WO2023134659A1 - 地图处理方法、自移动的园艺设备、以及自动割草机 - Google Patents

地图处理方法、自移动的园艺设备、以及自动割草机 Download PDF

Info

Publication number
WO2023134659A1
WO2023134659A1 PCT/CN2023/071525 CN2023071525W WO2023134659A1 WO 2023134659 A1 WO2023134659 A1 WO 2023134659A1 CN 2023071525 W CN2023071525 W CN 2023071525W WO 2023134659 A1 WO2023134659 A1 WO 2023134659A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
map
self
moving gardening
gardening equipment
Prior art date
Application number
PCT/CN2023/071525
Other languages
English (en)
French (fr)
Inventor
俞天宁
唐伟杰
李春红
Original Assignee
未岚大陆(北京)科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 未岚大陆(北京)科技有限公司 filed Critical 未岚大陆(北京)科技有限公司
Publication of WO2023134659A1 publication Critical patent/WO2023134659A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
    • G05D1/0274—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means using mapping information stored in a memory device

Definitions

  • the invention relates to the technical field of smart devices, in particular to a map processing method, self-moving gardening equipment, and an automatic lawnmower.
  • Existing self-moving gardening equipment also known as self-service gardening equipment or smart gardening equipment or automatic gardening equipment, etc.
  • self-service gardening equipment or smart gardening equipment or automatic gardening equipment, etc. can automatically mow grass without supervision or control, thereby reducing Occupying the user's time can also reduce the repetitive work of the user.
  • the existing self-moving gardening equipment relies on manual work to limit the working area, such as embedding the edges of the working area and the edges of obstacles, so that the self-moving gardening equipment can identify boundaries and obstacles by detecting embedding lines to prevent walking out of the work area. area or collision obstacles in the work area.
  • the problem with this thread embedding method is that it takes a lot of time for the user, or professional operators are required to embedding the thread, and it is expensive. For scenes with multiple working areas, there is a problem of complicated embedding.
  • an embodiment of the present invention which provides a map processing method, self-moving gardening equipment, and an automatic lawn mower, so as to at least solve the problem of poor mapping effect of existing self-moving gardening equipment.
  • One or more embodiments of the present invention provide a map processing method, including: controlling self-moving gardening equipment to collect a first environmental image in a local area and first pose information of the self-moving gardening equipment, the The local area is connected to the position of the charging pile; an initial map is generated according to the first environmental image of the local area and the first pose information of the self-moving gardening equipment, so that the self-moving gardening equipment can be based on The initial map is automatically returned to the pile; the self-moving gardening equipment is driven to collect the second environment image of the remaining area and the second pose information of the self-moving gardening equipment; the remaining area at least partially surrounds the local area ; According to the second environment image, the second pose information, and the initial map, determine a map of the first area, where the first area includes the partial area and the remaining area.
  • a map processing device including: a first driving module, configured to control self-moving gardening equipment to collect a first environmental image in a local area and a first image of the self-moving gardening equipment Pose information, the local area is connected to the position of the charging pile; a first determination module, configured to generate an initial map, so that the self-moving gardening equipment can automatically return to the pile based on the initial map; the second driving module is used to drive the self-moving gardening equipment to collect the second environmental image of the remaining area and the self-moving The second pose information of gardening equipment; the remaining area at least partially surrounds the local area; a second determination module, configured to determine according to the second environment image, the second pose information, and the initial map A map of the first area, where the first area includes the partial area and the remaining area.
  • self-propelled gardening equipment comprising a controller configured to perform the above method.
  • an automatic lawnmower which includes a controller configured to execute the above method.
  • the self-moving gardening equipment can realize the self-construction of the map, and in the self-construction of the map, first map the range where the charging pile is located to form the initial map, so as to ensure the self-moving gardening equipment
  • the device is conveniently returned to the pile for charging when the battery is low, thereby ensuring the reliability of self-moving gardening equipment.
  • the self-moving gardening equipment can move in the remaining area to collect the second environmental image and combine the initial map and the second environmental image to generate a map of the first area, so as to realize self-moving gardening equipment Self-contained mapping, thereby saving costs and reducing labor intensity.
  • FIG. 1A is a flow chart of the steps of the map processing method provided by Embodiment 1 of the present application;
  • FIG. 1B is a schematic diagram of a working area including multiple areas in Embodiment 1 of the present application;
  • FIG. 2 is a flow chart of the steps of the map processing method in Embodiment 1 of the present application;
  • FIG. 3A is a flow chart of the steps of the map processing method provided in Embodiment 4 of the present application.
  • Fig. 3B is a schematic diagram of moving within a certain range of the charging pile according to Embodiment 4 of the present application;
  • FIG. 3C is a schematic diagram of moving in the remaining area according to Embodiment 4 of the present application.
  • FIG. 3D is a schematic diagram of the original map of the first area in Embodiment 4 of the present application.
  • FIG. 3E is a schematic diagram of exploring an unexplored region in Embodiment 4 of the present application.
  • FIG. 3F is a schematic diagram of the first area and the target area of Embodiment 4 of the present application.
  • FIG. 4 is a structural block diagram of a map processing device according to Embodiment 4 of the present invention.
  • the mobile gardening equipment may be an automatic lawn mower, and of course, in other embodiments, the self-moving gardening equipment may also be other self-service gardening equipment, or other self-propelled equipment.
  • the automatic lawn mower can be used to trim the lawn to ensure that the height of the grass in the lawn meets the demand.
  • An automatic lawn mower includes a drive unit, a controller, a positioning component, and a mowing knife. Lawn mowers are used to cut grass.
  • the positioning component may include one or more of satellite positioning, ultra-wideband wireless communication positioning (UWB), inertial measurement unit (IMU), image acquisition device and wheel speedometer.
  • the pose of the automatic lawn mower at the current moment can be detected through the positioning component. Pose includes its position and attitude. The position can be represented by its coordinates on the X-axis, Y-axis and Z-axis of the positioning coordinate system. A pose can be determined using its angle relative to the X, Y, and Z axes.
  • the image acquisition device can collect the environment image at the location of the self-moving gardening equipment, and then can determine the pose of the self-moving gardening equipment according to the environment image.
  • the drive unit includes at least two sets of drive wheels and a drive assembly for controlling the rotation direction and rotational speed of the at least two sets of drive wheels.
  • a drive assembly may include a motor and a motor controller. Each group of driving wheels can be connected with a motor, and the rotation of the motor drives the driving wheels to rotate.
  • the motor controller can be a frequency converter or a PLC chip, etc., and the motor controller controls the rotation direction and speed of the motor.
  • the automatic lawn mower includes two sets of driving wheels, and each set of driving wheels includes at least one driving wheel.
  • the two sets of driving wheels rotate at the same speed to realize the forward or backward movement of the automatic lawn mower, and the differential rotation of the two groups of driving wheels realizes the turning of the automatic lawn mower.
  • the controller is connected with the motor controller and the positioning component with electric signals, generates a control signal according to the current pose of the automatic lawn mower detected by the positioning component, and sends the control signal to the motor controller to control the rotation of the motor.
  • the controller can also be connected to the control device, which can be integrated on the automatic lawn mower or can be independent of the automatic lawn mower.
  • the control device When the control device is integrated on the automatic lawn mower, it may include a display screen, control buttons and supporting circuit boards.
  • the control device When the control device is independent from the automatic lawn mower, the control device can be any suitable smart terminal, such as a smart phone, PAD, smart watch or computer.
  • the automatic lawn mower is connected with the smart terminal data to realize data interaction.
  • a control application program of an automatic lawn mower can be run on the smart phone, so as to control the automatic lawn mower through the control application program.
  • the following method can be used to carry out autonomous visual mapping of the working area, so that the self-moving gardening equipment can obtain a map of the working area, which can not only guide the self-moving
  • the self-moving gardening equipment moves in the working area to perform automatic work, and the self-moving gardening equipment can know the edges of the working area and the edges of obstacles. In this way, the self-moving gardening equipment can be moved automatically without additional embedding or marking stakes in the working area, reducing the use cost and workload.
  • FIG. 1A shows a schematic flowchart of the steps of the map processing method according to Embodiment 1 of the present application.
  • Step S102 Drive the self-moving gardening equipment to move within the range in the first area where the distance from the charging pile satisfies a set value, and collect environmental images within the range and pose information of the self-moving gardening equipment.
  • the working area of the self-moving gardening equipment may include one or more than one area, and charging piles may be set in some areas or in all areas. In this embodiment, if only one area is provided with charging posts, then the area where charging posts are provided is recorded as the first area. If charging piles are set in more than one area, one area can be selected as the first area.
  • the self-moving gardening equipment Before the self-moving gardening equipment builds a map, the self-moving gardening equipment cannot know the environment of the working area. In order to ensure the reliability, safety and stability of the mapping process, and avoid problems caused by insufficient power of the self-moving gardening equipment If the map building fails or the map building process is suspended, the user needs to manually operate or carry the self-moving gardening equipment.
  • the self-moving gardening equipment When building the map, the self-moving gardening equipment will be driven to move within the range of the distance from the charging pile to meet the set value, and the range will be collected.
  • the setting value can be determined according to needs, and there is no limitation on this.
  • the set value can be manually configured by the user or other personnel, for example, the set value is 10 meters to 15 meters.
  • the setting value can be selected and determined according to the area of the first area at a certain ratio. For example, if the area of the first area is 100 square meters, the value of the setting value can be such that the area of the range is 50 square meters.
  • the set value can be determined according to the total power of the self-moving gardening equipment. For example, if the total power is 10000mAh, the set value can be 10 meters to 15 meters.
  • the self-moving gardening equipment can use the charging pile as the starting point, randomly select a direction, move along this direction, and take an environmental image of the location at regular intervals, and use the wheel speedometer and IMU (Inertial navigation system) collects the pose information from the moving gardening equipment to determine the pose information corresponding to the environment image.
  • IMU Inertial navigation system
  • Step S104 Generate an initial map according to the environment image of the range and the pose information of the self-moving gardening equipment, so that the self-moving gardening equipment can automatically return to the stake based on the initial map.
  • an initial map can be established through motion reconstruction (SFM, Structure From Motion).
  • the initial map includes feature points of objects within the range, such as feature points of grass, feature points of trees, feature points of stones, feature points of pits, and so on.
  • the established initial map can enable the self-moving gardening equipment to perform path planning, navigation and positioning for automatic pile return. For example, if the self-moving gardening equipment detects that the power of the battery is low during the movement, the self-moving gardening equipment can be guided back to the charging pile for charging according to the initial map.
  • the remaining areas outside the range of the established initial map in the first area can be determined, and the remaining areas can be explored through steps S106 and S108, so as to complete the mapping of the first area.
