An unmanned aerial vehicle autonomous charging navigation method, device, equipment and medium

Abstract

This invention belongs to the field of power technology, specifically disclosing a method, device, equipment, and medium for autonomous charging and navigation of unmanned aerial vehicles (UAVs). It includes the following steps: scanning environmental information to obtain LiDAR data; identifying charging station landmarks from the LiDAR data and establishing a coordinate system; processing the coordinate system using the ICP method to obtain pose gain; acquiring the odometer information corresponding to the pose gain; fusing the odometer information and pose gain using the extended Kalman filter method to obtain an estimated pose; and using the estimated pose as input for a PID control method to complete navigation. This invention, by applying landmark recognition and positioning navigation technology, replaces the traditional manual charging method, greatly reducing the need for human resources, lowering UAV charging costs, and improving charging efficiency.

Description

Technical Field

[0001] This invention belongs to the field of power technology, specifically relating to a method, device, equipment, and medium for autonomous charging and navigation of unmanned aerial vehicles (UAVs). Background Technology

[0002] With the development of drone technology, the requirements for the autonomy and independence of drones are becoming increasingly stringent. Applications in industries and home services demand that drones operate autonomously for extended periods. Traditional manual charging methods not only hinder the intelligentization of drones but also waste human resources. Therefore, enabling drones to safely, quickly, and efficiently charge autonomously without human intervention is a key technology for achieving drone intelligence. Summary of the Invention

[0003] The purpose of this invention is to provide an autonomous charging and navigation method, device, equipment, and medium for unmanned aerial vehicles (UAVs) to solve the technical problem that existing UAV charging methods require human intervention and cannot charge autonomously.

[0004] To achieve the above objectives, the present invention employs the following technical solution:

[0005] Firstly, a method for autonomous charging and navigation of a drone includes the following steps:

[0006] Scan the environment to obtain LiDAR data;

[0007] Identify charging station landmarks from LiDAR data and establish a coordinate system;

[0008] The pose gain is obtained by processing the coordinate system using the ICP method.

[0009] Obtain odometry information corresponding to the pose gain;

[0010] The estimated pose is obtained by fusing odometry information and pose gain using an extended Kalman filter.

[0011] The estimated pose is used as the input to the PID control method to control the drone to move to the target charging position.

[0012] A further improvement of the present invention is that: before identifying charging pile road signs from lidar data, the lidar data is clustered, and the clustering process specifically includes the following steps:

[0013] Clustering is performed on the lidar data to find the breakpoints in a frame of lidar data and divide the lidar data into several clusters.

[0014] Connect the points in each cluster with lines, and use a split-merge algorithm. Take the clustered point set S as input, and divide it into n subsets {S1,...,S2}. n};

[0015] All points in each point set S are fitted to a straight line using the least squares method, such that the distance from all points in each point set S to the straight line fitted by this point set does not exceed a first preset value.

[0016] A further improvement of the present invention is that: when dividing the lidar into several clusters, the classification is performed according to the relationship between the Euclidean distance d between adjacent points in the lidar data and the first preset value threshold.

[0017] A further improvement of the present invention is that: when fitting the data to a straight line, the specific steps include:

[0018] A1. Take the point set S as input;

[0019] A2. Take the two endpoints p1 and p2 from the point set S to obtain the initial endpoint set E = {p1, p2}. N};

[0020] A3. Using the least squares method to divide (p i ,p i+1 Fit a straight line L to all points between ) and calculate point p. i and point p i+1 The distance d from the point between them to the line L m ;

[0021] A4. Determine d m The maximum value d in k The relationship between d and the first preset value, if d k If the value is greater than the first preset value, then insert point pk into the initialized endpoint set E, where E = {..., pk}. i ,p k ,p i+1 ,...}, then repeat A3 segmentation (p i ,p k ) and (p k ,p i+1 ) interval points, up to d k Less than or equal to a preset value.

[0022] A further improvement of the present invention is that: when obtaining the pose gain by processing the coordinate system using the ICP method, the specific steps include:

[0023] Obtain road sign template data;

[0024] The pose transformation relationship between the road sign template data and the lidar data is obtained by scanning and matching the road sign template data and lidar data.

[0025] The pose gain in the coordinate system is obtained based on the pose transformation relationship between the road sign template data and the lidar data.

[0026] A further improvement of the present invention is that the road sign template data is ideal data obtained by scanning charging pile road signs with a lidar at a preset reference pose.

