Positioning methods, devices, electronic equipment and storage media

Abstract

This disclosure relates to a positioning method, apparatus, electronic device, and storage medium. The method includes: filtering point cloud data to obtain filtered point cloud data; transforming the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain transformed point cloud data; segmenting the transformed point cloud data based on a preset segmentation rule to obtain segmented point cloud data; acquiring vehicle position information collected by a global positioning system at different times, as well as vehicle attitude information collected by an inertial measurement unit and wheel speedometers, and updating the vehicle's motion based on the position and attitude information to obtain multiple predicted poses of the vehicle; based on the multiple predicted poses, transforming the segmented point cloud data from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data; and registering the multiple target point cloud data with a likelihood map to obtain the target pose of the vehicle. This disclosure enables precise vehicle positioning.

Description

Technical Field

[0001] This disclosure relates to the field of positioning technology, and in particular to a positioning method, apparatus, electronic device and computer-readable storage medium. Background Technology

[0002] Open-pit mining is the process of removing the overburden from an ore body to obtain the desired minerals. Its production process includes drilling, blasting, loading, transportation, and dumping. With the development of autonomous driving technology, the application of autonomous vehicles in mining trucks has emerged. Currently, operations such as loading, transportation, and dumping are mainly completed by autonomous mining trucks.

[0003] As is well known, high-precision positioning technology is a prerequisite for realizing autonomous driving. Currently, the positioning technology suitable for autonomous mining trucks is Real-Time Kinematic (RTK) carrier phase differential technology (hereinafter referred to as "RTK technology"). RTK technology is not only highly accurate but also independent of external weather conditions, thus enabling uninterrupted all-weather, all-time operation. However, in practical applications, the implementation of RTK positioning function depends not only on network signals but also on satellite acquisition results. Therefore, RTK positioning function cannot be used in scenarios with no network access or indoor environments. Summary of the Invention

[0004] In view of this, embodiments of the present disclosure provide a positioning method, apparatus, electronic device, and computer-readable storage medium to solve the problem in the related art that the RTK positioning function cannot be used in offline or indoor scenarios.

[0005] A first aspect of this disclosure provides a positioning method applied to a vehicle. The method includes: acquiring point cloud data collected by a data acquisition device; filtering the point cloud data to obtain filtered point cloud data; converting the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain converted point cloud data; segmenting the converted point cloud data based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data; acquiring the vehicle's position information at different times collected by a global positioning system and the vehicle's attitude information collected by an inertial measurement unit and wheel speedometer, and updating the vehicle's motion based on the position information and attitude information to obtain multiple predicted poses of the vehicle; based on the multiple predicted poses, converting the segmented point cloud data from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data; and performing point cloud registration of the multiple target point cloud data with a likelihood map to obtain the target pose of the vehicle.

[0006] A second aspect of this disclosure provides a positioning device applied to a vehicle. The device includes: an acquisition module configured to acquire point cloud data collected by a data acquisition device, filter the point cloud data to obtain filtered point cloud data, and convert the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain converted point cloud data; a segmentation module configured to segment the converted point cloud data based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data; an update module configured to acquire vehicle position information collected by a global positioning system at different times and vehicle attitude information collected by an inertial measurement unit and wheel speedometer, and update the vehicle's motion based on the position and attitude information to obtain multiple predicted poses of the vehicle; and a registration module configured to convert the segmented point cloud data from the vehicle coordinate system to the world coordinate system based on the multiple predicted poses to obtain multiple target point cloud data, and register the multiple target point cloud data with a likelihood map to obtain the target poses of the vehicle.

[0007] A third aspect of this disclosure provides an electronic device including at least one processor; a memory for storing at least one processor-executable instruction; wherein the at least one processor is used to execute the instruction to implement the steps of the method described above.

[0008] A fourth aspect of this disclosure provides a computer-readable storage medium that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the steps of the above-described method.

[0009] The above-mentioned at least one technical solution adopted in the embodiments of this disclosure can achieve the following beneficial effects: Point cloud data collected by a data acquisition device is acquired, filtered to obtain filtered point cloud data, and the filtered point cloud data is transformed from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain transformed point cloud data; the transformed point cloud data is segmented based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data; the vehicle's position at different times is acquired from the global positioning system. The system collects vehicle attitude information from inertial measurement units and wheel speedometers, and updates the vehicle's motion based on the position and attitude information to obtain multiple predicted poses. Based on these multiple predicted poses, the segmented point cloud data is transformed from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data. These target point cloud data are then registered with the likelihood map to obtain the vehicle's target pose. This process ensures that the acquired vehicle target pose is more accurate and reliable, thus improving the accuracy and reliability of positioning, enabling precise vehicle control, ensuring vehicle driving continuity, and further improving vehicle operating efficiency. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart illustrating a positioning method provided as an exemplary embodiment of the present disclosure.

[0012] Figure 2 This is a flowchart illustrating another positioning method provided as an exemplary embodiment of the present disclosure.

