An indoor unmanned aerial vehicle flight path planning processing method and system

Abstract

The application discloses an indoor unmanned aerial vehicle flight path planning processing method and system, and the method comprises the following steps: scanning an indoor environment by using a laser radar to obtain real-time obstacle information; that is, the indoor environment is scanned by using a laser radar device, so that multiple obstacle point cloud information and obstacle position information in the environment can be obtained; a grid map is divided into multiple level partition regions according to the density of obstacles in the grid and the occupation surface probability value; based on the multiple level partition regions, an improved rapidly-exploring random tree algorithm is used to generate an initial flight path; the initial flight path is subjected to smoothing processing to obtain an optimized flight path; the improved rapidly-exploring random tree algorithm is used to generate the initial path, and the algorithm has a faster calculation speed and better path quality than a traditional algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of unmanned aerial vehicle flight control, in particular to an indoor unmanned aerial vehicle flight path planning processing method and system. BACKGROUND

[0002] The development of indoor unmanned aerial vehicle obstacle avoidance technology has undergone a transformation from simple mechanical obstacle avoidance to complex intelligent obstacle avoidance. Early obstacle avoidance technology mainly relies on mechanical sensors such as infrared sensors and ultrasonic sensors, which can provide limited obstacle avoidance information, but the detection range and accuracy are limited. With the development of computer vision technology, image recognition-based obstacle avoidance technology has emerged, which can provide more comprehensive environmental information, but the computational load is high and the hardware requirements are high.

[0003] Specifically, indoor unmanned aerial vehicle flight control and obstacle avoidance technology is also evolving, driving continuous innovation and progress in indoor unmanned aerial vehicle applications. In practical applications, it is often necessary to use a variety of technologies to achieve safe and efficient indoor unmanned aerial vehicle flight control and obstacle avoidance functions according to specific scenarios and requirements.

[0004] The obstacle avoidance method in the prior art is based on image recognition for flight obstacle detection and obstacle avoidance control. This control method has a large computational load, resulting in a long hang time and a large amount of computing power, which affects the work efficiency of indoor unmanned aerial vehicle transportation or indoor unmanned aerial vehicle safety operations. SUMMARY

[0005] The purpose of the present application is to provide an indoor unmanned aerial vehicle flight path planning processing method and system that solves the above technical problems identified in the prior art.

[0006] The present application proposes an indoor unmanned aerial vehicle flight path planning processing method, comprising the following operation steps:

[0007] Obtain three-dimensional map data of the indoor environment; the three-dimensional map data is a three-dimensional map data pre-constructed in the database;

[0008] Scan the indoor environment using a laser radar to obtain real-time obstacle information; that is, the indoor environment is scanned by a laser radar device to obtain multiple obstacle point cloud information and obstacle position information within the environment;

[0009] Construct a grid map based on the three-dimensional map data and the real-time obstacle information to obtain the density and occupancy surface probability value of the obstacles in each grid; divide the grid map into multiple hierarchical partition regions according to the density and occupancy surface probability value of the obstacles in the grid;

[0010] Based on the multiple hierarchical partition regions, an improved rapid expansion random tree algorithm is used to generate an initial flight path;

[0011] Smooth the initial flight path to obtain an optimized flight path;

[0012] Control the UAV to fly according to the optimized flight path.

[0013] Preferably, after controlling the UAV to fly according to the optimized flight path, it further comprises:

[0014] Collecting obstacle images on the flight path, judging the current obstacle type based on the obstacle images, screening dangerous obstacles according to the identified current obstacle type, marking the position nodes of dangerous obstacles, and re-adjusting the flight path.

[0015] Preferably, after controlling the UAV to fly according to the optimized flight path, it further comprises:

[0016] Collecting obstacle images on the flight path, judging the current obstacle type based on the obstacle images, screening dangerous obstacles according to the identified current obstacle type, and marking the position nodes of dangerous obstacles.

