WO2025201559A1 - 一种路径规划方法、自移动设备及存储介质 - Google Patents

一种路径规划方法、自移动设备及存储介质

Info

Publication number
WO2025201559A1
WO2025201559A1 PCT/CN2025/086261 CN2025086261W WO2025201559A1 WO 2025201559 A1 WO2025201559 A1 WO 2025201559A1 CN 2025086261 W CN2025086261 W CN 2025086261W WO 2025201559 A1 WO2025201559 A1 WO 2025201559A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
path
missing
areas
compensation
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
Application number
PCT/CN2025/086261
Other languages
English (en)
French (fr)
Inventor
郑思远
翁蒙
秦伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Positec Power Tools Suzhou Co Ltd
Original Assignee
Positec Power Tools Suzhou Co Ltd
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 Positec Power Tools Suzhou Co Ltd filed Critical Positec Power Tools Suzhou Co Ltd
Publication of WO2025201559A1 publication Critical patent/WO2025201559A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01D—HARVESTING; MOWING
    • A01D34/00—Mowers; Mowing apparatus of harvesters
    • 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

Definitions

  • the present disclosure relates to the technical field of path planning, and in particular to a path planning method, a self-mobile device, and a storage medium.
  • Autonomous devices can be devices with autonomous mobility capabilities that can move autonomously and perform lawn care tasks without human intervention.
  • autonomous devices include drones, robot vacuums, and robot lawn mowers.
  • an embodiment of the present disclosure provides a path planning method, which is applied to a self-moving device having a working component, wherein the self-moving device is configured to move along a preset working path and perform a lawn care task via the working component during movement.
  • the method includes:
  • the missed area is formed by the self-moving device deviating from the preset working path and/or the working component not covering it;
  • a compensation path covering the combined area is planned, wherein the compensation path is configured to guide the self-moving device to continuously move in the combined area, and the working component performs the lawn care task during the movement.
  • merging at least two of the missing regions into a combined region according to the spatial distribution characteristics includes:
  • spatial distribution characteristics include spatial distance, spatial density, and spatial direction
  • At least two of the missing regions are merged into one combined region, wherein the combined region covers each of the missing regions.
  • the spatial distribution conditions include:
  • merging at least two of the missing regions into one combined region comprises:
  • Boundary fitting is performed on the boundaries of each missing region to obtain the combined region.
  • performing boundary fitting on the boundaries of each missing region to obtain the combined region includes:
  • the area covered by the minimum circumscribed rectangle is used as the combined area.
  • selecting the minimum bounding rectangle corresponding to each missing area according to the geometric attributes includes:
  • the smallest circumscribed rectangle among all circumscribed rectangles is used as the minimum circumscribed rectangle.
  • the circumscribed rectangle with the smallest side length among all circumscribed rectangles is used as the minimum circumscribed rectangle;
  • the circumscribed rectangle with the largest aspect ratio among the circumscribed rectangles is used as the minimum circumscribed rectangle.
  • the compensation path includes a plurality of straight paths and a plurality of turning paths, and adjacent straight paths are connected by turning paths.
  • planning a compensation path covering the combined area includes:
  • the straight path is determined according to the long side.
  • determining the straight path according to the long side includes:
  • the direction of the straight path is determined according to the direction of the long side.
  • planning a compensation path covering the combined area includes:
  • the straight portion of the preset operation path within the minimum circumscribed rectangle is used as the straight path.
  • planning a compensation path covering the combined area includes:
  • the direction of the straight path of the compensation path is determined according to the straight direction of the straight portion of the preset working path.
  • the preset operation path guides the self-moving device to move within the lawn care area
  • the further step includes: determining a position of the circumscribed rectangle in the lawn care area;
  • a type of the turning path is determined based on the position.
  • the combined area is selected according to the compensation time.
  • selecting the combined area according to the compensation time includes:
  • the compensation times are sorted, and the combination area is selected according to the sorting result.
  • the blocking conditions include one or more of the following conditions:
  • the cutting parameters between the omitted regions are different.
  • the present disclosure provides a self-propelled device, comprising:
  • a work component configured to perform lawn care tasks
  • a moving component configured to drive the working component to move so that the working component performs the lawn care task while moving
  • a processor is configured to connect with the moving component and the working component to implement any one of the methods described.
  • the working assembly includes a cutting assembly
  • Embodiments of the present disclosure provide a path planning method, a self-mobile device, and a storage medium. Based on the path planning method provided herein, the self-mobile device can promptly identify missed areas and execute lawn care tasks, thereby improving job completion. Furthermore, a compensation path can be constructed based on the clustering results of missed areas, allowing each individual missed area to be processed simultaneously within a single lawn care task. This reduces operation time compared to executing lawn care tasks for each missed area individually, thereby improving overall operation efficiency.
  • FIG5 is a schematic diagram of determining a set of missing regions provided by an exemplary embodiment of the present disclosure.
