Ultra-wide-angle laser radar device and granary stitching mapping method

Abstract

The embodiment of the present invention discloses an ultra-wide-angle laser radar device and a granary stitching mapping method. The method is applied to a granary stitching mapping system that includes at least a plurality of radar devices disposed in a granary, each radar device includes at least two radar modules, and the method includes: obtaining the current echo data and the last echo data collected by each radar module; performing anti-interference processing using a preset anti-interference rule, the current echo data, and the last echo data to obtain the target echo data of each radar module; converting the target echo data of each radar module into the first point cloud data of each radar module; performing point cloud stitching processing based on the first point cloud data of multiple radar modules to obtain a complete granary environment map. In the above manner, multiple radar modules are used to increase the perception field of the radar device, and anti-interference processing is performed on the echo data to reduce dust interference, thereby improving the efficiency and accuracy of mapping.

Description

Technical Field

[0001] The present invention relates to the field of granary automation technology, and in particular to an ultra-wide-angle laser radar device and a granary splicing mapping method. Background Art

[0002] The grain storage and warehousing operation is one of the key links to ensure grain quality. In order to break through the limitations of traditional manual operations, domestic and foreign research institutions have been committed to the development of automated warehouse-cleaning robots in recent years, in order to achieve intelligent operations for grain surface leveling. It can be seen that promoting the technology related to automated warehouse-cleaning has important practical significance.

[0003] The ability to perceive the grain silo environment is crucial for the warehouse-closing robot. The first step to achieve automatic navigation is to build a grain surface environment map. In addition, the effectiveness of the robot's automatic warehouse-closing operation also depends largely on its perception of the grain silo environment.

[0004] However, at present, there are many shortcomings in the grain silo perception mapping technology. For example, it is impossible to construct the environment of the entire grain surface; there is a lot of dust in the silo, which has a serious impact on the mapping results, resulting in insufficient mapping efficiency and accuracy. Summary of the invention

[0005] The main purpose of the present invention is to provide an ultra-wide-angle laser radar device and a granary stitching mapping method, which can solve the problems of insufficient mapping efficiency and accuracy in the prior art.

[0006] To achieve the above-mentioned object, the first aspect of the present invention provides a granary stitching and mapping method, the method is applied to a granary stitching and mapping system, the system at least comprises a plurality of radar devices separately arranged in the granary, the radar device is an ultra-wide-angle laser radar device, each radar device comprises at least two radar modules, the method comprises:

[0007] Acquire the current echo data and the last echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary;

[0008] Use the preset anti-interference rules, current echo data and last echo data to perform anti-interference processing to obtain the target echo data of each radar module;

[0009] Convert the target echo data of each radar module into the first point cloud data of each radar module;

[0010] Point cloud stitching is performed based on the first point cloud data of multiple radar modules to obtain a complete granary environment map.

[0011] In a feasible implementation, the echo data at least includes the reflection intensity of the echo, and the anti-interference rule includes a preset reflection intensity ratio threshold. Then, the preset anti-interference rule, the current echo data, and the last echo data are used to perform anti-interference processing to obtain the target echo data of each radar module, including:

[0012] For each radar module:

[0013] determining a reflection intensity ratio between a first reflection intensity of a previous echo and a second reflection intensity of a current echo;

[0014] If the reflection intensity ratio is greater than a preset intensity ratio threshold, determining the target echo data as current echo data;

[0015] If the reflection intensity ratio is less than or equal to a preset intensity ratio threshold, it is determined that the target echo data is the previous echo data.

[0016] In a feasible implementation, the echo data further includes a propagation distance of the echo, and the propagation distance is used to reflect the distance between the radar module and the surface of the object; then the method further includes:

[0017] The reflection intensity of the echo is obtained by using the propagation distance of the echo and a preset reflection intensity algorithm.

[0018] In a feasible implementation, obtaining the reflection intensity of the echo by using the propagation distance of the echo and a preset reflection intensity algorithm includes:

[0019] Obtain the target grain type and current dust concentration stored in the current granary;

[0020] Determine the target reflectivity corresponding to the target grain type by using the correspondence between the preset grain type and the reflectivity of the grain surface and the target grain type;

[0021] Determine the target attenuation coefficient corresponding to the current dust concentration by using the preset correspondence between the dust concentration and the attenuation coefficient and the current dust concentration;

[0022] The reflection intensity of the echo is obtained according to the target reflectivity, target attenuation coefficient, propagation distance and reflection intensity algorithm.

[0023] In a feasible implementation, the first point cloud data generated by the multiple radar modules are spliced ​​to obtain a complete granary environment map, including:

[0024] Using two sets of first point cloud data of two radar modules in each radar device, a first rotation matrix between the two radar modules, and a first translation vector to perform point cloud registration between the radar modules, to obtain second point cloud data of each radar device;

[0025] Using the second point cloud data of each radar device, the second rotation matrix between the two radar devices, and the second translation vector, point cloud registration between the radar devices is performed to obtain target point cloud data;

[0026] Point cloud stitching is performed according to the target point cloud data and a preset weighted average algorithm to generate the granary environment map.