[0017] Preferably, collecting obstacle images on the flight path, judging the current obstacle type based on the obstacle images, screening dangerous obstacles according to the identified current obstacle type, and marking the position nodes of dangerous obstacles, specifically comprises:

[0018] During flight, set the collection frequency, and use the camera carried by the UAV to regularly take images of obstacles on the path;

[0019] Preprocess the collected images, analyze the preprocessed images using deep learning vision algorithms, identify whether the type of the current obstacle belongs to a dangerous obstacle, and if it does, mark the identified dangerous obstacle and record its position information.

[0020] Preferably, the grid map is constructed based on three-dimensional map data and real-time obstacle information, comprising:

[0021] Discretize the three-dimensional map data into three-dimensional grids;

[0022] Map the real-time obstacle information into the corresponding three-dimensional grids;

[0023] Determine the density of obstacles and the occupancy probability value of each grid.

[0024] Preferably, the density of obstacles and the occupancy probability value in each grid include:

[0025] Obtaining the number of obstacles in each grid in the three-dimensional grid, calculating the density of the obstacles in the grid; the density of the obstacles is the number of independent obstacles in a single grid;

[0026] Determining the occupancy state of the grid according to the density of the obstacles;

[0027] Simultaneously calculating the occupied surface probability value of the obstacles in the current grid, the occupied surface probability value of the obstacles being the ratio of the occupied obstacle area in the grid to the area of the current grid.

[0028] Preferably, the grid map is divided into multiple hierarchical partition regions according to the density of the obstacles in the grid and the occupied surface probability value, and specifically includes:

[0029] For each grid, calculating a score based on the density of the obstacles and the occupied surface probability value;

[0030] Score = w1 * the density of the obstacles + w2 * the occupied surface probability value; wherein w1 and w2 are preset first and second weights;

[0031] Defining multiple thresholds A1, A2, the score being less than or equal to A1 being a low-risk region, the score being greater than or equal to A2 being a high-risk region, and the score being between A1 and A2 being a medium-risk region;

[0032] Traversing the entire current grid map, calculating the score of each grid in the grid map, and dividing the risk levels according to the thresholds;

[0033] Clustering the adjacent grids of the same risk level to form a continuous risk region.

[0034] Preferably, based on the multiple hierarchical partition regions, an improved rapidly-exploring random tree algorithm is used to generate an initial flight path, and specifically includes:

[0035] Sampling nodes from the current multiple hierarchical partition regions, and when sampling, different sampling frequencies are given to regions of different risk levels to reduce the sampling probability of high-risk regions; wherein the sampling frequency gradually decreases from the low-risk region, the medium-risk region to the high-risk region;

[0036] Setting a starting point and an ending point, running the improved RRT algorithm to generate an initial flight path; specifically, starting from the node of the starting point as an initial candidate node, calculating the distance between the current candidate node and the existing sampled multiple tree nodes within a preset distance range;

[0037] Within the preset distance range, selecting n tree nodes closest to the current candidate node as parent nodes, and connecting the parent nodes and the candidate node by using a local planner;

[0038] checking whether the connection path intersects with the obstacle;

[0039] if not intersecting, adding the candidate node to the tree;

[0040] repeating the above steps until a feasible path is found or a maximum number of iterations is reached and the generation of the initial flight path is completed.

[0041] Preferably, after the generation of the initial flight path is completed, it further comprises:

[0042] judging the number of the current initial flight paths, if the current initial flight paths generated by the improved rapid expansion random tree algorithm are multiple, calculating the number of regions passed by each of the multiple initial flight paths through each risk level region, and selecting the initial flight path passing through the most number of low-risk regions as the final target flight path.

[0043] Preferably, the process of smoothing the initial flight path comprises:

[0044] using a cubic spline interpolation method to interpolate the path points;

[0045] then adjusting the path curvature of the interpolated path points by a local optimization algorithm.

