Laser positioning method, computer equipment and storage medium for port vehicles

Abstract

The present invention provides a laser positioning method, computer equipment and storage medium for port vehicles. The port vehicle includes a laser positioning sensor. The laser positioning method includes: determining the spatial coordinate system in which the port vehicle is located, and determining the initial value of the position and posture of the port vehicle in the spatial coordinate system. The laser point cloud of the port vehicle is collected in real time, and the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment are preprocessed. The two frames of preprocessed laser point clouds are matched between frames to obtain a position and posture conversion value, and the real-time position and posture initial value is obtained based on the position and posture conversion value and the initial value. Several laser point clouds are selected as key frame point clouds, a sub-map is generated, and the laser positioning result of the port vehicle is obtained. The present invention realizes laser positioning of the port vehicle by determining the real-time initial value of the position and posture of the port vehicle and its position in the sub-map, avoids the real-time maintenance of the global laser point cloud map, and improves the efficiency of laser positioning.

Description

Technical Field

[0001] The present invention relates to the technical field of laser positioning, and in particular to a laser positioning method, computer equipment and storage medium for port vehicles. Background Art

[0002] With the continuous development of port logistics, the demand for efficient container handling is gradually increasing. Traditional manned container trucks are no longer able to meet port container handling needs due to objective factors such as high labor costs and low work efficiency. Therefore, relevant unmanned driving technologies are needed for practical application in automated terminals. Among them, unmanned driving technology solutions, such as intelligent guided vehicles, have been widely used in automated terminal handling tasks.

[0003] Compared with unmanned driving technology scenarios such as smart warehousing, smart campuses, and indoor cleaning, the automated handling scenarios in ports have more stringent requirements for unmanned driving technology. The main factors include two points. First, in addition to engineering vehicles, the port also has handling vehicles from external merchants, so the port is an open environment. Second, the environmental characteristics of the terminal are dynamic (that is, the terminal has heavy mobile equipment such as quay cranes, rail cranes, and tire cranes), the operating area is large, and the working conditions are complex. In addition, the port climate is relatively harsh. For example, under conditions such as heavy rain, scorching sun, dust, and night work, it is a great challenge to the performance of vehicle-mounted equipment.

[0004] The current unmanned driving technology solution for ports is mainly based on simultaneous localization and mapping (SLAM). This solution requires the pre-collection of environmental data, offline generation of point cloud maps, and then matching the real-time scanned laser point cloud data with the point cloud map, and combining it with other sensor data to obtain the final positioning result. However, in the automated handling tasks of ports, the physical position and shape of quay cranes, rail cranes and containers will change, so the point cloud map needs to be maintained and updated in real time, which is complicated and inefficient. The open environment and large working area of ​​the port make the matching between laser data and point cloud maps poor, the positioning results are less accurate, and the stability is reduced. For the pre-map work, if the data collection and mapping calculation of the entire area are to be carried out at once, it will take a long time, seriously affecting the work efficiency of the container handling task. Summary of the Invention

[0005] In view of this, the present invention provides a laser positioning method, computer equipment and storage medium for port vehicles, which can solve the problem that laser positioning requires the establishment and real-time maintenance and updating of a global point cloud map, which is time-consuming and leads to low container handling efficiency.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The present invention provides a laser positioning method for a port vehicle, which is applied to the port vehicle. The port vehicle includes a laser positioning sensor. The laser positioning method includes:

[0008] Determine the spatial coordinate system where the port vehicle is located, and determine the initial value of the position and posture of the port vehicle in the spatial coordinate system;

[0009] Collect the laser point cloud of the port vehicle in the corresponding environment in real time, and pre-process the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment, wherein the time corresponding to the initial value is different from the current moment and the previous moment;

[0010] The two pre-processed laser point clouds are matched between frames to obtain the pose conversion value. Based on the pose conversion value and the initial value, the real-time pose initial value of the port vehicle at the current moment is obtained.