[0027] In a feasible implementation, the step of using the second point cloud data of each radar device, the second rotation matrix between the two radar devices, and the second translation vector to perform point cloud registration between the radar devices to obtain target point cloud data includes:

[0028] Using the second point cloud data of each radar device and a preset ICP objective function based on the warehouse wall plane constraint to perform an optimal transformation solution to obtain the second rotation matrix and the second translation vector;

[0029] Coordinate transformation is performed based on the second rotation matrix, the second translation vector and the second point cloud data of each radar device to obtain target point cloud data.

[0030] To achieve the above-mentioned object, the second aspect of the present invention provides an ultra-wide-angle laser radar device, wherein a plurality of the ultra-wide-angle laser radar devices are separately arranged in a granary, and the ultra-wide-angle laser radar device comprises at least two radar modules, wherein the radar modules are used to collect current echo data and previous echo data, and report them to a granary stitching and mapping system; the echo data is the data of the echo after the transmission signal of the radar module is reflected by the surface of an object in the granary;

[0031] The granary stitching and mapping system is used to receive the current echo data and the previous echo data, and execute the steps shown in the first aspect and any feasible implementation method.

[0032] To achieve the above-mentioned object, the third aspect of the present invention provides a granary stitching and mapping device, which is applied to a granary stitching and mapping system. The system at least includes a plurality of radar devices separately arranged in the granary, each radar device includes at least two radar modules, and the device includes:

[0033] Data acquisition module: used to obtain the current echo data and the last echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary;

[0034] Interference processing module: used to perform anti-interference processing using preset anti-interference rules, current echo data and last echo data to obtain target echo data of each radar module;

[0035] Data conversion module: used to convert the target echo data of each radar module into the first point cloud data of each radar module;

[0036] Point cloud modeling module: used to perform point cloud stitching processing based on the first point cloud data of multiple radar modules to obtain a complete granary environment map.

[0037] To achieve the above-mentioned purpose, the fourth aspect of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps shown in the first aspect and any feasible implementation method.

[0038] To achieve the above-mentioned purpose, the fifth aspect of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps shown in the first aspect and any feasible implementation method.

[0039] The embodiments of the present invention have the following beneficial effects:

[0040] The present invention provides a granary splicing mapping method, which is applied to a granary splicing mapping system. The system includes at least a plurality of radar devices disposed in the granary, each radar device includes at least two radar modules, and the method includes: obtaining the current echo data and the last echo data collected by each radar module, the echo data being the data of the echo after the transmission signal of the radar module is reflected by the surface of an object in the granary; performing anti-interference processing using a preset anti-interference rule, the current echo data, and the last echo data to obtain the target echo data of each radar module; converting the target echo data of each radar module into the first point cloud data of each radar module; performing point cloud splicing processing according to the first point cloud data of multiple radar modules to obtain a complete granary environment map. In the above manner, a radar device is arranged in the granary for granary splicing mapping, each radar device is arranged with at least two radar modules, the radar perception field of view is increased, and the radar echo data is subjected to anti-interference processing using an anti-interference rule to reduce dust interference, thereby improving the mapping efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] in:

[0043] Figure 1 This is a flow chart of a method for splicing and mapping a granary in an embodiment of the present invention;

[0044] FIG2 (a) is a schematic diagram of the installation of a radar module according to an embodiment of the present invention;

[0045] FIG2( b ) is a schematic diagram of the installation of a radar device according to an embodiment of the present invention;

[0046] Figure 3 Another flow chart of a method for splicing and mapping a granary in an embodiment of the present invention;

[0047] Figure 4 A schematic diagram showing a comparison of point clouds with and without anti-interference processing in an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of a granary environment map in an embodiment of the present invention;

[0049] Figure 6 This is a structural block diagram of a granary splicing and mapping device in an embodiment of the present invention;

[0050] Figure 7 2 is a structural block diagram of a computing power module in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0052] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for splicing and mapping a granary in an embodiment of the present invention. Figure 1 The method is applied to a granary stitching and mapping system, the system at least comprising a plurality of radar devices separately arranged in the granary, each radar device comprising at least two radar modules, and the method comprising the following steps:

[0053] 101. Acquire current echo data and previous echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary;

[0054] It should be noted that the method shown in the present application is applied to a granary splicing and mapping system, and the system at least includes a method applied to a granary splicing and mapping system, and the system at least includes a plurality of radar devices disposed in the granary, and each radar device includes at least two radar modules. Further, the system may also include a central control unit, and the radar modules are all communicated with the central control unit, and the central control unit is used to execute the steps of the method of the present application. The central control unit may be a terminal or a server. The terminal may specifically be a desktop terminal or a mobile terminal, and the mobile terminal may specifically be at least one of a mobile phone, a tablet computer, a laptop computer, and the like. The server may be implemented as an independent server or a server cluster consisting of multiple servers. The method can be applied to both terminals and servers, and this embodiment is illustrated by applying to a terminal.