[0046] Preferably, the process of adjusting the path curvature by the local optimization algorithm specifically comprises:

[0047] calculating the curvature of each point on the path;

[0048] for the point with curvature exceeding a threshold value, applying the following optimization formula:

[0049] P'i = Pi + λ * SPi-1 + Pi+1 - 2Pi+ μ * SPi+2 + Pi-2 - 2Pi);

[0050] wherein, P'i is the optimized point coordinate, Pi is the current point coordinate, Pi-1, Pi+1, Pi-2 and Pi+2 are adjacent point coordinates, and λ and μ are weight coefficients;

[0051] iteratively performing the above steps until the curvature meets the requirements or a maximum number of iterations is reached.

[0052] The present application provides an indoor unmanned aerial vehicle flight path planning processing system, comprising:

[0053] a three-dimensional map data acquisition module for acquiring three-dimensional map data of an indoor environment; the three-dimensional map data is three-dimensional map data pre-constructed in a database;

[0054] The laser radar scanning module is used for scanning an indoor environment by using a laser radar to obtain real-time obstacle information, that is, scanning the indoor environment by using a laser radar device to obtain a plurality of obstacle point cloud information and obstacle position information in the environment;

[0055] The grid map construction module is used for constructing a grid map based on the three-dimensional map data and the real-time obstacle information to obtain a density and an occupancy surface probability value of the obstacles in each grid, and dividing the grid map into a plurality of partition regions of different levels according to the density and the occupancy surface probability value of the obstacles in the grid.

[0056] The path generation module is used for generating an initial flight path by using an improved rapidly-exploring random tree algorithm based on the plurality of partition regions of different levels.

[0057] The path optimization module is used for performing smoothing processing on the initial flight path to obtain an optimized flight path.

[0058] The flight control module is used for controlling the unmanned aerial vehicle to fly according to the optimized flight path.

[0059] Compared with the prior art, the embodiments of the present application have at least the following technical advantages:

[0060] It can be known from the above-mentioned indoor unmanned aerial vehicle flight path planning processing method and system that three-dimensional map data of an indoor environment is obtained in specific application, the three-dimensional map data is three-dimensional map data constructed in advance in a database, and more accurate and timely environmental information can be obtained by combining the three-dimensional map data constructed in advance (S1) and obstacle information scanned in real time by a laser radar (S2). This is helpful to generate a more accurate flight path and adapt to dynamic changes of the environment.

[0061] The laser radar scanning module is used for scanning an indoor environment by using a laser radar to obtain real-time obstacle information, that is, scanning the indoor environment by using a laser radar device to obtain a plurality of obstacle point cloud information and obstacle position information in the environment;

[0062] The method using the grid map and the obstacle density / occupancy probability can more detailedly describe the obstacle distribution in the environment. Research finds that the division of the hierarchical region helps the unmanned aerial vehicle to better identify and avoid the regions with different risk levels. On the basis of the technology, the improved rapidly expanding random tree algorithm is used to generate an initial path, and the initial path has faster calculation speed and better path quality than the traditional algorithm. The initial flight path is smoothed to obtain an optimized flight path, and the unmanned aerial vehicle is controlled to fly according to the optimized flight path.

[0063] The indoor unmanned aerial vehicle flight path planning processing method and system provided by the application are particularly designed for indoor environments, and the characteristics of indoor environments (such as narrow space and multiple obstacles) are considered, so the method and system are particularly suitable for application in complex indoor environments, and high autonomy and intelligence of the system are embodied. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 A main flowchart of an indoor unmanned aerial vehicle flight path planning processing method of the first embodiment of the application;

[0065] Figure 2 A specific implementation process diagram of the indoor unmanned aerial vehicle flight path planning processing method of the first embodiment of the application;

[0066] Figure 3 Another specific implementation process diagram of the indoor unmanned aerial vehicle flight path planning processing method of the first embodiment of the application;

[0067] Figure 4 Still another specific implementation process diagram of the indoor unmanned aerial vehicle flight path planning processing method of the first embodiment of the application;

[0068] Figure 5 A whole architecture diagram of an indoor unmanned aerial vehicle flight path planning processing system of the second embodiment of the application.

[0069] The drawings show that: three-dimensional map data acquisition module 10, laser radar scanning module 20, grid map construction module 30, path generation module 40, path optimization module 50, flight control module 60. DETAILED DESCRIPTION

[0070] The technical solutions of the application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the application.