  • the control circuit is coupled to the mobile component, the sensor component, and the working component, and controls the movement of the mobile component and the working component based on sensor data output by the sensor component.
  • the sensor component may include, but is not limited to, at least one of a vision component, a positioning device, a collision sensor component, and an ultrasonic sensor component.
  • the mobile component generally includes a drive motor and a drive wheel connected to the drive motor. Optionally, it may also include a driven wheel that, together with the drive wheel, drives the self-propelled device.
  • the working component may include, but is not limited to, at least one of a bladed disc for mowing grass, a blower head for blowing away fallen leaves, and a snowplow head for sweeping snow.
  • the work components may be determined based on the type of lawn care required to be performed by the mobile device 110.
  • the work components may include a leaf blower head or a snow blower head.
  • the work components may include a cutter head.
  • the work components may include a spraying component.
  • the mobile component can be selected based on the actual operating environment of the autonomous device 110.
  • the mobile component can be a device that enables movement on the road, such as tires or tracks.
  • the mobile component can be a device that enables movement in the air, such as an air propeller or a jet engine.
  • Lawn care tasks for autonomous devices are generally mobile tasks, meaning they must operate while on the move.
  • a drone's lawn care task might involve flying within a target area and spraying pesticides or water.
  • a robotic lawn mower which can mow the lawn while moving, release insecticides or fragrances while moving, or even blow away fallen leaves or collect scattered grass.
  • lawn mowing robots As an example, the above-mentioned lawn mowing robots that can work autonomously are currently mainly used in the home field, and lawn mowing robots using intelligent control are rarely seen in the commercial lawn mowing field.
  • a full-coverage path planning algorithm can be used to pre-plan the robot's path within the lawn care area.
  • commercial robotic lawn mowers can easily miss areas during operation. For example, a commercial lawn mower robot moving across a lawn may miss areas if the lawn care area includes slopes, ditches, or irregular edges.
  • Another example is that when executing a pre-set path, the robot's motion may be unstable due to environmental factors (such as wind, terrain, and obstacles), resulting in missed areas.
  • the lawn care area can be divided into missed areas and non-missed areas. Missed areas are formed when the mowing robot deviates from the preset working path and/or is not covered by the working components. Conversely, non-missed areas are formed when the mowing robot follows the preset working path and is covered by the working components. In layman's terms, if the mowing robot moves along the preset working path, there will be no areas of the lawn care area that are not cut by the working components. However, if the mowing robot occasionally deviates and its actual path deviates from the preset working path, a missed area will be created between the actual path and the preset working path, i.e., the lawn in the missed area is not cut by the working components.
  • Missing areas directly impact the completion of mobile lawn care tasks. Missing areas indicate that portions of the lawn remain uncut, and the height of the lawn in these missed areas appears higher than in the remaining areas, affecting the overall lawn appearance. Specifically, because household lawn mowers are small and cut a small amount per session, they typically operate daily. The height difference between missed and remaining areas is minimal, making it unnoticeable to the user and not affecting the user's perception.
  • a commercial lawn mower robot can increase the amount of grass cut per time to reduce the number of times a commercial team is hired, specifically once every 1-2 weeks.
  • grass grows faster. Therefore, when there are missed areas, the height difference between the missed areas and the non-missed areas of the commercial lawn mower robot is large, sometimes even reaching more than five centimeters. The user can feel it strongly with the naked eye. If the missed areas are not mowed, it will affect the user's favorability towards the commercial team. Therefore, it is more important for commercial lawn mower robots to mow the missed areas.
  • the self-moving device is a robotic lawn mower
  • completion compensation where missed areas are compensated one by one after the lawn care task is completed.
  • the self-moving device is a robotic lawn mower
  • mowing missed areas can be considered a form of compensation.
  • irrigation machine watering missed areas can be considered a form of compensation.
  • blowing away fallen leaves can be considered a form of compensation.
  • the outdoor environment in which the lawn is located is very complex and can easily affect the commercial mower robot, causing its movement center to not strictly follow the preset operating path to perform the lawn care task.
  • missed areas may occur.
  • the movement center can be the geometric center of the commercial mower robot or the center of its positioning module.
  • the commercial lawn mower robot will solve the missed areas, but the current method is usually to solve the missed areas as soon as they appear. It is easy for the commercial lawn mower robot to turn around and retreat repeatedly, so the efficiency is low.
  • the present disclosure provides a path planning method, a self-mobile device and a storage medium.
  • the path planning method provided by the present disclosure can determine the missing areas in the lawn care area, and cluster the missing areas so that at least two missing areas are processed into a combined area, determine the compensation path of the combined area, and guide the self-mobile device to simultaneously perform lawn care tasks (compensation) for at least two missing areas in the combined area through the compensation path.
  • the processing means include aggregation, merging or selection, etc.
  • the self-mobile device can simultaneously identify at least two missing areas and perform lawn care tasks for at least two missing areas in a continuous compensation path to improve the operation coverage of the self-mobile device for the lawn care area.