[0055] In order to realize the perception of the granary environment, multiple radar devices are pre-installed in the granary. In order to increase the perception field of the radar device, at least two radar modules are set in each radar device, so that the perception field of each radar device can have an ultra-wide-angle field of view, and the granary environment can be perceived in a larger angle and a wide range, wherein the radar device includes but is not limited to a laser radar device, and the radar module includes but is not limited to a laser radar module. That is, the radar device shown in the present application is an ultra-wide-angle radar device, and multiple ultra-wide-angle laser radar devices are arranged in the granary. The ultra-wide-angle laser radar device includes at least two radar modules, and the radar module is used to collect the current echo data and the last echo data, and report them to the granary splicing mapping system; the echo data is the data of the echo after the transmission signal of the radar module is reflected by the surface of the object in the granary; the granary splicing mapping system is used to receive the current echo data and the last echo data, and execute the steps of the granary splicing mapping method shown in the present application.

[0056] For example, please refer to Figure 2 (a), which is a schematic diagram of the installation of a radar module in an embodiment of the present invention. As shown in Figure 2 (a), two radar modules 20 are installed in the radar device 10, wherein, as shown in Figure 2 (a) (i) is a top view of the radar device 10, as shown in Figure 2 (a) (i), the top horizontal FOV of the radar module 20 is 120°, the viewing angles of the two radar modules 20 overlap by 20°, and the top horizontal FOV of the radar device 10 is 200°, achieving an ultra-wide-angle field of view, and the A-direction view of the radar device is shown in Figure 2 (a) (ii), as shown in Figure 2 (a) (ii), the vertical field of view of the radar module 20 is 90°, that is, the vertical field of view of the radar device 10 is 90°, and finally the overall FOV of the laser radar can reach: horizontal 200°, vertical 90°, and the viewing angle of the radar module 20 overlaps by 20°; the downward tilt angle of the radar is , with a value range of 40~50°; the angle between the two radar modules in the radar device is , The value range is 60°~80°. In this way, the device can form a laser radar scanning range with a horizontal viewing angle of 200° and a vertical viewing angle of 90°, which is more than twice the viewing angle of the existing single radar, greatly improving the grain surface environment perception ability of the warehouse-closing robot and other types of grain warehouse robots, enabling the robot to have the ability to perceive the grain surface environment of the entire warehouse.

[0057] In addition, the upper and lower shells of the radar device are equipped with heat sinks, which can conduct heat from the laser radar through physical heat conduction to ensure the normal working temperature of the device. This avoids the dust explosion problem caused by using cooling fans and other methods. The top surface of the device is the installation location, which can be installed on the top or wall of the warehouse.

[0058] In order to realize the modeling of the granary, the radar device can be installed in the granary position such as the granary bulkhead and the granary roof, and the number of installed radar devices can be increased or decreased according to the adaptability of the granary area. For example, a standard bungalow (20 meters wide and 60 meters long) needs to install 4 laser radar devices, and a small bungalow (20 meters wide and 30 meters long) needs to install 2 laser radar devices. Please refer to Figure 2 (b). Figure 2 (b) is a schematic diagram of the installation of a radar device in an embodiment of the present invention. Figure 2 (b) exemplarily shows a feasible installation scheme. Figure 2 (b) shows that the width of the granary is 20 to 30 meters and the length is 60 meters. Four radar devices 10 can be installed in the granary with this area. In Figure 2 (b), a radar device 10 is set every 15 meters in the length direction.

[0059] It can be understood that the present application sets up a laser radar device in the granary whose field of view can cover the entire granary, and uses the laser radar device to model the environment. The laser radar determines the distance of the object by emitting laser pulses and measuring the time it takes for the reflected signal to return, thereby generating three-dimensional point cloud data containing rich geometric information. In order to obtain more accurate point cloud data, the present application collects echo data twice, and obtains the current echo data and the previous echo data collected by each radar module through step 101, wherein the echo data is the data of the echo after the transmitted signal of the radar module is reflected by the surface of the object in the granary. The transmitted signal can be a laser radar signal, wherein the signal will be reflected when it encounters the surface of the object and then received by the laser radar device, and the echo is collected to obtain the echo data. The echo data at least includes the propagation distance of the echo, the reflection intensity, etc., which are not limited here.

[0060] 102. Perform anti-interference processing using a preset anti-interference rule, current echo data, and last echo data to obtain target echo data of each radar module;

[0061] It should be noted that there is a lot of dust in the granary, which affects the radar echo. Therefore, after obtaining the current echo data and the previous echo data, the preset anti-interference rules, the current echo data and the previous echo data are used for anti-interference processing to obtain the target echo data of each radar module. The anti-interference rules are used to determine whether the echo is reflected from the surface of the granary object or the dust. Valid target echo data is obtained through the anti-interference rules. It can be understood that the echo data of each radar model needs to be processed with anti-interference to obtain the target echo data of each radar module.

[0062] 103. Convert the target echo data of each radar module into the first point cloud data of each radar module;

[0063] 104. Perform point cloud stitching processing based on the first point cloud data of multiple radar modules to obtain a complete granary environment map.