[0071] The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0072] Example 1

[0073] like Figure 1 As shown, the first embodiment of the present invention provides a method for indoor UAV flight path planning and processing, including the following steps:

[0074] S1. Acquire three-dimensional map data of an indoor environment; the three-dimensional map data is pre-constructed three-dimensional map data in a database;

[0075] S2. Use LiDAR to scan the indoor environment and obtain real-time obstacle information. That is, the indoor environment is scanned by LiDAR equipment to obtain point cloud information and obstacle location information of multiple obstacles in the environment.

[0076] S3. Construct a grid map based on the three-dimensional map data and real-time obstacle information, and obtain the density and occupancy probability value of the obstacles within each grid; divide the grid map into multiple levels of partitioned areas according to the density and occupancy probability values ​​of the obstacles within the grid;

[0077] S4, based on multiple levels of partitioned areas, an improved fast-expanding random tree algorithm is used to generate the initial flight path;

[0078] S5. Smoothing the initial flight path to obtain an optimized flight path;

[0079] S6. Control the drone to fly according to the optimized flight path.

[0080] An analysis of the indoor drone flight path planning method and system provided by the present invention reveals that, in specific applications, three-dimensional map data of the indoor environment is obtained; the three-dimensional map data is pre-constructed from a database. By combining the pre-constructed three-dimensional map data (S1) with obstacle information (S2) scanned in real time by a laser radar, the method can obtain more accurate and timely environmental information. This helps generate a more accurate flight path while adapting to dynamic changes in the environment.

[0081] The system uses LiDAR to scan the indoor environment and obtain real-time obstacle information. Specifically, the system uses LiDAR to scan the indoor environment, thereby obtaining point cloud information and location information of multiple obstacles within the environment. A grid map is constructed based on the 3D map data and real-time obstacle information to obtain the density and occupancy probability of obstacles within each grid. The grid map is then divided into multiple levels of partitioned areas based on the density and occupancy probability of obstacles within the grid. Based on these multiple levels of partitioned areas, an improved rapidly expanding random tree algorithm is used to generate the initial flight path.

[0082] The method using the grid map and the obstacle density / occupancy probability can more finely describe the obstacle distribution in the environment. Research finds that the division of such hierarchical regions helps the UAV to better identify and avoid regions of different risk levels. On the basis of this technology, an improved rapidly-exploring random tree algorithm is used to generate an initial path, which has faster calculation speed and better path quality than traditional algorithms. The initial flight path is smoothed to obtain an optimized flight path; the UAV is controlled to fly according to the optimized flight path.

[0083] The indoor UAV flight path planning processing method and system provided by the application are particularly designed for indoor environments and take into account the characteristics of indoor environments (such as narrow spaces and multiple obstacles), and are therefore particularly suitable for application in complex indoor environments, while embodying high autonomy and intelligence of the system.

[0084] Preferably, after the UAV is controlled to fly according to the optimized flight path, the method further comprises:

[0085] Collecting obstacle images on the flight path, performing image recognition based on the obstacle images to determine the current obstacle type, screening dangerous obstacles according to the recognized current obstacle type, and marking the position nodes of the dangerous obstacles.

[0086] Preferably, collecting obstacle images on the flight path, performing image recognition based on the obstacle images to determine the current obstacle type, screening dangerous obstacles according to the recognized current obstacle type, and marking the position nodes of the dangerous obstacles, specifically comprises:

[0087] During the flight, a collection frequency is set, and a camera carried by the UAV is used to regularly take obstacle images on the path;

[0088] A standard shape of the dangerous obstacle is obtained by presetting, the collected images are preprocessed, a deep learning vision algorithm is used to analyze and extract the outer contour shape of the current collected image after preprocessing, the outer contour shape is compared with the standard shape of the dangerous obstacle to determine the similarity (if the calculated similarity is greater than a standard threshold, it is determined that the matching is successful), and it is determined whether the type of the current obstacle belongs to the dangerous obstacle, if it belongs, the recognized dangerous obstacle is marked, and the position information (such as coordinates, relative height, etc.) is recorded.

[0089] Referring to Figure 2 , the grid map is constructed based on three-dimensional map data and real-time obstacle information, comprising:

[0090] S31, discretizing the three-dimensional map data into three-dimensional grids;

[0091] S32, mapping the real-time obstacle information into corresponding three-dimensional grids;

[0092] S33, determining the density of obstacles and the probability of occupied surface value for each grid.

[0093] Referring to Figure 3 , the obtaining of the density of obstacles and the probability of occupied surface value in each grid includes:

[0094] S331, obtaining the number of obstacles in each grid in the three-dimensional grid, and calculating the density of obstacles in the grid; the density of obstacles is the number of independent obstacles in a single grid;

[0095] S332, determining the occupancy state of the grid according to the density of obstacles;

[0096] S333, simultaneously calculating the probability of occupied surface value of the obstacles in the current grid, the probability of occupied surface value of the obstacles being the ratio of the occupied obstacle area in the grid to the current grid area.

[0097] The definitions of the occupancy state and the probability of occupied surface value are as follows: the occupancy state (Occupancy State): this is a discrete concept, usually divided into three states: occupied, idle and unknown. Occupied: indicating that there are obstacles in the grid. Idle: indicating that there are no obstacles in the grid. Unknown: indicating that it is impossible to determine whether the grid is occupied.

[0098] The probability of occupied surface value: this is a continuous concept, indicating the ratio probability of the occupied obstacle area in the grid to the current grid area. It is a numerical value between 0 and 1. The occupancy state is a discrete classification result. The probability of occupied surface value is a continuous numerical estimate. The grid map is divided into multiple levels of areas according to the density of obstacles and the probability of occupied surface value, such as high-risk area, medium-risk area and low-risk area.

[0099] Preferably, the density of obstacles and the probability of occupied surface value in the grid divide the grid map into multiple levels of partition areas, specifically including:

[0100] For each grid, calculating a score based on the density of obstacles and the probability of occupied surface value;

[0101] Score = w1 * obstacle density + w2 * probability of occupied surface value; wherein w1 and w2 are preset first weight and second weight;

[0102] Defining multiple threshold values A1, A2, the Score score less than or equal to A1 is a low-risk area, the Score score greater than or equal to A2 is a high-risk area, and the Score score between A1 and A2 is a medium-risk area;

[0103] Traverse the entire current grid map, calculate the score of each grid in the grid map, and divide the risk level according to the threshold value;

[0104] Cluster the adjacent grids of the same risk level to form a continuous risk area.

[0105] Referring to Figure 4 Based on the partition area of multiple levels, an improved rapid expansion random tree algorithm is used to generate an initial flight path:

[0106] S41, node sampling is performed on the current multiple level partition area, and during sampling, different sampling frequencies are given to areas of different risk levels to reduce the sampling probability of high-risk areas; wherein the sampling frequency gradually decreases from low-risk areas, medium-risk areas to high-risk areas;

[0107] S42, set the starting point and the ending point, run the improved RRT algorithm, and generate an initial flight path; during execution, the node of the starting point is used as the initial candidate node, and the distance between the current candidate node and the existing sampled multiple tree nodes within the preset distance range is calculated;

[0108] S43, within the preset distance range, select n tree nodes closest to the current candidate node as parent nodes, and connect the parent nodes and the candidate node using a local planner;

[0109] S44, check whether the connection path intersects with obstacles;

[0110] S45, if not, add the candidate node to the tree;

[0111] S46, repeat steps S41-S45 until a feasible path is found or the maximum number of iterations is reached and the generation of the initial flight path is completed.

[0112] Specific implementation case: a 10x10 indoor environment grid map, each grid size is 0.5m x 0.5m. Score function: Score = 0.6 * obstacle density + 0.4 * occupancy probability value;

[0113] Set threshold: A1 = 0.3 (low-risk and medium-risk boundary); A2 = 0.7 (medium-risk and high-risk boundary); traverse the grid and calculate the score:

[0114] Grid (2, 3): obstacle density = 0.2, occupancy probability = 0.15; Score(2, 3) = 0.6 * 0.2 + 0.4 * 0.15 = 0.18; risk level: low risk.

[0115] Grid (5, 6): Obstacle density = 0.5, Occupancy probability = 0.6; Score(5, 6) = 0.6 * 0.5 + 0.4 * 0.6 = 0.54; Risk level: medium risk.

[0116] Grid (8, 7): Obstacle density = 0.9, Occupancy probability = 0.8; Score(8, 7) = 0.6 * 0.9 + 0.4 * 0.8 = 0.86; Risk level: high risk.

[0117] Region clustering: Combining adjacent grids of the same risk level into a region.

[0118] Preferably, after the generation of the initial flight path is completed, the number of the current initial flight path is also judged, if the current initial flight path generated by the improved rapid expansion random tree algorithm is multiple, the number of the multiple current initial flight paths passing through each risk level region is calculated, and the initial flight path passing through the region with the largest number of low-risk regions is selected as the final target flight path.

[0119] It should be noted that, considering the risk level passed by the path, the path passing through the low-risk region is preferentially selected.

[0120] By assigning different sampling frequencies to different risk level regions, the probability of sampling in high-risk regions is reduced. This strategy effectively guides the path to avoid high-risk regions, thereby greatly improving the safety of the generated path. At the same time, the risk level differentiation sampling strategy adopted by the present technical solution not only improves safety, but also improves the efficiency of path planning. Because the algorithm tends to sample more in low-risk regions, which usually finds a feasible path faster.

[0121] In the specific embodiment process, by selecting multiple tree nodes close to each other as parent nodes instead of only selecting the nearest node, this strategy can improve the diversity and flexibility of the path, which helps to find a better path.

[0122] The above technical solution adopted by the embodiment of the present application checks whether the connection path intersects with obstacles when adding a new node (S44-S45), which ensures that the generated path is feasible and avoids the situation that the UAV crosses obstacles. The algorithm continuously expands the tree structure through repeated iteration (S46), which can achieve a good balance between global path exploration and local path optimization, so it balances global exploration and local optimization. By setting the maximum number of iterations (S46), the algorithm avoids the risk of falling into an infinite loop, improving the robustness and reliability of the system.

[0123] Preferably, the process of smoothing the initial flight path in step S5 comprises:

[0124] S51, interpolating the path points using a cubic spline interpolation method;

[0125] S52, then adjusting the path curvature of the interpolated path points using a local optimization algorithm.

[0126] Preferably, the process of adjusting the path curvature using a local optimization algorithm in step S52 comprises:

[0127] S521, calculating the curvature of each point on the path;

[0128] S522, for points with curvature exceeding a threshold, applying the following optimization formula:

[0129] P'i = Pi +λ* SPi-1 + Pi+1 - 2Pi+μ* SPi+2 + Pi-2 - 2Pi);

[0130] where P'i is the optimized point coordinate, Pi is the current point coordinate, Pi-1, Pi+1, Pi-2 and Pi+2 are adjacent point coordinates, and λ and μ are weight coefficients;

[0131] S523, iteratively performing steps S521 and S522 until the curvature meets the requirements or the maximum number of iterations is reached.

[0132] It can be seen from the above technical solution that the above S521-S523 technical solution adopted by the embodiment of the present application can generate a smooth curve between the original path points by using a cubic spline interpolation method. This method can significantly improve the continuity and smoothness of the path, making the flight trajectory of the unmanned aerial vehicle more smooth and natural. The smooth flight path can reduce the acceleration changes of the unmanned aerial vehicle during flight, thereby optimizing the flight dynamics performance. This not only improves the stability of flight, but also reduces energy consumption and prolongs flight time.

[0133] The cubic spline interpolation method can generate more intermediate path points while maintaining the original key path points. This increases the accuracy of the path, enabling the unmanned aerial vehicle to fly more accurately according to the planned path.