  • the compensation path can be constructed based on the clustering results of the missing areas, so that at least two missing areas can be processed simultaneously in one lawn care task. Compared with performing lawn care tasks on each missing area separately and sequentially, the operation time is reduced to improve the overall operation efficiency.
  • the self-mobile device 110 may be a robotic lawn mower for performing mowing operations, and the lawn care area 200 may be a lawn to be mowed.
  • the self-mobile device 110 may move and mow the lawn in the lawn care area 200 according to a set lawn care task.
  • missed areas unmowed areas
  • the self-mobile device 110 can generate a compensation path based on a clustering algorithm, thereby performing the lawn care task (mowing) in the missed areas.
  • a user can interact with a mobile device 110 through a user terminal 120.
  • the mobile device 110 can obtain lawn care tasks sent by the user terminal 120 through the network 130, and can also send the completion status of the lawn care tasks to the user terminal 120 through the network 130. Lawn care tasks include, but are not limited to, job types and lawn care areas 200.
  • the user terminal 120 is a terminal device that can establish a communication connection with the mobile device 110 to send control instructions, and the control instructions may include mobile lawn care tasks.
  • the user terminal 120 can be one of a mobile device 120-1, a tablet computer 120-2, a laptop computer 120-3, a desktop computer 120-4, or any combination thereof, having input and/or output functions.
  • the preset operation path planning process may involve multi-party interaction.
  • the aforementioned self-moving device 110 may include multiple devices, and cooperate with each other to complete the lawn care task.
  • the user terminal 120 may also be related to the path planning process (such as setting the operation boundary).
  • the aforementioned controller 210 can also be configured as a collection of computing devices involved in the path planning process.
  • the controller 210 may include relevant devices for setting the path planning area at the user terminal 120 and relevant controllers of each self-moving device 110.
  • a map of the lawn care area 200 may be pre-downloaded to the storage medium 220 or the controller 210 of the mobile device 100, and the mobile device 110 performs mobile operations in the lawn care area 200 based on the map of the lawn care area 200 and the set mobile lawn care tasks.
  • the controller 210 can be a computing device that executes the path planning method provided by the present disclosure.
  • the path planning method provided by the present disclosure is executed by the self-mobile device 110, and the controller 210 may include a controller (such as an embedded controller) that is set in the self-mobile device 110 and integrated with a path planning algorithm.
  • the path planning method provided by the present disclosure is executed by a cloud server, and the controller 210 may include a computing device (such as a CPU, SOC, etc.) in the cloud server.
  • the cloud server can configure the processing results of the path planning method provided by the present disclosure as control instructions for each controller in the self-mobile device 110, and send it to the self-mobile device 110 via the network 130, so that the self-mobile device 110 responds to the control instructions of the cloud processor.
  • the network 130 can be any one or more of a wired network or a wireless network.
  • the network 130 may include a cable network, a fiber optic network, a telecommunications network, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone network (PSTN), a Bluetooth network, a ZigBee network (ZigBee), a near field communication (NFC), an intra-device bus, an intra-device line, a cable connection, etc., or any combination thereof.
  • the network connection between each part may be in one of the above-mentioned ways, or in multiple ways.
  • the network may be a point-to-point, shared, centralized, or other topological structure or a combination of multiple topological structures.
  • the network 130 may include one or more network access points.
  • One or more components within the application scenario may be connected to the network 130 to exchange data and/or information.
  • the self-mobile device 110 may include but is not limited to a working component 111 , a moving component 112 , a sensor component 113 , and a controller 210 .
  • the specific composition of the working assembly 111, the mobile assembly 112, and the sensor assembly 113 can be tailored to the functionality of the self-mobile device 110 and adjusted based on actual operational requirements.
  • the working assembly 111 may include a blade for mowing.
  • the mobile assembly 112 typically includes a drive motor and a motion device (such as tires or tracks) connected to the drive motor.
  • the sensor assembly 113 may include at least one of a visual sensor, a positioning device, a collision sensor, and an ultrasonic sensor.
  • the aforementioned working component 111, moving component 112 and sensor component 113 can change the working state in response to the control instruction.
  • the control method of the working component 111, moving component 112 and sensor component 113 can be adjusted according to the actual situation.
  • the working component 111 and the moving component 112 can be driven by a motor.
  • a motor drive signal generally including a motor power supply signal and a Hall signal.
  • the working component 111, the moving component 112 and the sensor component 113 are driven by a corresponding controller (such as a domain controller, a domain controller, a microcontroller, etc.). Then, the corresponding controller can achieve control by generating a corresponding control signal.
  • the mobile device 110 may further include a storage medium 220 integrated into the mobile device.
  • the storage medium 220 may store instructions/data within the mobile device 110.
  • the storage medium 220 may store instructions related to the path planning method of the present disclosure, so that after the relevant instructions are called by the controller 210, the path planning method provided by the embodiment of the present disclosure is implemented.
  • the storage medium 220 may also cache relevant data (such as grid maps, sensor data, etc.) during the execution of the path planning method to ensure the execution of the path planning method.
  • the controller 210 may be integrated into the mobile device 110.
  • the mobile device 110 may be provided with a computing device such as a printed circuit board (PCB), a system on chip (SOC), or an electronic control unit (ECU).
  • the processing unit (such as a CPU) in the computing device may serve as the controller 210
  • the storage unit such as a RAM and a ROM
  • the storage medium 220 may serve as the storage medium 220.
  • the controller 210 can provide control signals to the working component 111, the moving component 112, and the sensor component 113 to control their operating states.
  • the controller 210 can generate motor drive signals based on a preset protocol to control at least one of the working component 111, the moving component 112, and the sensor component 113.
  • the controller 210 may include a path planning device 230, which provides control signals to the working component 111, the moving component 112, and the sensor component 113 to control their operating states.
  • the path planning device 230 may include a non-missing area determination component unit 231, a missing area determination component unit 232, a combined area determination component unit 233, and a path planning component unit.
  • the non-missing area determination component unit 231 may be used to control the mobile device to perform a lawn care task on the target lawn care area along a preset operation path, and determine the non-missing area in the target lawn care area.
  • the missing area determination component 232 may be used to determine missing areas in a lawn care area.
  • the path planning component unit 234 can be used to plan a compensation path for the self-moving device and perform the lawn care task on the combined area based on the compensation path.
  • the path planning device 230 determines a preset operating path for the self-moving device and generates corresponding control signals in real time based on the preset operating path.
  • the signals are sent to the working component 111 and the mobile component 112, so that the self-moving device, driven by the mobile component 112, moves along the preset operating path and performs mobile operations through the working component 111.
  • the controller 210 can determine the non-missing areas and the missing areas based on the execution of the preset operating path by the working component 111 and the mobile component 112.
  • the functions of the aforementioned missing area determination component unit 232 can be implemented by the sensor component 113, the controller 210, and the storage medium 220. That is, the sensor component 113 can sense the position of the self-mobile device 110 in real time and store it in the storage medium 220. The controller 210 can determine the missing area based on the real-time position of the self-mobile device 110 (for example, based on a grid map).
  • the functions of the aforementioned combined region determination component unit 233 may be implemented by the controller 210 and the storage medium 220. That is, the controller 210 may perform processing (such as clustering) based on the missing regions in the storage medium 220 to determine a combined region containing multiple missing regions.
  • the functions of the aforementioned path planning component unit 234 can be implemented by the working component 111, the moving component 112, and the controller 210. Specifically, the controller 210 can determine a compensation path based on the combined area, generate corresponding control signals based on the compensation path, and transmit them to the working component 111 and the moving component 112 to cause the self-moving device to move along the compensation path and perform a mobile operation, thereby completing the lawn care task for the missed area.
  • the division of functional units in the embodiments of the present disclosure can be adjusted based on actual needs. That is, the division of functional units in the embodiments of the present disclosure is illustrative and merely represents a logical functional division. In actual implementation, other division methods may be used. For example, two or more functional units may be integrated into a single unit. Furthermore, the aforementioned integrated functional units may be implemented in either hardware or software functional modules.
  • the controller 210 can independently control or control the mobile device 110 through the path planning device 230 to perform the following steps:
  • S310 determining a missing area, where the missing area is formed by the self-moving device deviating from the preset working path and/or the working component not covering.
  • S310 may be executed by the controller 210 or the missing area determining unit 232 .
  • S320 Merge at least two of the missing regions into a combined region according to the spatial distribution characteristics.
  • S320 may be executed by the controller 210 or the combined region determination component unit 233.
  • S330 Plan a compensation path covering the combined area, wherein the compensation path is configured to guide the self-moving device to continuously move within the combined area, and the working component performs the lawn care task during the movement.
  • S330 can be performed by the controller 210 or the path planning unit 234.
  • a lawn care area may refer to an area that a mobile device needs to cover to perform a lawn care task.
  • the mowing robot's actual path and the preset operating path cannot overlap, and the non-overlapping portion will form a missed area. It should be noted that, considering that the preset operating path can fully cover the target lawn care area, the lawn care area generally includes missed areas and non-missed areas.
  • the omitted region if the area of the omitted region is small enough, even smaller than a preset area threshold, it can be considered that the omitted region will not attract user attention and will not affect the appearance. Therefore, the omitted region with an area smaller than the area threshold can be ignored.
  • the specific omitted area threshold can be 0.5m2 .
  • the aforementioned S310 can be executed based on the actual execution of the self-moving device.
  • the missed area can be determined by comparing the difference between the actual path of the self-moving device and the preset working path.
  • the missed area can be determined by recording the coverage of the self-moving device based on the grid map.
  • the grid map For more information about the grid map, please refer to Figure 6 and its related description; in addition, the missed area can be determined by comparing the grass height near the moving path after moving along the preset working path. Since commercial lawn mowers have a large single-time cutting volume, when the grass height in a certain area is much higher than that in other areas, this area can be considered as a missed area.
  • the specific comparison method can be through pictures, outlines, etc.
  • the missing area determination component 232 can redetermine the missing areas in the lawn care area to prevent new missing areas from being generated after the lawn care task is performed on the combined area.
  • the preset algorithm may be an algorithm capable of clustering or classifying the missing regions. Based on the preset algorithm, at least two missing regions may be processed into a missing region set.
  • the missing region set is always a logical set formed by merging multiple adjacent missing regions based on a preset algorithm.
  • a combined area may refer to an area requiring lawn care tasks (re-mowing) based on the merged results.
  • a combined area may correspond one-to-one with a set of omitted areas. That is, omitted areas in the same omitted area set are located in the same combined area, and omitted areas in the same combined area belong to the same omitted area set. Considering that omitted areas are separated by non-missing areas, a combined area may include the set of omitted areas and the non-missing areas between any two omitted areas in the set of omitted areas.
  • S320 may be performed based on the aforementioned preset algorithm, i.e., the preset algorithm determines a set of missing regions reflecting the clustering result based on the distribution of each missing region, and then generates a combined region based on the missing region set. For more information on generating the combined region, see Figures 4 and 5 and related content.
  • a compensation path can refer to a planned, covering path within the combined area that can be executed by the autonomous mobile device. Specifically, the compensation path is used to guide the autonomous mobile device to perform lawn care tasks within the combined area, thereby ensuring that any missed areas are covered.
  • the aforementioned S330 can be performed based on a path planning algorithm.
  • a full coverage path planning algorithm can be executed on the combined area to determine a compensation path for the combined area; the compensation path guides the self-mobile device to perform full coverage cutting on the combined area.
  • a partial coverage path planning algorithm can be executed on the combined area to determine a compensation path for the combined area; the compensation path guides the self-mobile device to perform partial coverage cutting on the combined area.
  • the present disclosure does not limit full coverage and partial coverage; any method that can perform lawn care tasks on omitted areas in the combined area falls within the scope of protection of the present disclosure.
  • a movement path may be designed to guide the self-moving device from its current position to the combined area.
  • the working components of the self-moving device 110 may not be in an operating state.
  • the self-moving device 110 is a lawn mower robot
  • the working components may not be in an operating state while the lawn mower robot moves along the movement path, i.e., the lawn mower robot does not mow the grass.
  • the lawn mower robot moves along the compensation path
  • the working components of the lawn mower robot are in an operating state, i.e., the lawn mower robot mows the grass.
  • the execution order of the combined areas can be determined based on the relative distance between the current position of the mobile device 110 and the multiple combined areas, and steps S310 to S330 are repeated to determine the compensation path of each combined area and the movement path between each combined area.
  • merging at least two of the missing regions into a combined region according to the spatial distribution characteristics includes:
  • spatial distribution characteristics include spatial distance, spatial density, and spatial direction
  • the combined area is obtained according to the omitted area set, wherein the combined area covers each omitted area.
  • the distance between the centers of two missing areas can be determined, and the minimum distance between the boundaries of two missing areas can also be determined.
  • the corresponding spatial distribution condition is that the spatial distance between the two missing areas is less than the preset distance threshold. That is, when the distance between two missing areas is less than the preset distance threshold, they can be regarded as a set of missing areas.
  • the preset distance threshold is generally set to the width of a single operation. In the present disclosure, the width of a single operation can be 45cm. It should be noted that the present disclosure does not limit the statistical method of distance, and it can be specifically adaptively modified according to actual needs.
  • the distance can be the closest distance of the plane, the farthest distance of the plane, the closest distance of the projection at a specific angle/any angle, the farthest distance of the projection, etc.
  • missing region e.g., C
  • B Missing regions whose distances are less than the spatial distance threshold
  • the distances between the combined region and each missing region are then determined.
  • B, C, D, and E can be combined as a missing region set.
  • Missing region A can be processed independently and considered another missing region set. In this case, missing regions A-E are combined, resulting in two missing region sets.
  • the spatial density of each missing area can be determined by a preset algorithm.
  • the spatial distribution condition includes that the spatial density between two missing areas is less than a preset density threshold. When the spatial density between two missing areas is less than the preset density threshold, the two missing areas can be regarded as a missing area set.
  • the spatial distribution characteristics of the omitted regions can be taken into account, and different sets of omitted regions can be obtained using different spatial distribution characteristics and corresponding spatial distribution conditions.
  • different spatial distribution characteristics can be calculated to obtain different sets of omitted areas. For example, when using spatial distance, A can be obtained as one omitted area set, and B-E as another omitted area set. For distance, if using spatial density, A-C can be obtained as one omitted area set, and D-E as another omitted area set. In this case, the two spatial distribution characteristics can be compared to see which one has the shortest compensation time for A-E. The spatial distribution characteristic with the shortest compensation time is selected as the preferred method, and the corresponding combined area is then obtained by selecting the obtained spatial distribution characteristic.
  • the controller 210 selects different spatial distribution features to obtain at least two sets of missing areas; obtains at least two combined areas based on the at least two sets of missing areas; plans the compensation path of each combined area, and calculates the compensation time corresponding to each compensation path; and determines the spatial distribution feature based on the compensation time. Further, the compensation time of the candidate combined areas under different spatial distribution features is compared, and the corresponding spatial distribution conditions are selected according to the spatial distribution feature with the shorter compensation time, and several missing areas that meet the spatial distribution conditions are selected as the missing area set.
  • calculating the compensation time corresponding to each compensation path means calculating the time of the mobile device on the compensation path based on the compensation path under simulated conditions, wherein determining at least two sets of missing areas based on at least two preset spatial distribution conditions includes but is not limited to: a first set determination method and a second set determination method.
  • the first set determination method and the second set determination method can use two different spatial distribution features.
  • the first set determination method can be merged based on spatial distance.
  • the second set determination method can be merged based on spatial density.
  • S412 Calculate a first total time required for the self-moving device to perform lawn care along the plurality of first compensation paths, and use the first total time as a first compensation time.
  • a bounding rectangle is drawn along different directions along the boundaries of each missing area in the missing area set, and the bounding rectangle covers all missing areas in the missing area set;
  • the area covered by the minimum circumscribed rectangle is used as the combined area.
  • an initial direction may be selected, and 18 different circumscribed rectangles may be obtained in sequence with each 10° being a direction.
  • the present disclosure may use geometric features to determine a minimum bounding rectangle.
  • the minimum bounding rectangle refers to the minimum rectangle that can contain all the missing areas in the missing area set.
  • the minimum bounding rectangle can be generated based on a preset direction.
  • the minimum bounding rectangle can be generated based on an arbitrary direction.
  • multiple bounding rectangles can be generated. In the present disclosure, the areas of the multiple bounding rectangles are sorted, and the bounding rectangle with the smallest area is used as the minimum bounding rectangle. Similarly, the perimeters of the multiple bounding rectangles can also be sorted, and the bounding rectangle with the smallest perimeter is used as the minimum bounding rectangle.
  • a compensation path of the combined area needs to be planned.
  • the compensation path can cover the missed areas.
  • all the missed areas in the combined area can be compensated.
  • the self-mobile device can promptly identify missing areas and execute lawn care tasks, thereby improving the completion rate of the task.
  • a compensation path can be constructed based on the clustering results of the missing areas, so that each individual missing area can be processed in a single lawn care task. Compared with executing lawn care tasks for each missing area separately, this reduces the operation time and improves overall operation efficiency.
  • multiple circumscribed rectangles can be generated in any direction, and the circumscribed rectangle 71 and the circumscribed rectangle 72 both cover the omitted areas 73 and 74. Then, the area of each circumscribed rectangle is determined, and the circumscribed rectangle with the smallest area is identified as the minimum circumscribed rectangle. In Figure 7, the minimum circumscribed rectangle is circumscribed rectangle 72.
  • the perimeter of each circumscribed rectangle can also be determined, and the circumscribed rectangle with the smallest perimeter is identified as the minimum circumscribed rectangle. This disclosure does not limit this.
  • the above method requires many turns, which may cause the autonomous vehicle to wear out the lawn. Therefore, after obtaining a relatively appropriate combined area, it is necessary to plan a compensation path for the combined area.
  • This compensation path can cover the missed areas. When the autonomous vehicle moves along the compensation path, it can compensate for all the missed areas in the combined area.
  • the long side 721 and short side 722 of the bounding rectangle can be determined.
  • the long side 721 is longer than the short side 722.
  • a compensation path for performing the lawn care task on the combined area needs to be determined.
  • the compensation path guides the self-moving device to perform the lawn care task on the combined area.
  • a movement path can be designed, which is the shortest straight line from the current position of the self-moving device to the combined area. This ensures that the self-moving device enters the combined area as quickly as possible. This disclosure focuses on the compensation path.
  • the present disclosure takes into account that the speed of the self-moving device walking in a straight line is greater than the turning speed.
  • the compensation path in the present disclosure includes two parts.
  • the first part is a plurality of straight paths 75.
  • the spacing between the plurality of straight paths 75 can be fixed.
  • the second part is a turning path 76.
  • the turning path 76 connects two adjacent straight paths 75.
  • the present disclosure uses the direction parallel to the long side as the direction of the straight path 75.
  • the present disclosure does not limit the turning path 76. Any path that can connect the straight paths 75 to each other and can ensure that the missed area 73 or the missed area 74 near the short side can be cleared can be used as the turning path 76.
  • the compensation path of the present disclosure reduces the number of turns.
  • the self-mobile device can promptly identify missed areas and perform lawn care tasks to improve the completion rate of the task.
  • the compensation path can be constructed based on the clustering results of the missed areas, so that each individual missed area can be processed in a single lawn care task. Compared to performing lawn care tasks for each missed area separately, this reduces the operation time and the degree of wear on the lawn, thereby improving overall operation efficiency.
  • the present disclosure provides a method for determining a turning path, including: determining the position of a minimum bounding rectangle 72 in a lawn care area 7.
  • the minimum bounding rectangle 72 may be located in the middle of the lawn care area 7 or at the edge of the lawn care area 7. If the outside of the lawn care area 7 is an area inaccessible to a self-moving device, the position of the minimum bounding rectangle 72 in the lawn care area 7 will affect the type of the turning path 76. Specifically, if the minimum bounding rectangle 72 is located in the middle of the lawn care area 7, the turning path can exceed the minimum bounding rectangle 72 without worrying about the self-moving device colliding with the inaccessible area outside the lawn care area 7. If the minimum bounding rectangle 72 is located at the edge of the lawn care area 7, the turning path cannot exceed the minimum bounding rectangle 72, and collision with the inaccessible area outside the lawn care area 7 must be avoided.
  • the above method will cause the compensation path to intersect with the preset working path.
  • the pattern generated by the mobile device along the compensation path in the lawn care area is inconsistent with the pattern generated along the preset working path in the lawn care area.
  • a compensation path can also be designed.
  • the compensation path is the same as the above, except that the direction of the straight path needs to completely coincide with the preset working path. That is, the compensation path covering the combined area is planned, including:
  • a preset operation path encompassed by the minimum circumscribed rectangle is determined; the straight portion of the preset operation path within the minimum circumscribed rectangle is defined as the straight path. Since the present disclosure assumes that the preset operation path is a bow-shaped path, the preset operation path necessarily has a straight portion and a turning portion. To ensure a completely consistent pattern, the present disclosure can control the straight portion of the compensation path to completely overlap with the straight portion of the preset operation path. In this case, the pattern produced by the self-moving device along the compensation path in the lawn care area is consistent with the pattern produced along the preset operation path in the lawn care area.
  • a compensation path can also be designed.
  • the compensation path is the same as the above content, except that the direction of the straight path needs to partially overlap with the preset working path. That is, the compensation path covering the combined area is planned, including:
  • the straight direction of the straight portion of the preset working path is determined; and the direction of the straight path of the compensation path is determined based on the straight direction of the straight portion of the preset working path. Since the preset working path is assumed in the present disclosure to be a bow-shaped path, the preset working path necessarily has a straight portion and a turning portion. To ensure the direction of the pattern is consistent, the present disclosure can control the straight path of the compensation path to completely overlap with the straight portion of the preset working path. In this case, the pattern produced by the self-moving device along the compensation path in the lawn care area is consistent in direction with the pattern produced along the preset working path in the lawn care area.
  • a mobile device can determine the missed areas and non-missed areas in a lawn care area based on a raster map.
  • Raster map also known as raster image or raster data, is a pixel-based map representation method.
  • a raster map divides the geographic space into regular grid cells, each of which is called a pixel, and each pixel contains a specific value or attribute.
  • the state of a grid cell includes an un-lawn-care state and a lawn-care state, wherein the un-lawn-care state indicates that the working component is not covered, and the lawn-care state indicates that the working component is covered.
  • a grid map of a lawn care area may be first determined.
  • the status of multiple grid cells covered by a working component when the mobile device performs the lawn care task is updated from an uncared-for state to a cared-for state, and the area consisting of the multiple cared-for grid cells is defined as a cared-for area in the target lawn care area.
  • FIG6 is a schematic diagram of a grid map provided by an exemplary embodiment of the present disclosure.
  • the self-mobile device updates the state of the grid cells covered by the working component to the "lawn-tended” state based on its own positioning information.
  • the "lawn-tended” state is configured as a diagonal stripe pattern, while the "untended” state can be configured as a blank pattern (i.e., no pattern within the pixel).
  • multiple missing areas are determined based on the status of multiple grid cells in lawn care area 610. Specifically, if there are grid cells in an unmaintained state in areas 620 and 630 in lawn care area 610 shown in FIG6 , areas 620 and 630 can be configured as missing areas.
  • the controller 210 controls the mobile device 110 to perform the lawn care task for the next first sub-area, and then re-mowing on the missing area in the next first sub-area, repeating the above steps until the lawn care task for the entire lawn care area and re-mowing of the combined areas in the lawn care area are completed.
  • the above method by partitioning the lawn care area and sequentially performing lawn care tasks for each first sub-area and re-mowing the combined areas, can reduce the problem of large distance span when re-mowing the combined areas in the lawn care area.
  • the aforementioned first sub-area division can also be performed based on an island area.
  • the lawn care area can be divided into multiple first sub-areas along two circumferential tangent lines of the island area.
  • the two circumferential tangent lines of the island area are parallel to the directions of a portion of the preset operation path in the lawn care area, and the island area is configured to prohibit the autonomous mobile device from entering.
  • Island regions can be defined as a collection of grid cells in a raster map that represent obstacles.
  • the grid cells corresponding to island regions are completely filled with a solid black color.