[0064] After obtaining effective and accurate target echo data, the target echo data of each radar module can be converted into the first point cloud data of each radar module. The point cloud data is used to reflect the relative position relationship between the object in the granary and the above-mentioned radar module. Further, point cloud splicing processing is performed based on the first point cloud data of multiple radar modules to obtain a complete granary environment map. The granary environment map can be a grain surface environment map in the granary, which is not limited here. Among them, when the point cloud splicing processing is performed, the two sets of point cloud data in the coordinate system of the two radar modules in the same radar device can be first converted to the same radar module coordinate system to obtain the point cloud coordinate representation in the same coordinate system, and the point cloud coordinate representation in the same coordinate system is used as the point cloud data of the radar device corresponding to the two radar modules; then the point cloud data of each radar device is converted to the same radar device coordinate system to obtain the overall point cloud data of the granary, complete the point cloud registration, and use the registered point cloud for splicing to obtain the final complete granary environment map. The splicing method includes but is not limited to the weighted average of the point cloud.

[0065] The present invention provides a granary splicing mapping method, which is applied to a granary splicing mapping system. The system includes at least a plurality of radar devices disposed in the granary, each radar device includes at least two radar modules, and the method includes: obtaining the current echo data and the last echo data collected by each radar module, the echo data being the data of the echo after the transmission signal of the radar module is reflected by the surface of an object in the granary; performing anti-interference processing using a preset anti-interference rule, the current echo data, and the last echo data to obtain the target echo data of each radar module; converting the target echo data of each radar module into the first point cloud data of each radar module; performing point cloud splicing processing according to the first point cloud data of multiple radar modules to obtain a complete granary environment map. In the above manner, a radar device is arranged in the granary for granary splicing mapping, each radar device is arranged with at least two radar modules, the radar perception field of view is increased, and the radar echo data is subjected to anti-interference processing using an anti-interference rule to reduce dust interference, thereby improving the mapping efficiency and accuracy.

[0066] See also Figure 3 , Figure 3 FIG. 4 is another flow chart of a method for mapping a granary using splicing in an embodiment of the present invention. Figure 3 The method is applied to a granary stitching and mapping system, the system at least comprising a plurality of radar devices separately arranged in the granary, each radar device comprising at least two radar modules, and the method comprising the following steps:

[0067] 301. Acquire current echo data and previous echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary;

[0068] It should be noted that step 301 and Figure 1 The content of step 101 is similar, so it will not be described here to avoid repetition. For details, please refer to Figure 1 Step 101 content.

[0069] In a feasible implementation, in order to reduce dust interference, the present application judges dust interference by the reflection intensity of the echo to reduce dust interference, wherein the echo data at least includes the reflection intensity of the echo, and the anti-interference rule includes a preset reflection intensity ratio threshold. Then, the preset anti-interference rule, the current echo data and the previous echo data are used to perform anti-interference processing to obtain the target echo data of each radar module, including the following steps 302 to 304.

[0070] It is understandable that each radar module will receive an echo, and the echo data of each radar module will be processed through the following steps 302 to 304 to reduce dust interference.

[0071] 302. Determine a reflection intensity ratio between a first reflection intensity of a previous echo and a second reflection intensity of a current echo;

[0072] The first reflection intensity of the last echo of the radar module and the second reflection intensity of the current echo are used to obtain the ratio of the reflection intensity between the last echo and the current echo. This ratio is used as the reflection intensity ratio. The reflection intensity ratio is used to reflect the degree of change of the current echo compared with the previous echo. If the change is large, it means that the current echo has encountered dust. Otherwise, a small change means that the possibility of encountering dust is small.

[0073] In a feasible implementation, in order to further improve the anti-interference effect, the reflection intensity of the echo can be calculated by the propagation distance of the echo. Therefore, the echo data can also include the propagation distance of the echo. The propagation distance is used to reflect the distance between the radar module and the surface of the object. The method also includes: using the propagation distance of the echo and a preset reflection intensity algorithm to obtain the reflection intensity of the echo.

[0074] The reflection intensity of the echo is obtained through the reflection intensity algorithm and the propagation distance of the echo. It can be understood that the corresponding reflection intensity of each echo can be calculated in the above way. No matter it is the current echo or the previous echo, the reflection intensity of the echo can be obtained by using the propagation distance of the echo and the preset reflection intensity algorithm.

[0075] In a feasible implementation, the radar transmission signal is reflected by different object surfaces with different intensities and is also affected by dust. Therefore, in order to further improve the accuracy of the reflection intensity calculation result and reduce the impact of the reflection environment, the propagation distance of the echo and the preset reflection intensity algorithm are used to obtain the reflection intensity of the echo, including steps A01 to A04:

[0076] A01. Obtain the target grain type and current dust concentration stored in the current granary;

[0077] It should be noted that the radar transmission signal is reflected by different grain surfaces with different intensities. Grains with smooth surfaces (such as corn) have high reflection intensity, while those with rough surfaces (such as wheat) have low reflection intensity. The dust concentration in the granary will also affect the reflection intensity. Therefore, in order to improve the accuracy of the reflection intensity calculation results, it is necessary to obtain the target grain type and current dust concentration stored in the current granary. The correspondence between each granary and the grain type can be obtained in the preset granary information database, from which the target grain type stored in the granary being modeled can be obtained. In addition, a dust concentration sensor can be installed on the radar device to collect the current dust concentration in real time through the sensor to obtain the current dust concentration.