[0011] Based on the laser point clouds within a preset time period, a number of laser point clouds are selected as key frame point clouds, and a submap is generated based on the key frame point clouds, wherein the preset time period is from the spatial coordinate system where the port vehicle is located to the current moment, or a period of time from the spatial coordinate system where the port vehicle is located to the current moment;

[0012] Based on the laser point cloud, real-time initial pose value and sub-map at that moment, the laser positioning result of the port vehicle at the current moment is obtained.

[0013] In one embodiment of the present invention, determining a spatial coordinate system in which a port vehicle is located and determining an initial value of the position and posture of the port vehicle in the spatial coordinate system includes:

[0014] The initial position of the port vehicle in the spatial coordinate system is determined based on global satellite positioning technology.

[0015] In one embodiment of the present invention, a laser point cloud of a port vehicle in an environment is collected in real time, and the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment are pre-processed, including:

[0016] Based on the preset conditions, the laser point cloud at the current moment and the laser point cloud at the previous moment are range filtered, and a second filtering is performed through the voxel grid. Then, the outliers of the laser point cloud are removed to obtain two frames of preprocessed laser point cloud. The preset condition is that it is less than the predetermined distance corresponding to each coordinate axis of the port vehicle in the spatial coordinate system.

[0017] In one embodiment of the present invention, inter-frame matching is performed on two pre-processed laser point clouds to obtain a pose conversion value, including:

[0018] The pose transformation value includes the spatial translation and rotation transformation, which are used to determine the spatial translation distance and rotation transformation angle of the port vehicle.

[0019] In one embodiment of the present invention, based on the laser point clouds within a preset time period, a plurality of laser point clouds are selected as key frame point clouds, including:

[0020] Directly use the two frames of laser point cloud input at the initial moment as key frame point cloud; and,

[0021] Based on the preset time, preset translation distance and preset rotation angle, the laser point cloud that meets the requirements at other times is determined as the key frame point cloud.

[0022] In one embodiment of the present invention, based on the laser point cloud, real-time initial pose value and sub-map at that moment, the laser positioning result of the port vehicle at the current moment is obtained, including:

[0023] The laser point cloud at the current moment is frame-matched with the sub-map to obtain the position data of the port vehicle on the sub-map at the current moment. The position data is matrix multiplied with the real-time initial pose value to obtain the laser positioning result of the port vehicle in the spatial coordinate system at the current moment.

[0024] In one embodiment of the present invention, the laser positioning method further includes:

[0025] The posture data is obtained through the inertial measurement unit, and based on the laser positioning result and the posture data, the final laser positioning result of the port vehicle in the spatial coordinate system at the current moment is obtained.

[0026] In one embodiment of the present invention, based on the laser positioning result and the posture data, the final laser positioning result of the port vehicle at the current moment in the spatial coordinate system is obtained, including:

[0027] The posture data is integrated, and the result of the posture data integration is fused with the posture value in the laser positioning result to obtain the final laser positioning result.

[0028] The present invention also provides a computer device, comprising:

[0029] A memory, used for storing instructions for port vehicles when they are working;

[0030] The processor is used to execute the above-mentioned laser positioning method for port vehicles based on the instructions.

[0031] The present invention also provides a computer-readable storage medium for storing a computer program, which implements the above-mentioned laser positioning method for port vehicles when executed by a processor.

[0032] The above technical solution of the present invention has at least the following beneficial effects:

[0033] The laser positioning method for port vehicles of the present invention realizes the laser positioning of port vehicles by determining the real-time initial value of the position and posture of the port vehicle and the position of the port vehicle on the sub-map. Compared with the traditional laser positioning method for port vehicles, which requires the establishment, real-time maintenance and updating of a global laser point cloud map, the present invention only needs to maintain the sub-map generated by the laser point cloud in real time during the laser positioning process, which reduces the time for map construction and data collection and reduces the difficulty of laser positioning operations. In addition, the laser positioning result is determined according to the real-time initial value of the position and posture of the port vehicle in the spatial coordinate system and the position of the port vehicle on the sub-map. The positioning result of the port vehicle is calculated through two matchings, which ensures the accuracy of laser matching positioning and helps to improve the efficiency of laser positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A scene diagram of a laser positioning method for port vehicles according to an embodiment of the present invention;

