CN118068840A - Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product - Google Patents
Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product Download PDFInfo
- Publication number
- CN118068840A CN118068840A CN202410358830.5A CN202410358830A CN118068840A CN 118068840 A CN118068840 A CN 118068840A CN 202410358830 A CN202410358830 A CN 202410358830A CN 118068840 A CN118068840 A CN 118068840A
- Authority
- CN
- China
- Prior art keywords
- boundary
- obstacle
- mowing
- path planning
- path
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000000007 visual effect Effects 0.000 claims abstract description 41
- 230000008569 process Effects 0.000 claims abstract description 15
- 230000004888 barrier function Effects 0.000 claims abstract description 10
- 238000005516 engineering process Methods 0.000 claims abstract description 10
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 238000004590 computer program Methods 0.000 claims description 22
- 238000013135 deep learning Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 5
- 230000002068 genetic effect Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000002922 simulated annealing Methods 0.000 claims description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000003898 horticulture Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a path planning method, equipment, medium and product of a mowing robot, and relates to the field of path planning of mowing robots, wherein the method comprises the following steps: traversing the block boundary by using a down-looking binocular passive vision system carried by the unmanned aerial vehicle to generate a block boundary coordinate graph; identifying an obstacle in the block boundary based on the block boundary coordinate graph, and correcting the block boundary; the obstacle comprises the number, the category and the GPS coordinates of the obstacle; dividing the corrected land block boundary by adopting a visual identification and segmentation technology, determining a movable area and barrier boundary information, and updating the movable area and the barrier boundary information in real time; in the working process of the mowing robot, covering an initial mowing path according to the real-time updated drivable area and obstacle boundary information, generating a new mowing path, and performing a mowing task according to the new mowing path. The invention can improve the path planning efficiency and the adaptability.
Description
Technical Field
The invention relates to the field of path planning of mowing robots, in particular to a path planning method, equipment, medium and product of a mowing robot.
Background
Currently, automation and robotics are gradually penetrating into the field of agriculture and horticulture management. Particularly in large-scale, open-environment work tasks, such as mowing, there is a significant need for automation. Although conventional automated mowing robots have been widely used, they generally suffer from low path planning efficiency, poor adaptability, and the like, particularly in complex and diverse outdoor environments.
Disclosure of Invention
The invention aims to provide a path planning method, equipment, medium and product for a mowing robot, which are used for solving the problems of low path planning efficiency and poor adaptability.
In order to achieve the above object, the present invention provides the following solutions:
A robot lawnmower path planning method, comprising:
Traversing the block boundary by using a down-looking binocular passive vision system carried by the unmanned aerial vehicle to generate a block boundary coordinate graph; the plot boundary coordinate graph is an accurate plot boundary coordinate graph;
identifying an obstacle in the block boundary based on the block boundary coordinate graph, and correcting the block boundary; the obstacle comprises the number, the category and the GPS coordinates of the obstacle;
dividing the corrected land block boundary by adopting a visual identification and segmentation technology, determining a movable area and barrier boundary information, and updating the movable area and the barrier boundary information in real time;
in the working process of the mowing robot, covering an initial mowing path according to the real-time updated drivable area and obstacle boundary information, generating a new mowing path, and performing a mowing task according to the new mowing path.
Optionally, traversing the block boundary by using a down-looking binocular passive vision system on board the unmanned aerial vehicle to generate a block boundary coordinate graph, which specifically comprises:
selecting a block boundary, acquiring a connection relation between rough GPS coordinates and coordinate points of the block boundary, storing the connection relation as a doubly linked list data structure, and generating a rough block boundary coordinate graph of the block boundary; the rough block boundary coordinate graph comprises rough GPS coordinates, a connection relation among coordinate points and a doubly linked list data structure;
Inputting the block boundary coordinate graph to an unmanned plane, starting from a head node, selecting the direction of the next node as the basis of a block boundary route, and establishing a flight task for traversing the block boundary;
In the flight task, continuously acquiring ground images by using a down-looking binocular passive vision system on board an unmanned aerial vehicle, and determining a visual boundary of a land block by means of a deep learning network; the visual boundaries of the land block comprise ridges, ravines, paved roadsides and water bodies;
Combining the boundary point of the visual boundary of the land block and the relative position of the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary point of the visual boundary of the land block based on the GPS coordinates of the unmanned aerial vehicle;
and updating the doubly linked list data structure according to the boundary points of the new visual boundary of the land parcel, the GPS coordinates and the connection relation among the coordinate points, and generating an accurate coordinate diagram of the boundary of the land parcel.