[0027] A further improvement of the present invention is that: when fusing odometry information and pose gain using the extended Kalman filter to obtain the estimated pose, it includes pose prediction and pose update. The pose prediction obtains observation information by calculating the Jacobian matrix of the motion model and updating the covariance matrix of the UAV pose state. The pose update obtains the expression form of Kalman gain by calculating the Jacobian matrix of the observation model and combining it with the covariance. The pose estimate is obtained based on the Kalman gain and the observation information.

[0028] Secondly, an autonomous charging and navigation device for unmanned aerial vehicles (UAVs) includes:

[0029] LiDAR: Used to scan environmental information and obtain LiDAR data;

[0030] Road sign recognition module: used to identify charging piles from LiDAR data and establish a coordinate system;

[0031] Pose gain calculation module: used to process the coordinate system using the ICP method to obtain pose gain;

[0032] Odometry: Used to obtain odometry information corresponding to pose gain;

[0033] Pose estimation module: used to fuse odometry information and pose gain using extended Kalman filtering to obtain estimated pose;

[0034] Control module: Used to complete navigation by using the estimated pose as input for PID control.

[0035] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the above-described autonomous charging and navigation method for unmanned aerial vehicles.

[0036] Fourthly, a computer-readable storage medium stores a computer program that, when executed by a processor, implements the aforementioned autonomous charging and navigation method for unmanned aerial vehicles.

[0037] Compared with the prior art, the present invention has at least the following beneficial effects:

[0038] 1. This invention replaces the traditional manual charging method by applying road sign recognition and positioning navigation technology, which greatly reduces the use of human resources, reduces the charging cost of drones, and improves charging efficiency;

[0039] 2. This invention ensures that laser scanning points in the same cluster belong to the same continuous object in the environment by performing clustering processing on lidar data, thereby increasing the accuracy of road sign identification;

[0040] 3. The algorithm steps used in this invention are simple, the entire charging process is relatively easy to implement, navigation is fast and accurate, and charging efficiency is greatly improved. Attached Figure Description

[0041] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0042] In the attached diagram:

[0043] Figure 1 This is a flowchart of an autonomous charging and navigation method for unmanned aerial vehicles (UAVs) according to the present invention;

[0044] Figure 2 This is a first schematic diagram of the ICP method in the autonomous charging and navigation method for unmanned aerial vehicles of the present invention;

[0045] Figure 3 This is a second schematic diagram of the ICP method in the autonomous charging and navigation method for unmanned aerial vehicles of the present invention;

[0046] Figure 4 This is a logic block diagram of the PID navigation control method in an autonomous charging and navigation method for unmanned aerial vehicles (UAVs) according to the present invention.

[0047] Figure 5 This is a structural block diagram of an autonomous charging and navigation device for unmanned aerial vehicles (UAVs) according to the present invention. Detailed Implementation

[0048] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0049] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0050] Example 1

[0051] A method for autonomous charging and navigation of unmanned aerial vehicles, such as Figure 1 As shown, it includes the following steps:

[0052] S1. Scan the environment to obtain lidar data;

[0053] When the drone's battery level drops below the preset level, the drone will charge itself. While traveling to the charging station, it will scan the external environment using lidar.

[0054] S2. Identify charging piles from lidar data and establish a coordinate system;

[0055] The charging station markers are manual markers placed near the drone's work site. The drone's lidar scan data identifies the charging stations, and a rectangular coordinate system is established using the intersection of the two legs of the identified isosceles triangle charging station as the origin of the local coordinate system.

[0056] After acquiring a frame of LiDAR data, the first step is to identify charging station landmarks from the laser data and extract the laser data belonging to those landmarks. Clustering is then performed on the scanned LiDAR data to identify breakpoints within a single frame. If the interval between adjacent points is greater than or equal to a first preset value, these two points are considered to belong to different clusters, thus dividing the LiDAR data into different clusters. Line fitting is recursively performed on points within the same cluster using the least squares method, fitting the points in the same cluster into line segments. The shape, side length, and included angle between adjacent sides of the charging station landmarks are known data. After obtaining the line segments, a segment-finding operation is performed. Segment length and included angle are two conditions for searching line segments. The segment length must be less than 10cm compared to the charging station reference data, and the included angle must be within a tolerance range of ±10°. Multiple segments may meet the conditions; the segment connected by the nearest hinge point to the LiDAR is selected as the final landmark, completing the identification of the manually marked charging station.

[0057] Before identifying charging station road signs from LiDAR data in S2, the LiDAR data is clustered. The clustering process includes the following steps:

[0058] S21. Perform clustering processing on the lidar data, determine whether the Euclidean distance d between adjacent points in the lidar data exceeds the first preset value threshold, find the breakpoint in a frame of lidar data, and divide the lidar data into several clusters.