[0035] Figure 2 This is a schematic structural diagram of a port vehicle according to an embodiment of the present invention;

[0036] Figure 3 This is a flow chart of a laser positioning method for port vehicles according to one embodiment of the present invention;

[0037] Figure 4 A flowchart of determining an initial value of a position and posture of a port vehicle in a spatial coordinate system according to an embodiment of the present invention;

[0038] Figure 5 This is a flow chart of preprocessing the laser point cloud of a port vehicle at the current moment and the laser point cloud at the previous moment according to one embodiment of the present invention;

[0039] Figure 6 This is a flow chart of selecting a plurality of laser point clouds as key frame point clouds based on laser point clouds within a preset time period according to an embodiment of the present invention;

[0040] Figure 7 A schematic diagram of a computer device according to an embodiment of the present invention.

[0041] Reference numerals:

[0042] 100. Automated terminal;

[0043] 200, IGV;

[0044] 300. Satellite. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of 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 described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0046] The following first describes in detail the application scenario of the laser positioning method according to the embodiment of the present invention with reference to the accompanying drawings.

[0047] It should be noted that the port vehicle may be a movable vehicle such as a transporter or a navigation vehicle. The present invention takes an intelligence guide vehicle (IGV) working in an automated terminal as an example for explanation.

[0048] refer to Figure 1 , Figure 1 FIG1 shows a scene diagram of a laser positioning method for a port vehicle according to an embodiment of the present invention. Figure 1 As shown, numerous IGVs 200 are constantly operating in and out of quay cranes at the automated terminal 100. The IGVs 200 communicate with satellites 300 to obtain initial location information. The automated terminal 100 is relatively open and crowded with external port vehicles, creating a complex environment. Therefore, data collection and mapping are time-consuming, resulting in low positioning efficiency for the IGVs 200 within the automated terminal 100.

[0049] refer to Figure 2 , Figure 2 FIG. 1 shows a schematic structural diagram of a port vehicle according to an embodiment of the present invention. Figure 2 As shown, a port vehicle includes a control system and sensors. The control system can be a console for controlling the port vehicle. The sensor can be a laser positioning sensor for acquiring position data. The sensor can obtain position data according to the control system's instructions and send the position data to the control system, which processes and integrates the data to ultimately obtain the port vehicle's positioning results.

[0050] The laser positioning method for port vehicles according to an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] refer to Figure 3 , Figure 3 A flow chart of a laser positioning method for a port vehicle according to an embodiment of the present invention is shown.

[0052] like Figure 1 and Figure 3As shown, the laser positioning method for port vehicles includes S110-S160, which is applied to port vehicles. The port vehicles include laser positioning sensors, wherein the laser positioning sensor is a sensor that uses laser technology for non-contact measurement, usually based on optical triangulation or laser echo analysis to measure the position and displacement of objects.

[0053] In S110 , the space coordinate system where the port vehicle is located is determined, and the initial value of the position and posture of the port vehicle in the space coordinate system is determined.

[0054] For example, a spatial coordinate system is determined based on the automated terminal 100 where the IGV 200 is located, and an initial value of the position of the IGV 200 in the spatial coordinate system is determined. The initial value of the position is a matrix. The coordinate axes of the spatial coordinate system are in three different and mutually perpendicular directions. The origin of the spatial coordinate system can be any position on the automated terminal 100. For example, the initial position of the IGV 200 is used as the origin of the spatial coordinate system, and the three mutually perpendicular directions of the IGV 200 are used as the directions of the coordinate axes of the spatial coordinate system.

[0055] In S120 , the laser point cloud of the port vehicle in the corresponding environment is collected in real time.

[0056] A laser point cloud refers to a collection of three-dimensional points measured by the IGV200's sensor by emitting laser pulses and receiving light pulses reflected from a target. For example, a laser positioning sensor can collect real-time laser point cloud data while the IGV200 is operating on the automated terminal 100.

[0057] In S130, the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment are preprocessed, wherein the initial value corresponds to a different moment than the current moment and the previous moment.

