment S of the pulse emission time series T extracted from the preprocessing t Match, that is, P i =(x i ,y i , z i , S t ).

[0190] sequence Record and Object P i For q adjacent real objects , whose real object distance is equal to G j ; Use the minimum value of q real object distances as object P i Ground anchor point AP i ;

[0191]

[0192]

[0193] Step 2: In the pulse emission time sequence T, t Take the starting point as the starting point and search forward one by one for the potential correct pulse emission time of the isolated point; if the found time is marked as "-1" or the number of forward searches is greater than M, return to step 1 and find the next marked isolated point; calculate the object P according to the following formula i Virtual echo FP in MTA zone k ik Distance from laser scanner FR ik , go to step 3;

[0194]

[0195] Step 3: Submit FR under SOCS ik With AP i In comparison.

[0196] (1) If AP i -FR ik >uH, then virtual echo FP ik There is a high probability that it is due to distance ambiguity and will not be in a reasonable spatial position; return to step 2 and continue to look forward for the pulse emission time;

[0197] (2) If -dH≤AP i -FR ik ≤uH, then the virtual echo FP ik There is a high probability that it comes from the unambiguous distance and will be in a reasonable spatial position; that is, the position of the virtual echo is very likely to be the correct position of the isolated point; the correct MTA interval of this echo pulse is recorded at the corresponding position of the mapping array;

[0198] (3) If AP i -FR ik <-dH, then the virtual echo FP ik It has obviously sunk below the surface, which is impossible in a LiDAR system, indicating that the virtual echo is in the wrong spatial position. Since a smaller pulse emission time will only make the virtual echo farther from the scanner, return to step 1 to find the next marked isolated point.

[0199] Among them, calculating the upward tolerance threshold and the downward tolerance threshold of the survey area object can include: for a high repetition rate, wide field of view laser radar system, obtaining the distance error limit factor as T, the standard height of the overhead transmission line; and obtaining the upward tolerance threshold and the downward tolerance threshold according to the following calculation. That is, uH = 1.5 × H s , dH=T.

[0200] 4. Process the echo information points of the scan lines other than the first one one by one according to the set method.

[0201] First, obtain the pulse emission time sequence of a scan line outside the first scan line, and use the pulse repetition frequency to calculate the pulse time interval. If the time difference between adjacent pulse emission times is a multiple of the pulse time interval, use a marker to fill in the unrecorded pulse emission time to ensure the strict correctness of the distance calculation. Then, according to the set point cloud filtering, all echo information points of the first scan line are divided into: echo information points with high spatial continuity and echo information points with low spatial continuity. The specific implementation process can refer to the above related instructions, which will not be repeated here.

[0202] Then, for echo information points with high spatial continuity, use the terrain reference array as the terrain reference. Use a one-way DP meter to trace back from the final position with the MTA interval with the lowest noise energy as the starting point to obtain a mapping sequence. Use the mapping sequence combined with the pulse emission time sequence to calculate the correct position of each echo with low spatial dispersion on this scan line, and use the array of this correct position as the terrain reference array. Among them, the noise energy algorithm in the one-way DP meter involves terrain reference, which is different from the first line. The mapping sequence contains an array of the MTA interval to which each echo belongs, and each time a set of terrain reference data is determined, the previous terrain reference array will be updated.

[0203] Specifically, the echo information points with high spatial continuity are traversed, and according to the set dynamic programming algorithm, the POS data subset and the terrain reference of the previous scan line are used to solve the pulse emission time corresponding to the echo information points with high spatial continuity, and the echo pulse signal with correct ranging is obtained, and the terrain reference is updated to the echo pulse position record with high spatial continuity of the scan line that eliminates distance ambiguity.

[0204] Specifically, assuming that for a scan line containing N echoes and a maximum MTA interval of M, a terrain reference array G of a previous scan line containing points is obtained, and a pulse emission time series T and point cloud data P are obtained. The implementation method of determining the terrain reference may include:

[0205] Step 1: Declare and initialize a distance table R with N columns and M rows; a noise energy table E with N columns and M rows; a pointer table P with N columns and M rows; and a mapping array A with a length of N.

[0206] Step 2: Use the double cursor method to move the echo information point P of the point cloud data P i The moment S of the pulse emission time series T extracted from the preprocessing t Match, that is, P i =(x i ,y i , z i , S t ).