Optionally, identifying an obstacle in the block boundary based on the block boundary coordinate graph, and correcting the block boundary specifically includes:
Based on the plot boundary coordinate graph, identifying obstacles appearing in ground images acquired by a downward-looking binocular passive vision system on the unmanned aerial vehicle according to an obstacle identification model based on deep learning network training while the unmanned aerial vehicle flies along the plot boundary, and storing the numbers, types and GPS coordinates of the obstacles in a dictionary data structure;
And identifying and adjusting areas which cannot be directly accessed due to terrains or obstacles based on the dictionary data structure, and automatically correcting the land parcel boundaries.
Optionally, the visual recognition and segmentation technology is adopted to divide the corrected land block boundary, determine the driving area and the obstacle boundary information, and update the driving area and the obstacle boundary information in real time, which specifically includes:
removing the movable obstacle from the dictionary data structure according to the category of the obstacle, renumbering the rest obstacle, and generating a new dictionary data structure;
Establishing a flight task for traversing the obstacle according to the new dictionary data structure;
according to the flight task, carrying out circumferential flight around the obstacle by using an unmanned aerial vehicle according to the outline edge of the obstacle, continuously acquiring an obstacle image by using a downward-looking binocular passive vision system, and searching the boundary of the obstacle by means of a deep learning network;
Collecting the relative positions of the boundary points of the obstacle and the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary points of the boundary of the obstacle based on the GPS coordinates of the unmanned aerial vehicle;
And updating the dictionary data structure according to the boundary points of the new obstacle boundary, the GPS coordinates and the connection relation among the coordinate points, and updating the drivable area and the obstacle boundary information in real time.
Optionally, updating the dictionary data structure according to the boundary point of the new obstacle boundary, the GPS coordinates, and the connection relationship between the coordinate points, and updating the drivable area and the obstacle boundary information in real time, and then further including:
Outputting a 3D point cloud by utilizing a visual matching algorithm according to the downward-looking binocular passive visual system, and separating ground data and non-ground data; the points in the 3D point cloud are provided with GPS coordinate information;
Drawing points in the 3D point cloud into different columnar body data structures through a deep learning network, and representing occupation conditions in the columnar body data structures;
constructing a data set with ground elevation information, and establishing an autoregressive ground elevation deep learning model according to the data set;
Processing the columnar data structure by using the autoregressive high Cheng Shendu learning model, removing point clouds and shapes which do not belong to the ground in the 3D point clouds, and generating a ground elevation map;
Calculating curvature information of point cloud in the ground elevation map;
Fitting a 3D ground according to the 3D point cloud, and performing terrain analysis, sighting marks and marking terrain features according to the 3D ground and the curvature information; the terrain analysis includes calculating terrain attributes including slope and slope direction; the topographical features include ravines and ridges.
Optionally, the initial mowing path is generated by a random sampling method.
Optionally, during the working process of the mowing robot, covering the initial mowing path according to the real-time updated drivable area and the obstacle boundary information to generate a new mowing path, which specifically includes:
In the working process of the mowing robot, according to the real-time updated drivable area and obstacle boundary information, the mowing path is iteratively optimized by combining a genetic algorithm and a simulated annealing algorithm, the initial mowing path is covered, and a new mowing path is generated.
A computer device, comprising: the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the path planning method of the mowing robot.