[0078] A02. Determine the target reflectivity corresponding to the target grain type by using the preset correspondence between the grain type and the reflectivity of the grain surface and the target grain type;

[0079] After obtaining the target grain type, the preset grain type and grain surface reflectivity can be used to ρ The corresponding relationship is used to obtain the target reflectivity corresponding to the target food type. ρ Specifically, the preset grain type and grain surface reflectivity can be ρ Find the target reflectivity corresponding to the target food type in the corresponding relationship ρ .

[0080] For example, please refer to Table 1, which shows the corresponding relationship between several types of grains and the reflectivity of the grain surface:

[0081] Table 1

[0082]

[0083] A03. Determine the target attenuation coefficient corresponding to the current dust concentration by using the preset correspondence between the dust concentration and the attenuation coefficient and the current dust concentration;

[0084] After the current dust concentration is obtained, the target attenuation coefficient corresponding to the current dust concentration can be obtained by using the preset correspondence between the dust concentration and the attenuation coefficient.

[0085] A04. Obtain the reflection intensity of the echo according to the target reflectivity, target attenuation coefficient, propagation distance and reflection intensity algorithm.

[0086] Finally, according to the target reflectivity , target attenuation coefficient , Propagation distance And the reflection intensity algorithm is used to obtain the reflection intensity of the echo. It can be understood that the reflection intensity of each echo can be calculated according to steps A01 to A04. The reflection intensity algorithm is an echo model based on physical characteristics, which is established based on the significant differences in the reflection and penetration characteristics of different grain types to lasers. The algorithm parameters can be dynamically adjusted according to different dust concentrations and grain types to improve the accuracy of the calculation results.

[0087] Exemplarily, the reflection intensity algorithm is as follows:

[0088] ;

[0089] In the formula, is the reflection intensity, which can also be expressed as I ; is the system gain coefficient; is the grain surface reflectivity; is the echo propagation distance, which is used to indicate the distance between the LiDAR module and the target surface. is the attenuation coefficient (related to dust concentration).

[0090] 303. If the reflection intensity ratio is greater than a preset intensity ratio threshold, determining the target echo data as current echo data;

[0091] 304. If the reflection intensity ratio is less than or equal to a preset intensity ratio threshold, determining that the target echo data is the previous echo data;

[0092] Furthermore, the principle of the enhanced anti-dust interference method is that when the laser radar receives multiple echoes, the intensity difference between the dust (first echo) and the real target (subsequent echo) is significant. By comparing the echo intensity ratio, dust interference is eliminated. Specifically, the reflection intensity ratio of the echo is compared with the preset intensity ratio threshold. If the reflection intensity ratio is greater than the preset intensity ratio threshold, it is considered that the current echo data is the real target, and the target echo data is determined to be the current echo data; conversely, if the reflection intensity ratio is less than or equal to the preset intensity ratio threshold, it is considered that the current echo data is not the real target, and the last echo is continued as the current echo, that is, the target echo data is determined to be the last echo data.

[0093] Exemplarily, the anti-interference rule includes the following anti-interference algorithm:

[0094] ;

[0095] In the formula, is the intensity ratio threshold, which can be referred to in Table 1: Different types of grains have different intensity ratio intervals, and the intensity ratio threshold can be designed according to the intensity ratio interval; is the reflection intensity of the current echo; is the reflection intensity of the last echo; is the propagation distance of the current echo.

[0096] For example, Figure 4 is a schematic diagram of comparison of point clouds with and without anti-interference processing in an embodiment of the present invention, Figure 4 The middle picture (1) is the point cloud without the anti-dust method, that is, the point cloud without anti-interference processing. Figure 4 The middle figure (2) shows the point cloud with the anti-dust method, that is, the point cloud with anti-interference processing. Through the lidar scanning test in a high dust environment, the results show that the lidar can still maintain 85% point cloud efficiency when the dust coverage rate reaches 70%.

[0097] 305. Convert the target echo data of each radar module into the first point cloud data of each radar module;

[0098] 306. Perform point cloud stitching processing based on the first point cloud data of multiple radar modules to obtain a complete granary environment map.

[0099] It should be noted that steps 305 and 306 are Figure 1 Steps 103 and 104 are similar in content and will not be described here to avoid repetition. Figure 1 Steps 103 and 104 content.

[0100] In a feasible implementation, step 306 may include steps B01 to B03:

[0101] B01. performing point cloud registration between the radar modules using two sets of first point cloud data of two radar modules in each radar device, a first rotation matrix between the two radar modules, and a first translation vector to obtain second point cloud data of each radar device;

[0102] It can be understood that the point cloud data in different coordinate systems need to be aligned and converted to the same coordinate system. The present application first performs point cloud alignment between radar modules in the same radar device, and uses two sets of first point cloud data of two radar modules in each radar device, the first rotation matrix between the two radar modules, and the first translation vector to perform point cloud alignment between the radar modules to obtain the second point cloud data of each radar device.