[0207] Read S in T in sequence t ,S t-1 ,S t-2 ,…,S t-M+1 There are M moments in total. If the moment contains the missing mark, the M elements in the i-th column of the distance table R are set to "-1" as a mark; if the M moments do not contain special marks, the i-th column of the table R is filled according to the following formula:

[0208]

[0209] Step 3: For all echo information points P of i∈[0,N) i , the position of P0 in different MTA intervals is the starting point of the dynamic programming process, and the first column of Table E and P is filled in using the following formula: E 0j =0,P 0j =j.

[0210] Next, process them in the order of increasing i; set a natural number g to record the distance between the previous unmarked column and the current column; for columns whose elements in the distance table R are all "-1", set the corresponding columns in tables E and P to "-1", and g increases by 1; for unmarked column i in R, if i <N l :ΔE ijk =(R ij -R (i-g)k ) 2 +(R ij -G i ) 2 +E (i-g)k .

[0211] If i ≥ N l :ΔE ijk =(R ij -R (i-g)k ) 2 +E (i-g)k .

[0212] Use the formula to fill in the table E, P:E ij =min(ΔE ij0 ΔE ij1 ,…,ΔE ijM ), P ij = k where ΔE ijk =E ij .

[0213] Element E in the noise energy table E ij Record the lowest noise energy when passing through this position, and P ij Record E ij The noise energy value inherited from any position of the previous unmarked echo; after the calculation of the non-"-1" column is completed, jumpStep is reset to "1".

[0214] Step 4: After the E and P tables are filled, the path is traced back to obtain the correct mapping between the echo and the emission time, and filled into the mapping array A; from the N-1th column of the noise energy table E, look forward to the nearest non-"-1" column, find the position with the lowest noise energy in the column, and use the same position in the pointer table P as the starting point of the backtracking. According to the pointer, the correct mapping relationship is continuously read forward and filled into the corresponding position of the mapping table A, and it stops after reading to the first column of the table P; for the point cloud data with time St The object P i , use the following formula group to calculate the correct position of the echo:

[0215]

[0216] Correcting the laser pulse signal using the calculated Cartesian coordinates and the correct pulse emission time;

[0217]

[0218] Fill the corrected data into the terrain reference array G, and record the length N of the terrain reference array G. l =N. Finally, for the echo information points with low spatial continuity, the updated terrain reference array is used as the terrain reference. For each echo information point with high spatial dispersion, the maximum value of the scanner distance is found in the field of its index (array subscript) as the ground object anchor point of the echo, and the potential correct virtual echo position of the echo is compared with the ground object anchor point position to calculate the correct position of the echo with high spatial dispersion. The detailed implementation process can refer to the above-mentioned related description, which will not be repeated here.

[0219] The following will describe in detail the laser radar range ambiguity elimination device of one or more embodiments of the present invention. Those skilled in the art will appreciate that these devices can be configured using commercially available hardware components through the steps taught in this solution. Figure 8 A schematic diagram of the structure of a laser radar distance ambiguity elimination device provided by an embodiment of the present invention, such as Figure 8 As shown, the device includes: an acquisition module 11, a first determination module 12, a second determination module 13, a third determination module 14, and an adjustment module 15.

[0220] The acquisition module 11 is used to acquire point cloud data to be processed, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments.

[0221] The first determination module 12 is used to determine the pulse repetition frequency corresponding to the point cloud data.

[0222] The second determination module 13 is used to determine the pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency.

[0223] The third determination module 14 is used to determine a mapping sequence between the multiple echo information points and pulse emission intervals if the multiple echo information points are non-isolated data, and the pulse emission interval is determined based on the pulse repetition frequency.

[0224] The adjustment module 15 is used to adjust the first mapping relationship according to the mapping sequence and the pulse emission time sequence to obtain target point cloud data for realizing distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

[0225] Optionally, the second determination module 13 can be specifically used to: determine the original pulse emission time sequence based on the point cloud data; determine the pulse time interval according to the pulse repetition frequency; determine the pulse emission time sequence corresponding to the point cloud data according to the original pulse emission time sequence and the pulse time interval.

[0226] Optionally, the second determination module 13 can be specifically used to: determine the time difference between adjacent original pulse emission moments; if the time difference is a multiple of the pulse time interval, determine that there is an unrecorded pulse emission moment in the original pulse emission moment sequence, and determine the pulse emission moment sequence corresponding to the point cloud data based on the original pulse emission moment sequence and the unrecorded pulse emission moment. If the time difference is the pulse time interval, determine the original pulse emission moment sequence as the pulse emission moment sequence corresponding to the point cloud data.