  • Step S106 Driving the self-moving gardening equipment to move in the remaining area outside the range of the first area, and collecting environmental images of the remaining area and pose information of the self-moving gardening equipment .
  • the self-moving gardening equipment is driven to move in the remaining area, and the environmental image of the location of the self-moving gardening equipment can be collected at regular intervals, and the corresponding pose of the environmental image can be determined through the wheel speedometer, IMU or satellite positioning module, etc. information.
  • Step S108 Determine the map of the first area according to the environment image of the remaining area, the pose information of the self-moving gardening equipment, and the initial map.
  • a map of the remaining area is established based on the environment image and its corresponding pose information, and the map of the remaining area is merged with the initial map to obtain a map of the first area.
  • the autonomous mapping of the self-moving gardening equipment can be realized, and in the autonomous mapping, the range where the charging pile is located is firstly mapped to form an initial map, so as to ensure that the self-moving gardening equipment is in low power. It is convenient to return to the pile for charging, thereby ensuring the reliability of self-moving gardening equipment.
  • the self-moving gardening equipment can move in the remaining area to collect the environmental image of the remaining area and combine the initial map and the environmental image of the remaining area to generate a map of the first area, so as to realize self-moving Independent mapping of gardening equipment, thereby saving costs and reducing labor intensity.
  • FIG. 2 shows a schematic flowchart of the steps of the map processing method according to Embodiment 2 of the present application.
  • Step S202 Control the self-moving gardening equipment to collect a first environment image and first pose information of the self-moving gardening equipment in a local area, the local area is adjacent to the location of the charging pile.
  • the working area of the self-moving gardening implement may comprise one or more than one area.
  • the area where the charging pile corresponding to the self-moving gardening equipment is located may be referred to as the first area.
  • the first area a part of the area that is in contact with the location of the charging pile is recorded as a local area.
  • the local area may be an area within five meters around the charging pile.
  • the first environmental image can be collected by traveling, etc., and the corresponding first pose information can be collected by the mounted sensor.
  • Step S204 Generate an initial map according to the first environmental image of the local area and the first pose information of the self-moving gardening equipment, so that the self-moving gardening equipment can automatically return to the post based on the initial map .
  • An initial map can be established through motion reconstruction (SFM, Structure From Motion) according to the first environment image and the corresponding first pose information.
  • the initial map includes feature points of objects within the range, such as feature points of grass, feature points of trees, feature points of stones, feature points of pits, and so on.
  • the established initial map can enable the self-moving gardening equipment to perform path planning, navigation and positioning for automatic pile return. For example, if the self-moving gardening equipment detects that the power of the battery is low during the movement, the self-moving gardening equipment can be guided back to the charging pile for charging according to the initial map.
  • Step S206 Drive the self-moving gardening equipment to collect a second environment image of a remaining area and second pose information of the self-moving gardening equipment; the remaining area at least partially surrounds the local area.
  • the remaining area may be an area other than the partial area in the first area.
  • the self-moving gardening equipment is driven to move in the remaining area, so that the second environment image and the corresponding second pose information can be collected during its movement.
  • Step S208 Determine the map of the first area according to the second environment image, the second pose information, and the initial map, where the first area includes the local area and the remaining area.
  • a map is established for the remaining area based on the second environment image and its corresponding second pose information, and the map of the remaining area is merged with the initial map to obtain the first area. map.
  • the map of the first area can be conveniently established, and when there are problems such as insufficient power during the map building process, it can also return to the charging pile for charging in time.
  • the method can realize self-construction of self-moving gardening equipment, thereby saving cost and reducing labor intensity.
  • FIG. 3A shows a schematic flowchart of the steps of the map processing method according to Embodiment 3 of the present application.
  • the method includes the following steps:
  • Step S302 controlling the self-moving gardening equipment to collect a first environment image and first pose information of the self-moving gardening equipment in a local area, the local area being adjacent to the location of the charging pile.
  • the working area of the self-moving gardening equipment includes two mutually independent and disconnected areas as an example for illustration, and the number of independent areas is more than two. The method is similar, so it is not repeated here.
  • the area where the charging pile is installed may be selected as the first area.
  • the self-moving gardening equipment can be set on the charging pile.
  • Self-moving gardening equipment is withdrawn from the charging station. The exit process may be: if the front of the self-moving gardening equipment is facing the charging pile, retreat from the charging pile and rotate 180 degrees on the spot.
  • the self-moving gardening equipment can be in a circular area with a radius of 10-15 meters centered on the charging pile (that is, a local area, the distance between the local area and the charging pile meets the set value, that is, Figure 3B Random driving within the initial exploration area shown in ), as shown in Figure 3B.
  • the self-moving gardening equipment collects the first environmental image through the image acquisition device mounted on it, and detects the first pose information of the self-moving gardening equipment through the wheel speedometer or IMU.
  • the first environmental image can generate an initial map in subsequent steps;
  • the neural network model can be used to perform image recognition on the first environmental image to identify whether there is an obstacle, and judge whether the obstacle is moving relative to itself. The orientation and distance of the equipment for obstacle avoidance from moving gardening equipment.
  • Step S304 Generate an initial map according to the first environmental image of the local area and the first pose information of the self-moving gardening equipment, so that the self-moving gardening equipment can automatically return to the pile based on the initial map .
  • the first pose information it can be determined whether the self-moving gardening equipment has traversed the local area. If it has traversed, the first environmental image can be processed based on the first pose information corresponding to the first environmental image in the collected local area. Processing, such as obtaining the initial map corresponding to the local area through SFM (visual 3D reconstruction) or visual SLAM.
  • SFM visual 3D reconstruction
  • SLAM visual SLAM
  • Step S306 Drive the self-moving gardening equipment to collect a second environment image of a remaining area and second pose information of the self-moving gardening equipment; the remaining area at least partially surrounds the local area.
  • mapping for an independent area can be referred to as single-area mapping, which refers to autonomously moving and drawing a passable map of the grass and visual feature map.
  • the mapping of the first region may be referred to as single-region mapping. Since the initial map has been established for the local area within a certain range of the charging pile, during the map building process, for the remaining area outside the local area in the first area, self-moving gardening equipment can be driven to move in the remaining area, and every The second environment image of the location is collected once in a period of time, and the second pose information corresponding to the second environment image is located, as shown in FIG. 3C .
  • a neural network model capable of obstacle identification can be used to identify the collected second environment image, so as to determine whether there is an obstacle, and if there is an obstacle, the self-moving gardening equipment can avoid the obstacle and go around.
  • the self-moving gardening equipment can be driven to move within the remaining area using appropriate strategies.
  • Random strategy turn the angle randomly; or drive in a straight line, if you encounter an obstacle, turn a certain angle randomly in the direction away from the obstacle to avoid the obstacle, and continue to drive in a straight line.
  • Edge-following strategy According to the identified obstacle edge, drive along the edge.
  • Random strategy mixed edge strategy execution For example, according to a certain time ratio, mix the above two strategies.
  • the self-moving gardening equipment In the process of collecting the second environmental image in the remaining area, if it is detected that the battery of the self-moving gardening equipment is insufficient, the self-moving gardening equipment will suspend the autonomous movement and return to the pile for charging according to the initial map. After the battery is sufficient, continue to explore and move independently.
  • Step S308 Determine the map of the first area according to the second environment image, the second pose information, and the initial map, where the first area includes the local area and the remaining area.
  • the stake after the exploration is determined to be terminated, the stake can be automatically returned according to the initial map. After being piled up, the self-moving gardening equipment can be charged and obtain the map of the first area through the following process.
  • Process A1 Determine an original map of the first area according to the initial map, the second environment image, and the second pose information.
  • the process A1 can be implemented as follows: according to the second environment image and its corresponding second pose information, use a three-dimensional reconstruction (SFM) method to establish a visual map of the remaining area, and combine the initial map and the visual map of the remaining area Fusion is performed to obtain the original visual map of the first region.
  • SFM three-dimensional reconstruction
  • obstacle recognition is performed on the second environment image and the first environment image corresponding to the initial map through the obstacle recognition neural network model, and a passable map corresponding to the first area is obtained based on the recognized obstacles and their positions.
  • the second pose information includes the pose determined based on the wheel speedometer and the IMU, but does not include the position of the satellite positioning obtained based on the satellite positioning, the above method is adopted.
  • the original map of the first region includes an original visual map and a passable map, and the original map of the first region can be generated according to the original visual map and the passable map.
  • the original visual map can be positioned and navigated in the subsequent automatic work process, and since it includes the initial map, it can also be automatically piled back.
  • the passable map can carry out path planning and obstacle avoidance because it contains obstacle information.
  • the process A1 can be implemented as: According to the satellite positioning position of the self-moving gardening equipment, determine the mapping pose of the pose of the image acquisition device in the geographic coordinate system; according to the initial map, the second environment image and the mapping position pose, determine the original map of the first region.
  • the position of satellite positioning can be obtained based on RTK technology.
  • satellite positioning position for example, longitude and latitude
  • the orthogonal (eg, due east and north as coordinate axes) GNSS track positions are obtained.
  • the conversion relationship between the GNSS trajectory position and the pose of the image acquisition device can be determined, and then the mapping pose for converting the pose of the image acquisition device into the UTM coordinate system can be determined.
  • the pose information of the environmental image carries the latitude and longitude and the covariance information of the latitude and longitude (the covariance information indicates the confidence degree of the latitude and longitude, if the satellite signal strength is good, the confidence degree indicated by the co-defense difference information is high, indicating that the latitude and longitude more reliable), and then based on the pose information, the reliability of the original map of the first region obtained by using the SFM method is better. This improves the robustness of the original visual map and the navigable map in the original map.
  • the original map contains multiple feature points, which can be feature points in the original map, or feature points of reference objects (such as grass, pits, trees, etc.) in the second environment image, or some feature points are the feature points of the initial map, and the other part of the feature points are the feature points of at least one reference object in the second environment image.
  • feature points can be feature points in the original map, or feature points of reference objects (such as grass, pits, trees, etc.) in the second environment image, or some feature points are the feature points of the initial map, and the other part of the feature points are the feature points of at least one reference object in the second environment image.