[0103] The point cloud data in the radar module No. 1 coordinate system can be converted to the radar module No. 2 coordinate system, and the point cloud data in the radar module No. 2 coordinate system can also be converted to the radar module No. 1 coordinate system. There is no limitation here. For example, the point cloud data in the radar module No. 1 coordinate system can be converted to the radar module No. 2 coordinate system. The following content can be referred to:

[0104] The two radars in a single device are stitched together, and the point cloud of radar 2 is converted to radar 1 through rotation matrix and translation. The rotation matrix It is expressed as:

[0105] ;

[0106] In the formula, is the downward tilt angle of the radar module, is the angle between the two radar modules in the device;

[0107] Translation Vector for:

[0108] ;

[0109] In the formula, l is the relative displacement between radar modules.

[0110] Finally, the fusion of the two radars in the device is achieved by the following formula:

[0111] ;

[0112] Where P 1 It is the converted point cloud of radar module No. 2 converted to the coordinate system of radar module No. 1; P 2 This is the original point cloud of radar module No.2.

[0113] B02. performing point cloud registration between the radar devices using the second point cloud data of each radar device, the second rotation matrix between the two radar devices, and the second translation vector to obtain target point cloud data;

[0114] It should be noted that after obtaining the point cloud of a single radar device, multiple radar point clouds are spliced, and a high-precision three-dimensional grain surface of the entire warehouse is constructed through the warehouse wall plane constraint. In order to improve the registration accuracy, the present application uses an ICP registration algorithm based on the warehouse wall plane constraint. Exemplarily, step B02 includes: using the second point cloud data of each radar device and the preset ICP objective function based on the warehouse wall plane constraint to perform the optimal transformation solution to obtain the second rotation matrix and the second translation vector; based on the second rotation matrix, the second translation vector and the second point cloud data of each radar device, coordinate transformation is performed to obtain target point cloud data.

[0115] That is, the optimal second rotation matrix and second translation vector are solved by the ICP objective function based on the silo plane constraint. For example, as shown in FIG2(b), the length of the silo is L and the width is D. Taking any two radar devices as an example, the coordinate systems of radar device A and radar device B in the silo are { O A , X A, Y A , Z A}and{ O B , X B , Y B , Z B}, through the rotation matrix and translation vectors Convert the point cloud of radar B to the coordinate system of radar A:

[0116] Rotation Matrix It can be expressed as:

[0117] ;

[0118] Translation Vector for:

[0119] ;

[0120] The point cloud transformation between devices is expressed as:

[0121] ;

[0122] Among them, P B is the point cloud of radar B, P A This is the converted point cloud. The point clouds of different radars in the granary are converted to the same coordinate system.

[0123] The plane constraint of the warehouse wall is further used to align the radar point cloud and fit the plane equations of each warehouse wall:

[0124] The radar device installed on the roof of the silo can be used to scan the silo wall to obtain point cloud data , the plane equation of the silo wall is directly determined by the geometric dimensions of the silo (length, width, and height).

[0125] ( k =12,…, K );

[0126] Where:

[0127] is the normal vector (unit vector) of the kth warehouse wall, pointing into the warehouse.

[0128] is the offset of the kth bin wall, which is related to the bin size.

[0129] Finally, the joint optimization formula based on the warehouse wall constraint is as follows:

[0130]

[0131] In the formula is the point cloud registration error, is the plane constraint error;

[0132] The above formula is the ICP objective function based on the plane constraint of the warehouse wall. It is the objective function of an optimization problem used for point cloud registration (i.e. aligning two point cloud data sets). The objective function consists of two parts: point cloud registration error and plane constraint error. The goal is to find a set of rotation matrices R i and the translation vector T i , so that the weighted sum of these two parts is minimized. With the goal of minimizing the error, the rotation matrix with the smallest error is obtained and translation vector , as the optimal rotation matrix and the optimal translation vector By minimizing this objective function, a set of rotation matrices and translation vectors can be found so that the transformed point cloud is aligned with the target point cloud and complies with the given plane constraint.

[0133] The goal of the point cloud registration error part is to minimize the distance between the two point clouds. Specifically, it calculates the point cloud P after the rotation and translation transformation. i Points in Corresponding points in the target point cloud The sum of the squares of the Euclidean distances between them. Where:

[0134] N is the number of point cloud datasets of the radar device;

[0135] P i is the i-th source point cloud. The purpose of point cloud registration is to transform the source point cloud into the coordinate system of the target point cloud. Taking the two radar devices AB as an example, A can be used as the source point cloud and B as the target point cloud. Conversely, B can be used as the source point cloud and A as the target point cloud. This is not limited here.

[0136] It is the point cloud P i The jth point in ;

[0137] R i is the rotation matrix of the i-th source point cloud, which is used to rotate and align the point cloud;

[0138] T i is the translation vector of the i-th source point cloud, which is used to translate and align the point cloud;

[0139] The target point cloud The corresponding point of .

[0140] The goal of the plane constraint error part is to ensure that the transformed point cloud conforms to the given plane constraint as much as possible, which is achieved by calculating the sum of the squares of the distances from the transformed points to the plane.