A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the above described path planning method for a robot lawnmower.
A computer program product comprising a computer program which, when executed by a processor, implements the steps of the robot lawnmower path planning method described above.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the invention identifies obstacles in the block boundary through integrating the unmanned aerial vehicle exploration and downward-looking binocular passive vision system, divides the corrected block boundary by adopting the vision identification and segmentation technology, determines and updates the movable area and the obstacle boundary information in real time, covers the initial mowing path according to the real-time updated movable area and the obstacle boundary information based on the intelligent path planning optimization technology, generates a new mowing path, and performs mowing tasks according to the new mowing path, and provides a high-efficiency and flexible mowing robot path planning solution, which greatly improves the working efficiency and accuracy of the mowing robot in a changeable environment, thereby promoting the development of the agricultural automation technology.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a path planning method of a robot lawnmower provided in embodiment 1 of the present invention;
fig. 2 is a flowchart of a path planning method for a robot lawnmower according to embodiment 2 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a path planning method, equipment, medium and product for a mowing robot, which improve path planning efficiency and adaptability.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in fig. 1, the path planning method of the mowing robot of the invention comprises the following steps:
Step 101: traversing the block boundary by using a down-looking binocular passive vision system carried by the unmanned aerial vehicle to generate a block boundary coordinate graph; the plot boundary coordinate graph is an accurate plot boundary coordinate graph.
In practical application, the step 101 specifically includes: selecting a block boundary, acquiring a connection relation between rough GPS coordinates and coordinate points of the block boundary, storing the connection relation as a doubly linked list data structure, and generating a rough block boundary coordinate graph of the block boundary; the rough block boundary coordinate graph comprises rough GPS coordinates, a connection relation among coordinate points and a doubly linked list data structure; inputting the block boundary coordinate graph to an unmanned plane, starting from a head node, selecting the direction of the next node as the basis of a block boundary route, and establishing a flight task for traversing the block boundary; in the flight task, continuously acquiring ground images by using a down-looking binocular passive vision system on board an unmanned aerial vehicle, and determining a visual boundary of a land block by means of a deep learning network; the visual boundaries of the land block comprise ridges, ravines, paved roadsides and water bodies; combining the boundary point of the visual boundary of the land block and the relative position of the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary point of the visual boundary of the land block based on the GPS coordinates of the unmanned aerial vehicle; and updating the doubly linked list data structure according to the boundary points of the new visual boundary of the land parcel, the GPS coordinates and the connection relation among the coordinate points, and generating an accurate coordinate diagram of the boundary of the land parcel.
Step 102: identifying an obstacle in the block boundary based on the block boundary coordinate graph, and correcting the block boundary; the obstacle includes the number, category, and GPS coordinates of the obstacle.
In practical applications, step 102 specifically includes: based on the plot boundary coordinate graph, identifying obstacles appearing in ground images acquired by a downward-looking binocular passive vision system on the unmanned aerial vehicle according to an obstacle identification model based on deep learning network training while the unmanned aerial vehicle flies along the plot boundary, and storing the numbers, types and GPS coordinates of the obstacles in a dictionary data structure; and identifying and adjusting areas which cannot be directly accessed due to terrains or obstacles based on the dictionary data structure, and automatically correcting the land parcel boundaries.
Step 103: dividing the corrected land block boundary by adopting a visual identification and segmentation technology, determining a movable area and barrier boundary information, and updating the movable area and the barrier boundary information in real time.
In practical applications, the step 103 specifically includes: removing the movable obstacle from the dictionary data structure according to the category of the obstacle, renumbering the rest obstacle, and generating a new dictionary data structure; establishing a flight task for traversing the obstacle according to the new dictionary data structure; according to the flight task, carrying out circumferential flight around the obstacle by using an unmanned aerial vehicle according to the outline edge of the obstacle, continuously acquiring an obstacle image by using a downward-looking binocular passive vision system, and searching the boundary of the obstacle by means of a deep learning network; collecting the relative positions of the boundary points of the obstacle and the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary points of the boundary of the obstacle based on the GPS coordinates of the unmanned aerial vehicle; and updating the dictionary data structure according to the boundary points of the new obstacle boundary, the GPS coordinates and the connection relation among the coordinate points, and updating the drivable area and the obstacle boundary information in real time.