  • Process B1 Determine whether there is an unexplored part in the original map of the first region, and if so, perform supplementary mapping on the first region to obtain a map of the first region.
  • process B11 Since there may be unexplored parts in the first area due to obstacles and other reasons during the exploration process, it can be determined whether there are unexplored parts in the first area after obtaining the original map of the first area, and if there is, the process B11 is executed; otherwise, If it does not exist, process B12 is executed.
  • the determining whether there is an unexplored part in the original map of the first region may be implemented as: determining each Distribution information of feature points; according to the distribution information, determine whether there is an unexplored part in the original map of the first region.
  • the original visual map in the original map is divided into blocks, and the number of feature points in each block is counted as distribution information.
  • abnormal blocks whose number of feature points is less than a set number (which can be determined according to needs, such as 10, 20, etc.) can be determined. If multiple abnormal blocks are connected, and the connected area is greater than or equal to the set area (it can be determined according to needs, such as greater than or equal to the area of a self-moving gardening equipment), then it will be determined as an unexplored area (including unexplored areas).
  • the original map of the first area of the explored area is shown in Figure 3D), and the position of the unexplored area can also be determined at the same time. Otherwise, exception blocks are ignored.
  • Process B11 Perform supplementary mapping on the first region.
  • the process B11 can be implemented as: driving the self-moving gardening equipment to move in the unexplored area, and collecting a third environment image of the unexplored area and a third pose of the self-moving gardening equipment information; using the third environment image and the third pose information to update the original map of the first area, so as to obtain the map of the first area.
  • the self-moving gardening equipment can be driven to move into the unexplored area and automatically explore it.
  • the exploration strategy can be consistent with the aforementioned strategy, so it will not be described again.
  • the schematic diagram of the process of moving in the unexplored area is shown in Figure 3E.
  • the third environment image of the unexplored area is collected, and the third pose information corresponding to the third environment image is determined.
  • the third pose information can be based on the existing The original visual map is determined, so that the coordinate system of the third pose information of the newly collected environment image is consistent with the coordinate system of the original visual map.
  • the self-moving gardening equipment can be automatically returned to the pile for charging, and the original visual map and the passable map can be updated by SFM to obtain the map of the first area.
  • the above process can be repeated for each unexplored area until the supplementary mapping of all unexplored areas is completed.
  • Process B12 Complete single-region mapping.
  • the method further includes the following steps:
  • Step S310 Determine whether there is a second area other than the existing map area in the working area, and if there is a second area, create a map corresponding to the second area.
  • the second area may be an area independent of the area of the existing map.
  • the working area includes area A and area B, wherein the first area is area A, which has been mapped. Then area B can be used as the second area.
  • the working area includes area A, area B, and area C, where the first area is area A, which has completed mapping, and area B as the second area has also completed mapping, then area C can be used as a new Mapping of the second region.
  • a feasible way of determining the second area may be: asking the user through the display device whether the second area exists, and if the user indicates that it exists, creating a map corresponding to the second area. Or, if the user indicates that it does not exist, the mapping is completed.
  • Process A2 Determine the target area from the second area.
  • the second area it is directly determined as the second area. Or, if there are more than one second area, one of them can be randomly determined as the target area, or, according to the existing map, the second area closest to the position of the self-moving gardening equipment can be selected as the target area, or, it can also be selected by The user designates a second area as the target area.
  • step S310 may include process B2.
  • Process B2 determining the passageway between the area where the self-moving gardening equipment is located and the target area.
  • the map of the area where the self-moving gardening equipment is located contains at least part of the target area, then according to the map of the area where the self-moving gardening equipment is located, Determine the passing channel, as shown in Figure 3F.
  • the map of the first area can be converted into an image or a two-dimensional grid form, where grass areas and non-grass areas are divided.
  • the grass area at the location and all visible grass areas are extracted by means of image morphology, wherein the grass area that is visible but not at the location may be the grass in the target area.
  • Use the image morphology method to divide the isolated area between two grasslands, and then select a passage from the isolated area and mark it on the map.
  • the selection method passage can be: cross the shortest path in the isolation area; or, when the self-moving gardening equipment includes a depth sensor that can collect three-dimensional information, calculate a smooth passage according to the terrain of the isolation area; or, display the collected map to the user and terrain, the user draws a passage or selects one from recommended passages.
  • the passageway can be set in a way specified by the user, and the user can remotely control the self-moving gardening equipment to the target area to be mapped, and the self-moving gardening equipment will automatically Identify impassable areas between grass fields and record the paths traversed as passages.
  • process C2 After moving to the target area through the passageway, process C2 can be performed.
  • the process B2 can be omitted, and the self-moving gardening equipment can be moved to the target area in other adaptive ways.
  • Process C2 collecting a fourth environment image of the target area, and determining a map of the target area according to the fourth environment image.
  • the self-moving gardening equipment can explore in the target area according to the strategy, and collect a fourth environmental image of the location at regular intervals, and determine the fourth digit corresponding to the fourth environmental image Posture information. Then, after the fourth environment image is collected, a map of the target area is generated based on the fourth environment image and the fourth pose information.
  • the map includes a visual map and a traversable map.
  • the aforementioned passable map can be generated on the basis of the pose of the image acquisition device acquired by the visual map and combined with the obstacle information obtained by recognition.
  • the passable area, obstacles, and unknown areas are expressed in three different colors in the passable map.
  • Process D2 Determine whether the termination condition is satisfied, and if the termination condition is not satisfied, return to determine the target region from the second region to continue execution.
  • the termination condition may be that the difference between the area of the mapped area and the area of the work area designated by the user is less than or equal to the area threshold (which can be determined as required, for example, 0, or other values greater than 0). If the termination condition is satisfied, the mapping is completed. If the termination condition is not satisfied, a new target region is determined from the remaining second region, and the map is constructed until the termination condition is met, and the map construction is completed.
  • the area threshold which can be determined as required, for example, 0, or other values greater than 0.
  • FIG. 4 it shows a structural block diagram of a map processing device according to Embodiment 4 of the present application.
  • the unit includes:
  • the first driving module 402 is used to control the self-moving gardening equipment to collect the first environmental image and the first pose information of the self-moving gardening equipment in a local area, the local area is connected to the position of the charging pile;
  • the first determining module 404 is configured to generate an initial map according to the first environmental image of the local area and the first pose information of the self-moving gardening equipment, so that the self-moving gardening equipment can be based on the The initial map is automatically returned to the pile;
  • the second driving module 406 is configured to drive the self-moving gardening equipment to collect a second environment image of the remaining area and the second pose information of the self-moving gardening equipment; the remaining area at least partially surrounds the local area ;
  • the second determining module 408 is configured to determine a map of the first area according to the second environment image, the second pose information, and the initial map, and the first area includes the local area and the initial map. the rest of the area.
  • the second determination module 408 is configured to determine the original map of the first area according to the initial map, the second environment image, and the second pose information; determine the first area Whether there is an unexplored part in the original map of , and if so, supplementary mapping is performed on the first region to obtain a map of the first region.
  • the original map contains a plurality of feature points
  • the feature points include at least one of feature points in the initial map and feature points of at least one reference object in the second environment image
  • the The second determination module 408 is used to determine the distribution information of each feature point in the original map of the first area when determining whether there is an unexplored part in the original map of the first area; according to the distribution information, determine the Whether there are unexplored parts of the original map of the first area.
  • the second determining module 408 is configured to drive the self-moving gardening equipment in the unexplored area when supplementary mapping is performed on the first area to obtain a map of the first area. and collect a third environment image of the unexplored area and third pose information of the self-moving gardening equipment; use the third environment image and the third pose information to update the first The original map of the region to obtain the map of the first region.
  • the pose information of the self-moving gardening equipment includes the pose of the image acquisition device of the self-moving gardening equipment and the position of the satellite positioning of the self-moving gardening equipment; the second determination module 408 uses When determining the original map of the first area according to the initial map, the second environment image, and the second pose information, determine the position according to the satellite positioning of the self-moving gardening equipment.
  • the device also includes:
  • the third determination module 410 is configured to determine whether there is a second area other than the area of the existing map in the working area, and if there is a second area, create a map corresponding to the second area.
  • the third determination module 410 is configured to determine the target area from the second area when establishing the map corresponding to the second area; collect a fourth environmental image of the target area, and The fourth environment image determines the map of the target area; determines whether the termination condition is satisfied, and if the termination condition is not satisfied, return to determine the target area from the second area to continue execution.
  • the third determination module 410 is further configured to determine the passageway between the area where the self-moving gardening equipment is located and the target area when the map corresponding to the second area is established.
  • the third determination module 410 is configured to determine the passage between the area where the self-moving gardening equipment is located and the target area, if the map of the area where the self-moving gardening equipment is located contains at least part of The target area, then according to the map of the area where the self-moving gardening equipment is located, the passageway is determined.
  • the device can realize the effects corresponding to the above methods, so details are not described again.
  • a self-moving gardening equipment which includes a controller, and the controller is configured to execute the above method and achieve corresponding effects, which will not be repeated here.
  • an automatic lawn mower which includes a controller, and the controller is configured to execute the above method and achieve corresponding effects, which will not be repeated here.
  • first and second are only used to describe different components or names conveniently, and shall not be understood as indicating or implying a sequence relationship, relative importance or implicit indication The number of technical characteristics indicated.
  • the features defined as “first” and “second” may explicitly or implicitly include at least one of these features.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Remote Sensing (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Electromagnetism (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

一种地图处理方法、自移动的园艺设备、以及自动割草机。地图处理方法包括:控制自移动的园艺设备采集局部区域内的第一环境图像和自移动的园艺设备的第一位姿信息，局部区域与充电桩的位置相接(S202);根据局部区域的第一环境图像和自移动的园艺设备的第一位姿信息，生成初始地图，以使自移动的园艺设备能够基于初始地图自动回桩(S204)；驱动自移动的园艺设备采集剩余区域的第二环境图像和自移动的园艺设备的第二位姿信息；剩余区域至少部分包围局部区域(S206)；根据第二环境图像、第二位姿信息和初始地图，确定第一区域的地图，第一区域包括局部区域和剩余区域(S208)。本方法能够快速、方便地建立地图。

Description

地图处理方法、自移动的园艺设备、以及自动割草机 技术领域
本发明涉及智能设备技术领域，尤其涉及一种地图处理方法、自移动的园艺设备、以及自动割草机。
背景技术
现有的自移动的园艺设备(也可以称为自助园艺设备或者智能园艺设备或者自动园艺设备等)，如自动割草机，可以在无人照看或者控制的情况下进行自动割草，从而减少对使用者时间的占用，也可以减少使用者的重复劳动。
现有的自移动的园艺设备依赖人工对工作区域进行限定，如，对工作区域的边沿以及障碍物的边沿埋线，使自移动园艺设备通过检测埋线识别边界和障碍物，以防止走出工作区域或者碰撞工作区域中的障碍物。这种埋线的方式存在的问题在于需要用户消耗大量时间，或需要专业的作业人员进行埋线，而且价格昂贵。对于存在多片工作区域的场景，存在埋线复杂的问题。
发明内容
鉴于上述问题，提出了本发明实施例，其提供一种地图处理方法、自移动的园艺设备、以及自动割草机，以至少解决现有的自移动的园艺设备建图效果不好的问题。
本发明的一个或者多个实施例提供一种地图处理方法，包括：控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接；根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩；驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域；根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。根据本发明的另一方面，提供一种地图处理装置，包括：第一驱动模块，用于控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接；第一确定模块，用于根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩；第二驱动模块，用于驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域；第二确定模块，用于根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
根据本发明的另一方面，提供一种自移动的园艺设备，其包括控制器，所述控制器用于执行上述的方法。
根据本发明的另一方面，提供一种自动割草机，其包括控制器，所述控制器用于执行上述的方法。
通过本实施例，通过这种方式可以实现自移动的园艺设备的自主建图，而且在自主建图时，先对充电桩所在的范围进行建图，以形成初始地图，从而保证自移动的园艺设备在低电量时方便地回桩充电，从而保证了自移动的园艺设备的可靠性。在建立完初始地图后，自移动的园艺设备可以在剩余区域内移动，以采集第二环境图像并结合初始地图和第二环境图像生成第一区域的地图，以此实现对自移动的园艺设备的自主建图，从而节省成本，降低劳动强度。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1A为本申请的实施例一提供的地图处理方法的步骤流程图；
图1B为本申请的实施例一的包含多个区域的工作区域的示意图；
图2为本申请的实施例一的地图处理方法的步骤流程图；
图3A为本申请的实施例四提供的地图处理方法的步骤流程图；
图3B为本申请的实施例四的在充电桩一定范围内移动的示意图；
图3C为本申请的实施例四的在剩余区域内移动的示意图；
图3D为本申请的实施例四的第一区域的原始地图的示意图；
图3E为本申请的实施例四的探索未探索区域的示意图；
图3F为本申请的实施例四的第一区域和目标区域的示意图；
图4为本发明的实施例四的地图处理装置的结构框图。
具体实施方式
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述。显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
移动的园艺设备可以是自动割草机，当然，在其他实施例中，该自移动的园艺设备还可以是其他自助园艺设备，或者，其他可以实现自行走的设备。
自动割草机可以用于对草坪进行修剪，以保证草坪内的草的高度满足需求。自动割 草机包括驱动单元、控制器、定位组件和割草刀等。割草刀用于割草。定位组件可以包括卫星定位、超宽带无线通信定位(UWB)、惯性测量单元(IMU)、图像采集装置和轮速计中的一种或几种。通过定位组件可以检测出自动割草机当前时刻的位姿。位姿包括其位置和姿态。位置可以采用其在定位坐标系的X轴、Y轴和Z轴的坐标表示。姿态可以采用其相对X轴、Y轴和Z轴的角度确定。
图像采集装置可以采集自移动的园艺设备所在位置处的环境图像，进而可以根据环境图像确定自移动的园艺设备的位姿。
驱动单元包括至少两组驱动轮和控制所述至少两组驱动轮的转动方向和转速的驱动组件。驱动组件可以包括电机和电机控制器。每组驱动轮可以连接一个电机，通过电机的转动带动驱动轮转动。电机控制器可以是变频器或者PLC芯片等，电机控制器控制电机的转动方向和转速。
以自动割草机包括两组驱动轮，每组驱动轮包括至少一个驱动轮。两组驱动轮同速转动实现自动割草机的前进或者后退，两组驱动轮的差速转动实现自动割草机的转弯。
控制器与电机控制器、以及定位组件电信号连接，根据定位组件检测的自动割草机当前时刻的位姿，生成控制信号，并将控制信号发送给电机控制器，以使其控制电机转动。
除此之外，控制器还可以与控制设备连接，控制设备可以集成在自动割草机上，也可以独立于自动割草机。当控制设备集成在自动割草机上时，其可以包括显示屏、控制按钮及配套的电路板。当控制设备独立于自动割草机上时，控制设备可以是任何适当的智能终端，如智能手机、PAD、智能手表或者计算机等。
自动割草机与智能终端数据连接，从而实现数据交互。以智能手机为例，智能手机上可以运行自动割草机的控制应用程序，以通过控制应用程序操控自动割草机。
为了保证自移动的园艺设备能够自动在工作区域内移动，可以通过下述方法对工作区域进行自主视觉建图，从而使得自移动的园艺设备可以获得工作区域的地图，该地图不仅能够指导自移动的园艺设备在工作区域内移动从而进行自动工作，而且可以使自移动的园艺设备知道工作区域的边沿和障碍物的边沿。这样在不额外增加工作区域内埋线或者设置标识桩的情况下，使自移动的园艺设备可以自动移动，降低了使用成本和工作量。
下面对该方法的实现过程进行说明如下：
实施例一
参照图1A，示出了本申请的实施例一的地图处理方法的步骤流程示意图。
步骤S102：驱动自移动的园艺设备在第一区域的与充电桩的距离满足设定值的范围内移动，并采集所述范围内的环境图像和所述自移动的园艺设备的位姿信息。
如图1B所示，自移动的园艺设备的工作区域可以包括一个或一个以上的区域，充电桩可以设置在部分区域内或者设置在全部区域内。在本实施例中，若仅有一个区域内设置有充电桩，则设置充电桩的区域记作第一区域。若一个以上的区域内设置有充电桩， 可以任选一个区域作为第一区域。
在自移动的园艺设备建图之前，自移动的园艺设备无法获知工作区域的环境，为了保证建图过程的可靠性、安全性和稳定性，避免由于自移动的园艺设备的电量不足等问题导致建图失败或者建图过程中止需要用户手动操作或者搬运自移动的园艺设备，在建图时，驱动自移动的园艺设备在与充电桩的距离满足设定值的范围内移动，并采集该范围内的环境图像，以及确定每个环境图像对应的位姿信息。这样后续可以先建立该范围内的初始地图，以在自移动的园艺设备的电量较低时基于该初始地图自动回桩充电，进而保证可靠性。
其中，设定值可以根据需要确定，对此不作限制。例如，设定值可以由用户或者其他人员手动配置，如设定值为10米到15米等。或者，设定值可以根据第一区域的面积，以一定比例选择确定，如第一区域的面积100平米，则设定值的取值可以为使该范围的面积为50平米的取值。又或者，设定值可以根据自移动的园艺设备的总电量确定，如总电量为10000mAh，则设定值可以为10米到15米等。
一种示例中，自移动的园艺设备可以以所在的充电桩作为起点，随机选取一个方向，沿该方向移动，并每隔一段时间拍摄一张所在位置的环境图像，并通过轮速计和IMU(惯性导航系统)采集自移动的园艺设备的位姿信息，以确定环境图像对应的位姿信息。
步骤S104：根据所述范围的环境图像和所述自移动的园艺设备的位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩。
根据该范围的环境图像和对应位姿信息可以通过运动重建(SFM，Structure From Motion)建立初始地图。初始地图包括了范围内物体的特征点，如草的特征点、树木的特征点、石头的特征点、和凹坑的特征点等等。
建立的初始地图可以使自移动的园艺设备依据其进行自动回桩的路径规划和导航、定位。例如，若自移动的园艺设备在移动过程中检测到电池的电量较低，则可以根据初始地图引导自移动的园艺设备回到充电桩充电。
在建立完初始地图后可以确定第一区域内已经建立初始地图的范围之外的剩余区域，并通过步骤S106和步骤S108对剩余区域进行探索，以便完成第一区域的建图。
步骤S106：驱动所述自移动的园艺设备在所述第一区域的所述范围之外的剩余区域内移动，并采集所述剩余区域的环境图像和所述自移动的园艺设备的位姿信息。
通过驱动自移动的园艺设备在剩余区域内移动，且每隔一段时间可以采集自移动的园艺设备所在位置的环境图像，并通过轮速计、IMU或者卫星定位模块等确定环境图像对应的位姿信息。
步骤S108：根据所述剩余区域的环境图像、所述自移动的园艺设备的位姿信息和所述初始地图，确定所述第一区域的地图。
在采集完成剩余区域的环境图像后，基于环境图像及其对应的位姿信息对剩余区域建立地图，并经剩余区域的地图和初始地图进行合并，从而获得第一区域的地图。
通过这种方式可以实现自移动的园艺设备的自主建图，而且在自主建图时，先对充 电桩所在的范围进行建图，以形成初始地图，从而保证自移动的园艺设备在低电量时方便地回桩充电，从而保证了自移动的园艺设备的可靠性。在建立完初始地图后，自移动的园艺设备可以在剩余区域内移动，以采集剩余区域的环境图像并结合初始地图和剩余区域的环境图像生成第一区域的地图，以此实现对自移动的园艺设备的自主建图，从而节省成本，降低劳动强度。
实施例二
参照图2，示出了本申请实施例二的地图处理方法的步骤流程示意图。
步骤S202：控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接。
如前一实施例中所述，自移动的园艺设备的工作区域可以包括一个或一个以上的区域。与自移动的园艺设备对应的充电桩所在的区域可以记作第一区域。而第一区域中与充电桩位置相接一部分区域记作局部区域。例如，局域区域可以是充电桩周围五米范围内的区域。
自移动的园艺设备在局部区域内移动的过程中可以通过行进等采集第一环境图像，通过搭载的传感器可以采集对应的第一位姿信息。
步骤S204：根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩。
根据第一环境图像和对应的第一位姿信息可以通过运动重建(SFM，Structure From Motion)建立初始地图。初始地图包括了范围内物体的特征点，如草的特征点、树木的特征点、石头的特征点、和凹坑的特征点等等。
建立的初始地图可以使自移动的园艺设备依据其进行自动回桩的路径规划和导航、定位。例如，若自移动的园艺设备在移动过程中检测到电池的电量较低，则可以根据初始地图引导自移动的园艺设备回到充电桩充电。
步骤S206：驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域。
剩余区域可以是第一区域中除局部区域之外的区域。通过驱动自移动的园艺设备在剩余区域内移动，使得可以在其移动过程中采集第二环境图像和对应的第二位姿信息。
步骤S208：根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
在采集完成剩余区域的第二环境图像后，基于第二环境图像及其对应的第二位姿信息对剩余区域建立地图，并将剩余区域的地图和初始地图进行合并，从而获得第一区域的地图。
通过这种方式可以方便地建立第一区域的地图，且在建图过程中出现电量不足等问题时也可以及时回到充电桩充电。该方法能够实现自移动的园艺设备的自主建图，从而节省成本，降低劳动强度。
实施例三
参照图3A，示出了本申请的实施例三的地图处理方法的步骤流程示意图。
在本实施例中，该方法包括以下步骤：
步骤S302：控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接。
在本实施例中，以自移动的园艺设备的工作区域包括两个相互独立且不连通的区域为例进行说明，独立的区域的数量多于2个的处理方式与2个区域的数量的处理方式类似，故不再赘述。
在存在2个独立的区域的情况下，可以选取设置有充电桩的区域作为第一区域。初始状态下，自移动的园艺设备可以设置在充电桩上。自移动的园艺设备从充电桩退出。退出过程可以为：若自移动的园艺设备的车头朝向充电桩，则从充电桩后退，并原地旋转180度。
从充电桩退出后，自移动的园艺设备可以在以充电桩为圆心，半径10-15米的圆形区域(即局部区域，该局部区域与充电桩的距离满足设定值，也即图3B中所示初始探索区域)内随机行驶，如图3B所示。
在行驶过程中，自移动的园艺设备通过其上搭载的图像采集装置采集第一环境图像，并通过轮速计或者IMU等检测自移动的园艺设备的第一位姿信息。该第一环境图像一方面可以在后续步骤中生成初始地图，另一方面，可以通过神经网络模型对第一环境图像进行图像识别，以识别是否存在障碍物，并判断障碍物相对自移动的园艺设备的方位和距离，以便自移动的园艺设备进行避障。
步骤S304：根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩。
根据第一位姿信息可以确定自移动的园艺设备是否遍历了局部区域，若已遍历，则可以基于采集的该局部区域内的第一环境图像对应的第一位姿信息对第一环境图像进行处理，如通过SFM(视觉三维重建)或者视觉SLAM方式获得该局部区域对应的初始地图。
后续若需要回桩充电，则可以采集新的环境图像，将采集的环境图像中的特征点和初始地图中的特征点进行匹配，进而根据匹配到的特征点确定自移动的园艺设备的位姿，从而对其进行导航，使其能够自动回桩。
步骤S306：驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域。
在本实施例中，针对一个独立区域的建图可以称为单区域建图，其指在草地外围区域均有边界、草地内部区域相互联通的场景下进行自主移动并绘制草地可通行地图以及视觉特征地图。
在本示例中，对第一区域建图就可以称为单区域建图。由于已经针对充电桩一定范围内的局部区域建立了初始地图，因此在建图过程中，对于第一区域内局部区域范围外的剩余区域，可以驱动自移动的园艺设备在剩余区域内移动，每隔一段时间采集一次所 在位置的第二环境图像并定位出该第二环境图像对应的第二位姿信息，如图3C所示。针对采集的第二环境图像可以使用能够进行障碍物识别的神经网络模型对其进行识别，从而确定出是否有障碍物，若有障碍物，自移动的园艺设备可以避障绕行。
在一种可行方式中，可以采用适当的策略驱动自移动的园艺设备在剩余区域内移动。
策略例如为：
1、随机策略：随机转动角度；或者沿直线行驶，若遇到障碍物，则向远离障碍物方向随机转动一定角度，规避障碍物，并继续沿直线行驶。
2、循边策略：根据识别出的障碍物边沿，沿着边沿行驶。
3、随机策略混合循边策略执行。如根据一定时间比例，混合上述两种策略。
在采集剩余区域内的第二环境图像的过程中若检测到自移动的园艺设备电量不足，自移动的园艺设备将暂停自主移动，并根据初始地图回桩充电。在电量充足后，继续自主探索移动。
在移动过程中，结合用户输入的草地面积以及检测的自移动的园艺设备的第二位姿信息，确定是否停止单区域探索。例如，若确定已经遍历了剩余区域，则可以停止单区域探索。
步骤S308：根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
在本示例中，确定终止探索后可以根据初始地图自动回桩。上桩后，自移动的园艺设备可以进行充电并通过下述过程获取第一区域的地图。
过程A1：根据所述初始地图、所述第二环境图像和所述第二的位姿信息，确定所述第一区域的原始地图。
一种情况中，过程A1可以实现为：根据第二环境图像及其对应的第二位姿信息，使用三维重建(SFM)方式建立剩余区域的视觉地图，并将初始地图和剩余区域的视觉地图进行融合，从而获得第一区域的原始视觉地图。此外，通过障碍物识别的神经网络模型对第二环境图像和初始地图对应的第一环境图像进行障碍物识别，基于识别出的障碍物及其位置，获得第一区域对应的可通行地图。例如，在第二位姿信息中包含基于轮速计和IMU确定的位姿，而未包括基于卫星定位获得的卫星定位的位置的情况下，采用上述方式实现。
第一区域的原始地图包括原始视觉地图和可通行地图，可根据原始视觉地图和可通行地图生成第一区域的原始地图。原始视觉地图可以在后续自动工作过程中进行定位和导航，由于其包括了初始地图，因此也可以进行自动回桩。可通行地图由于包含了障碍物信息，因此可以进行路径规划和避障。
或者，在另一情况中，若第二位姿信息包括所述自移动的园艺设备的图像采集装置的位姿和所述自移动的园艺设备的卫星定位的位置，则过程A1可以实现为：根据所述自移动的园艺设备的卫星定位的位置，确定所述图像采集装置的位姿在地理坐标系中的映射位姿；根据所述初始地图、所述第二环境图像和所述映射位姿，确定所述第一区域的 原始地图。
例如，卫星定位的位置可以基于RTK技术获得。在采集的环境图像中具有对应的卫星定位的位置(例如为经纬度)。通过将经纬度转换至UTM坐标系，得到正交的(例如，以正东正北为坐标轴)GNSS轨迹位置。基于该GNSS轨迹位置和图像采集装置的位姿可以确定GNSS轨迹位置与图像采集装置的位姿的转换关系，进而就可以确定将图像采集装置的位姿转换到UTM坐标系内的映射位姿。这样使得环境图像的位姿信息中携带了经纬度及该经纬度的协方差信息(协方差信息指示经纬度的置信度，若卫星信号强度较好，则协防差信息指示的置信度高，表示该经纬度更加可靠)，进而基于该位姿信息对使用SFM方式获得的第一区域的原始地图的可靠性更好。这样提高了原始地图中的原始视觉地图和可通行地图的鲁棒性。
原始地图中包含多个特征点，这些特征点可以是原始地图中的特征点，或者是第二环境图像中的参照对象(如草、凹坑、树木等)的特征点，也可以一部分特征点是初始地图的特征点，另一部分特征点是第二环境图像中至少一个参照对象的特征点。
过程B1：确定所述第一区域的原始地图是否存在未探索部分，若存在，则对所述第一区域进行补充建图，以获得所述第一区域的地图。
由于在探索过程中可能由于障碍物等原因导致第一区域中有未探索部分，因此，在获得第一区域的原始地图后可以确定其是否存在未探索部分，若存在则执行过程B11，反之，若不存在则执行过程B12。
其中，为了能够准确地确定是否存在未探索部分，以此保证可靠性，所述确定所述第一区域的原始地图是否存在未探索部分可以实现为：确定所述第一区域的原始地图中各特征点的分布信息；根据所述分布信息，确定所述第一区域的原始地图是否存在未探索部分。
例如，将原始地图中原始视觉地图进行分块，统计每块中特征点的数量作为分布信息。基于该分布信息可以确定特征点的数量小于设定数量(其可以根据需要确定，如10个、20个等等)的异常分块。若多个异常分块连通，且连通面积大于或等于设定面积(其可以根据需要确定，如大于或等于1个自移动的园艺设备的面积)，则将其确定为未探索区域(包含未探索区域的第一区域的原始地图如图3D所示)，同时也可以确定出未探索区域的位置。反之，则忽略异常分块。
