Roadway obstacle avoidance method and device based on laser point cloud, terminal equipment and medium

Abstract

The invention relates to a roadway obstacle avoidance method and device based on a laser point cloud, terminal equipment and a medium, and the method comprises the steps: obtaining original point cloud data of a roadway environment through employing an explosion-proof three-dimensional laser radar, and each piece of point cloud data in the original point cloud data comprises a three-dimensional coordinate of a point cloud corresponding position; performing extraction processing on the original point cloud data to obtain a point cloud data set of a preset area; processing the point cloud data set by adopting a principal component analysis method to obtain a curvature value of each point cloud; determining obstacle information data in the roadway according to the curvature values of all the point clouds; and the mobile device performs path planning and obstacle avoidance control according to the obstacle information data. According to the roadway obstacle avoidance method based on the laser point cloud, the obstacle avoidance object in the roadway can be quickly and accurately determined.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of roadway obstacle avoidance method, device, terminal equipment and medium based on laser point cloud. BACKGROUND

[0002] In the roadway operating environment such as mine, tunnel, low pipe gallery, self-operated or remote-controlled equipment (such as mine truck, heading machine, inspection robot, etc.) faces complex and dynamic obstacles. Obstacles can be rock falling from the roof, protrusions from the sidewall, temporarily placed equipment or supporting structure, etc.

[0003] Traditional roadway obstacle avoidance technology mainly relies on two-dimensional laser radar, ultrasonic sensor, vision camera; among them, two-dimensional laser radar can provide two-dimensional plane profile information, but cannot perceive three-dimensional space obstacles such as roof and sidewall, and is prone to miss detection of suspended or low obstacles; ultrasonic sensor is easily disturbed by environmental noise, temperature and airflow, has low ranging accuracy and poor resolution, and cannot accurately identify the shape and size of obstacles; vision camera has serious imaging quality decline in poor lighting conditions and large dust and smoke in the roadway, resulting in low identification reliability; traditional roadway obstacle avoidance technology has insufficient perception ability for three-dimensional space information or is severely restricted by the environment, and it is difficult to accurately quantify (size, position) obstacles in the roadway in unstructured environment.

[0004] To improve the perception effect of three-dimensional space information, roadway obstacle avoidance technology proposes obstacle detection technology based on three-dimensional laser point cloud, i.e. first obtaining full point cloud data of the environment through laser radar, and then separating the ground and obstacles from the point cloud using methods such as grid method, clustering algorithm or plane segmentation method; i.e. the current obstacle detection technology based on three-dimensional laser point cloud still has the following problems:

[0005] (1) Since full scene, massive three-dimensional point cloud data is directly processed, extremely high computing resources are required, and the computing power requirement of embedded computing unit on the device is extremely high, resulting in slow algorithm response, long detection period, and difficulty in meeting the urgent need for real-time obstacle avoidance when the device is running at high speed, i.e. poor real-time performance;

[0006] (2) Roadway arm is usually uneven, with various pipelines, supports and other non-obstacle structures, and general point cloud clustering or segmentation algorithm is prone to misjudge these structures as obstacles to be avoided, resulting in a large number of false alarms; in other words, it has poor adaptability to unstructured environments such as roadways.

[0007] Therefore, it is urgent to solve the problems of poor real-time performance and low accuracy in the obstacle avoidance process in the roadway environment. SUMMARY

[0008] The application aims to provide a tunnel obstacle avoidance method and device based on laser point cloud, a terminal device and a medium, so as to solve the problems of poor real-time performance and low accuracy in the obstacle avoidance process in the prior art.

[0009] To solve the above problems, the tunnel obstacle avoidance method based on laser point cloud provided by the application adopts the following technical solutions:

[0010] An explosion-proof three-dimensional laser radar is used to obtain raw point cloud data of a tunnel environment, and each point cloud data in the raw point cloud data includes three-dimensional coordinates of a position corresponding to the point cloud.

[0011] The raw point cloud data is extracted to obtain a point cloud data set of a preset area; the preset area includes a roof and a sidewall of the tunnel.