[0227] Optionally, the device may further include a judgment module which may be specifically used to: determine the degree of spatial continuity corresponding to each echo information point in the point cloud data; and determine whether each echo information point is non-isolated data based on the degree of spatial continuity.

[0228] Optionally, the judgment module can be specifically used to: if the spatial continuity corresponding to the echo information point is greater than or equal to a first preset value, determine that the echo information point is non-isolated data; or, if the spatial continuity corresponding to the echo information point is less than the first preset value, determine that the echo information point is isolated data.

[0229] Optionally, the third determination module 14 can be specifically used to: segment the point cloud data according to the scanning angle in the point cloud data, determine the scanning lines corresponding to each echo information point in the point cloud data, and the scanning lines include the first scanning line; determine the target data point corresponding to the scanning angle in the first scanning line, and the scanning angle is 0° or close to 0°; determine the initial mapping sequence between the target data point and the pulse emission interval according to the pulse repetition frequency corresponding to the point cloud data; use the dynamic programming algorithm to backtrack with the initial mapping sequence as the starting point to determine the first sub-mapping sequence between the multiple echo information points in the first scanning line and the pulse emission interval. According to the first sub-mapping sequence and the dynamic programming algorithm, determine the second sub-mapping sequence between the multiple echo information points outside the first scanning line and the pulse emission interval; based on the first sub-mapping sequence and the second sub-mapping sequence, determine the mapping sequence between the multiple echo information points and the pulse emission interval.

[0230] Optionally, the third determination module 14 can be specifically used to: determine the current pulse emission time corresponding to each of the multiple echo information points in the first scan line in the pulse emission time sequence; obtain multiple historical pulse emission times before the current pulse emission time; determine the two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to each of the multiple echo information points, the current pulse emission time and the multiple historical pulse emission times; determine the two-dimensional noise energy information corresponding to the multiple echo information points according to the two-dimensional distance information; determine the two-dimensional pointer information corresponding to the multiple echo information points according to the two-dimensional noise energy information; detect Whether the multiple historical pulse emission moments include unrecorded pulse emission moments; if not, backtracking from the initial mapping sequence as the starting point based on the two-dimensional pointer information and the two-dimensional noise energy information to determine the first sub-mapping sequence between the multiple echo information points in the first scan line and the pulse emission interval; or, if unrecorded pulse emission moments are included, the column information corresponding to the echo information points in the two-dimensional distance information, the column information corresponding to the echo information points in the two-dimensional noise energy information, and the column information corresponding to the echo information points in the two-dimensional pointer information are all set to a second preset value.

[0231] Optionally, the third determination module 14 can also be specifically used to: determine the distance information corresponding to each of the multiple echo information points in each pulse emission interval according to the measured distances corresponding to each of the multiple data points, the current pulse emission time and the multiple historical pulse emission times; determine the two-dimensional distance information corresponding to the multiple echo information points according to the distance information corresponding to each of the multiple echo information points.

[0232] Optionally, the third determination module 14 can also be specifically used to: obtain the pulse emission time in the interval corresponding to the multiple echo information points in the pulse emission interval; obtain the time difference between the current pulse emission time and the pulse emission time in the interval; determine the spatial distance corresponding to each of the multiple echo information points based on the time difference and the laser propagation speed; obtain the distance sum between the measured distance corresponding to the echo information point and the spatial distance; and determine the distance sum as the distance information corresponding to each of the multiple echo information points in each pulse emission interval.

[0233] Optionally, the third determination module 14 can also be specifically used to: obtain the data relationship between the multiple echo information points and the target data point; if the multiple echo information points are data obtained before the target data point, then according to the two-dimensional distance information, starting with the first echo information point, determine the two-dimensional noise energy information corresponding to each of the multiple echo information points in a preset ascending order; if the echo information point is data obtained after the target echo information point, then according to the two-dimensional distance information, starting with the last echo information point, determine the two-dimensional noise energy information corresponding to each of the multiple echo information points in a preset descending order.