In practical application, the dictionary data structure is updated according to the boundary point of the new obstacle boundary, the GPS coordinates and the connection relationship between the coordinate points, and the drivable area and the obstacle boundary information are updated in real time, and then the method further includes: outputting a 3D point cloud by utilizing a visual matching algorithm according to the downward-looking binocular passive visual system, and separating ground data and non-ground data; the points in the 3D point cloud are provided with GPS coordinate information; drawing points in the 3D point cloud into different columnar body data structures through a deep learning network, and representing occupation conditions in the columnar body data structures; constructing a data set with ground elevation information, and establishing an autoregressive ground elevation deep learning model according to the data set; processing the columnar data structure by using the autoregressive high Cheng Shendu learning model, removing point clouds and shapes which do not belong to the ground in the 3D point clouds, and generating a ground elevation map; calculating curvature information of point cloud in the ground elevation map; fitting a 3D ground according to the 3D point cloud, and performing terrain analysis, sighting marks and marking terrain features according to the 3D ground and the curvature information; the terrain analysis includes calculating terrain attributes including slope and slope direction; the topographical features include ravines and ridges.
Step 104: in the working process of the mowing robot, covering an initial mowing path according to the real-time updated drivable area and obstacle boundary information, generating a new mowing path, and performing a mowing task according to the new mowing path.
In practical applications, the initial mowing path is generated by using a random sampling method.
In practical application, in the working process of the mowing robot, covering an initial mowing path according to the real-time updated drivable area and obstacle boundary information to generate a new mowing path, which specifically comprises the following steps: in the working process of the mowing robot, according to the real-time updated drivable area and obstacle boundary information, the mowing path is iteratively optimized by combining a genetic algorithm and a simulated annealing algorithm, the initial mowing path is covered, and a new mowing path is generated.
According to the invention, through optimized path planning, time required for completing mowing tasks and energy consumption of the robot are obviously reduced.
The invention can adapt to various terrains and environmental conditions, and automatically adjusts path planning to cope with the terrains and the newly-appearing obstacles.
The invention utilizes advanced visual recognition technology to improve the recognition accuracy of the land block boundary and the obstacle and ensure the whole coverage of the mowing range.
The real-time updating function of the invention enables the mowing robot to self-adjust the path in the working process, thereby enhancing the adaptability to uncertain environments.
Example 2
Based on the path planning method of the robot lawnmower of embodiment 1, in the actual operation, the execution process of the path planning method of the robot lawnmower is as follows, as shown in fig. 2.
Step one: unmanned aerial vehicle parcel boundary traversal and obstacle marking.
And (3) performing aerial reconnaissance on the appointed range by using the unmanned aerial vehicle, and automatically traversing the land block boundary and identifying the obstacle by using the data of public map services (such as Google map).
In public map services (such as google satellite map), rough GPS coordinates of boundaries are provided by selecting the boundaries of a land block, and the connection relation between coordinate points through edges is stored as a data structure of a doubly linked list to form a rough map of the boundaries of the land block.
Through unmanned aerial vehicle operation platform, with the rough map of parcel boundary, including GPS coordinate and the relation of connection between the coordinate, with the data structure input to unmanned aerial vehicle of two-way linked list, from the beginning node, select the direction of next node as the basis of parcel boundary route, establish the flight task of traversing parcel boundary.