[0012] A principal component analysis method is used to process the point cloud data set to obtain a curvature value of each point cloud.

[0013] Obstacle information data in the tunnel is determined according to the curvature values of all point clouds.

[0014] A mobile device performs path planning and obstacle avoidance control according to the obstacle information data.

[0015] In some embodiments, a 64-line explosion-proof three-dimensional laser radar is installed on the top of a mine truck to perform 360° scanning on the tunnel environment to obtain the raw point cloud data.

[0016] In some embodiments, the raw point cloud data is extracted to obtain the point cloud data set of the preset area according to the installation position of the explosion-proof three-dimensional laser radar and the geometric structure characteristics of the tunnel.

[0017] In some embodiments, the process of obtaining the curvature value of each point cloud includes:

[0018] A neighborhood of the point cloud is determined; the neighborhood includes all point clouds in a spherical domain with a set radius value.

[0019] A covariance matrix is constructed based on the point cloud data corresponding to all point clouds in the neighborhood, and an eigenvalue of the covariance matrix is calculated.

[0020] The curvature value of the point cloud is calculated according to the eigenvalue of the covariance matrix.

[0021] In some embodiments, the process of determining the obstacle information data in the tunnel according to the curvature values of all point clouds includes:

[0022] All curvature values corresponding to all point clouds are filtered to obtain potential point clouds based on a preset curvature threshold.

[0023] performing cluster analysis on the potential point cloud to determine parameter information of each obstacle;

[0024] encapsulating the parameter information of all obstacles into obstacle information data in the form of a list.

[0025] In some embodiments, the preset process of the curvature threshold value includes:

[0026] mapping the curvature values of all point clouds into a curvature intensity map;

[0027] determining an obstacle virtual position according to the curvature intensity map;

[0028] determining an obstacle actual position according to the on-site survey;

[0029] determining the curvature threshold value according to all curvature values corresponding to the obstacle actual position and all curvature values corresponding to the obstacle virtual position.

[0030] In some embodiments, the cluster analysis on the potential point cloud is performed using a Euclidean clustering algorithm to obtain at least one cluster, each cluster is taken as an obstacle, and the parameter information of the corresponding obstacle is determined according to the three-dimensional coordinates of all point clouds in each cluster.

[0031] To solve the above problems, a roadway obstacle avoidance device based on laser point cloud according to the present application comprises a collection module, an extraction module, a processing module and an information module;

[0032] The collection module is used to obtain raw point cloud data of a roadway environment using an explosion-proof three-dimensional laser radar, and each point cloud data in the raw point cloud data comprises three-dimensional coordinates of a position corresponding to the point cloud;

[0033] The extraction module is used to perform extraction processing on the raw point cloud data to obtain a point cloud data set of a preset area; the preset area comprises a roof and a sidewall of the roadway;

[0034] The processing module is used to process the point cloud data set using a principal component analysis method to obtain a curvature value of each point cloud;

[0035] The information module is used to determine obstacle information data in the roadway according to the curvature values of all point clouds; so that a mobile device performs path planning and obstacle avoidance control according to the obstacle information data.

[0036] To solve the above problems, a terminal device according to the present application comprises a processor and a memory, the memory stores a computer program, and the processor is used to execute the computer program to implement the above-mentioned roadway obstacle avoidance method based on laser point cloud.

[0037] In order to solve the above problems, the computer readable storage medium of the present application stores a computer program, which, when executed on a processor, implements the above-mentioned laser point cloud-based tunnel obstacle avoidance method.