[0013] Figure 3 This is a schematic diagram of a positioning device provided for an exemplary embodiment of the present disclosure.

[0014] Figure 4 A schematic diagram of the structure of an electronic device provided for an exemplary embodiment of this disclosure.

[0015] Figure 5 This is a schematic diagram of the structure of a computer system provided for an exemplary embodiment of the present disclosure. Detailed Implementation

[0016] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0017] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0018] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0019] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0020] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0021] A positioning method and apparatus according to an embodiment of the present disclosure will now be described in detail with reference to the accompanying drawings.

[0022] Figure 1 This is a flowchart illustrating a positioning method provided as an exemplary embodiment of the present disclosure. Figure 1 The positioning method can be executed by a server or electronic device in an autonomous driving system. For example... Figure 1 As shown, the positioning method includes:

[0023] S101: Acquire point cloud data collected by the data acquisition device, filter the point cloud data to obtain filtered point cloud data, and transform the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain transformed point cloud data.

[0024] S102, the converted point cloud data is segmented based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflective column point cloud data, and other point cloud data.

[0025] S103: Obtain the vehicle's position information at different times collected by the global positioning system and the vehicle's attitude information collected by the inertial measurement unit and wheel speedometer, and update the vehicle's motion based on the position information and attitude information to obtain multiple predicted poses of the vehicle.

[0026] S104: Based on multiple predicted poses, the segmented point cloud data is transformed from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data. The multiple target point cloud data are then registered with the likelihood map to obtain the target pose of the vehicle.

[0027] Specifically, taking the server in an autonomous driving system as an example, during the vehicle's operation, the server collects point cloud data around the vehicle in real time through data acquisition devices installed on the vehicle, and uses a filtering algorithm to filter the collected point cloud data to obtain filtered point cloud data. Further, the server transforms the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system, that is, it projects the filtered point cloud data into the vehicle coordinate system to obtain transformed point cloud data.

[0028] Here, an autonomous driving system refers to a system composed of hardware and software capable of continuously performing some or all of the dynamic driving tasks. Dynamic driving tasks refer to the perception, decision-making, and execution required to complete vehicle driving; that is, they include all real-time operational and tactical functions when driving a road vehicle, but exclude planning functions such as trip planning, destination and route selection. For example, dynamic driving tasks may include, but are not limited to, controlling the vehicle's lateral movement, controlling the vehicle's longitudinal movement, monitoring the driving environment and preparing responses by detecting, identifying, and classifying targets and events, executing responses, making driving decisions, and controlling vehicle lighting and signaling devices.

[0029] The server can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms. This disclosure does not limit the scope of the embodiments.

[0030] The vehicle can be a driverless vehicle or an autonomous vehicle. A driverless vehicle is an intelligent vehicle that achieves driverless operation through a computer system. It perceives the surrounding environment through an onboard sensor system and controls the vehicle's steering and speed based on the perceived road, vehicle position, and obstacle information, thereby enabling the vehicle to travel safely and reliably on the road. In this embodiment, the driverless vehicle may include, but is not limited to, wide-body vehicles, large mining trucks, data collection vehicles, and forklifts equipped with data acquisition devices.

[0031] Data acquisition devices can collect data from various points on surrounding target objects by transmitting and receiving signals; that is, they collect point data of the target points. The collection of point data acquired by the data acquisition device within one sampling period constitutes a frame of point cloud data. Data acquisition devices may include, but are not limited to, lidar, millimeter-wave radar, ultrasonic radar, cameras, etc. In this embodiment of the disclosure, the data acquisition device is a lidar. A lidar is a radar system that uses emitted laser beams to detect the position, velocity, and other characteristics of a target. The measurement range of a lidar is between 10 centimeters and 200 meters.

[0032] A point cloud is a collection of points that represent the spatial distribution and surface characteristics of a target object within a given spatial reference frame. Point cloud data is a set of vectors in a three-dimensional coordinate system. These vectors are typically represented by three-dimensional coordinates (X, Y, Z) and are generally used to represent the outer surface shape of an object. In addition to the geometric position information represented by three-dimensional coordinates, point cloud data can also represent the color (R, G, B) information of a point or the intensity information of an object's reflective surface. In the embodiments of this disclosure, point cloud data is used to characterize the three-dimensional coordinates of each point in the point cloud in a spatial reference coordinate system, which can be the lidar coordinate system corresponding to the lidar. In practical applications, the lidar can be triggered to scan the driving area at a preset scanning frequency (e.g., 10Hz) to obtain a frame of point cloud data corresponding to each scanning frequency.