[0058] For example, preprocessing is performed on two frames of laser point cloud data from the IGV200 at the current moment and the previous moment to remove noise and outliers. This makes the laser point cloud data more accurate and facilitates subsequent calculations and processing. The initial value corresponds to the starting moment, which is different from the current moment and the previous moment. The IGV200 laser point cloud data comes from the laser positioning sensor located on the IGV200.

[0059] In S140, inter-frame matching is performed on the two pre-processed laser point clouds to obtain a posture conversion value, and the real-time posture initial value corresponding to the port vehicle at the current moment is obtained based on the posture conversion value and the initial value.

[0060] For example, the posture conversion value is a matrix. Based on the initial value obtained in S110, the initial value in S110 is matrix-multiplied with the posture conversion value obtained in S140 to obtain the real-time posture initial value of IGV200 in the spatial coordinate system, which is also a matrix.

[0061] In S150, a preset time period is determined, and based on the laser point clouds within the preset time period, a plurality of laser point clouds are selected as keyframe point clouds, and a submap is generated based on the keyframe point clouds. The preset time period is a period from the time when the spatial coordinate system where the port vehicle is located is determined to the current time, or a period from the time when the spatial coordinate system where the port vehicle is located is determined to the current time.

[0062] For example, from all the laser point clouds acquired by the IGV 200, several laser point clouds are selected as keyframe point clouds using adaptive rules. The adaptive rule can be a preset range smaller than the current location of the IGV 200. Because the IGV 200 is constantly moving within the automated terminal 100, the keyframe point clouds selected based on the adaptive rule are dynamic, helping to accurately reflect the environment of the automated terminal 100.

[0063] Generate a submap of the IGV200 at the current moment based on the keyframe point cloud. Since the keyframe point cloud is dynamically changing, the submap generated based on the keyframe point cloud is also dynamic. Generating submaps avoids the need to maintain and update the global map in real time, reduces the workload of data collection, and eases the difficulty of map generation.

[0064] In S160, based on the laser point cloud, real-time initial pose value and sub-map at that moment, the laser positioning result of the port vehicle at the current moment is obtained.

[0065] For example, the current laser point cloud data of the IGV200 is first frame-matched with the submap to obtain the current position of the IGV200 on the submap. The frame-matched result is then matrix-multiplied with the initial real-time pose value to obtain the laser positioning result of the IGV200 in the global spatial coordinate system. The frame-matching process will be explained in detail in the following examples.

[0066] Compared with the laser positioning method of the prior art, it is necessary to establish and maintain a global laser point cloud map in real time, which is time-consuming and leads to low efficiency in container handling. The laser positioning method of the port vehicle of the present invention realizes the laser positioning of the port vehicle by determining the real-time initial value of the position and posture of the port vehicle and the position of the port vehicle on the sub-map. During the laser positioning process, the present invention only needs to maintain the sub-map generated by the laser point cloud in real time, which reduces the time for map construction and data collection and reduces the difficulty of laser positioning operation. Furthermore, the present invention determines the laser positioning result based on the real-time initial value of the position and posture of the port vehicle in the spatial coordinate system and the position of the port vehicle in the sub-map, and calculates the positioning result of the port vehicle through two matchings, thereby ensuring the accuracy of laser matching positioning and helping to improve the efficiency of laser positioning.

[0067] The following is a detailed description of the embodiments of the present invention. Figure 3 Each step is described in detail.

[0068] First, S110, determining the spatial coordinate system where the port vehicle is located, and determining the initial value of the position and posture of the port vehicle in the spatial coordinate system, will be described.

[0069] refer to Figure 4 , Figure 4 A flow chart of determining the initial value of the position and posture of a port vehicle in a spatial coordinate system according to one embodiment of the present invention is shown.

[0070] like Figure 4 As shown, in S110, determining the initial value of the position and posture of the port vehicle in the spatial coordinate system may include: determining the initial value of the position and posture of the port vehicle in the spatial coordinate system based on global satellite positioning technology. The global satellite positioning technology can be implemented by the Global Navigation Satellite System (GNSS). Specifically, the following steps may be included: S111-S114.

[0071] Step S111: Acquire GNSS data of port vehicles through the Global Navigation Satellite System.

[0072] Step S112: Perform status check on the GNSS data.

[0073] Step S113: Determine whether the differential signal corresponding to the GNSS data is normal.

[0074] If the result of step S113 is yes, step S114 is executed to use the GNSS data as the initial value of the port vehicle in the spatial coordinate system. In other words, obtaining the initial value of the spatial coordinate system where the IGV 200 is located through GNSS can ensure the accuracy and efficiency of the initial value of the IGV 200.

[0075] If the result of the determination in step S113 is no, step S111 is executed to reacquire the GNSS data of the port vehicle through the global navigation satellite system to avoid using GNSS data with abnormal differential signals, thereby ensuring the validity of the GNSS data.

[0076] The following describes the pre-processing of the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment in S130.

[0077] refer to Figure 5 , Figure 5 A flowchart of preprocessing the laser point cloud of a port vehicle at the current moment and the laser point cloud at the previous moment according to one embodiment of the present invention is shown.

[0078] like Figure 5 As shown, in S130, the laser point cloud of the port vehicle at the current moment and the laser point cloud at the previous moment are pre-processed, which may include the following steps S131-S135.

[0079] Step S131: Determine that the preset condition is less than the predetermined distance corresponding to each coordinate axis of the port vehicle in the spatial coordinate system.

[0080] Step S132: performing range filtering on the two frames of laser point clouds based on preset conditions.

[0081] Step S133: Determine the voxel grid model.

[0082] Step S134: performing a second filtering on the two frames of laser point clouds using a voxel grid.

[0083] Step S135: removing outliers from the laser point cloud to obtain two frames of pre-processed laser point clouds.

[0084] By preprocessing the two frames of laser point cloud data through the above steps, the points around the IGV200 body, namely, outliers and outliers, can be removed to make the laser point cloud data more accurate.

[0085] In one embodiment of the present invention, in S140, inter-frame matching is performed on the two pre-processed frames of laser point clouds to obtain a posture conversion value, including: the posture conversion value includes a spatial translation amount and a rotation transformation amount, which is used to determine the spatial translation distance and rotation transformation angle of the port vehicle.

[0086] For example, the two pre-processed laser point clouds are matched between frames using the scan-context method (i.e., a laser radar-based environment recognition and positioning method) to obtain a spatial translation relationship T and a rotation transformation relationship R.

[0087] Define the laser point cloud at the current moment as L t , the laser point cloud at the last moment is Lt-1 , then the translation transformation relationship T and rotation transformation relationship R of the current frame in the radar space coordinate system are:

[0088] T=arg min Sim(L t ,L t-1 )

[0089] R=2π / (T*N)

[0090] Among them, N is the number of ring divisions of the current frame point cloud. This value can be set according to the amount of data from the laser radar. Sim(L t ,L t-1 ) is the similarity calculation function in scan-context, defined as:

[0091]

[0092] in, It's L t The matrix consisting of the data in column i, It's L t-1 The matrix consisting of the i-th column numbers in .

[0093] That is, in the spatial coordinate system, the IGV200 vehicle translates and rotates from point A (previous moment) to point B (current moment). Spatial translation is the difference between the coordinates of points A and B. Rotational transformation is the difference in attitude angles between points A and B, such as heading, pitch, and roll. This facilitates recording changes in the IGV200's laser point cloud.

[0094] In one embodiment of the present invention, reference Figure 6 , Figure 6 A flow chart is shown for selecting a number of laser point clouds as key frame point clouds based on laser point clouds within a preset time period.

[0095] like Figure 6 As shown, in S150, based on the laser point clouds within a preset time length, a number of laser point clouds are selected as key frame point clouds, which may include the following steps S151-S154.

[0096] Step S151: Determine whether the two frames of laser point cloud are input at the initial moment.

[0097] For example, all laser point clouds acquired by the sensor are marked in chronological order. The mark can be a timestamp, that is, each frame of the laser point cloud has a timestamp representing the time it was acquired. The control system of the port vehicle selects the two laser point cloud frames input at the initial moment based on the corresponding mark.

[0098] If the determination result in step S151 is yes, step S152 is executed.

[0099] Step S152: The two frames of laser point cloud input at the initial moment are used as key frame point clouds.

[0100] If the determination result in step S151 is no, step S153 is executed.

[0101] Step S153: Determine whether each frame of laser point cloud at other times is within the range of a preset time, a preset translation distance, and a preset rotation angle.

[0102] If the determination result in step S153 is yes, step S154 is executed.

[0103] Step S154: Use the laser point cloud at other times as the key frame point cloud.

[0104] For example, define τ as the time threshold, D TH is the translation distance threshold, ΔD is the translation distance between two adjacent frames of laser point cloud data, R TH is the rotation angle threshold, ΔR is the angle increment between two frames of data, and the key frame point cloud is defined as P key When the following formula is satisfied, the laser point cloud at the current moment is the key frame at that moment. The formula is:

[0105]

[0106] On the one hand, this formula improves the computational efficiency of the overlapping parts of two adjacent laser point clouds when the port vehicle, i.e., IGV200, is at low speed or stationary. On the other hand, it also ensures the feature matching of distant laser point clouds when the port vehicle is running at high speed. The combination of the two minimizes the drift of the global map, provides the positioning system with more proportional environmental features, and improves the accuracy of global consistency matching.

[0107] The key frame is updated by setting an adaptive distance threshold to update the key frame point cloud, eliminating old key frame point clouds that are too far away and have no computational value, and obtaining continuously updated key frame point cloud data.

[0108] The present invention defines the key frame translation threshold adaptability as:

[0109] sig t =asig t-1 +bD EU .

[0110] Among them, D EU is the three-dimensional Euclidean distance from the origin of the world coordinate system to the laser point cloud position at the current moment. a and b are two constants with a value range of a∈[0.90,1),b∈(0.01,0.10]. t It is used to update the key frame threshold K at time tTH The smooth signal, threshold K TH Defined as:

[0111]

[0112] By selecting and updating keyframe point clouds, the accuracy of the sub-map system can be guaranteed.

[0113] In one embodiment of the present invention, in S160, based on the laser point cloud, the real-time initial pose value and the sub-map at that moment, the laser positioning result of the port vehicle is obtained, which may include two steps.

[0114] The first step is to perform frame matching between the laser point cloud at the current moment and the sub-map to obtain the position data of the port vehicle on the sub-map at the current moment.

[0115] For example, the GICP algorithm can be selected as the frame image matching algorithm. Define the current time as t, the previous time as t-1, the global space coordinate system as g, the radar coordinate system as l, the sub-map system as sub, and the IGV200 pose obtained by GICP as The point cloud is P, so the GICP objective function is defined as:

[0116]

[0117] where ε is the residual in GICP, defined as:

[0118]

[0119] Among them, c t is the covariance matrix at the current moment, c t-1 is the covariance matrix of the previous moment, j is the laser point sequence index in the point cloud P, sum is the total number of laser points in the point cloud P, d j is the jth laser point in the point cloud P.

[0120] The above algorithm is used to calculate the current position data of IGV200 on the sub-map. The position data is a matrix, which facilitates more accurate positioning data of IGV200 in the spatial coordinate system.

[0121] In the second step, the position data is matrix multiplied with the initial value of the real-time posture to obtain the laser positioning result of the port vehicle in the spatial coordinate system at the current moment.

[0122] In another embodiment of the present invention, the laser positioning method may further include the following steps after step S160: acquiring posture data through an inertial measurement unit, and obtaining the final laser positioning result of the port vehicle in the spatial coordinate system at the current moment based on the laser positioning result and the posture data.

[0123] It should be noted that an inertial measurement unit (IMU) is a device that measures an object's three-axis attitude angle (or angular rate) and acceleration. An IMU typically consists of three single-axis accelerometers and three single-axis gyroscopes. The accelerometers detect the object's acceleration signals along three independent axes in the carrier's coordinate system, while the gyroscopes detect the carrier's angular velocity signals relative to the navigation coordinate system. Together, they measure the object's angular velocity and acceleration in three-dimensional space and use this to calculate the object's attitude.