[0038] The beneficial effects of the present application are as follows:

[0039] The laser point cloud-based tunnel obstacle avoidance method of the present application can greatly reduce the amount of point cloud data that needs to be processed through extraction processing, thereby reducing the computational load and meeting the rapid response requirements of real-time obstacle avoidance. In addition, the present application can effectively distinguish between tunnel flat walls and truly convex obstacles through curvature, thereby improving the accuracy of obstacle identification. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments:

[0041] Figure 1 The flowchart of the laser point cloud-based tunnel obstacle avoidance method in the embodiments of the present application;

[0042] Figure 2 The schematic diagram of the installation position of the laser radar in the embodiments of the present application;

[0043] Figure 3 The schematic diagram of the original point cloud data in the embodiments of the present application;

[0044] Figure 4 The local schematic diagram of the obstacle in the point cloud data in the embodiments of the present application;

[0045] Figure 5 The flowchart of determining the curvature value of the point cloud in the embodiments of the present application;

[0046] Figure 6 The flowchart of determining the obstacle information data in the embodiments of the present application;

[0047] Figure 7 The preset flowchart of the curvature threshold in the embodiments of the present application;

[0048] Figure 8 The curvature intensity schematic diagram of all point clouds in the point cloud data in the embodiments of the present application;

[0049] Figure 9 The curvature intensity schematic diagram of the potential point cloud in the embodiments of the present application. DETAILED DESCRIPTION

[0050] To make the technical objectives, technical solutions, and beneficial effects of the present invention clearer, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention; that is, the described embodiments are merely some embodiments of the present invention, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0051] In the following text, the terms "comprising," "having," and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more combinations thereof. Furthermore, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0052] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0053] The following is in conjunction with the appendix Figures 1-9 This application provides a detailed description of some embodiments. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0054] The following examples illustrate the obstacle avoidance method for alleyways based on laser point clouds.

[0055] like Figure 1 As shown, a specific embodiment of the lane obstacle avoidance method based on laser point clouds involved in this invention is described, the method comprising:

[0056] The S100 uses an explosion-proof 3D LiDAR to acquire raw point cloud data of the tunnel environment.

[0057] In this embodiment of the invention, the explosion-proof three-dimensional lidar employs, but is not limited to, 16-line, 32-line, or 64-line lidars. The explosion-proof three-dimensional lidar is installed at a preset position; in some embodiments, this preset position is the top of the mining truck. As other implementations, the preset position can also be, for example... Figure 2 Within the box shown.

[0058] The explosion-proof 3D lidar in this embodiment of the invention performs a 360° scan of the tunnel environment to acquire the original point cloud data of the tunnel environment, such as... Figure 3 As shown, each point cloud data in the original point cloud data has the LiDAR as the origin 0, and each point cloud data in the original point cloud data includes the three-dimensional coordinates (X, Y, Z) of the corresponding position of the point cloud.

[0059] S200: Extract and process the raw point cloud data to obtain a point cloud dataset for a preset area.

[0060] In this embodiment of the invention, the original point cloud data is extracted and processed to obtain a point cloud dataset of a preset area based on the installation location of the explosion-proof three-dimensional lidar and the geometric characteristics of the tunnel; the preset area includes the roof and sidewalls of the tunnel.

[0061] Specifically, the original point cloud data is processed using a preset algorithm (such as curvature) to obtain the region of interest, i.e., the preset region. The preset region mainly focuses on the roof and upper edge of the sidewall of the tunnel. By extracting and processing, the ground points and the lower part of the sidewall, which account for most of the data volume, are directly filtered out, which can reduce the amount of original point cloud data processing by more than 50% and greatly improve the efficiency of subsequent data processing.

[0062] S300 uses principal component analysis to process the point cloud dataset to obtain the curvature value of each point cloud.

[0063] Curvature is a mathematical quantity that characterizes the degree of curvature of a surface. On a flat tunnel roof or sidewall, the curvature value of a point cloud is close to 0; for example... Figure 4 As shown in the multiple boxes, when there is a protruding obstacle, the point cloud corresponding to the obstacle will form a convex curved surface, resulting in a significant increase in curvature value. Therefore, this application utilizes the characteristic that obstacles will inevitably cause the top or side wall contours to convex and deform, and calculates the curvature of the point cloud to accurately and efficiently lock the obstacle.