[0234] Optionally, the third determination module 14 can also be specifically used to: set the column information corresponding to the first echo information point in the two-dimensional noise energy information to an initial state; starting from the first echo information point, successively obtain the distance values ​​corresponding to the echo information points at each position in the two-dimensional distance information, and the previous distance value corresponding to the previous echo information point in ascending order; obtain the previous noise energy value corresponding to the previous echo information point; determine the noise energy values ​​of the multiple echo information points at each position based on the distance values ​​corresponding to each echo information point, the previous distance value and the previous noise energy value; determine the two-dimensional noise energy information corresponding to the multiple echo information points based on the noise energy values ​​of the multiple echo information points at each position.

[0235] Optionally, the third determination module 14 can also be specifically used to: set the column information corresponding to the last echo information point in the two-dimensional noise energy information to an initial state; starting from the last echo information point, successively obtain the distance values ​​corresponding to the echo information points at each position in the two-dimensional distance information, and the subsequent distance value corresponding to the subsequent echo information point in descending order; obtain the subsequent noise energy value corresponding to the subsequent echo information point; determine the noise energy values ​​of the multiple echo information points at each position according to the distance values ​​corresponding to each echo information point, the subsequent distance value and the subsequent noise energy value; determine the two-dimensional noise energy information corresponding to the multiple echo information points according to the noise energy values ​​of the multiple echo information points at each position.

[0236] Optionally, the third determination module 14 can also be specifically used to: determine the pointer value corresponding to each echo information point according to the noise energy value at the location of each echo information point in the two-dimensional noise energy information; determine the two-dimensional pointer information corresponding to each echo information point according to the pointer value corresponding to each echo information point.

[0237] Optionally, the third determination module 14 can also be specifically used to: obtain the data relationship between the multiple echo information points and the target data points; if the multiple echo information points are data obtained before the target data points, then according to the two-dimensional pointer information and the two-dimensional noise energy information, backtrack in descending order from the initial mapping sequence to determine the mapping sequence corresponding to the multiple echo information points; if the multiple echo information points are data obtained after the target data points, then according to the two-dimensional pointer information and the two-dimensional noise energy information, backtrack in ascending order from the initial mapping sequence to determine the mapping sequence corresponding to the multiple data points.

[0238] Optionally, the third determination module 14 can also be specifically used to: determine the current pulse emission time corresponding to each of the multiple echo information points outside the first scanning line in the pulse emission time sequence; obtain multiple historical pulse emission times before the current pulse emission time; determine the two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to each of the multiple echo information points, the current pulse emission time and the multiple historical pulse emission times; determine the two-dimensional noise energy information corresponding to the multiple echo information points according to the two-dimensional distance information and the first sub-mapping sequence; determine the two-dimensional pointer information corresponding to the multiple echo information points according to the two-dimensional noise energy information; detect the multiple historical pulse emission times; Whether the moment includes an unrecorded pulse emission moment; if the unrecorded pulse emission moment is not included, then determine the target mapping sequence between the last echo information point and the pulse emission interval, and based on the two-dimensional pointer information and the two-dimensional noise energy information, backtrack with the target mapping sequence as the starting point to determine the first sub-mapping sequence between multiple echo information points outside the first scanning line and the pulse emission interval; or, if the unrecorded pulse emission moment is included, then the column information corresponding to the echo information point in the two-dimensional distance information, the column information corresponding to the echo information point in the two-dimensional noise energy information, and the column information corresponding to the echo information point in the two-dimensional pointer information are all set to the second preset value.

[0239] Optionally, the adjustment module 15 can also be specifically used to: determine the actual pulse emission time corresponding to each of the multiple echo information points according to the mapping sequence and the pulse emission time sequence; update the first mapping relationship based on the multiple echo information and the actual emission time corresponding to each echo information, and obtain target point cloud data for realizing distance detection.

[0240] Optionally, the device may also include an isolated data determination module which may be specifically used for: if the multiple echo information points are isolated data, determining the virtual echo positions corresponding to each of the multiple echo information points; determining the ground object target points corresponding to each of the multiple echo information points; and adjusting the first mapping relationship according to the ground object target points and the virtual echo positions to obtain target point cloud data for distance detection.

[0241] Optionally, the isolated data determination module can also be specifically used to: obtain the measured distance corresponding to each of the multiple echo information points; obtain the current pulse emission time corresponding to each of the multiple echo information points; obtain the pulse emission time in the interval corresponding to the multiple echo information points in the pulse emission interval; obtain the time difference between the current pulse emission time and the pulse emission time in the interval; determine the spatial distance corresponding to each of the multiple echo information points based on the time difference and the laser propagation speed; obtain the distance sum between the measured distance corresponding to the echo information point and the spatial distance; determine the distance sum as the virtual echo position corresponding to each of the multiple echo information points in each pulse emission interval.