[0033] Point cloud filtering refers to the process of removing invalid points from point cloud data. Typically, during the acquisition of point cloud data by LiDAR, due to factors such as the product's own system, the surface of the object being measured, and the scanning environment, some noise points and / or outliers inevitably appear in the point cloud data. These noise points and / or outliers can affect the results of subsequent point cloud segmentation, point cloud registration, and other processing. Therefore, they need to be removed directly or processed in a smoothing manner. Here, noise points refer to point cloud data that is useless for subsequent positioning, i.e., it cannot provide valuable information for subsequent positioning; noise points can include, but are not limited to, gravel, dust, water mist, and dynamic obstacles (e.g., vehicles, people, flying insects). Outliers refer to point cloud data that are outside the measurement range of the LiDAR. In the field of autonomous driving, common filtering algorithms include, but are not limited to, Kalman filtering, average value filtering, and median filtering. By filtering the point cloud data, noise points and / or outliers introduced during the acquisition process can be removed to form more accurate point cloud data, thus improving the quality of the point cloud data.

[0034] The coordinate system of the data acquisition device is a polar coordinate system established with the location of the data acquisition device as the origin. In this embodiment of the disclosure, the coordinate system of the data acquisition device is the lidar coordinate system. The lidar coordinate system is a three-dimensional coordinate system established with the lidar as the center. Typically, the origin of the lidar coordinate system is the laser emission center of the lidar. When the lidar is set horizontally, the plane containing the X-axis and Y-axis is parallel to the ground, and the Z-axis is perpendicular to the ground.

[0035] A Vehicle Coordinate System (VCS) is a special moving coordinate system used to describe the motion of a vehicle. The origin of the VCS is fixed relative to the vehicle and coincides with the vehicle's center of mass. The vehicle's center of mass can include, but is not limited to, the center of the front of the vehicle, the center of the front axle, and the center of the rear axle. Preferably, in this embodiment, the vehicle's center of mass is the center of the rear axle; that is, the VCS is established with the rear axle center as the origin. The VCS can be established using either a left-handed or right-handed system, and this embodiment does not impose any limitations on this. For example, if the VCS is established as a left-handed system, when the vehicle is stationary on a level road, the X-axis of the VCS is parallel to the ground and points forward, the Y-axis points to the driver's left, and the Z-axis is perpendicular to the ground and points upward. For example, the vehicle coordinate system is established as a right-hand system. When the vehicle is stationary on a level road, the X-axis of the vehicle coordinate system is parallel to the ground and points forward of the vehicle, the Y-axis points to the driver's right, and the Z-axis is perpendicular to the ground and points upward of the vehicle. It should be understood that at different times during the vehicle's movement, the established vehicle coordinate system will differ due to the vehicle's different positions.

[0036] Next, the server segments the converted point cloud data based on preset segmentation rules to obtain segmented point cloud data, which includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data. Further, the server acquires the vehicle's position information collected by the Global Positioning System at different times, as well as the vehicle's attitude information collected by the Inertial Measurement Unit and wheel speedometer, and updates the vehicle's motion based on the position and attitude information to obtain multiple predicted poses of the vehicle.

[0037] Here, the preset segmentation rules can be set according to actual needs, and this embodiment of the disclosure does not impose any restrictions on them. In this embodiment of the disclosure, the type of point cloud points may include ground type, slope type, reflective column type, and other types. Here, ground type refers to point cloud points in the point cloud data that represent the ground, that is, point cloud points collected by the laser emitted by the lidar to the ground and reflected by the ground; slope type refers to point cloud points in the point cloud data that represent the slope; reflective column type refers to point cloud points in the point cloud data that represent reflective columns; other types refer to point cloud points in the point cloud data that are not represented as ground, slope, or reflective columns. These point cloud points may be collected by the laser emitted by the lidar after being reflected by moving vehicles, pedestrians, rocks, etc.

[0038] The process of segmenting point cloud data is called point cloud segmentation. The purpose of point cloud segmentation is to divide the points in the point cloud data into different types, so that points of each type belong to the same layer. In this embodiment of the disclosure, a random consistency sampling algorithm can be used to segment and classify the converted point cloud data to separate ground point cloud data, slope point cloud data, reflector point cloud data, and / or other point cloud data. By segmenting the point cloud data, subsequent calculations can be performed separately for each type of point cloud data. Therefore, the amount of data computation is reduced, the computational efficiency is improved, and the accuracy of the calculation results is increased.

[0039] It should be noted that in this embodiment of the disclosure, the point cloud data is segmented at a resolution of 0.1 meters, and point cloud segmentation is performed specifically for static obstacles.

[0040] The Global Positioning System (GPS) is used to acquire the vehicle's position and velocity information. The Inertial Measurement Unit (IMU) is used to measure the vehicle's angular velocity and acceleration, among other physical information. Wheel speed sensors are used to collect the vehicle's wheel speed information. In this embodiment, based on the position information acquired by GPS and the attitude information acquired by the IMU and wheel speed sensors, the vehicle's pose can be updated using the Extended Kalman Filter (EKF) algorithm to obtain a predicted value for each pose, thereby obtaining multiple predicted poses of the vehicle at different times.

[0041] Pose is used to describe position and attitude. Position can be represented by coordinates in various coordinate systems, while attitude refers to orientation in various coordinate systems and can be represented by a rotation matrix. Pose data can include position and attitude information to characterize changes in the vehicle's position and attitude at different times. Pose data can include longitude, latitude, and attitude angles, where attitude angles can include one or more of pitch, yaw (also called heading), and roll. Optionally, position changes can be represented by translation vectors, and pose changes can be represented by rotation matrices. For example, the rotation matrix can be used to represent changes in the vehicle's yaw, pitch, and roll angles. In some embodiments, pose data can also include altitude information.

[0042] Furthermore, after obtaining multiple predicted poses, the server transforms the segmented point cloud data from the vehicle coordinate system to the world coordinate system based on the multiple predicted poses. That is, the segmented point cloud data is projected into the world coordinate system to obtain multiple target point cloud data. The server performs point cloud registration between the multiple target point cloud data and the likelihood map, and determines the target pose of the vehicle based on the registration results, so as to locate the vehicle based on the target pose.

[0043] Here, the World Coordinate System (WCS) is the system's absolute coordinate system. Before establishing other coordinate systems, the coordinates of all points on the screen are determined by the origin of this coordinate system. The World Coordinate System is used to describe absolute and relative positions on Earth. The World Coordinate System is a coordinate system fixed relative to an object; therefore, regardless of how the object rotates, translates, or scales, the position and orientation of the World Coordinate System will not change. In this embodiment of the disclosure, the World Coordinate System is a Cartesian coordinate system with a custom origin. Exemplarily, the World Coordinate System could be a Cartesian coordinate system with the base station as its origin.

[0044] A likelihood map is generated by segmenting laser point clouds into ground point clouds, slope point clouds, reflector point clouds, and other point clouds according to their types. For slope point clouds, an altitude above the ground is assigned to each point cloud based on the direction of its normal vector. Then, a likelihood map is generated from the segmented point clouds according to the maximum likelihood probability. The likelihood map can be a pre-built likelihood map stored on a cloud server, or it can be a likelihood map generated on the vehicle based on a point cloud map downloaded from a cloud server; this disclosure does not limit this. The likelihood map can include multiple layers of different types, each layer carrying semantic information to characterize the type of each point cloud point. Here, a layer is thematic data that divides spatial information according to its geometric features and attributes. Layers can be divided into vector layers and raster layers. In this disclosure, a layer refers to different types of raster layers, such as ground layers, slope layers, reflector layers, and other layers.

[0045] It should be noted that the number of layers can be set according to actual needs, and this embodiment does not impose any limitations on this. Furthermore, it should be noted that when there are two or more layers, the layer weights from highest to lowest are: slope layer, ground layer, reflector column layer, and other layers.

[0046] Point cloud registration refers to the process of transforming a set of 3D point clouds from two or more different coordinate systems into the same coordinate system through transformation relationships. Point cloud registration can generally be divided into two stages: coarse registration and fine registration. Coarse registration involves registering point clouds when their relative poses are completely unknown, finding a rotation and translation transformation matrix that makes the two point clouds relatively approximate, and then transforming the point cloud data to be registered into a unified coordinate system. Coarse registration provides good initial values ​​for fine registration. Fine registration, based on coarse registration, aims to minimize the spatial positional differences between point clouds, obtaining a more accurate rotation and translation transformation matrix. Point cloud registration algorithms can include, but are not limited to, the Iterative Closest Point (ICP) algorithm, the Normal Distribution Transform (NDT) algorithm, the Kernel Correlation (KC) algorithm, and the Robust Point Matching (RPM) algorithm.

[0047] According to the technical solution provided in this disclosure, point cloud data collected by a data acquisition device is obtained, and the point cloud data is filtered to obtain filtered point cloud data. The filtered point cloud data is then transformed from the coordinate system of the data acquisition device to the coordinate system of the vehicle to obtain transformed point cloud data. The transformed point cloud data is then segmented based on a preset segmentation rule to obtain segmented point cloud data. The segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data. The system also acquires vehicle position information at different times from a global positioning system and inertial measurement data. The vehicle's attitude information is collected by the unit of measurement and wheel speed sensor, and the vehicle's motion is updated based on the position and attitude information to obtain multiple predicted poses. Based on the multiple predicted poses, the segmented point cloud data is transformed from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data. The multiple target point cloud data are then registered with the likelihood map to obtain the vehicle's target pose. This ensures that the obtained vehicle target pose is more accurate and reliable, thus improving the accuracy and reliability of positioning, enabling accurate control of the vehicle, ensuring the continuity of vehicle driving, and further improving the vehicle's operating efficiency.

[0048] In some embodiments, multiple target point cloud data are respectively registered with a likelihood map to obtain the target pose of the vehicle, including: determining the type of each point in each target point cloud data in the multiple target point cloud data, and calculating the likelihood probability of each type based on the layer corresponding to each type; summing all likelihood probabilities of each type to obtain the sum of likelihood probabilities of each type, and summing the sum of likelihood probabilities of all types to obtain the type likelihood probability of the vehicle in the current predicted pose; selecting the predicted pose corresponding to the largest type likelihood probability from the multiple type likelihood probabilities of the multiple predicted poses as the target pose of the vehicle.