[0064] In this embodiment of the invention, for the extracted point cloud dataset of the top plate and the upper edge of the sidewalls, the curvature value of each point cloud is calculated. For example... Figure 5 As shown, the calculation method for each point cloud curvature value includes:

[0065] S310, Determine the neighborhood of the point cloud; the neighborhood includes all point clouds within a sphere of a set radius value;

[0066] S320: Construct a covariance matrix based on the point cloud data corresponding to all point clouds in the neighborhood, and calculate the eigenvalues ​​of the covariance matrix.

[0067] S330, calculate the curvature value of the point cloud based on the eigenvalues ​​of the covariance matrix.

[0068] Specifically, all points within a sphere with a radius of 10cm are selected as the neighborhood of the point cloud. Then, a covariance matrix is ​​constructed based on all point cloud data within the neighborhood. Since the point cloud data in this application is based on three-dimensional coordinates, three eigenvalues ​​are calculated, namely λ1, λ2, and λ3. Assuming λ1≥λ2≥λ3, the curvature value of the point cloud is σ=λ3 / (λ1+λ2+λ3). Among them, the smaller λ3 is, the closer the neighborhood of the point cloud is to a plane (small curvature); the larger λ3 is, the more curved the neighborhood of the point cloud (large curvature).

[0069] S400 determines obstacle information data within the tunnel based on the curvature values ​​of all point clouds.

[0070] In embodiments of the present invention, such as Figure 6 As shown, the process of determining obstacle information data within the alleyway based on the curvature values ​​of all point clouds includes:

[0071] S410, based on a preset curvature threshold, filters the curvature values ​​corresponding to all point clouds to obtain potential point clouds.

[0072] This invention filters out all points with curvature values ​​higher than the curvature threshold to form a new point cloud set containing only high curvature values. This point cloud set is essentially the potential point cloud of all obstacles protruding from the normal contour of the alley. By filtering through the threshold, flat surfaces and noise points can be effectively ignored.

[0073] In some embodiments, such as Figure 7 As shown, the process of setting the curvature threshold includes:

[0074] S411, map the curvature values ​​of all point clouds to a curvature intensity map;

[0075] S412, Determine the virtual position of the obstacle based on the curvature intensity map;

[0076] S413, Determine the actual location of the obstacle based on the on-site survey;

[0077] S414, Determine the curvature threshold based on all curvature values ​​corresponding to the actual position of the obstacle and all curvature values ​​corresponding to the virtual position of the obstacle.

[0078] Specifically, the curvature intensity map obtained by mapping the curvature values ​​of all point clouds is as follows: Figure 8 As shown, Figure 8The image shown is a grayscale image, but in reality, blue represents low curvature, green and yellow transition, and red represents high curvature. The curvature intensity map provides a visual representation of the location and color of obstacle areas. During the process of determining the curvature threshold based on all curvature values ​​corresponding to the actual and virtual locations of obstacles, the threshold can be dynamically adjusted manually, determined based on extensive adjustment experience. For example, a curvature threshold of 0.05; other implementations include 0.06, 0.055, etc. After filtering, the potential point cloud is obtained as follows: Figure 9 As shown.

[0079] S420 performs cluster analysis on the potential point cloud to determine the parameter information of each obstacle; the parameter information includes length, width, height and three-dimensional centroid coordinates.

[0080] In this embodiment of the invention, the Euclidean clustering algorithm is used to perform cluster analysis on the potential point cloud to obtain at least one cluster (e.g., Figure 9 The method uses a rectangle with a background color to represent each cluster as an obstacle, and determines the parameter information of the corresponding obstacle based on the three-dimensional coordinates of all point clouds in each cluster. Other clustering algorithms can also be used as alternative implementations; therefore, this invention includes, but is not limited to, the Euclidean clustering algorithm.

[0081] Specifically, clustering analysis groups points that are close in location in space into the same cluster, with each cluster representing an independent obstacle. When two obstacles are very close, their high-curvature point clouds will merge into one, and the clustering algorithm will automatically merge the two obstacles into a large obstacle cluster for processing, which conforms to the "worst-case" principle of safe obstacle avoidance.