In the flight task, an unmanned aerial vehicle airborne computing platform and a downward-looking binocular passive vision system are utilized to continuously collect ground images, and a deep learning scheme is relied on to find visual boundaries of land plots, such as ridges, ravines, paved roadsides, water bodies and the like. And through the passive visual system of binocular, uninterruptedly range finding to the boundary point of visual system output, through the result of range finding, combine the relative position of boundary point and unmanned aerial vehicle body to based on the GPS coordinate of unmanned aerial vehicle body, carry out GPS coordinate assignment to the parcel visual boundary point.
The obtained visual boundary points, along with the GPS coordinates and the connection relation between the coordinates, are updated into the aforementioned doubly linked list data structure.
And generating an accurate block boundary coordinate graph with the doubly linked list as a data structure.
The unmanned aerial vehicle is allowed to process image data in real time in the flight process, the areas which cannot be directly accessed due to the terrain or obstacles are identified and adjusted, and the land block boundaries are automatically corrected.
During the process of flying along the land parcel boundary of the unmanned aerial vehicle, the object appearing in the downlooking binocular passive vision system is identified through a pre-trained object identification model, and the number, the category and the GPS coordinates of the object are stored in a dictionary data structure. To further refine the information about the obstacle in the next step.
Step two: visual recognition and demarcation of the travelable region.
And by adopting a visual recognition and segmentation technology, analyzing images collected by the unmanned aerial vehicle, and accurately dividing a drivable area and obstacles.
And extracting the dictionary data structure stored in the first step, wherein the dictionary data structure comprises object numbers, categories and GPS coordinates identified in the block boundaries.
The class of the object is first determined and if it is determined to be an immovable object, it is retained in the dictionary data structure. If a movable object is determined, it is removed from the dictionary data structure and the numbering of the objects is reordered. Finally, a dictionary data structure is obtained, which covers the number, category and GPS coordinates of all immovable objects identified in the block boundary.
According to the numbers of the objects in the dictionary data structure and the corresponding GPS coordinates, inputting the numbers to an unmanned plane platform, and establishing a flight task of traversing the obstacle.
After the unmanned aerial vehicle reaches the GPS coordinates of the object, the unmanned aerial vehicle flies circumferentially around the object according to the outline edge of the object, and the unmanned aerial vehicle airborne computing platform and the downward-looking binocular passive vision system are utilized to continuously collect the image of the object, and the boundary of the object is found by means of a deep learning scheme. And the object boundary points output by the visual system are continuously measured by the binocular passive visual system, and the object visual boundary points are assigned with GPS coordinates based on the GPS coordinates of the unmanned aerial vehicle body by combining the relative positions of the boundary points and the unmanned aerial vehicle body through the measurement result. And establishing a two-way linked list data structure for storing object boundary points and connection relations between the object boundary points.
After the observation of the object boundary is completed, the obtained doubly linked list data structure is added into the dictionary structure, and finally the dictionary data structure taking the object number, the category and the GPS coordinates as keys and taking the doubly linked list data structure taking the object boundary point GPS coordinates as the content as a value is formed.
After the objects stored in all dictionaries are observed, the numbers, the types, the GPS coordinates and the coordinates of object boundary points of all objects in the block boundaries are finally obtained, and all information is stored in a data structure of the dictionary for subsequent path planning.
Aiming at different types of ground and barriers, multiple visual recognition models are designed, and accuracy and efficiency of area division are improved. On the basis of separating the barriers in the steps, other travelable areas in the block boundary need to be further analyzed and identified for subsequent path planning.
Ground and non-ground data separation: and outputting a 3D point cloud through a visual matching algorithm by depending on a down-looking binocular passive visual system of the unmanned aerial vehicle. And gives the GPS coordinates of the unmanned aerial vehicle body and the inherent distance information of the 3D point cloud, and the points in the 3D point cloud are provided with the GPS coordinate information.
Through PointPillar deep learning networks, points in the 3D point cloud acquired by the unmanned aerial vehicle are respectively drawn into different columnar body data structures to represent occupation conditions in the columnar bodies. A dataset with ground elevation information is constructed using SEMANTICKITTI and used to build an autoregressive ground elevation deep learning model. And processing the columnar body data structure acquired by the unmanned aerial vehicle by using the autoregressive ground elevation deep learning model, so as to remove point clouds and columnar bodies which do not belong to the ground in the 3D point clouds, and obtain a ground elevation map.