[0049] Specifically, after obtaining multiple target point cloud data sets, the server determines the type of each point in each target point cloud data set and performs point cloud registration between each point belonging to each type and its corresponding layer, i.e., calculates the likelihood probability of each type. Further, the server sums all likelihood probabilities for each type to obtain a total likelihood probability for each type, and then sums the total likelihood probabilities for all types to obtain the type likelihood probability of the vehicle in its current predicted pose. Then, the server selects the predicted pose corresponding to the highest type likelihood probability from the multiple predicted poses as the vehicle's target pose, i.e., the optimal position of the vehicle. The formula for calculating the total likelihood probability of each type is:

[0050]

[0051] Here, P represents the sum of likelihood probabilities for the i-th type, and N represents the number of point clouds in each target point cloud data, ∈ i Let z represent the weight of the i-th type, z represent the likelihood probability of the i-th type, x represent the state of the vehicle, and m represent the weight of the i-th type. i This represents the map information of the layer corresponding to the i-th type.

[0052] The following example illustrates the calculation process of type likelihood probability.

[0053] Specifically, assume the target point cloud data includes 10,000 point points, of which 3,000 belong to the ground type, 3,000 belong to the slope type, and 4,000 belong to the reflector type. First, for the 3,000 point points belonging to the ground type, calculate the likelihood probability of each point point individually, and then sum the calculated likelihood probabilities of each point point to obtain the total likelihood probability of the 3,000 point points. Since the likelihood probability is calculated in the same way for each type, we can obtain the total likelihood probability of the 3,000 point points belonging to the slope type and the total likelihood probability of the 4,000 point points belonging to the reflector type. Furthermore, by summing the likelihood probabilities of the 3,000 point cloud points belonging to the ground type, the 3,000 point cloud points belonging to the slope type, and the 4,000 point cloud points belonging to the reflector type, we can obtain the type likelihood probabilities of 10,000 point cloud points, which is the type likelihood probability of the vehicle in the current predicted pose.

[0054] According to the technical solution provided in the embodiments of this disclosure, by calculating the likelihood probability of each type based on the type to which each point cloud point belongs, calculating the type likelihood probability of the vehicle in the current predicted pose based on the sum of the likelihood probabilities of each type, and selecting the predicted pose corresponding to the largest type likelihood probability from multiple predicted poses as the target pose of the vehicle, the accuracy and reliability of the obtained target pose of the vehicle can be guaranteed. Therefore, the precise positioning of the vehicle is achieved, the driving continuity of the vehicle is guaranteed, and the operating efficiency of the vehicle is further improved.

[0055] In some embodiments, multiple target point cloud data are respectively registered with a likelihood map to obtain the target pose of the vehicle, including: for each point cloud point belonging to the slope type, calculating the ground height of each point cloud point, and calculating the height likelihood probability of the vehicle in the current predicted pose based on the ground height; summing the type likelihood probability and the height likelihood probability to obtain the overall likelihood probability of the vehicle in the current predicted pose; and selecting the predicted pose corresponding to the largest overall likelihood probability from the multiple overall likelihood probabilities of multiple predicted poses as the target pose of the vehicle.

[0056] Specifically, when the type of each point cloud point includes a slope type, the server can calculate the ground clearance of each point cloud point belonging to the slope type, and calculate the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance. Further, the server sums the calculated type likelihood probability and height likelihood probability to obtain the overall likelihood probability of the vehicle in the current predicted pose, and selects the predicted pose corresponding to the highest overall likelihood probability from multiple predicted poses as the vehicle's target pose. The formula for calculating the height likelihood probability of the vehicle in the current predicted pose is:

[0057]

[0058] Here, ρ represents the height likelihood probability, σ represents the standard deviation of each type, and d represents the ground elevation of each point cloud point.

[0059] The calculation process of the overall likelihood probability will be explained below, using the examples mentioned above.

[0060] Specifically, since 3,000 of the 10,000 point cloud points belong to the slope type, the height likelihood probability of the 3,000 point cloud points belonging to the slope type can be calculated. Furthermore, by summing the calculated type likelihood probability and height likelihood probability of the vehicle in the current predicted pose, the overall likelihood probability of the 10,000 point cloud points can be obtained, that is, the overall likelihood probability of the vehicle in the current predicted pose.

[0061] According to the technical solution provided in this disclosure, by calculating the ground clearance of each point cloud point belonging to the slope type, and calculating the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance, the height likelihood probability can be combined with the type likelihood probability to ensure that the obtained target pose of the vehicle is more accurate and reliable. Therefore, the accuracy and reliability of positioning are improved, the situation of pose perception error is reduced, accurate control of the vehicle is achieved, and the operating efficiency of the vehicle is further improved.