[0082] The three-dimensional centroid coordinates (X_c, Y_c, Z_c) of each obstacle are calculated based on the point cloud data of the corresponding obstacle cluster. The maximum and minimum values ​​of the point cloud of the corresponding obstacle cluster in the X, Y, and Z axes are obtained based on the point cloud data of the obstacle cluster, and the length, width, and height (ΔX, ΔY, ΔZ) of the obstacle are calculated.

[0083] S430 encapsulates all obstacle parameter information into a list to obtain obstacle information data.

[0084] In practice, there are several obstacles in the tunnel. To facilitate the processing and management of obstacle information data by mobile devices, the parameter information of all obstacles is encapsulated in a set format to obtain obstacle information data. The set format includes, but is not limited to, lists, charts, etc.

[0085] The S500 mobile device performs path planning and obstacle avoidance control based on obstacle information data.

[0086] In this embodiment of the invention, the mobile device reads the position (i.e., three-dimensional centroid coordinates) and dimensions (i.e., length, width, and height) of each obstacle according to the obstacle information data in the list; and transmits the obstacle information data to the path control system of the mobile device. The path control system calculates the horizontal distance, azimuth angle, and height of the obstacle relative to the mobile device based on the three-dimensional centroid coordinates of the obstacle, and determines whether to detour around the obstacle, slow down, or stop moving in the travel path based on the size information.

[0087] Compared with the prior art, the present invention has the following advantages:

[0088] Improved efficiency and real-time performance: By extracting and processing the raw point cloud data, the amount of point cloud data that needs to be processed can be significantly reduced, thereby reducing the computational load and meeting the response requirements for real-time obstacle avoidance.

[0089] Improved accuracy and robustness: By introducing the key geometric feature of "curvature", it is possible to effectively distinguish between smooth roadway walls and real protruding obstacles, significantly reducing the false alarm rate.

[0090] Achieving precise quantification of obstacles: By using obstacle parameter information, the three-dimensional dimensions (length, width, height) and precise location (distance, azimuth) of obstacles in the tunnel space can be accurately calculated, providing reliable data support for path planning and obstacle avoidance decisions.

[0091] Enhance the algorithm's adaptability in complex scenarios: It can reliably identify and process obstacles even when they are close together or connected to the tunnel walls.

[0092] This invention targets application scenarios with distinct spatial structural features, such as alleyways (tunnels, utility tunnels). It proposes extracting and segmenting raw point cloud data, then using curvature to achieve efficient and accurate obstacle identification. Practical verification shows an accuracy rate as high as 95%, with a significantly reduced false alarm rate. Furthermore, the 3D dimensions and precise locations of obstacles obtained from the point cloud data provide more comprehensive decision-making support for intelligent obstacle avoidance.

[0093] The following describes the laser point cloud-based obstacle avoidance device using specific embodiments.

[0094] An embodiment of the present invention provides a tunnel obstacle avoidance device based on laser point clouds, comprising a data acquisition module, an extraction module, a processing module, and an information module. The data acquisition module uses an explosion-proof 3D laser radar to acquire raw point cloud data of the tunnel environment, where each point cloud data point includes the 3D coordinates of its corresponding location. The extraction module processes the raw point cloud data to obtain a point cloud dataset for a preset region, including the tunnel's roof and sidewalls. The processing module uses principal component analysis to process the point cloud dataset to obtain the curvature value of each point cloud. The information module determines obstacle information data within the tunnel based on the curvature values ​​of all point clouds, enabling the mobile device to perform path planning and obstacle avoidance control based on the obstacle information data.

[0095] This application also provides a terminal device, exemplary of which includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to enable the terminal device to perform the above-described alleyway obstacle avoidance method based on laser point clouds.

[0096] The processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), 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, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0097] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory is used to store computer programs, and the processor can execute the computer programs accordingly after receiving execution instructions.