Ground 3D curved surface curvature calculation: curvature information in the ground elevation map point cloud is calculated by constructing Voronoi covariance measurements (Voronoi Covariance Measure, VCM). In the point cloud set K, a point x i is selected, and within the radius R, VCM is calculated by the following formula: Wherein V K,R({xi) is the normal cone of the point x i in the point cloud set, based on the measured value of the point x i with the radius R, p i -1(xi) is the normal cone of the point x i in the point cloud set/> For the outer product operation, y belongs to the point set in the Voronoi Cell centered on x i, the definition of Voronoi Cell is Vor (x i)＝{y:dk(y)＝||xi-y||};Vor(xi) is an element based on x i, and d k (y) is the closest distance around x i in the K set to xi.
Based on the fitted 3D ground and the 3D curvature of the ground, terrain analysis is performed, such as calculating terrain attributes of gradient, slope direction and the like, and specific terrain features (such as ravines, ridges and the like) are identified and marked for subsequent path planning.
Step three: path planning and optimization.
An initial mowing path is generated using a random sampling method, such as a fast-explored random tree (Rapidly-exploring RandomTrees, RRT).
And combining genetic algorithm, simulated annealing and other optimization algorithms, and performing iterative optimization on the path to improve the coverage rate and efficiency of the path and reduce repeated mowing and missed mowing areas.
Updating and path re-planning in real time.
And in the working process of the mowing robot, repeatedly calculating the land block boundaries and the obstacle recognition algorithm in the first step and the second step on data acquired by the robot when the robot works on the ground, updating the land block boundary information and updating the obstacle boundary information.
And regenerating a coverage path according to the land block boundary information and the obstacle boundary information data updated in real time, so as to ensure that the mowing task is completed efficiently and comprehensively.
Example 3
A computer device, comprising: the memory, the processor, and the computer program stored on the memory and executable on the processor, the processor executing the computer program to implement the steps of the path planning method of the robot lawnmower of embodiment 1.
Example 4
A computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the path planning method of the robot lawnmower of embodiment 1.
Example 5
A computer program product comprising a computer program which, when executed by a processor, implements the steps of the robot lawnmower path planning method of embodiment 1.
Example 6
A computer device may be a database. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store the pending transactions. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by the processor to implement the path planning method of the robot lawnmower in embodiment 1.
It should be noted that, the object information (including, but not limited to, object device information, object personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) related to the present invention are both information and data authorized by the object or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as Static Random access memory (Static Random access memory AccessMemory, SRAM) or dynamic Random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor according to the embodiments of the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, or the like, but is not limited thereto.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (10)
1. A method for path planning for a robot lawnmower, comprising:
Traversing the block boundary by using a down-looking binocular passive vision system carried by the unmanned aerial vehicle to generate a block boundary coordinate graph; the plot boundary coordinate graph is an accurate plot boundary coordinate graph;
identifying an obstacle in the block boundary based on the block boundary coordinate graph, and correcting the block boundary; the obstacle comprises the number, the category and the GPS coordinates of the obstacle;
dividing the corrected land block boundary by adopting a visual identification and segmentation technology, determining a movable area and barrier boundary information, and updating the movable area and the barrier boundary information in real time;
in the working process of the mowing robot, covering an initial mowing path according to the real-time updated drivable area and obstacle boundary information, generating a new mowing path, and performing a mowing task according to the new mowing path.