[0242] Optionally, the isolated data determination module can also be specifically used to: obtain multiple adjacent echo information points adjacent to each of the multiple data points, wherein the multiple adjacent echo information points are non-isolated data; determine the target adjacent echo information point corresponding to the maximum measurement distance among the multiple adjacent echo information points; and respectively determine the target adjacent echo information point corresponding to each of the multiple echo information points as the ground target point corresponding to each of the multiple echo information points.

[0243] Optionally, the isolated data determination module can also be used to: obtain a tolerance for distance detection operations; obtain the difference between the measured distance of the target point and the virtual echo position; and adjust the first mapping relationship according to the tolerance and the difference to obtain target point cloud data for distance detection.

[0244] Figure 8 The device shown can perform the aforementioned Figures 1 to 7 The laser radar distance ambiguity elimination method provided in the illustrated embodiment, the detailed execution process and technical effects can be found in the description of the aforementioned embodiment, which will not be repeated here.

[0245] In one possible design, the above Figure 8 The structure of the target detection device shown can be implemented as an electronic device, such as Fig. 9 As shown, the electronic device may include: a processor 21 and a memory 22. The memory 22 stores executable code. When the executable code is executed by the processor 21, the processor 21 can at least implement the above-mentioned Figures 1 to 7 The laser radar distance ambiguity elimination method provided in the illustrated embodiment.

[0246] Optionally, the electronic device may further include a communication interface 23 for communicating with other devices.

[0247] In addition, an embodiment of the present invention provides a non-transitory machine-readable storage medium, wherein an executable code is stored on the non-transitory machine-readable storage medium. When the executable code is executed by a processor of an electronic device, the processor can at least implement the above-mentioned Figures 1 to 7 The laser radar distance ambiguity elimination method provided in the illustrated embodiment.

[0248] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by adding a necessary general hardware platform, and of course, it can also be implemented by a combination of hardware and software. Based on such an understanding, the above technical solution can essentially or in other words, contribute to the prior art in the form of a computer product, and the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes.

[0249] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A laser radar range ambiguity elimination method, characterized in that: include: Acquire point cloud data to be processed, wherein the point cloud data includes a first mapping relationship between a plurality of echo information points and a plurality of original pulse emission moments, and a scanning angle; Determining a pulse repetition frequency corresponding to the point cloud data; Determining a pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency; If the plurality of echo information points are non-isolated data, segmenting the point cloud data according to the scanning angle in the point cloud data to determine the scanning lines corresponding to the respective echo information points in the point cloud data, wherein the scanning lines include the first scanning line; Determine a target data point in the first scan line corresponding to the scan angle, wherein the scan angle is 0° or close to 0°; Determining an initial mapping sequence of the target data points and pulse emission intervals according to a pulse repetition frequency corresponding to the point cloud data; Using a dynamic programming algorithm, backtracking from the initial mapping sequence as a starting point, to determine a first sub-mapping sequence of a plurality of echo information points in the first scanning line and the pulse emission interval; Determine, according to the first sub-mapping sequence and a dynamic programming algorithm, a second sub-mapping sequence of a plurality of echo information points outside the first scanning line and the pulse emission interval; Determine, based on the first sub-mapping sequence and the second sub-mapping sequence, a mapping sequence of the plurality of echo information points and pulse transmission intervals, wherein the pulse transmission interval is determined based on the pulse repetition frequency; According to the mapping sequence and the pulse emission time sequence, the first mapping relationship is adjusted to obtain target point cloud data for realizing distance detection, wherein the target point cloud data includes a second mapping relationship between multiple echo information points and multiple actual emission times, and the first mapping relationship is different from the second mapping relationship.

2. The method according to claim 1, characterized in that The step of determining a pulse emission time sequence corresponding to the point cloud data according to the pulse repetition frequency includes: Based on the point cloud data, determining the original pulse emission time sequence; Determining a pulse time interval according to the pulse repetition frequency; A pulse emission time sequence corresponding to the point cloud data is determined according to the original pulse emission time sequence and the pulse time interval.

3. The method according to claim 1, characterized in that The point cloud data includes a measured distance, and the dynamic programming algorithm is used to backtrack from the initial mapping sequence as a starting point to determine a plurality of echo information points in the first scanning line and a first sub-mapping sequence of the pulse emission interval, including: Determining, in the pulse emission time sequence, current pulse emission times corresponding to each of a plurality of echo information points in the first scanning line; Acquire multiple historical pulse emission moments before the current pulse emission moment; Determine the two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to the multiple echo information points, the current pulse emission time and the multiple historical pulse emission times; Determining two-dimensional noise energy information corresponding to the plurality of echo information points according to the two-dimensional distance information; Determining two-dimensional pointer information corresponding to the plurality of echo information points according to the two-dimensional noise energy information; Detecting whether the plurality of historical pulse emission moments include unrecorded pulse emission moments; If the unrecorded pulse emission time is not included, then based on the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed with the initial mapping sequence as the starting point to determine a plurality of echo information points in the first scanning line and a first sub-mapping sequence of the pulse emission interval; or, If the unrecorded pulse emission time is included, the column information corresponding to the echo information point in the two-dimensional distance information, the column information corresponding to the echo information point in the two-dimensional noise energy information, and the column information corresponding to the echo information point in the two-dimensional pointer information are all set to the second preset value.

4. The method according to claim 3, characterized in that The determining of the two-dimensional distance information corresponding to the plurality of echo information points according to the measured distances respectively corresponding to the plurality of echo information points, the current pulse emission time and the plurality of historical pulse emission times comprises: Determine the distance information corresponding to each of the plurality of echo information points in each pulse emission interval according to the respective corresponding measured distances of the plurality of data points, the current pulse emission time and the plurality of historical pulse emission times; According to the distance information corresponding to each of the plurality of echo information points, the two-dimensional distance information corresponding to the plurality of echo information points is determined.

5. The method according to claim 3, characterized in that: The step of determining the two-dimensional noise energy information corresponding to the plurality of echo information points according to the two-dimensional distance information comprises: Acquire the data relationship between the plurality of echo information points and the target data point; If the multiple echo information points are data obtained before the target data point, then according to the two-dimensional distance information, starting from the first echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset increasing order; If the echo information point is data obtained after the target echo information point, the two-dimensional noise energy information corresponding to each of the multiple echo information points is determined in a preset descending order starting from the last echo information point according to the two-dimensional distance information.

6. The method according to claim 5, characterized in that The method of determining the two-dimensional noise energy information corresponding to each of the plurality of echo information points according to the two-dimensional distance information, starting from the first echo information point, in a preset increasing order, includes: Setting the column information corresponding to the first echo information point in the two-dimensional noise energy information to an initial state; Starting from the first echo information point, sequentially acquiring the distance values ​​corresponding to the echo information points at various positions in the two-dimensional distance information and the previous distance value corresponding to the previous echo information point in ascending order; Obtaining a previous noise energy value corresponding to the previous echo information point; Determining the noise energy value of the plurality of echo information points at each position according to the distance value corresponding to each echo information point, the previous distance value and the previous noise energy value; According to the noise energy values ​​of the multiple echo information points at various positions, the two-dimensional noise energy information corresponding to the multiple echo information points is determined.

7. The method according to claim 5, characterized in that The method of determining the two-dimensional noise energy information corresponding to each of the plurality of echo information points according to the two-dimensional distance information, starting from the last echo information point, in a preset descending order, includes: Setting the column information corresponding to the last echo information point in the two-dimensional noise energy information to an initial state; Starting from the last echo information point, sequentially obtaining the distance values ​​corresponding to the echo information points at various positions in the two-dimensional distance information and the next distance value corresponding to the next echo information point in descending order; Obtaining a subsequent noise energy value corresponding to the subsequent echo information point; Determine the noise energy value of the plurality of echo information points at each position according to the distance value corresponding to each echo information point, the subsequent distance value and the subsequent noise energy value; According to the noise energy values ​​of the multiple echo information points at various positions, the two-dimensional noise energy information corresponding to the multiple echo information points is determined.

8. The method according to claim 3, characterized in that The step of determining the two-dimensional pointer information corresponding to the plurality of echo information points according to the two-dimensional noise energy information comprises: Determining the pointer value corresponding to each echo information point according to the noise energy value at the location of each echo information point in the two-dimensional noise energy information; According to the pointer value corresponding to each echo information point, the two-dimensional pointer information corresponding to each echo information point is determined.

9. The method according to claim 3, characterized in that: The step of backtracking based on the two-dimensional pointer information and the two-dimensional noise energy information and taking the initial mapping sequence as a starting point to determine a first sub-mapping sequence between a plurality of echo information points in the first scanning line and the pulse emission interval includes: Acquire the data relationship between the plurality of echo information points and the target data point; If the multiple echo information points are data obtained before the target data point, then according to the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed in descending order from the initial mapping sequence to determine the mapping sequence corresponding to the multiple echo information points; If the multiple echo information points are data obtained after the target data point, then according to the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed in ascending order starting from the initial mapping sequence to determine the mapping sequence corresponding to the multiple data points.

10. The method according to claim 1, characterized in that The determining, according to the first sub-mapping sequence and a dynamic programming algorithm, a second sub-mapping sequence of a plurality of echo information points outside the first scanning line and the pulse emission interval comprises: Determining, in the pulse emission time sequence, current pulse emission times corresponding to each of a plurality of echo information points outside the first scanning line; Acquire multiple historical pulse emission moments before the current pulse emission moment; Determine the two-dimensional distance information corresponding to the multiple echo information points according to the measured distances corresponding to the multiple echo information points, the current pulse emission time and the multiple historical pulse emission times; Determining two-dimensional noise energy information corresponding to the plurality of echo information points according to the two-dimensional distance information and the first sub-mapping sequence; Determining two-dimensional pointer information corresponding to the plurality of echo information points according to the two-dimensional noise energy information; Detecting whether the plurality of historical pulse emission moments include unrecorded pulse emission moments; If the unrecorded pulse emission time is not included, a target mapping sequence between the last echo information point and the pulse emission interval is determined, and based on the two-dimensional pointer information and the two-dimensional noise energy information, backtracking is performed with the target mapping sequence as the starting point to determine a second sub-mapping sequence between multiple echo information points outside the first scanning line and the pulse emission interval; or, If the unrecorded pulse emission time is included, the column information corresponding to the echo information point in the two-dimensional distance information, the column information corresponding to the echo information point in the two-dimensional noise energy information, and the column information corresponding to the echo information point in the two-dimensional pointer information are all set to the second preset value.

11. The method according to claim 1, characterized in that The step of adjusting the first mapping relationship according to the mapping sequence and the pulse emission time sequence to obtain target point cloud data for distance detection includes: Determining actual pulse emission times corresponding to each of the plurality of echo information points according to the mapping sequence and the pulse emission time sequence; The first mapping relationship is updated based on the multiple echo information and the actual emission time corresponding to each echo information to obtain target point cloud data for realizing distance detection.

12. The method according to claim 1, characterized in that The method further comprises: If the plurality of echo information points are isolated data, determining virtual echo positions corresponding to the plurality of echo information points; Determine the ground object target point corresponding to each of the plurality of echo information points; The first mapping relationship is adjusted according to the ground object target point and the virtual echo position to obtain target point cloud data for realizing distance detection.

13. The method according to claim 12, characterized in that The determining of the virtual echo positions corresponding to the plurality of echo information points comprises: Obtaining the measurement distances corresponding to each of the plurality of echo information points; Obtaining the current pulse emission time corresponding to each of the plurality of echo information points; Obtaining pulse emission times in the interval corresponding to the plurality of echo information points in the pulse emission interval; Obtaining the time difference between the current pulse emission time and the pulse emission time within the interval; Determining the spatial distances corresponding to the plurality of echo information points according to the time difference and the laser propagation speed; Obtaining a distance and a value between the measured distance corresponding to the echo information point and the spatial distance; The distance and value are determined as virtual echo positions corresponding to each of the plurality of echo information points in each pulse transmission interval.

14. The method according to claim 12, characterized in that The echo information point includes a measured distance, and the determining of the ground object target point corresponding to each of the plurality of echo information points includes: Acquire a plurality of adjacent echo information points adjacent to each of the plurality of data points, wherein the plurality of adjacent echo information points are non-isolated data; Determine, among the plurality of adjacent echo information points, a target adjacent echo information point corresponding to a maximum measurement distance; The target adjacent echo information points corresponding to each of the plurality of echo information points are respectively determined as the ground object target points corresponding to each of the plurality of echo information points.

15. The method according to claim 12, characterized in that The step of adjusting the first mapping relationship according to the ground object target point and the virtual echo position to obtain target point cloud data for distance detection includes: Get the tolerance for distance detection operation; Obtaining the difference between the measured distance of the ground object target point and the virtual echo position; The first mapping relationship is adjusted according to the tolerance and the difference to obtain target point cloud data for distance detection.

Images