[0059] The steps in S21 ensure that the laser scanning points in the same cluster belong to the same continuous object in the environment, thereby increasing the accuracy of road sign identification.

[0060] S22. Connect the points in each cluster with lines, using a split-merge algorithm. Take the clustered point set S as input and divide it into n subsets {S1,...,S2}.n};

[0061] S23. All points in each point set are fitted to a straight line using the least squares method, ensuring that the distance from all points in each point set to the straight line fitted by this point set does not exceed the first preset value.

[0062] When fitting a straight line in S23, the specific steps include:

[0063] S231. Take the point set S as input;

[0064] S232. Take the two endpoints p1 and p2 from the point set S to obtain the initial endpoint set E = {p1, p2, p3}. N};

[0065] S233, Using the least squares method to divide (p i ,p i+1 Fit a straight line L to all points between ) and calculate point p. i and point p i+1 The distance d from the point between them to the line L m ;

[0066] S234, Determine d m The maximum value d in k The relationship between d and the first preset value, if d k If the value is greater than the threshold, then insert point p into the initialization endpoint set E. k E = {..., p i ,p k ,p i+1 ,...}, then repeat step S233 to segment (p i ,p k ) and (p k ,p i+1 ) interval points, up to d k Less than or equal to the threshold.

[0067] As an example of the present invention, the first preset value is set to 20cm in this solution.

[0068] The charging station road signs used in this plan are generally in the shape of an isosceles triangle with a side length of 30cm and an included angle of 120° between adjacent sides.

[0069] S3. Obtain the pose gain by processing the coordinate system using the ICP method; such as... Figure 2-3 As shown;

[0070] S31. Obtain road sign template data;

[0071] The road sign template data is the standard for the ICP matching algorithm. The road sign template data is ideal data obtained by scanning human road signs at a manually set reference pose using a lidar. This data can be calculated and saved offline in advance.

[0072] S32. Perform scanning and matching based on the road sign template data and the lidar data to obtain the pose transformation relationship between the road sign template data and the lidar data;

[0073] Coordinate system X L O L Y L The origin is fixed at the intersection of the two legs of the artificial road sign (i.e., the isosceles triangle charging station);

[0074] Coordinate system X L O L Y L The pose q = (x, y, θ) = (t, θ) is given. The corresponding second transformation matrix is ​​expressed as:

[0075]

[0076] Where R(θ) is the orthogonal rotation matrix, and the homogeneous transformation matrix of the reference pose is denoted as... The homogeneous transformation matrix of the current pose of the UAV is denoted as: The homogeneous transformation matrix of the current pose relative to the reference pose is denoted as The pose transformation relationship between road sign template data and road sign scan data is denoted as: They satisfy the following:

[0077]

[0078]

[0079] S33. Obtain the pose gain in the coordinate system based on the pose transformation relationship between the road sign template data and the lidar data;

[0080] Pose gain in coordinate system:

[0081]

[0082] The reference pose is manually selected, therefore It is a known quantity.

[0083] S4. Obtain the odometry information corresponding to the pose gain;

[0084] The odometer information represents the distance traveled by the drone from the moment it began searching for a charging station until the attitude gain occurred.

[0085] S5. The estimated pose is obtained by fusing the odometer information and pose gain using the extended Kalman filter method.

[0086] By fusing pose gain and odometry information using an extended Kalman filter, the local pose of the UAV in the coordinate system established by the charging station is estimated. This process mainly consists of two parts: pose prediction and pose update. The pose prediction part primarily calculates the Jacobian matrix of the motion model and updates the covariance matrix of the UAV pose state estimation to obtain the observation information. The pose update part primarily calculates the Jacobian matrix of the observation model, and combined with the covariance, obtains the expression for the Kalman gain. Finally, the pose estimate is obtained using the Kalman gain and the observation information.

[0087] S6, such as Figure 4 As shown, the estimated pose is used as the input of the PID control method to control the drone to move to the target charging position.

[0088] The principle of PID controller is as follows:

[0089]

[0090] Among them, K p T is the proportionality coefficient. i Let T be the integration time constant. d Let e(t) be the differential time constant, e(t) be the input of the PID controller, and u(t) be the output of the PID controller.

[0091] PID control feeds back the deviation between the given value (in this case, the drone's local pose) and the measured actual value, and uses proportional (P), integral (I), and derivative (D) operations to correct the deviation, thereby controlling the drone to move precisely to the target position for charging.