[0124] Furthermore, based on the laser positioning results and the posture data, the final laser positioning result of the port vehicle in the spatial coordinate system at the current moment is obtained, which can also include: integrating the posture data of the inertial measurement unit, and fusing the integrated posture data with the posture value in the laser positioning result to obtain the final laser positioning result.

[0125] The accuracy of laser positioning can be further improved by correcting the attitude value in the laser positioning result through the attitude data measured by the inertial measurement unit.

[0126] It should be noted that the specific calculation process of the posture data in the inertial measurement unit can refer to the calculation method in the prior art, and this application will not describe it in detail.

[0127] The present invention also provides a computer device comprising a memory and a processor. The memory is used to store instructions (including software programs and data logs) for the operation of a port vehicle. The processor is used to execute the above-mentioned laser positioning method for the port vehicle based on the instructions.

[0128] refer to Figure 7 , Figure 7 FIG. 1 is a schematic diagram showing a computer device according to an embodiment of the present invention. Figure 7 As shown, the computer device of the embodiment of the present invention may also include a sensor device, an input / output interface, a bus, a memory, and a processor. The sensor device is connected to the bus via the input / output interface. The memory and the processor are respectively connected to the bus.

[0129] Sensor equipment refers to external sensors installed on port vehicles, including but not limited to lidar, inertial measurement unit, global positioning system (GPS), etc.

[0130] The input and output interface is the interface device that directly transmits sensor data to the bus, including but not limited to a network port with a speed of 100M or above, a Universal Serial Bus (USB) interface, etc.

[0131] A bus is used to transfer and share information between devices. A bus can be one of several types of buses, including address bus, data bus, control bus, etc., serial bus, parallel bus, etc.

[0132] The present invention also provides a computer-readable storage medium for storing a computer program, which implements the above-mentioned laser positioning method for port vehicles when executed by a processor.

[0133] The working principle of the laser positioning method for port vehicles is explained below in conjunction with the steps.

[0134] Based on the spatial coordinate system of the automated terminal 100 where the IGV 200 is located, the initial value of the position and posture of the IGV 200 in the spatial coordinate system is determined. The laser positioning sensor is used to collect laser point cloud data of the IGV 200 in real time when it is operating on the automated terminal 100. It is determined whether each frame of the laser point cloud is the laser point cloud of the current moment or the laser point cloud of the previous moment. If so, the two frames of laser point cloud data of the IGV 200 at the current moment and the previous moment are preprocessed to remove noise points and abnormal points in the laser point cloud data. The two frames of preprocessed laser point cloud are matched between frames to obtain the pose conversion value. The pose conversion value and the initial value are matrix multiplied to obtain the real-time pose initial value corresponding to the IGV 200 at the current moment. A preset time length is determined, and it is determined whether each frame of the laser point cloud is within the preset time length and whether it is a key frame point cloud. If so, a submap of the IGV 200 at the current moment is generated based on the key frame point cloud. Based on the laser point cloud, real-time initial pose value and sub-map at that moment, the laser positioning result of IGV200 at the current moment is obtained, realizing the laser positioning of port vehicles.

[0135] In summary, the present invention's laser positioning method for port vehicles achieves laser positioning of port vehicles by determining the port vehicle's initial real-time pose and its position on a submap. This reduces the time required for map creation and data collection, and reduces the difficulty of laser positioning operations. Furthermore, the present invention determines the laser positioning result based on the port vehicle's initial real-time pose in the spatial coordinate system and its position on the submap. The positioning result of the port vehicle is calculated through two matching operations, ensuring the accuracy of laser matching positioning and facilitating improved laser positioning efficiency.