[0098] This application also provides a computer-readable storage medium for storing the computer program used in the aforementioned terminal device. For example, the computer-readable storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0099] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. 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 alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive 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 diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0100] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0101] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a smartphone, personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0102] Finally, it should be noted that the above embodiments are only for illustration and not for limiting the technical solutions of the present invention. Any equivalent substitutions, modifications or partial substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for tunnel obstacle avoidance based on laser point cloud, characterized in that, The method comprises the following steps: acquiring original point cloud data of a roadway environment by using an explosion-proof three-dimensional laser radar, each point cloud data in the original point cloud data comprising three-dimensional coordinates of a position corresponding to the point cloud; extracting the original point cloud data to obtain a point cloud data set of a preset region; the preset region comprising a roof and a sidewall of the roadway; processing the point cloud data set by using a principal component analysis method to obtain a curvature value of each point cloud; determining obstacle information data in the roadway according to the curvature values of all point clouds; controlling path planning and obstacle avoidance of a mobile device according to the obstacle information data.

2. The laser point cloud based tunnel obstacle avoidance method of claim 1, wherein, A 64-line explosion-proof three-dimensional laser radar is installed on the top of a mine truck to perform 360° scanning on a roadway environment to acquire the original point cloud data.

3. The laser point cloud based tunnel obstacle avoidance method of claim 2, wherein, The original point cloud data is extracted to obtain the point cloud data set of the preset region according to the installation position of the explosion-proof three-dimensional laser radar and the geometric structure characteristics of the roadway.

4. The laser point cloud based tunnel obstacle avoidance method of claim 1, wherein, The process of obtaining the curvature value of each point cloud comprises the following steps: determining a neighborhood of the point cloud; the neighborhood comprising all point clouds in a spherical domain with a set radius value; constructing a covariance matrix based on the point cloud data corresponding to all point clouds in the neighborhood, and calculating eigenvalues of the covariance matrix; calculating the curvature value of the point cloud according to the eigenvalues of the covariance matrix.

5. The laser point cloud based tunnel obstacle avoidance method according to claim 1 or 4, characterized in that, The process of determining the obstacle information data in the roadway according to the curvature values of all point clouds comprises the following steps: filtering the curvature values corresponding to all point clouds according to a preset curvature threshold to obtain potential point clouds; performing clustering analysis on the potential point clouds to determine parameter information of each obstacle; encapsulating the parameter information of all obstacles into obstacle information data in the form of a list.

6. The laser point cloud based tunnel obstacle avoidance method of claim 5, wherein, The preset process of the curvature threshold comprises the following steps: mapping the curvature values of all point clouds into a curvature intensity map; determining a virtual position of an obstacle according to the curvature intensity map; determining an actual position of the obstacle according to a field survey; determining the curvature threshold according to all curvature values corresponding to the actual position of the obstacle and all curvature values corresponding to the virtual position of the obstacle.

7. The laser point cloud based tunnel obstacle avoidance method of claim 5, wherein, Performing clustering analysis on the potential point clouds by using a Euclidean clustering algorithm to obtain at least one cluster, taking each cluster as an obstacle, and determining parameter information of a corresponding obstacle according to three-dimensional coordinates of all point clouds in each cluster.

8. A roadway obstacle avoidance device based on laser point cloud, characterized in that, The method comprises a collection module, an extraction module, a processing module and an information module. The collection module is configured to acquire original point cloud data of a roadway environment by using an explosion-proof three-dimensional laser radar, each point cloud data in the original point cloud data comprising three-dimensional coordinates of a position corresponding to the point cloud. The extraction module is configured to extract the original point cloud data to obtain a point cloud data set of a preset region; the preset region comprising a roof and a sidewall of the roadway. The processing module is configured to process the point cloud data set by using a principal component analysis method to obtain a curvature value of each point cloud. The information module is configured to determine obstacle information data in the roadway according to the curvature values of all point clouds. The mobile device is controlled to perform path planning and obstacle avoidance according to the obstacle information data.

9. A terminal device, comprising: The terminal device comprises a processor and a memory, the memory stores a computer program, and the processor is configured to execute the computer program to implement the laser point cloud based tunnel obstacle avoidance method in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer program is stored in the memory and is configured to, when executed on the processor, implement the laser point cloud based tunnel obstacle avoidance method in any one of claims 1-7.

Images