过程B11：对所述第一区域进行补充建图。
其中，过程B11可以实现为：驱动所述自移动的园艺设备在所述未探索区域内移动，并采集所述未探索区域的第三环境图像和所述自移动的园艺设备的第三位姿信息；使用所述第三环境图像和所述第三位姿信息更新所述第一区域的原始地图，以获得所述第一区域的地图。
基于未探索区域的位置可以驱动自移动的园艺设备移动到未探索区域内，并对其进行自动探索，探索策略可以与前述策略一致，故不再赘述。在未探索区域内移动过程中的示意图如图3E所示，采集未探索区域的第三环境图像，并确定第三环境图像对应的第 三位姿信息，该第三位姿信息可以基于已有的原始视觉地图确定，这样使得新采集的环境图像的第三位姿信息的坐标系与原始视觉地图的坐标系一致。完成第三环境图像的采集后自移动的园艺设备可以自动回桩充电，并可以采用SFM方式对原始视觉地图和可通行地图进行更新，从而获得第一区域的地图。
通过这种方式解决了探索策略的随机性以及图像采集装置采集范围的限制导致容易出现未探索区域的问题，保证了地图的可靠性。
在未探索区域包括一个以上时，可以针对每个未探索区域都重复上述过程，直至完成所有未探索区域的补充建图。
过程B12：完成单区域建图。
可选地，为了提升该方法的适应性，对于存在一个以上单区域的工作区域，该方法还包括下述步骤：
步骤S310：确定所述工作区域是否存在已有地图的区域之外的第二区域，若存在第二区域，则建立所述第二区域对应的地图。
第二区域可以是独立于已有地图的区域之外的区域。例如，工作区域包括区域A和区域B，其中，第一区域为区域A，其已经完成建图。则区域B可以作为第二区域。或者，又例如，工作区域包括区域A、区域B和区域C，其中，第一区域为区域A，其已经完成建图，区域B作为第二区域也已经完成建图，则区域C可以作为新的第二区域进行建图。
一种可行的确定第二区域的方式可以为：通过显示装置向用户询问是否存在第二区域，若用户指示存在，则建立第二区域对应的地图。或者，若用户指示不存在，则完成建图。
当然，也可以采用其他可行方式确定是否存在第二区域，对此不作限制。
若存在第二区域，则建立所述第二区域对应的地图可以通过下述过程实现：
过程A2：从所述第二区域中确定目标区域。
若第二区域仅为1个，则直接将其确定为第二区域。或者，若第二区域为一个以上，则可以从中随机确定一个作为目标区域，或者，根据已有的地图选取与自移动的园艺设备所在位置最近的第二区域作为目标区域，或者，也可以由用户指定一个第二区域作为目标区域。
可选地，为了能够方便地引导自移动的园艺设备运动到目标区域，步骤S310可以包括过程B2。
过程B2：确定所述自移动的园艺设备所在区域与所述目标区域之间的通行通道。
对于已经采集到至少部分目标区域的环境图像的情况，若所述自移动的园艺设备所在区域的地图中包含至少部分的所述目标区域，则根据所述自移动的园艺设备所在区域的地图，确定所述通行通道，如图3F所示。例如，第一区域的地图可转化为图像或者二维网格形式，其中划分了草地区域和非草地区域。通过图像形态学方式提取所在位置的草地区域和所有可见草地区域，其中可见但非所在位置的草地区域可以是目标区域的草地。使用图像形态学方式划分出在两块草地之间的隔离区域，再从隔离区域中选择一条 通行通道并在地图上标记出来。
选择方法通行通道可以是：穿越隔离区域最短路径；或者，在自移动的园艺设备包括深度传感器可以采集立体信息时，根据隔离区域的地形，计算一条平整通行通道；或者，向用户展示采集的地图和地形，由用户画出通行通道或从推荐通行通道中选择一条。
对于未采集到目标区域的环境图像的情况，可以通过用户指定的方式设定通行通道，用户可通过遥控方式将自移动的园艺设备遥控至需建图的目标区域，自移动的园艺设备将自动识别草地间不可通行区域，并记录通过的路径作为通行通道。
在通过通行通道移动到目标区域内，并可以执行过程C2。当然，在其他实施例中，可以省略过程B2，自移动的园艺设备可以通过其他适应的方式移动到目标区域。
过程C2：采集所述目标区域的第四环境图像，并根据所述第四环境图像确定所述目标区域的地图。
在到达目标区域内之后，自移动的园艺设备可以按照策略在目标区域内探索，且每隔一段时间就采集一张所在位置的第四环境图像，并确定该第四环境图像对应的第四位姿信息。然后，在采集完第四环境图像后，基于第四环境图像和第四位姿信息生成目标区域的地图。该地图包括视觉地图和可通行地图。
基于视觉地图和可通行地图也可以确定是否有未探索区域，若有未探索区域也可以进行重复探索，从而对视觉地图和可通行地图进行更新，直至完成对目标区域的建图。
前述的可通行地图可以在视觉地图获取的图像采集装置的位姿的基础上，结合识别得到的障碍物信息生成。其中，可通行地图中将可通行区域、障碍物、未知区域表示为三种不同的颜色。使用图像形态学的方法提取可通行区域的最大边界作为草地外边界，如果边界上或者边界内存在未知区域，并且未知区域与可通行区域连通，则作为需要继续探索的未知区域。
过程D2：确定是否满足终止条件，若未满足所述终止条件，则返回从所述第二区域中确定目标区域继续执行。
终止条件可以是已建图的区域的面积与用户指定的工作区域的面积的差值小于或等于面积阈值(其可以根据需要确定，例如为0，或者其他大于0的值)。若满足终止条件，则完成建图。若未满足终止条件，则从剩余的第二区域内确定新的目标区域，并对其进行建图，直至满足终止条件，完成建图。
基于该方法，可以免除人工部署埋线，实现自移动的园艺设备的自动建图，其能够自动探索未知区域，并且确定需要进一步探索的区域，在用户确认后自动完成未知区域的探索，提高多块草地建图的效率。
实施例四
参照图4，示出了本申请实施例四的地图处理装置的结构框图。
该装置包括：
第一驱动模块402，用于控制自移动的园艺设备采集局部区域内的第一环境图像和 所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接；
第一确定模块404，用于根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩；
第二驱动模块406，用于驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域；
第二确定模块408，用于根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
可选地，所述第二确定模块408用于根据所述初始地图、所述第二环境图像和所述第二位姿信息，确定所述第一区域的原始地图；确定所述第一区域的原始地图是否存在未探索部分，若存在，则对所述第一区域进行补充建图，以获得所述第一区域的地图。
可选地，所述原始地图中包含多个特征点，所述特征点包括所述初始地图中的特征点和所述第二环境图像中至少一个参照对象的特征点中至少之一，所述第二确定模块408用于在确定所述第一区域的原始地图是否存在未探索部分时，确定所述第一区域的原始地图中各特征点的分布信息；根据所述分布信息，确定所述第一区域的原始地图是否存在未探索部分。
可选地，所述第二确定模块408用于在对所述第一区域进行补充建图，以获得所述第一区域的地图时，驱动所述自移动的园艺设备在所述未探索区域内移动，并采集所述未探索区域的第三环境图像和所述自移动的园艺设备的第三位姿信息；使用所述第三环境图像和所述第三位姿信息更新所述第一区域的原始地图，以获得所述第一区域的地图。
可选地，所述自移动的园艺设备的位姿信息包括所述自移动的园艺设备的图像采集装置的位姿和所述自移动的园艺设备的卫星定位的位置；第二确定模块408用于在根据所述初始地图、所述第二环境图像和所述第二位姿信息，确定所述第一区域的原始地图时，根据所述自移动的园艺设备的卫星定位的位置，确定所述图像采集装置的位姿在地理坐标系中的映射位姿；根据所述初始地图、所述第二环境图像和所述映射位姿，确定所述第一区域的原始地图。
可选地，所述装置还包括：
第三确定模块410，用于确定所述工作区域是否存在已有地图的区域之外的第二区域，若存在第二区域，则建立所述第二区域对应的地图。
可选地，第三确定模块410用于在所述建立所述第二区域对应的地图时，从所述第二区域中确定目标区域；采集所述目标区域的第四环境图像，并根据所述第四环境图像确定所述目标区域的地图；确定是否满足终止条件，若未满足所述终止条件，则返回从所述第二区域中确定目标区域继续执行。
可选地，第三确定模块410还用于在所述建立所述第二区域对应的地图时，确定所述自移动的园艺设备所在区域与所述目标区域之间的通行通道。
可选地，第三确定模块410用于在确定所述自移动的园艺设备所在区域与所述目标 区域之间的通行通道时，若所述自移动的园艺设备所在区域的地图中包含至少部分的所述目标区域，则根据所述自移动的园艺设备所在区域的地图，确定所述通行通道。
该装置能够实现上述方法对应的效果，故不再赘述。
实施例五
本实施例中，提供一种自移动的园艺设备，其包括控制器，所述控制器用于执行上述的方法，并实现相应的效果，对此不再赘述。
实施例六
本实施例中，提供一种自动割草机，其包括控制器，所述控制器用于执行上述的方法，并实现相应的效果，对此不再赘述。
需要说明的是，在本发明的描述中，术语“第一”、“第二”仅用于方便描述不同的部件或名称，而不能理解为指示或暗示顺序关系、相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。
需要说明的是，虽然结合附图对本发明的具体实施例进行了详细地描述，但不应理解为对本发明的保护范围的限定。在权利要求书所描述的范围内，本领域技术人员不经创造性劳动即可做出的各种修改和变形仍属于本发明的保护范围。
本发明实施例的示例旨在简明地说明本发明实施例的技术特点，使得本领域技术人员能够直观了解本发明实施例的技术特点，并不作为本发明实施例的不当限定。
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (13)

  1. 一种地图处理方法，包括：
    控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接；
    根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩；
    驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域；
    根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
  2. 根据权利要求1所述的方法，所述根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，包括：
    根据所述初始地图、所述第二环境图像和所述第二位姿信息，确定所述第一区域的原始地图；
    确定所述第一区域的原始地图是否存在未探索部分，若存在，则对所述第一区域进行补充建图，以获得所述第一区域的地图。
  3. 根据权利要求2所述的方法，所述原始地图中包含多个特征点，所述特征点包括所述初始地图中的特征点和所述第二环境图像中至少一个参照对象的特征点中至少之一，所述确定所述第一区域的原始地图是否存在未探索部分，包括：
    确定所述第一区域的原始地图中各特征点的分布信息；
    根据所述分布信息，确定所述第一区域的原始地图是否存在未探索部分。
  4. 根据权利要求2所述的方法，所述对所述第一区域进行补充建图，以获得所述第一区域的地图，包括：
    驱动所述自移动的园艺设备在所述未探索区域内移动，并采集所述未探索区域的第三环境图像和所述自移动的园艺设备的第三位姿信息；
    使用所述第三环境图像和所述第三位姿信息更新所述第一区域的原始地图，以获得所述第一区域的地图。
  5. 根据权利要求2所述的方法，所述第二位姿信息包括所述自移动的园艺设备的图像采集装置的位姿和所述自移动的园艺设备的卫星定位的位置；
    根据所述初始地图、所述第二环境图像和所述第二位姿信息，确定所述第一区域的原始地图，包括：
    根据所述自移动的园艺设备的卫星定位的位置，确定所述图像采集装置的位姿在地 理坐标系中的映射位姿；
    根据所述初始地图、所述第二环境图像和所述映射位姿，确定所述第一区域的原始地图。
  6. 根据权利要求2所述的方法，所述根据所述初始地图、所述第二环境图像和所述第二位姿信息，确定所述第一区域的原始地图，包括：
    根据所述第二环境图像及所述第二位姿信息，确定所述第一区域的原始视觉地图；
    对所述第二环境图像和所述初始地图进行障碍物识别，确定所述第一区域的可通行地图；
    根据所述原始视觉地图和所述可通行地图，生成所述第一区域的原始地图。
  7. 根据权利要求1-6中任一项所述的方法，所述方法还包括：
    确定所述工作区域是否存在已有地图的区域之外的第二区域，若存在第二区域，则建立所述第二区域对应的地图。
  8. 根据权利要求7所述的方法，所述建立所述第二区域对应的地图，包括：
    从所述第二区域中确定目标区域；
    采集所述目标区域的第四环境图像，并根据所述第四环境图像确定所述目标区域的地图；
    确定是否满足终止条件，若未满足所述终止条件，则返回从所述第二区域中确定目标区域的步骤继续执行。
  9. 根据权利要求8所述的方法，所述建立所述第二区域对应的地图，还包括：
    确定所述自移动的园艺设备所在区域与所述目标区域之间的通行通道。
  10. 根据权利要求9所述的方法，所述确定所述自移动的园艺设备所在区域与所述目标区域之间的通行通道，包括：
    若所述自移动的园艺设备所在区域的地图中包含至少部分的所述目标区域，则根据所述自移动的园艺设备所在区域的地图，确定所述通行通道。
  11. 一种地图处理装置，包括：
    第一驱动模块，用于控制自移动的园艺设备采集局部区域内的第一环境图像和所述自移动的园艺设备的第一位姿信息，所述局部区域与充电桩的位置相接；
    第一确定模块，用于根据所述局部区域的第一环境图像和所述自移动的园艺设备的第一位姿信息，生成初始地图，以使所述自移动的园艺设备能够基于所述初始地图自动回桩；
    第二驱动模块，用于驱动所述自移动的园艺设备采集剩余区域的第二环境图像和所述自移动的园艺设备的第二位姿信息；所述剩余区域至少部分包围所述局部区域；
    第二确定模块，用于根据所述第二环境图像、所述第二位姿信息和所述初始地图，确定所述第一区域的地图，所述第一区域包括所述局部区域和所述剩余区域。
  12. 一种自移动的园艺设备，包括控制器，所述控制器用于执行权利要求1-10中任一项所述的方法。
  13. 一种自动割草机，包括控制器，所述控制器用于执行权利要求1-10中任一项所述的方法。
PCT/CN2023/071525 2022-01-11 2023-01-10 地图处理方法、自移动的园艺设备、以及自动割草机 WO2023134659A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210029187.2A CN116466693A (zh) 2022-01-11 2022-01-11 地图处理方法、自移动的园艺设备、以及自动割草机
CN202210029187.2 2022-01-11

Publications (1)

Publication Number Publication Date
WO2023134659A1 true WO2023134659A1 (zh) 2023-07-20

Family

ID=87177560

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/071525 WO2023134659A1 (zh) 2022-01-11 2023-01-10 地图处理方法、自移动的园艺设备、以及自动割草机

Country Status (2)

Country Link
CN (1) CN116466693A (zh)
WO (1) WO2023134659A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017133707A1 (zh) * 2016-02-06 2017-08-10 苏州宝时得电动工具有限公司 自动工作系统，自移动设备及其控制方法
WO2018014838A1 (zh) * 2016-07-19 2018-01-25 苏州宝时得电动工具有限公司 自移动园艺机器人及其系统
CN109682368A (zh) * 2018-11-30 2019-04-26 上海肇观电子科技有限公司 机器人及地图构建方法、定位方法、电子设备、存储介质
CN112578779A (zh) * 2019-09-29 2021-03-30 苏州宝时得电动工具有限公司 地图建立方法、自移动设备、自动工作系统
CN113625701A (zh) * 2020-05-09 2021-11-09 苏州宝时得电动工具有限公司 一种割草机器人路径规划方法及割草机器人

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017133707A1 (zh) * 2016-02-06 2017-08-10 苏州宝时得电动工具有限公司 自动工作系统，自移动设备及其控制方法
WO2018014838A1 (zh) * 2016-07-19 2018-01-25 苏州宝时得电动工具有限公司 自移动园艺机器人及其系统
CN108541308A (zh) * 2016-07-19 2018-09-14 苏州宝时得电动工具有限公司 自移动园艺机器人及其系统
CN109682368A (zh) * 2018-11-30 2019-04-26 上海肇观电子科技有限公司 机器人及地图构建方法、定位方法、电子设备、存储介质
CN112578779A (zh) * 2019-09-29 2021-03-30 苏州宝时得电动工具有限公司 地图建立方法、自移动设备、自动工作系统
CN113625701A (zh) * 2020-05-09 2021-11-09 苏州宝时得电动工具有限公司 一种割草机器人路径规划方法及割草机器人

Also Published As

Publication number Publication date
CN116466693A (zh) 2023-07-21

Similar Documents

Publication Publication Date Title
CN112584697B (zh) 使用视觉系统的自主机器导航和训练
EP3119178B1 (en) Method and system for navigating an agricultural vehicle on a land area
CN106647765B (zh) 一种基于割草机器人的规划平台
WO2021093474A1 (zh) 一种割草机导航方法、装置和割草机
US10598793B2 (en) Navigation for a robotic work tool
CN111256702B (zh) 一种电力杆塔巡检用无人机自主巡检方法
EP3366100A1 (en) Method and system for navigating a robotic garden tool
CN113128747B (zh) 智能割草系统及其自主建图方法
CN111199677B (zh) 一种室外区域的工作地图自动建立方法，装置，存储介质及工作设备
US20230236604A1 (en) Autonomous machine navigation using reflections from subsurface objects
CN111103886A (zh) 通行窄道的识别方法、装置、设备及计算机可读存储介质
WO2023134659A1 (zh) 地图处理方法、自移动的园艺设备、以及自动割草机
CN114995444A (zh) 建立虚拟工作边界的方法、装置、远程终端及存储介质
CN116430838A (zh) 自移动设备、及其控制方法
CN114937258A (zh) 割草机器人的控制方法、割草机器人以及计算机存储介质
JP7462642B2 (ja) 自律作業機、制御装置、自律作業機の制御方法、制御装置の動作方法及びプログラム
CN114296463A (zh) 一种作业区域间路径生成方法及园林系统
CN113110411A (zh) 视觉机器人回基站控制方法、装置及割草机器人
US20230320263A1 (en) Method for determining information, remote terminal, and mower
CN116088533B (zh) 信息确定方法、远程终端、设备、割草机及存储介质
EP4235336A1 (en) Method and system for robot automatic charging, robot, and storage medium
WO2024022204A1 (zh) 园林工具及其控制方法和系统
US20230206647A1 (en) Automatic Robotic Lawn Mowing Boundary Detection Using 3D Semantic Segmentation
WO2021014586A1 (ja) 自律作業機、自律作業機の制御方法及びプログラム
CN116736865A (zh) 信息确定方法、远程终端、设备、割草机及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23740001

Country of ref document: EP

Kind code of ref document: A1