[0136] Unless otherwise defined, the technical or scientific terms used in the present invention shall have the usual meanings understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0137] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A laser positioning method for a port vehicle, applied to a port vehicle, wherein the port vehicle includes a laser positioning sensor, characterized in that: The laser positioning method comprises: Determining a spatial coordinate system in which the port vehicle is located, and determining an initial value of the position and posture of the port vehicle in the spatial coordinate system; collecting a laser point cloud in a corresponding environment of the port vehicle in real time, and preprocessing the laser point cloud of the port vehicle at a current moment and a laser point cloud at a previous moment, wherein the moment corresponding to the initial value is different from the current moment and the previous moment; Performing inter-frame matching on the two pre-processed laser point clouds to obtain a posture conversion value, and obtaining a real-time posture initial value corresponding to the port vehicle at the current moment based on the posture conversion value and the initial value; Based on the laser point clouds within a preset time period, a plurality of the laser point clouds are selected as key frame point clouds, and a submap is generated based on the key frame point clouds, wherein the preset time period is a period from the determination of the spatial coordinate system where the port vehicle is located to the current moment, or a period from the determination of the spatial coordinate system where the port vehicle is located to the current moment; Obtaining a laser positioning result of the port vehicle at the current moment based on the laser point cloud, the real-time initial pose value, and the sub-map at the current moment; The step of selecting a plurality of laser point clouds as key frame point clouds based on the laser point clouds within a preset time period includes: Directly use the two frames of laser point clouds input at the initial moment as the key frame point clouds; and, Based on the preset time, the preset translation distance and the preset rotation angle, the laser point cloud that meets the requirements at other moments is determined as the key frame point cloud.

2. The laser positioning method for port vehicles according to claim 1, characterized in that: The determining of the spatial coordinate system in which the port vehicle is located and determining the initial value of the position and posture of the port vehicle in the spatial coordinate system includes: The initial value of the position and posture of the port vehicle in the spatial coordinate system is determined based on global satellite positioning technology.

3. The laser positioning method for port vehicles according to claim 1, characterized in that: The real-time acquisition of the laser point cloud in the environment corresponding to the port vehicle and the pre-processing of the laser point cloud at the current moment and the laser point cloud at the previous moment of the port vehicle include: Based on a preset condition, the laser point cloud at the current moment and the laser point cloud at the previous moment are range filtered, and a second filtering is performed through a voxel grid, and then outliers in the laser point cloud are removed to obtain two frames of preprocessed laser point clouds, wherein the preset condition is less than a predetermined distance corresponding to each coordinate axis of the port vehicle in the spatial coordinate system.

4. The laser positioning method for port vehicles according to claim 1, characterized in that: The inter-frame matching of the two pre-processed laser point clouds to obtain a pose conversion value includes: The posture conversion value includes a spatial translation amount and a rotation transformation amount, which are used to determine the spatial translation distance and rotation transformation angle of the port vehicle.

5. The laser positioning method for port vehicles according to claim 1, characterized in that: The laser positioning result of the port vehicle at the current moment is obtained based on the laser point cloud at the current moment, the real-time initial pose value and the sub-map, including: The laser point cloud at the current moment is frame-matched with the sub-map to obtain the position data of the port vehicle on the sub-map at the current moment, and the position data is matrix-multiplied with the real-time initial pose value to obtain the laser positioning result of the port vehicle in the spatial coordinate system at the current moment.

6. The laser positioning method for port vehicles according to claim 5, characterized in that: The laser positioning method further comprises: The posture data is acquired through an inertial measurement unit, and based on the laser positioning result and the posture data, a final laser positioning result of the port vehicle in the spatial coordinate system at the current moment is obtained.

7. The laser positioning method for port vehicles according to claim 6, characterized in that: The step of obtaining a final laser positioning result of the port vehicle in the spatial coordinate system at a current moment based on the laser positioning result and the posture data includes: The posture data is integrated, and the result of the posture data integration is fused and calculated with the posture value in the laser positioning result to obtain the final laser positioning result.

8. A computer device, characterized in that: include: A memory, used for storing instructions for port vehicles when they are working; The processor is configured to execute the laser positioning method for a port vehicle according to any one of claims 1 to 7 based on the instructions.

9. A computer-readable storage medium for storing a computer program, characterized in that: When the computer program is executed by a processor, the laser positioning method for port vehicles according to any one of claims 1 to 7 is implemented.

Images