[0141] K is the number of planes, that is, the number of warehouse wall planes;

[0142] W k is the set of points belonging to the k-th warehouse wall plane;

[0143] n k is the normal vector of the kth warehouse wall plane;

[0144] d k is the distance from the kth warehouse wall plane to the origin;

[0145] is a weight coefficient used to control the strength of the plane constraint (usually λ=0.5∼1.0).

[0146] B03. Perform point cloud stitching according to the target point cloud data and a preset weighted average algorithm to generate the granary environment map.

[0147] After the point cloud is registered, the point cloud can be spliced ​​to produce the granary environment map. Specifically, the point cloud is spliced ​​according to the target point cloud data and the preset weighted average algorithm to generate the granary environment map. The environment map can be a grain surface environment map, and it is three-dimensional. Figure 5 , Figure 5 It is a schematic diagram of a granary environment map in an embodiment of the present invention, showing a radar stitching grain surface map (the grain surface is 20 meters wide and 60 meters long).

[0148] For example, weighted averaging is performed according to the quality or confidence of the point cloud to fuse the registered point clouds into a whole. The weighted averaging algorithm is as follows:

[0149] ;

[0150] in, is the weight, M is the number of radars, P j is the target point cloud data, P fused It is the point cloud data after weighted average.

[0151] The method was tested by LiDAR scanning in a high dust environment and the results showed that:

[0152] 1) LiDAR can still maintain 85% point cloud efficiency when the dust coverage rate reaches 70%.

[0153] 2) The heat sink’s thermal conductive design effectively prevents equipment overheating caused by dust accumulation.

[0154] Please refer to Table 2, which shows the advantages of the present invention compared with the traditional method:

[0155] Table 2

[0156]

[0157] The present invention provides a grain silo splicing mapping method, and proposes an ultra-wide-angle laser radar device and a method for splicing and mapping multiple devices for the perception and mapping technology of the grain surface of the grain silo by the warehouse-leveling robot, so as to construct a three-dimensional grain surface and provide a high-precision grain surface map for the warehouse-leveling robot. The advantages of this method are as follows: 1) The device can form a laser radar scanning range with a horizontal viewing angle of 200° and a vertical viewing angle of 90°, which is more than twice the viewing angle of the existing single radar; 2) It best adapts to the space where the top of the silo is close to the grain surface (the top of the silo is about 1.5 meters to 2 meters away from the grain surface); 3) The grain surface environment perception ability of the warehouse-leveling robot and other types of grain silo robots is greatly improved, so that the robot has the ability to perceive the grain surface environment of the entire warehouse; 4) The grain silo wall constraint equation is used to fuse the radar point clouds to accurately obtain the distribution of grain surface and grain piles, and provide an accurate target working area for the robot's warehouse-leveling operation; 5) The echo model is constructed based on the physical properties of the grain, and the dust filtering algorithm of multiple echoes (that is, the above-mentioned anti-interference rule) is constructed to enhance the anti-dust interference ability of the radar device in the grain silo environment.

[0158] See also Figure 6 , Figure 6 : is a structural block diagram of a granary splicing mapping device according to an embodiment of the present invention, such as Figure 6 The device is applied to a granary stitching and mapping system, the system comprises at least a plurality of radar devices separately arranged in the granary, each radar device comprises at least two radar modules, and the device comprises:

[0159] Data acquisition module 601: used to obtain the current echo data and the last echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary;

[0160] Interference processing module 602: used to perform anti-interference processing using preset anti-interference rules, current echo data and previous echo data to obtain target echo data of each radar module;

[0161] Data conversion module 603: used to convert the target echo data of each radar module into the first point cloud data of each radar module;

[0162] Point cloud modeling module 604: used to perform point cloud stitching processing based on the first point cloud data of multiple radar modules to obtain a complete granary environment map.

[0163] It should be noted that Figure 6 The functions of each module in the device shown are Figure 1 The contents of each step are similar, so we will not go into details here to avoid repetition. For details, please refer to Figure 1 The content of each step.

[0164] The present invention provides a granary splicing mapping device, which is applied to a granary splicing mapping system. The system includes at least a plurality of radar devices arranged in the granary, each radar device includes at least two radar modules, and the device includes: a data acquisition module: used to obtain the current echo data and the last echo data collected by each radar module, and the echo data is the data of the echo after the transmission signal of the radar module is reflected by the surface of the object in the granary; an interference processing module: used to perform anti-interference processing using a preset anti-interference rule, the current echo data and the last echo data to obtain the target echo data of each radar module; a data conversion module: used to convert the target echo data of each radar module into the first point cloud data of each radar module; a point cloud modeling module: used to perform point cloud splicing processing according to the first point cloud data of multiple radar modules to obtain a complete granary environment map. Through the above method, a radar device is set in the granary for granary splicing mapping, each radar device is set with at least two radar modules, the radar perception field is increased, and the radar echo data is subjected to anti-interference processing using anti-interference rules to reduce dust interference, thereby improving the mapping efficiency and accuracy.