[0092] Example 2

[0093] like Figure 5 As shown, an autonomous charging and navigation device for unmanned aerial vehicles (UAVs), based on the aforementioned autonomous charging and navigation method for UAVs, includes:

[0094] LiDAR: Used to scan environmental information and obtain LiDAR data;

[0095] Road sign recognition module: used to identify charging station road signs from LiDAR data and establish a coordinate system;

[0096] Pose gain calculation module: used to process the coordinate system using the ICP method to obtain pose gain;

[0097] Odometry: Used to obtain odometry information corresponding to pose gain;

[0098] Pose estimation module: used to fuse odometry information and pose gain using extended Kalman filtering to obtain estimated pose;

[0099] Control module: Used to complete navigation by using the estimated pose as input for PID control.

[0100] Example 3

[0101] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements an autonomous charging and navigation method for a drone according to Embodiment 1.

[0102] Example 4

[0103] A computer-readable storage medium storing a computer program that, when executed by a processor, implements an autonomous charging and navigation method for a drone as described in Embodiment 1.

[0104] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative in all respects and are not the only ones. All modifications within the scope of this invention or its equivalents are included in this invention.

[0105] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0107] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0108] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent 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 autonomous charging and navigation of unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: Scan the environment to obtain LiDAR data; Charging station landmarks are identified from LiDAR data, and a local coordinate system is established based on these landmarks. Each landmark is an isosceles triangle, and the origin of the local coordinate system is located at the intersection of the two legs of the isosceles triangle. The LiDAR data is clustered to identify breakpoints in a frame, dividing the data into clusters. Points within each cluster are connected by lines, and a segmentation-merging algorithm is used to obtain the clustered point sets. As input, it is divided into Subset All points in each point set S are fitted to a straight line using the least squares method, ensuring that the distance from each point in S to the fitted straight line does not exceed a first preset value; when dividing the lidar into several clusters, the distance is determined based on the Euclidean distance between adjacent points in the lidar data. Compared with the first preset value Classify the size relationships between them; Clustering and line fitting are performed on LiDAR data to identify charging station road signs. The line fitting uses a segmentation-merging algorithm, specifically including: A1, taking the point set S as input; A2, taking the last two points p1 and p2 in the point set S to obtain the initialized endpoint set. A3. Using the least squares method to... Fit a straight line L to all points between them, and calculate the points. and points Distance from the point between them to line L A4, Judgment The maximum value in The relationship between the magnitude of the value and the first preset value, if If the value is greater than the first preset value, then insert point pk into the initialized endpoint set E. Repeat the A3 segmentation. and Interval points, until Less than or equal to a preset value; Acquire road sign template data; perform scanning and matching based on road sign template data and LiDAR data to obtain the pose transformation relationship between road sign template data and LiDAR data; obtain the pose gain in the coordinate system based on the pose transformation relationship between road sign template data and LiDAR data; the road sign template data is the ideal data obtained by LiDAR scanning charging pile road signs at a preset reference pose. Obtain the odometry information corresponding to the pose gain; based on the odometry information and pose gain, use the extended Kalman filter method to fuse them to obtain the estimated pose of the UAV in the local coordinate system; use the estimated pose as the input of the PID control method to control the UAV to move to the target charging position. When estimating the pose using the extended Kalman filter method based on odometry information and pose gain, the process includes pose prediction and pose update. Pose prediction obtains observation information by calculating the Jacobian matrix of the motion model and updating the covariance matrix of the UAV pose state. Pose update obtains the expression of the Kalman gain by calculating the Jacobian matrix of the observation model and combining it with the covariance. The pose estimate is then obtained based on the Kalman gain and the observation information.

2. An autonomous charging and navigation device for unmanned aerial vehicles (UAVs), used to implement the autonomous charging and navigation method for UAVs as described in claim 1, characterized in that, include: LiDAR: Used to scan environmental information and obtain LiDAR data; Road sign recognition module: used to identify charging station road signs from LiDAR data and establish a coordinate system; Pose gain calculation module: used to process the coordinate system using the ICP method to obtain pose gain; Odometry: Used to obtain odometry information corresponding to pose gain; Pose estimation module: used to fuse odometry information and pose gain using extended Kalman filtering to obtain estimated pose; Control module: Used to complete navigation by using the estimated pose as input for PID control.

3. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the autonomous charging and navigation method for unmanned aerial vehicles as described in claim 1.

4. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the autonomous charging and navigation method for unmanned aerial vehicles as described in claim 1.

Images