[0134] Finally, it is necessary to implement the improvement of the local characteristics of the path; that is, by adjusting the path curvature using a local optimization algorithm (S52), the local characteristics of the path can be further improved. This optimization can eliminate sharp turns or unreasonable curvature changes in the path, making the path more reasonable and executable.

[0135] It is obvious that a smooth and reasonable curvature path can reduce the violent motion of the unmanned aerial vehicle during flight, thereby reducing the risk of loss of control or collision and improving flight safety.

[0136] Embodiment two

[0137] Referring to Figure 5 The application provides an indoor unmanned aerial vehicle flight path planning processing system, comprising: a three-dimensional map data acquisition module 10, a laser radar scanning module 20, a grid map construction module 30, a path generation module 40, a path optimization module 50, and a flight control module 60.

[0138] The three-dimensional map data acquisition module 10 is used for acquiring three-dimensional map data of an indoor environment; the three-dimensional map data is three-dimensional map data constructed in advance in a database;

[0139] The laser radar scanning module 20 is used for scanning the indoor environment by using a laser radar to acquire real-time obstacle information; that is, the indoor environment is scanned by using a laser radar device, so that multiple obstacle point cloud information and obstacle position information in the environment can be obtained;

[0140] The grid map construction module 30 is used for constructing a grid map based on the three-dimensional map data and the real-time obstacle information, obtaining the density and the occupancy surface probability value of obstacles in each grid, and dividing the grid map into multiple hierarchical partition regions according to the density and the occupancy surface probability value of the obstacles in the grid;

[0141] The path generation module 40 is used for generating an initial flight path by using an improved rapidly-exploring random tree algorithm based on the multiple hierarchical partition regions;

[0142] The path optimization module 50 is used for performing smoothing processing on the initial flight path to obtain an optimized flight path;

[0143] The flight control module 60 is used for controlling the unmanned aerial vehicle to fly according to the optimized flight path.

[0144] In summary, the indoor unmanned aerial vehicle flight path planning processing method and system provided by the application are particularly designed for indoor environments and take into account the characteristics of indoor environments (such as narrow spaces and multiple obstacles), and are therefore particularly suitable for application in complex indoor environments, while also embodying high autonomy and intelligence of the system.

[0145] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the application, but not to limit them; the ordinary skilled in the art can modify the technical solutions described in the above embodiments, or make equivalent replacement for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the application.

Claims

1. A method for indoor UAV flight path planning, characterized in that: The steps are as follows: Acquire three-dimensional map data of the indoor environment; the three-dimensional map data is pre-constructed three-dimensional map data in a database; Use LiDAR to scan the indoor environment and obtain real-time obstacle information; That is, by scanning the indoor environment through the laser radar equipment, the point cloud information and obstacle location information of multiple obstacles in the environment can be obtained; A grid map is constructed based on 3D map data and real-time obstacle information to obtain the density and occupancy probability of obstacles within each grid. The grid map is then divided into multiple levels of partitioned areas based on the density and occupancy probability of obstacles within the grid. Based on multiple levels of partitioned areas, an improved fast-expanding random tree algorithm is used to generate the initial flight path; Smoothing the initial flight path to obtain the optimized flight path; Control the drone to fly according to the optimized flight path; The grid map is constructed based on the three-dimensional map data and the real-time obstacle information, including: Discretize 3D map data into 3D grids; Mapping real-time obstacle information to the corresponding three-dimensional grid; Determine the density of obstacles and the probability of occupancy for each grid; Obtaining the density and occupancy probability of obstacles within each grid includes: Obtain the number of obstacles in each grid of the three-dimensional grid and calculate the density of obstacles in the grid; the density of obstacles is the number of independent obstacles in a single grid; Determine the occupancy status of the grid based on the obstacle density; At the same time, the occupancy probability value of the obstacle in the current grid is calculated, where the occupancy probability value of the obstacle is the ratio of the occupied obstacle area in the grid to the current grid area; The grid map is divided into multiple levels of partition areas according to the density of obstacles and the probability of occupancy within the grid, including: For each grid, a score is calculated based on the obstacle density and the occupancy probability value; Score = w1 * obstacle density + w2 * occupied surface probability value; where w1 and w2 are the preset first weight and second weight; Define multiple thresholds A1 and A2. Score scores less than or equal to A1 are considered low-risk areas, Score scores greater than or equal to A2 are considered high-risk areas, and Score scores between A1 and A2 are considered medium-risk areas. Traverse the entire current grid map, calculate the score for each grid in the grid map, and divide the risk level according to the threshold; Adjacent grids of the same risk level are clustered to form continuous risk areas.