[0062] All the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this disclosure, and will not be described in detail here. In addition, the sequence number of each step in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this disclosure.

[0063] Figure 2 This is a flowchart illustrating another positioning method provided as an exemplary embodiment of the present disclosure. Figure 2 The positioning method can be executed by a server or electronic device in an autonomous driving system. For example... Figure 2 As shown, the positioning method may include:

[0064] S201: Acquire point cloud data collected by the data acquisition device, filter the point cloud data, and obtain filtered point cloud data.

[0065] S202, the filtered point cloud data is transformed from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain the transformed point cloud data;

[0066] S203, The converted point cloud data is segmented based on the preset segmentation rules to obtain the segmented point cloud data;

[0067] S204, acquire vehicle position information collected by the global positioning system at different times, as well as vehicle attitude information collected by the inertial measurement unit and wheel speedometer;

[0068] S205 updates the vehicle's motion based on position and attitude information to obtain multiple predicted poses of the vehicle;

[0069] S206, based on multiple predicted poses, transforms the segmented point cloud data from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data;

[0070] S207, determine the type of each point in each target point cloud data in multiple target point cloud data, and calculate the likelihood probability of each type;

[0071] S208, sum all likelihood probabilities for each type to obtain the total likelihood probability for each type;

[0072] S209, sum the likelihood probabilities of all types to obtain the type likelihood probability of the vehicle in the current predicted pose;

[0073] S210, determine whether the type of each point cloud point includes the slope type. If yes, execute S211; otherwise, execute S214.

[0074] S211, calculate the ground clearance of each point cloud point belonging to the slope type, and calculate the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance.

[0075] S212, sum the type likelihood probability and the height likelihood probability to obtain the overall likelihood probability of the vehicle in the current predicted pose;

[0076] S213, Select the prediction pose corresponding to the largest overall likelihood probability from multiple predicted poses as the target pose of the vehicle.

[0077] S214, Select the predicted pose corresponding to the largest type likelihood probability from multiple predicted poses as the target pose of the vehicle.

[0078] According to the technical solution provided in the embodiments of this disclosure, by calculating the likelihood probability of each type based on the type to which each point cloud point belongs, calculating the type likelihood probability of the vehicle in the current predicted pose based on the sum of the likelihood probabilities of each type, and selecting the predicted pose corresponding to the largest type likelihood probability from multiple predicted poses as the target pose of the vehicle, the accuracy and reliability of the obtained target pose of the vehicle can be guaranteed. Therefore, the precise positioning of the vehicle is achieved, ensuring the driving continuity and safety of the vehicle.

[0079] Furthermore, by calculating the ground clearance of each point cloud point belonging to the slope type, and calculating the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance, the height likelihood probability can be combined with the type likelihood probability to ensure that the obtained target pose of the vehicle is more accurate and reliable. Therefore, the accuracy and reliability of positioning are improved, the situation of pose perception error is reduced, accurate control of the vehicle is achieved, and the operating efficiency of the vehicle is further improved.

[0080] In the case of dividing each functional module according to its corresponding functions, this disclosure provides a positioning device, which can be a server or a chip applied to a server. Figure 3 This is a schematic diagram of a positioning device provided for an exemplary embodiment of this disclosure. Figure 3 As shown, the positioning device 300 includes:

[0081] The acquisition module 301 is configured to acquire point cloud data collected by the data acquisition device, filter the point cloud data to obtain filtered point cloud data, and transform the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain transformed point cloud data.

[0082] The segmentation module 302 is configured to segment the converted point cloud data based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data.

[0083] The update module 303 is configured to acquire the vehicle's position information collected by the global positioning system at different times and the vehicle's attitude information collected by the inertial measurement unit and wheel speedometer, and to update the vehicle's motion based on the position information and attitude information to obtain multiple predicted poses of the vehicle.

[0084] The registration module 304 is configured to transform the segmented point cloud data from the vehicle coordinate system to the world coordinate system based on multiple predicted poses to obtain multiple target point cloud data. The multiple target point cloud data are then registered with the likelihood map to obtain the target pose of the vehicle.

[0085] According to the technical solution provided in this disclosure, point cloud data collected by a data acquisition device is obtained, and the point cloud data is filtered to obtain filtered point cloud data. The filtered point cloud data is then transformed from the coordinate system of the data acquisition device to the coordinate system of the vehicle to obtain transformed point cloud data. The transformed point cloud data is then segmented based on a preset segmentation rule to obtain segmented point cloud data. The segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data. The system also acquires vehicle position information at different times from a global positioning system and inertial measurement data. The vehicle's attitude information is collected by the unit of measurement and wheel speed sensor, and the vehicle's motion is updated based on the position and attitude information to obtain multiple predicted poses. Based on the multiple predicted poses, the segmented point cloud data is transformed from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data. The multiple target point cloud data are then registered with the likelihood map to obtain the vehicle's target pose. This ensures that the obtained vehicle target pose is more accurate and reliable, thus improving the accuracy and reliability of positioning, enabling accurate control of the vehicle, ensuring the continuity of vehicle driving, and further improving the vehicle's operating efficiency.

[0086] In some embodiments, the likelihood map includes multiple layers of different types, including a ground layer, a slope layer, a reflector layer, and other layers. Each of the multiple layers carries semantic information to characterize the type to which each point cloud point belongs, including ground type, slope type, reflector type, and other types.

[0087] In some embodiments, Figure 3 The registration module 304 determines the type of each point in each target point cloud data in multiple target point cloud data, and calculates the likelihood probability of each type based on the layer corresponding to each type; sums all likelihood probabilities of each type to obtain the sum of likelihood probabilities of each type, and sums the sum of likelihood probabilities of all types to obtain the type likelihood probability of the vehicle in the current predicted pose; and selects the predicted pose corresponding to the largest type likelihood probability from multiple predicted poses as the target pose of the vehicle.

[0088] In some embodiments, Figure 3 The registration module 304 calculates the sum of likelihood probabilities for each type using the following formula: Where P represents the sum of likelihood probabilities for the i-th type, and N represents the number of point clouds in each target point cloud data, ∈ i Let z represent the weight of the i-th type, z represent the likelihood probability of the i-th type, x represent the state of the vehicle, and m represent the weight of the i-th type. iThis represents the map information of the layer corresponding to the i-th type.

[0089] In some embodiments, for each point cloud point belonging to the slope type Figure 3 The registration module 304 calculates the ground clearance of each point cloud point and, based on the ground clearance, calculates the height likelihood probability of the vehicle in the current predicted pose; it sums the type likelihood probability and the height likelihood probability to obtain the overall likelihood probability of the vehicle in the current predicted pose; and it selects the predicted pose corresponding to the largest overall likelihood probability from the multiple overall likelihood probabilities of multiple predicted poses as the target pose of the vehicle.

[0090] In some embodiments, Figure 3 The registration module 304 calculates the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance using the following formula: Where ρ represents the height likelihood probability, σ represents the standard deviation of each type, and d represents the ground elevation of each point cloud point.

[0091] In some embodiments, the weights of the multiple layers, from highest to lowest, are: slope layer, ground layer, reflector layer, and other layers.

[0092] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0093] This disclosure also provides an electronic device, including: at least one processor; and a memory for storing at least one processor-executable instruction; wherein the at least one processor is used to execute the instruction to implement the corresponding steps in the method disclosed in this disclosure.

[0094] Figure 4 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this disclosure. For example... Figure 4 As shown, the electronic device 400 includes at least one processor 401 and a memory 402 coupled to the processor 401, which can perform the methods disclosed in the embodiments of this disclosure.

[0095] The processor 401 described above can also be called a Central Processing Unit (CPU), which can be an integrated circuit chip with signal processing capabilities. Each step in the method disclosed in this embodiment can be implemented by the integrated logic circuitry in the processor 401 or by software instructions. The processor 401 can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this embodiment can be directly implemented by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software modules can be located in the memory 402, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor 401 reads information from the memory 402 and, in conjunction with its hardware, completes the steps of the method described above.

[0096] Furthermore, various operations / processes according to this disclosure, implemented via software and / or firmware, can be transmitted from a storage medium or network to a computer system with a dedicated hardware architecture, for example, Figure 5 The computer system 500 shown is equipped with the programs that constitute the software. When various programs are installed, the computer system is able to perform various functions, including functions such as those described above. Figure 5 This is a schematic diagram of the structure of a computer system provided for an exemplary embodiment of the present disclosure.

[0097] Computer system 500 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of this disclosure described and / or claimed herein.

[0098] like Figure 5As shown, the computer system 500 includes a computing unit 501, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the computer system 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0099] Multiple components in the computer system 500 are connected to the I / O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 can be any type of device capable of inputting information into the computer system 500. The input unit 506 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 507 can be any type of device capable of presenting information and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. The storage unit 508 may include, but is not limited to, a hard disk and an optical disk. The communication unit 509 allows the computer system 500 to exchange information / data with other devices via a network such as the Internet, and may include, but is not limited to, a modem, network card, infrared communication device, wireless communication transceiver, and / or chipset, such as Bluetooth. TM Devices, WiFi devices, WiMax devices, cellular communication devices and / or the like.

[0100] The computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above. For example, in some embodiments, the methods disclosed in this disclosure can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on the computer system 500 via ROM 502 and / or communication unit 509. In some embodiments, the computing unit 501 can be used to perform the methods disclosed in this disclosure by any other suitable means (e.g., by means of firmware).

[0101] This disclosure also provides a computer-readable storage medium, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to perform the methods disclosed in this disclosure.

[0102] The computer-readable storage medium in this disclosure can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The aforementioned computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specifically, the aforementioned computer-readable storage medium may include electrical connections based on one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0103] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0104] This disclosure also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, it implements the methods disclosed in the embodiments of this disclosure.