[0165] Figure 7 FIG. 1 shows an internal structure diagram of a computing power module in an embodiment, and the computing power module may include a computer device. The computer device may be a terminal or a server. Figure 7 As shown, the computing power module includes a processor, a memory and a network interface connected via a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computing power module stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the above method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the above method. Those skilled in the art can understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computing power module to which the solution of the present application is applied. The specific computing power module may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0166] In one embodiment, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes Figure 1 or Figure 3 steps.

[0167] In one embodiment, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the processor executes Figure 1 or Figure 3 steps.

[0168] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0169] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0170] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for granary splicing and mapping, characterized in that: The method is applied to a granary stitching and mapping system, the system comprising at least a plurality of radar devices separately arranged in the granary, the radar devices being ultra-wide-angle laser radar devices, each radar device comprising at least two radar modules, and the method comprising: Acquire the current echo data and the last echo data collected by each radar module, wherein the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary; Performing anti-interference processing using a preset anti-interference rule, current echo data, and last echo data to obtain target echo data of each radar module, wherein the echo data at least includes the reflection intensity of the echo, and the anti-interference rule includes a preset reflection intensity ratio threshold, then performing anti-interference processing using the preset anti-interference rule, current echo data, and last echo data to obtain target echo data of each radar module includes: for each radar module: determining a reflection intensity ratio between a first reflection intensity of the last echo and a second reflection intensity of the current echo; if the reflection intensity ratio is greater than a preset intensity ratio threshold, determining that the target echo data is current echo data; if the reflection intensity ratio is less than or equal to a preset intensity ratio threshold, determining that the target echo data is last echo data; Wherein, the echo data also includes the propagation distance of the echo, and the propagation distance is used to reflect the distance between the radar module and the surface of the object; then the method also includes: using the propagation distance of the echo and a preset reflection intensity algorithm to obtain the reflection intensity of the echo, including: obtaining the target grain type and the current dust concentration stored in the current granary; using the preset correspondence between the grain type and the reflectivity of the grain surface and the target grain type, determining the target reflectivity corresponding to the target grain type; using the preset correspondence between the dust concentration and the attenuation coefficient and the current dust concentration, determining the target attenuation coefficient corresponding to the current dust concentration; according to the target reflectivity, target attenuation coefficient, propagation distance and reflection intensity algorithm, obtaining the reflection intensity of the echo; Convert the target echo data of each radar module into the first point cloud data of each radar module; Point cloud stitching processing is performed based on the first point cloud data of multiple radar modules to obtain a complete granary environment map, including: using two groups of first point cloud data of two radar modules in each radar device, a first rotation matrix between the two radar modules, and a first translation vector to perform point cloud registration between the radar modules to obtain second point cloud data of each radar device; using the second point cloud data of each radar device, a second rotation matrix between the two radar devices, and a second translation vector to perform point cloud registration between the radar devices to obtain target point cloud data; performing point cloud stitching based on the target point cloud data and a preset weighted average algorithm to generate the granary environment map.

2. The method according to claim 1, characterized in that: The method of using the second point cloud data of each radar device, the second rotation matrix and the second translation vector between the two radar devices to perform point cloud registration between the radar devices to obtain target point cloud data includes: Using the second point cloud data of each radar device and a preset ICP objective function based on the warehouse wall plane constraint to perform an optimal transformation solution to obtain the second rotation matrix and the second translation vector; Coordinate transformation is performed based on the second rotation matrix, the second translation vector and the second point cloud data of each radar device to obtain target point cloud data.

3. The method according to claim 1, characterized in that: The reflection intensity algorithm includes the following mathematical expression: ； In the formula, is the reflection intensity; is the system gain coefficient; is the grain surface reflectivity; is the echo propagation distance, which is used to indicate the distance between the LiDAR module and the target surface. is the attenuation coefficient.

4. The method according to claim 2, characterized in that: The second point cloud data includes a source point cloud and a target point cloud, and the ICP objective function includes the following mathematical expression: ； In the formula is the point cloud registration error, is the plane constraint error; Where N is the number of point cloud datasets of the radar device; P i is the i-th source point cloud; It is the point cloud P i The jth point in ; R i is the rotation matrix of the i-th source point cloud; T i is the translation vector of the i-th source point cloud; The target point cloud The corresponding point of Where K is the number of warehouse wall planes; W k is the set of points belonging to the k-th warehouse wall plane; n k is the normal vector of the kth warehouse wall plane; d k is the distance from the kth warehouse wall plane to the origin; is the weight coefficient.

5. The method according to claim 1, characterized in that: The weighted average algorithm includes the following mathematical expression: ； In the formula, is the weight, M is the number of radars in the radar device, P j is the target point cloud data, P fused It is the point cloud data after weighted average.

6. An ultra-wide-angle laser radar device, characterized in that: A plurality of ultra-wide-angle laser radar devices are separately arranged in the granary, and the ultra-wide-angle laser radar device includes at least two radar modules, and the radar modules are used to collect current echo data and previous echo data, and report them to the granary stitching and mapping system; the echo data is the echo data after the transmission signal of the radar module is reflected by the surface of the object in the granary; The granary stitching and mapping system is used to receive the current echo data and the previous echo data, and execute the steps of the granary stitching and mapping method as described in any one of claims 1 to 5.

Images