2. The path planning method of a robot lawnmower of claim 1, wherein traversing the parcel boundary with the down-looking binocular passive vision system onboard the unmanned aerial vehicle generates a parcel boundary graph, comprising:
selecting a block boundary, acquiring a connection relation between rough GPS coordinates and coordinate points of the block boundary, storing the connection relation as a doubly linked list data structure, and generating a rough block boundary coordinate graph of the block boundary; the rough block boundary coordinate graph comprises rough GPS coordinates, a connection relation among coordinate points and a doubly linked list data structure;
Inputting the block boundary coordinate graph to an unmanned plane, starting from a head node, selecting the direction of the next node as the basis of a block boundary route, and establishing a flight task for traversing the block boundary;
In the flight task, continuously acquiring ground images by using a down-looking binocular passive vision system on board an unmanned aerial vehicle, and determining a visual boundary of a land block by means of a deep learning network; the visual boundaries of the land block comprise ridges, ravines, paved roadsides and water bodies;
Combining the boundary point of the visual boundary of the land block and the relative position of the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary point of the visual boundary of the land block based on the GPS coordinates of the unmanned aerial vehicle;
and updating the doubly linked list data structure according to the boundary points of the new visual boundary of the land parcel, the GPS coordinates and the connection relation among the coordinate points, and generating an accurate coordinate diagram of the boundary of the land parcel.
3. The robot lawnmower path planning method of claim 1, wherein identifying an obstacle within the parcel boundary based on the parcel boundary graph, correcting the parcel boundary, specifically comprises:
Based on the plot boundary coordinate graph, identifying obstacles appearing in ground images acquired by a downward-looking binocular passive vision system on the unmanned aerial vehicle according to an obstacle identification model based on deep learning network training while the unmanned aerial vehicle flies along the plot boundary, and storing the numbers, types and GPS coordinates of the obstacles in a dictionary data structure;
And identifying and adjusting areas which cannot be directly accessed due to terrains or obstacles based on the dictionary data structure, and automatically correcting the land parcel boundaries.
4. The path planning method of a robot lawnmower of claim 3, wherein the visual recognition and segmentation technique is used to divide the corrected parcel boundary, determine the drivable area and the obstacle boundary information, and update the drivable area and the obstacle boundary information in real time, and specifically comprising:
removing the movable obstacle from the dictionary data structure according to the category of the obstacle, renumbering the rest obstacle, and generating a new dictionary data structure;
Establishing a flight task for traversing the obstacle according to the new dictionary data structure;
according to the flight task, carrying out circumferential flight around the obstacle by using an unmanned aerial vehicle according to the outline edge of the obstacle, continuously acquiring an obstacle image by using a downward-looking binocular passive vision system, and searching the boundary of the obstacle by means of a deep learning network;
Collecting the relative positions of the boundary points of the obstacle and the unmanned aerial vehicle, and carrying out GPS coordinate assignment on the boundary points of the boundary of the obstacle based on the GPS coordinates of the unmanned aerial vehicle;
And updating the dictionary data structure according to the boundary points of the new obstacle boundary, the GPS coordinates and the connection relation among the coordinate points, and updating the drivable area and the obstacle boundary information in real time.
5. The path planning method of claim 4, wherein updating the dictionary data structure according to the boundary point of the new obstacle boundary, the GPS coordinates, and the connection relationship between the coordinate points, updating the travelable area and the obstacle boundary information in real time, and further comprising:
Outputting a 3D point cloud by utilizing a visual matching algorithm according to the downward-looking binocular passive visual system, and separating ground data and non-ground data; the points in the 3D point cloud are provided with GPS coordinate information;
Drawing points in the 3D point cloud into different columnar body data structures through a deep learning network, and representing occupation conditions in the columnar body data structures;
constructing a data set with ground elevation information, and establishing an autoregressive ground elevation deep learning model according to the data set;
Processing the columnar data structure by using the autoregressive high Cheng Shendu learning model, removing point clouds and shapes which do not belong to the ground in the 3D point clouds, and generating a ground elevation map;
Calculating curvature information of point cloud in the ground elevation map;
Fitting a 3D ground according to the 3D point cloud, and performing terrain analysis, sighting marks and marking terrain features according to the 3D ground and the curvature information; the terrain analysis includes calculating terrain attributes including slope and slope direction; the topographical features include ravines and ridges.