[0105] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer.

[0106] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0107] The modules, components, or units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules, components, or units do not necessarily constitute a limitation on the module, component, or unit itself.

[0108] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0109] The above description is merely an embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0110] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A positioning method, characterized in that, Applied to vehicles, the method includes: Acquire point cloud data collected by a data acquisition device, filter the point cloud data to obtain filtered point cloud data, and transform the filtered point cloud data from the coordinate system of the data acquisition device to the coordinate system of the vehicle to obtain transformed point cloud data. The converted point cloud data is segmented based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data. The system acquires the vehicle's position information at different times from the global positioning system and the vehicle's attitude information from the inertial measurement unit and wheel speedometer, and updates the vehicle's motion based on the position information and attitude information to obtain multiple predicted poses of the vehicle. Based on the multiple predicted poses, the segmented point cloud data is transformed from the vehicle coordinate system to the world coordinate system to obtain multiple target point cloud data. The multiple target point cloud data are then registered with the likelihood map to obtain the target pose of the vehicle.

2. The method according to claim 1, characterized in that, The likelihood map includes multiple layers of different types, including a ground layer, a slope layer, a reflector layer, and other layers. Each of the multiple layers carries semantic information, which is used to characterize the type to which each point cloud point belongs. The types include ground type, slope type, reflector type, and other types.

3. The method according to claim 2, characterized in that, The step of performing point cloud registration between the multiple target point cloud data and the likelihood map to obtain the target pose of the vehicle includes: Determine the type of each point in each of the multiple target point cloud data, and calculate the likelihood probability of each type based on the layer corresponding to each type; Summing all likelihood probabilities for each type yields the total likelihood probability for each type, and summing the total likelihood probabilities for all types yields the type likelihood probability of the vehicle in the current predicted pose. The predicted pose corresponding to the largest type likelihood probability among the multiple predicted poses is selected as the target pose of the vehicle.

4. The method according to claim 3, characterized in that, The summation of all likelihood probabilities for each type to obtain the total likelihood probability for each type includes: The sum of the likelihood probabilities for each type is calculated using the following formula: Where P represents the sum of likelihood probabilities for the i-th type, and N represents the number of point cloud points in each target point cloud data, ∈ i Let z represent the weight of the i-th type, z represent the likelihood probability of the i-th type, x represent the state of the vehicle, and m represent the weight of the i-th type. i This represents the map information of the layer corresponding to the i-th type.

5. The method according to claim 3, characterized in that, The step of performing point cloud registration between the multiple target point cloud data and the likelihood map to obtain the target pose of the vehicle includes: For each point cloud point belonging to the slope type, calculate the ground height of each point cloud point, and based on the ground height, calculate the height likelihood probability of the vehicle in the current predicted pose. The summation of the type likelihood probability and the height likelihood probability yields the overall likelihood probability of the vehicle in the current predicted pose. The predicted pose corresponding to the largest overall likelihood probability among the multiple predicted poses is selected as the target pose of the vehicle.

6. The method according to claim 5, characterized in that, The step of calculating the height likelihood probability of the vehicle in the current predicted pose based on the ground clearance includes: Based on the ground clearance, the height likelihood probability of the vehicle in the current predicted pose is calculated using the following formula: Where ρ represents the height likelihood probability, σ represents the standard deviation of each type, and d represents the ground height of each point cloud point.

7. The method according to any one of claims 2 to 6, characterized in that, The weights of the multiple layers, from highest to lowest, are: the slope layer, the ground layer, the reflector layer, and the other layers.

8. A positioning device, characterized in that, Applied to vehicles, the device includes: The acquisition module is configured to acquire point cloud data collected by the data acquisition device, filter the point cloud data to obtain filtered point cloud data, and transform the filtered point cloud data from the coordinate system of the data acquisition device to the vehicle coordinate system to obtain transformed point cloud data. The segmentation module is configured to segment the converted point cloud data based on a preset segmentation rule to obtain segmented point cloud data, wherein the segmented point cloud data includes at least one of ground point cloud data, slope point cloud data, reflector point cloud data, and other point cloud data. The update module is configured to acquire the vehicle's position information collected by the global positioning system at different times and the vehicle's attitude information collected by the inertial measurement unit and wheel speedometer, and to perform motion updates on the vehicle based on the position information and the attitude information to obtain multiple predicted poses of the vehicle. The registration module is configured to transform the segmented point cloud data from the vehicle coordinate system to the world coordinate system based on the multiple predicted poses to obtain multiple target point cloud data, and to perform point cloud registration of the multiple target point cloud data with the likelihood map to obtain the target pose of the vehicle.

9. An electronic device, characterized in that, include: At least one processor; Memory for storing the at least one processor-executable instruction; The at least one processor is configured to execute the instructions to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is enabled to perform the method as described in any one of claims 1 to 7.

Images