2. The indoor UAV flight path planning and processing method according to claim 1 is characterized in that: After controlling the drone to fly according to the optimized flight path, it also includes: Obstacle images on the flight path are collected, image recognition is performed based on the obstacle images to determine the current obstacle type, dangerous obstacles are screened according to the identified current obstacle type, and dangerous obstacle location nodes are marked.

3. The indoor UAV flight path planning and processing method according to claim 2 is characterized in that: Collecting obstacle images on the flight path, performing image recognition based on the obstacle images to determine the current obstacle type, screening dangerous obstacles according to the identified current obstacle type, and marking dangerous obstacle location nodes, specifically including: During the flight, set the acquisition frequency and use the camera on the drone to regularly capture images of obstacles on the path; The collected images are preprocessed and analyzed using deep learning visual algorithms to identify and determine whether the current obstacle type is a dangerous obstacle. If it is, the identified dangerous obstacle is marked and its location information is recorded.

4. The indoor UAV flight path planning and processing method according to claim 3 is characterized in that: Based on multiple levels of partitioned areas, an improved rapidly expanding random tree algorithm is used to generate the initial flight path, including: Perform node sampling on the current multiple levels of partitioned areas. During sampling, different sampling frequencies are assigned to areas of different risk levels to reduce the sampling probability of high-risk areas. The sampling frequency gradually decreases from low-risk areas to medium-risk areas and then to high-risk areas. Set the starting point and end point, run the improved RRT algorithm, and generate the initial flight path. During the specific execution, the node at the starting point is used as the initial candidate node, and the distance between the current candidate node and multiple existing sampled tree nodes is calculated within the preset distance range. Within the preset distance range, select n tree nodes that are closer to the current candidate node as parent nodes, and use the local planner to connect the parent nodes and candidate nodes; Check whether the connection path intersects with obstacles; If they do not intersect, the candidate node is added to the tree; Repeat the above steps until a feasible path is found or the maximum number of iterations is reached and the generation of the initial flight path is completed.

5. The indoor UAV flight path planning and processing method according to claim 4 is characterized in that: After the initial flight path is generated, it also includes: The number of current initial flight paths is determined. If the improved rapidly expanding random tree algorithm generates multiple current initial flight paths, the number of areas of each risk level passed by the multiple current initial flight paths is calculated, and the initial flight path with the largest number of low-risk areas is selected as the final target flight path.

6. An indoor UAV flight path planning and processing system, which adopts the indoor UAV flight path planning and processing method according to any one of claims 1 to 5, and is characterized in that: include: A three-dimensional map data acquisition module is used to acquire three-dimensional map data of an indoor environment; the three-dimensional map data is pre-constructed three-dimensional map data stored in a database; The LiDAR scanning module is used to scan the indoor environment using LiDAR to obtain real-time obstacle information. That is, the indoor environment is scanned by LiDAR equipment to obtain point cloud information and obstacle location information of multiple obstacles in the environment. The grid map construction module is used to construct a grid map based on 3D map data and real-time obstacle information, obtain the density and occupancy probability value of obstacles within each grid, and divide the grid map into multiple levels of partitioned areas based on the density and occupancy probability values ​​of obstacles within the grid. A path generation module is used to generate an initial flight path based on multiple levels of partitioned areas using an improved rapidly expanding random tree algorithm; The path optimization module smoothes the initial flight path to obtain the optimized flight path; The flight control module is used to control the drone to fly according to the optimized flight path.

Images