6. The robot lawnmower path planning method of claim 1, wherein the initial mowing path is generated using a random sampling method.
7. The method for planning a path of a mowing robot according to claim 1, wherein the step of generating a new mowing path by covering the initial mowing path according to the real-time updated travelable area and obstacle boundary information during the operation of the mowing robot comprises the steps of:
In the working process of the mowing robot, according to the real-time updated drivable area and obstacle boundary information, the mowing path is iteratively optimized by combining a genetic algorithm and a simulated annealing algorithm, the initial mowing path is covered, and a new mowing path is generated.
8. A computer device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the robot lawnmower path planning method of any of claims 1-7.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the path planning method of a robot lawnmower according to any one of claims 1-7.
10. A computer program product comprising a computer program, characterized in that the computer program/instructions, when executed by a processor, realizes the steps of the robot lawnmower path planning method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410358830.5A CN118068840A (en) | 2024-03-27 | 2024-03-27 | Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410358830.5A CN118068840A (en) | 2024-03-27 | 2024-03-27 | Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118068840A true CN118068840A (en) | 2024-05-24 |
Family
ID=91099116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410358830.5A Pending CN118068840A (en) | 2024-03-27 | 2024-03-27 | Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118068840A (en) |
-
2024
- 2024-03-27 CN CN202410358830.5A patent/CN118068840A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019238711B2 (en) | Method and apparatus for acquiring boundary of area to be operated, and operation route planning method | |
| CN108763287B (en) | Construction method of large-scale passable regional driving map and unmanned application method thereof | |
| CN110222626B (en) | Unmanned scene point cloud target labeling method based on deep learning algorithm | |
| Zhang et al. | Hierarchical topic model based object association for semantic SLAM | |
| CN110770791A (en) | Image boundary acquisition method and device based on point cloud map and aircraft | |
| CN109255302A (en) | Object recognition methods and terminal, mobile device control method and terminal | |
| Hu et al. | Research on a single-tree point cloud segmentation method based on UAV tilt photography and deep learning algorithm | |
| CN113706710A (en) | Virtual point multi-source point cloud fusion method and system based on FPFH (field programmable gate flash) feature difference | |
| CN113110455B (en) | Multi-robot collaborative exploration method, device and system for unknown initial state | |
| US11861855B2 (en) | System and method for aerial to ground registration | |
| CN116630403A (en) | Lightweight semantic map construction method and system for mowing robot | |
| CN112949407A (en) | Remote sensing image building vectorization method based on deep learning and point set optimization | |
| Aguiar et al. | Localization and mapping on agriculture based on point-feature extraction and semiplanes segmentation from 3D LiDAR data | |
| CN113310488A (en) | Orchard robot navigation method based on SLAM | |
| CN114120067A (en) | Object identification method, device, equipment and medium | |
| CN114758086A (en) | Method and device for constructing urban road information model | |
| CN116739739A (en) | Loan amount evaluation method and device, electronic equipment and storage medium | |
| CN118068840A (en) | Mowing robot path planning method, mowing robot path planning equipment, mowing robot path planning medium and mowing robot path planning product | |
| CN114397894B (en) | Mobile robot target searching method imitating human memory | |
| CN113947665B (en) | Method for constructing map of spherical hedge trimmer based on multi-line laser radar and monocular vision | |
| CN114998539A (en) | Smart city sensor terrain positioning and mapping method | |
| Mao et al. | An adaptive localization system for image storage and localization latency requirements | |
| CN109901589B (en) | Mobile robot control method and device | |
| Ma et al. | Road Curbs Extraction from Mobile Laser Scanning Point Clouds with Multidimensional Rotation‐Invariant Version of the Local Binary Pattern Features | |
| Hroob et al. | Learned long-term stability scan filtering for robust robot localisation in continuously changing environments |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |