An algorithm for eliminating interference from strongly reflective objects in 2D laser reflective column positioning

Abstract

The present application discloses a 2D laser reflective column positioning strong reflective object interference elimination algorithm, including establishing an electronic map of reflective columns, judging whether the scanned reflective columns are formed by strong reflectors, and if so, marking them as abnormal reflective columns; taking the radar as the center of the circle O, the front of the radar as 0 degrees, setting the starting direction A and the ending direction B counterclockwise, and the radius R, in the sector OAB with a radius of R, if an abnormal reflective column is identified, the abnormal reflective column is eliminated. The present application provides a 2D laser reflective column positioning strong reflective object interference elimination algorithm, according to the reflective column formation principle, marking strong reflectors as abnormal points and setting a range to eliminate abnormal reflective columns, effectively preventing strong reflectors from interfering with the reflective column positioning, greatly improving the reflective column positioning adaptability to the application environment, effectively eliminating strong reflectors, stabilizing the reflective column positioning posture output, and improving the accuracy of AGV trolley positioning.

Description

Technical Field

[0001] The present application relates to the technical field of laser reflective column positioning, and in particular to an algorithm for eliminating interference from strongly reflective objects in 2D laser reflective column positioning. Background Art

[0002] With the development of industrial automation and intelligence, the use of automatic guided vehicles (AGVs) in the field of logistics and transportation is becoming more and more widespread. The positioning methods of AGVs are also becoming more and more mature. For scenes with irregular environments and large changes, the use of reflective columns for positioning is more advantageous. However, strongly reflective objects will cause great interference to the positioning of reflective columns. Therefore, the present invention is used to solve the interference caused by strongly reflective objects to the positioning of reflective columns, and improve the adaptability of reflective column positioning to the application environment. Summary of the invention

[0003] The present application proposes an interference elimination algorithm for 2D laser reflective column positioning of strongly reflective objects, which has the advantage of eliminating strong reflective objects and is used to solve the problems raised by the above-mentioned background technology.

[0004] To achieve the above purpose, the present application adopts the following technical solution: a 2D laser reflective column positioning strong reflective object interference elimination algorithm, including:

[0005] (1) Abnormal marking

[0006] Establish an electronic map of reflective columns to determine whether the scanned reflective columns are formed by strong reflectors. If so, mark them as abnormal reflective columns;

[0007] (2) Area elimination

[0008] With the radar as the center O and the direction directly in front of the radar as 0 degrees, the starting direction A and the ending direction B are set counterclockwise, and the radius is R. In the sector OAB with a radius of R, if an abnormal reflective column is identified, the abnormal reflective column will be eliminated.

[0009] Furthermore, the specific steps of the region elimination are as follows:

[0010] Get radar point cloud data;

[0011] Mark the point cloud according to the set parameters R, A, B;

[0012] Obtain reflective columns through radar point cloud;

[0013] Scanned reflective columns are matched with map landmarks;

[0014] The matched landmarks are outliers, and the scanned reflective columns are removed;

[0015] The positioning pose is obtained based on the matching result.

[0016] Further, the algorithm flow is:

[0017] Step 1: Set the maximum scanning range max_dis, and mark the point cloud with a distance greater than max_dis as abnormal;

[0018] Step 2: Set the effective reflection intensity range of the laser point cloud, the minimum reflection intensity l1, the maximum reflection intensity l2, and mark the point cloud that exceeds the reflection intensity range as abnormal;

[0019] Step 3: Set the starting direction A, the ending direction B, and the radius R, and mark the point cloud data within the set sector range as abnormal;

[0020] Step 4: Assume that the relationship function between the distance of the reflective column and the reflection intensity is dis=f(l), where dis is the distance from the reflective column to the laser radar, and l is the reflection intensity;

[0021] Step 5: Calculate the reflective column based on the relationship between the distance and reflective intensity of the point cloud in the laser point cloud and f(l);

[0022] Step 6: Mark the misidentified reflective column as an abnormal reflective column.

[0023] Furthermore, it is applied to reflective columns and radars, wherein a reflective layer is provided on the outer wall of the reflective column, wherein the reflective layer includes a strong reflective surface and a weak reflective surface, a signal receiver is provided on the top of the reflective column, a swivel seat is connected to the bottom end of the reflective column, and an identification action device is connected to the bottom end of the swivel seat.

[0024] Furthermore, the identification action device includes a housing, a motor and an identification controller, the motor and the identification controller are fixedly connected to the inside of the housing, the signal receiver and the motor are both connected to the identification controller, the swivel seat is a bearing, the outer ring of the swivel seat is connected to the reflective column, and the inner ring of the swivel seat is connected to the housing.

[0025] Furthermore, the control system of the radar includes a positioning module and an identification module, the identification module includes a fixed strong reflector identification unit, a mobile strong reflector identification unit and a reflection intensity comparison unit, the positioning module obtains the positioning posture of the radar according to the position of the normal reflective column, and the identification module is used to identify and eliminate strong reflectors; the fixed strong reflector identification unit is used to judge that the reflector is a strong reflector at a fixed position, the mobile strong reflector identification unit is used to judge that the reflector is a mobile strong reflector, and the reflection intensity comparison unit is used to judge the reflection intensity of the reflector according to the reflected laser.

[0026] Furthermore, the specific steps of the region elimination are as follows:

[0027] Get radar point cloud data;

[0028] Mark the point cloud according to the set parameters R, A, B;

[0029] Obtain reflective columns through radar point cloud;

[0030] Scanned reflective columns are matched with map landmarks;

[0031] Determine whether the matched landmark is an abnormal point. If so, determine that the reflective column is a fixed strong reflective column and remove the scanned reflective column;

[0032] If not, the radar will send an identification signal in the direction of the reflective column, and then the radar will send the laser again to obtain the reflected laser and compare it with the intensity of the first reflected laser; if the difference in reflection intensity does not meet the preset threshold, the reflective column is judged to be a moving strong reflective column and the scanned reflective column is removed; if not, the reflective column is judged to be a normal reflective column;

[0033] Finally, the positioning pose is obtained based on the matching results.

[0034] 1. The present application provides an algorithm for eliminating interference from strongly reflective objects in 2D laser reflective column positioning. According to the formation principle of reflective columns, strong reflectors are marked as abnormal points and a set range is used to eliminate abnormal reflective columns, thereby effectively preventing interference from strong reflectors, aluminum alloys, glass, etc. on reflective column positioning, greatly improving the adaptability of reflective column positioning to the application environment, effectively eliminating strong reflectors, stabilizing the output of reflective column positioning posture, and improving the accuracy of AGV vehicle positioning.

[0035] 2. The present application provides an interference elimination algorithm for locating strongly reflective objects using a D laser reflective column. Through the unique structure of the reflective column and radar, the radar transmits an identification signal to the signal receiver, and the motor controls the rotation of the reflective column to change the positions of the strong reflective surface and the weak reflective surface, so that the reflected intensity of the reflected laser received by the radar is different and the difference in reflection intensity is fixed, thereby judging that the reflector is a moving strong reflector or a normal reflective column, thereby realizing the function of identifying moving strong reflectors and further improving the accuracy of AGV positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which constitute a part of the specification, illustrate embodiments disclosed in the present application and, together with the description, serve to explain the principles disclosed in the present application.

[0037] The present disclosure may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0038] Figure 1 This is an algorithm flow chart of Embodiment 1 of the present invention;

[0039] Figure 2 This is a schematic diagram of the recognition principle of the first embodiment of the present invention;

[0040] Figure 3 This is a schematic diagram of the structure of the reflective column and the radar in the second embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of a reflective layer according to a second embodiment of the present invention;

[0042] Figure 5 is a system diagram of a radar in Embodiment 2 of the present invention;

[0043] Figure 6 This is an algorithm flow chart of Embodiment 2 of the present invention.

[0044] In the figure: 1. reflective column; 2. radar; 201. positioning module; 202. identification module; 221. fixed strong reflector identification unit; 222. mobile strong reflector identification unit; 223. reflection intensity comparison unit; 3. reflective layer; 301. strong reflective surface; 302. weak reflective surface; 4. signal receiver; 5. swivel seat; 6. identification action device; 601. box; 602. motor; 603. identification controller. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] Embodiment 1

[0047] A 2D laser reflective column positioning strong reflective object interference elimination algorithm, including:

[0048] (1) Abnormal marking

[0049] Establish an electronic map of reflective columns to determine whether the scanned reflective columns are formed by strong reflectors. If so, mark them as abnormal reflective columns;

[0050] (2) Area elimination

[0051] See attached Figure 2 , with the radar as the center O, the front of the radar as 0 degrees, the starting direction A and the ending direction B are set counterclockwise, and the radius is R. In the sector OAB with a radius of R, if an abnormal reflective column is identified, the abnormal reflective column will be removed. Figure 2 In the figure, the rectangular block on the left within the scanning range is a strong reflector, and the others are normal reflective columns.

[0052] See attached Figure 1 , the specific steps of region elimination are as follows:

[0053] Get radar point cloud data;

[0054] Mark the point cloud according to the set parameters R, A, B;

[0055] Obtain reflective columns through radar point cloud;

[0056] Scanned reflective columns are matched with map landmarks;

[0057] The matched landmarks are outliers, and the scanned reflective columns are removed;

[0058] The positioning pose is obtained based on the matching result.

[0059] The elimination algorithm process is:

[0060] Step 1: Set the maximum scanning range max_dis, and mark the point cloud with a distance greater than max_dis as abnormal;

[0061] Step 2: Set the effective reflection intensity range of the laser point cloud, the minimum reflection intensity l1, the maximum reflection intensity l2, and mark the point cloud that exceeds the reflection intensity range as abnormal;

[0062] Step 3: Set the starting direction A, the ending direction B, and the radius R, and mark the point cloud data within the set sector range as abnormal;

[0063] Step 4: Assume that the relationship function between the distance of the reflective column and the reflection intensity is dis=f(l), where dis is the distance from the reflective column to the laser radar, and l is the reflection intensity;

[0064] Step 5: Calculate the reflective column based on the relationship between the distance and reflective intensity of the point cloud in the laser point cloud and f(l);

[0065] Step 6: Mark the misidentified reflective column as an abnormal reflective column.

[0066] Embodiment 2

[0067] Embodiment 1 can solve the problem of identifying and eliminating strong reflectors that are located in fixed positions for a long time in the venue. However, in actual applications, there may be movable objects that exist briefly in the venue. Since the movable strong reflectors are not marked on the electronic map, the movable strong reflectors cannot be eliminated. Therefore, the present application proposes embodiment 2 to solve the problem of eliminating movable strong reflectors.

[0068] See attached Figure 3 The algorithm is applied to the reflective column 1 and the radar 2. The outer wall of the reflective column 1 is provided with a reflective layer 3. Figure 3 and attached Figure 4The reflective layer 3 includes a strong reflective surface 301 and a weak reflective surface 302, which are set according to the areas where the strong reflective surface 301 and the weak reflective surface 302 are laid, and the angle between the strong reflective surface 301 and the weak reflective surface 302 in the fixed direction of the reflective column 1 under normal circumstances. That is, assuming that the strong reflective surface 301 and the weak reflective surface 302 each occupy half of the outer surface area of ​​the reflective column 1, under normal circumstances, the middle of the strong reflective surface 301 faces the road side, then the reflective column 1 rotates 45 degrees to make the middle of the weak reflective surface 302 face the road side, if the radar 2 is on the road at this time, the radar 2 emits lasers twice respectively, then before and after the reflective column 1 rotates, the laser emitted by the radar 2 is reflected by the strong reflective surface 301 and the weak reflective surface 302 respectively, because the strong reflective surface 301 and the weak reflective surface 302 have different abilities to reflect lasers, the intensity of the reflected laser obtained by the radar 2 is also different, and the difference between the two should be fixed. A signal receiver 4 is disposed on the top of the reflective column 1. The signal receiver 4 has the function of receiving signals in all directions. The signal receiver 4 may also be a laser receiver. A rotating seat 5 is connected to the bottom of the reflective column 1. An identification action device 6 is connected to the bottom of the rotating seat 5.

[0069] The identification action device 6 includes a box body 601, a motor 602 and an identification controller 603. The motor 602 and the identification controller 603 are fixedly connected inside the box body 601. The signal receiver 4 and the motor 602 are both connected to the identification controller 603. The swivel seat 5 is a bearing. The outer ring of the swivel seat 5 is connected to the reflective column 1, and the inner ring of the swivel seat 5 is connected to the box body 601. When the rotating shaft of the motor 602 rotates, it can drive the reflective column 1 to rotate. The swivel seat 5 connects and supports the reflective column 1 and the identification action device 6.

[0070] See attached Figure 5 The control system of the radar 2 includes a positioning module 201 and an identification module 202. The identification module 202 includes a fixed strong reflector identification unit 221, a mobile strong reflector identification unit 222 and a reflection intensity comparison unit 223. The positioning module 201 obtains the positioning posture of the radar 2 according to the position of the normal reflective column. The identification module 202 is used to identify and eliminate strong reflectors; the fixed strong reflector identification unit 221 is used to determine that the reflector is a strong reflector at a fixed position, the mobile strong reflector identification unit 222 is used to determine that the reflector is a mobile strong reflector, and the reflection intensity comparison unit 223 is used to determine the reflection intensity of the reflector according to the reflected laser.

[0071] See attached Figure 6 ,The specific steps of region elimination are as follows:

[0072] Get radar point cloud data;

[0073] Mark the point cloud according to the set parameters R, A, B;

[0074] Obtain reflective columns through radar point cloud;

[0075] Scanned reflective columns are matched with map landmarks;

[0076] Determine whether the matched landmark is an abnormal point. If so, determine that the reflective column is a fixed strong reflective column and remove the scanned reflective column;

[0077] If not, the radar will send an identification signal in the direction of the reflective column, and then the radar will send the laser again to obtain the reflected laser and compare it with the intensity of the first reflected laser; if the difference in reflection intensity does not meet the preset threshold, the reflective column is judged to be a moving strong reflective column and the scanned reflective column is removed; if not, the reflective column is judged to be a normal reflective column;

[0078] Finally, the positioning pose is obtained based on the matching results.

[0079] The function of the fixed strong reflector identification unit 221 in this embodiment is the function of identifying and removing fixed strong reflectors in Embodiment 1. The idea of ​​identifying the mobile strong reflector by the mobile strong reflector identification unit 222 is that when the current reflective column is not a fixed strong reflector, it is necessary to determine whether the reflective column is a mobile strong reflector, and transmit an identification signal in the direction of the reflective column 1 through the radar 2. The identification signal is received by the signal receiver 4 and then transmitted to the identification controller 603. The identification controller 603 controls the rotation shaft of the motor 602 to rotate a preset angle so that the positions of the strong reflective surface 301 and the weak reflective surface 302 are swapped. Then the radar 2 transmits the laser signal again. The signal is reflected by the swapped reflective surface and received by the radar 2. The reflection intensity comparison unit 223 determines and compares the intensity of the reflected laser before the first swap of the reflective surface. If the laser intensity difference is in the preset area, it indicates that the reflector is a normal reflective column. In summary, in this application, a normal reflector can rotate, so that before and after the reflector rotates, the same laser emitted by the radar has different reflection intensities twice, and the difference in reflection intensity is a fixed value, which is a normal reflector. When the reflector is neither a fixed strong reflector nor a normal reflector, the reflector is a moving strong reflector and is removed.

Claims

1. A 2D laser reflective column positioning strong reflective object interference elimination algorithm, characterized in that: include: 1) Abnormal marking: Establish an electronic map of reflective columns to determine whether the scanned reflective columns are formed by strong reflectors. If so, mark them as abnormal reflective columns; 2) Area elimination: With the radar as the center O, the front of the radar as 0 degrees, the starting direction A and the ending direction B are set counterclockwise, and the radius is R. In the sector OAB with a radius of R, if an abnormal reflective column is identified, the abnormal reflective column will be eliminated; The 2D laser reflective column positioning strong reflective object interference elimination algorithm is applied to a reflective column (1) and a radar (2); a reflective layer (3) is provided on the outer wall of the reflective column (1); the reflective layer (3) comprises a strong reflective surface (301) and a weak reflective surface (302); a signal receiver (4) is provided on the top of the reflective column (1); a rotating seat (5) is connected to the bottom end of the reflective column (1); and an identification action device (6) is connected to the bottom end of the rotating seat (5); The identification action device (6) comprises a housing (601), a motor (602) and an identification controller (603); the motor (602) and the identification controller (603) are fixedly connected inside the housing (601); the signal receiver (4) and the motor (602) are both connected to the identification controller (603); the rotating seat (5) is a bearing; the outer ring of the rotating seat (5) is connected to the reflective column (1); and the inner ring of the rotating seat (5) is connected to the housing (601); The control system of the radar (2) comprises a positioning module (201) and an identification module (202); the identification module (202) comprises a fixed strong reflector identification unit (221), a mobile strong reflector identification unit (222) and a reflection intensity comparison unit (223); the positioning module (201) obtains the positioning posture of the radar (2) according to the position of a normal reflective column; the identification module (202) is used to identify and remove strong reflectors; the fixed strong reflector identification unit (221) is used to judge that the reflector is a strong reflector at a fixed position; the mobile strong reflector identification unit (222) is used to judge that the reflector is a mobile strong reflector; and the reflection intensity comparison unit (223) is used to judge the reflection intensity of the reflector according to the reflected laser.

2. According to claim 1, a 2D laser reflective column positioning strong reflective object interference elimination algorithm is characterized in that: The specific steps of the region removal are as follows: Get radar point cloud data; Mark the point cloud according to the set parameters R, A, B; Obtain reflective columns through radar point cloud; Scanned reflective columns are matched with map landmarks; The matched landmarks are outliers, and the scanned reflective columns are removed; The positioning pose is obtained based on the matching result.

3. According to claim 1, a 2D laser reflective column positioning strong reflective object interference elimination algorithm is characterized in that: The algorithm flow is: Step 1: Set the maximum scanning range max_dis, and mark the point cloud with a distance greater than max_dis as abnormal; Step 2: Set the effective reflection intensity range of the laser point cloud, the minimum reflection intensity l1, the maximum reflection intensity l2, and mark the point cloud that exceeds the reflection intensity range as abnormal; Step 3: Set the starting direction A, the ending direction B, and the radius R, and mark the point cloud data within the set sector range as abnormal; Step 4: Assume that the relationship function between the distance of the reflective column and the reflection intensity is dis=f(l), where dis is the distance from the reflective column to the laser radar, and l is the reflection intensity; Step 5: Calculate the reflective column based on the relationship between the distance and reflective intensity of the point cloud in the laser point cloud and f(l); Step 6: Mark the misidentified reflective column as an abnormal reflective column.

4. The interference elimination algorithm for 2D laser reflective column positioning with strong reflective objects according to claim 1, characterized in that: The specific steps of the region removal are as follows: Get radar point cloud data; Mark the point cloud according to the set parameters R, A, B; Obtain reflective columns through radar point cloud; Scanned reflective columns are matched with map landmarks; Determine whether the matched landmark is an abnormal point. If so, determine that the reflective column is a fixed strong reflective column and remove the scanned reflective column; If not, the radar will send an identification signal in the direction of the reflective column, and then the radar will send the laser again to obtain the reflected laser and compare it with the intensity of the first reflected laser; if the difference in reflection intensity does not meet the preset threshold, the reflective column is judged to be a moving strong reflective column and the scanned reflective column is removed; if not, the reflective column is judged to be a normal reflective column; Finally, the positioning pose is obtained based on the matching results.

Images