  • grid cells can be further divided into foreground cells, which represent obstacles, and background cells, which represent non-obstacles.
  • Connected regions composed of foreground pixels can be extracted using algorithms such as connected component analysis. These connected regions are called island regions, representing the location and shape of obstacles.
  • a circumtangent line is a line that is tangent to a given shape (such as a circle, ellipse, or polygon) and tangent to the shape's curve at the point of tangency.
  • a circumtangent line is tangent to the shape's curve at only one point and does not pass through the shape.
  • the circumtangent lines of an island region are line segments extending along the edge of the island region.
  • the lawn care area may include an island area 640 .
  • Two tangent lines AA and BB can be drawn along the vertical direction of the figure to form three first sub-areas: a first sub-area 650 , a first sub-area 660 , and a first sub-area 670 enclosed by AA and BB.
  • the first sub-area 650 is located to the left of AA
  • the second sub-area 660 is located to the right of BB.
  • the first sub-area can also be divided based on the height difference between grid cells. Specifically, the height difference between each two grid cells can be determined based on a grid map of the lawn care area. When the height difference is greater than a preset value, the two grid cells are divided into different first sub-areas. In some embodiments, the height difference can be determined based on the height adjustment capability of the mobile device.
  • the first sub-area can be partitioned into multiple levels to ensure accurate identification of missed areas.
  • multiple first sub-areas can be determined within the lawn care area, each of which can be divided into at least two second sub-areas according to a preset path width.
  • the self-moving device can be controlled to move within the second sub-areas along a preset operation path. After completing a second sub-area, missed areas within the second sub-area are determined. For example, upon completing a lawn care task in a current second sub-area, missed areas within the second sub-area are determined to avoid excessively long movement paths during the compensation process.
  • the second sub-area can be determined based on a preset path width. For example, when the width of the first sub-area is an integer multiple of the preset path width, the first sub-area is divided into multiple second sub-areas, each with a width equal to the preset path width. For another example, when the width of the first sub-area is not an integer multiple of the preset path width, the first sub-area is divided into at least one second sub-area with a width equal to the preset path width and one second sub-area with a width less than the preset path width.
  • the preset path width can be an integer multiple of the width of a single operation of the self-moving device.
  • the size of the serial numbers of the above-mentioned processes does not mean the order of execution.
  • the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.
  • Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the objectives of the disclosed solution based on actual needs.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Environmental Sciences (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Guiding Agricultural Machines (AREA)

Abstract

本公开提供一种路径规划方法、自移动设备及存储介质,方法包括确定遗漏区域,遗漏区域由自移动设备偏移预设作业路径和/或工作组件未覆盖所形成;按照空间分布特征将至少两个遗漏区域合并为一个组合区域;规划覆盖组合区域的补偿路径,补偿路径被配置为引导自移动设备在所述组合区域中连续移动,工作组件在移动过程中执行草坪护理任务。由此,基于本公开提供的路径规划方法,自移动设备可以及时识别遗漏区域并执行草坪护理任务,以提高作业的完成度。此外,补偿路径可以基于遗漏区域的聚类结果构建,从而使各个单独的遗漏区域可以在一个补偿路径中同时被处理,相对于对各个遗漏区域单独执行草坪护理任务,减少了作业时间,以提高整体作业效率。

Description

一种路径规划方法、自移动设备及存储介质 技术领域
本公开涉及路径规划技术领域,具体涉及一种路径规划方法、自移动设备及存储介质。
背景技术
自移动设备可以指具备自主移动能力的设备,可以在无人操作的情况下自主地移动并执行草坪护理任务。例如,自移动设备可以包括无人机、扫地机器人以及割草机器人等。
在自移动设备执行草坪护理任务时,通过移动实现对草坪护理区域的完全覆盖,但受实际执行环境的影响,会不可避免的出现遗漏区域,从而影响对草坪护理任务的执行情况。因此,如何实现消除遗漏区域是本领域技术人员亟待解决的技术问题。
发明内容
有鉴于此,本公开实施例提供了一种路径规划方法,应用于具有工作组件的自移动设备,所述自移动设备被配置为沿预设作业路径移动,并在移动过程中通过所述工作组件执行草坪护理任务,所述方法包括:
确定遗漏区域,所述遗漏区域由所述自移动设备偏移所述预设作业路径和/或所述工作组件未覆盖所形成;
按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域;
规划覆盖所述组合区域的补偿路径,所述补偿路径被配置为引导所述自移动设备在所述组合区域中连续移动,所述工作组件在移动过程中执行所述草坪护理任务。
在一些实施例中,所述按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域,包括:
确定所述遗漏区域之间的空间分布特征,其中所述空间分布特征包括空间距离、空间密度、空间方向;
选择空间分布特征满足预设空间分布条件的至少两个所述遗漏区域;
将至少两个所述遗漏区域合并为一个所述组合区域,其中所述组合区域覆盖各遗漏区域。
在一些实施例中,所述空间分布条件包括:
两个遗漏区域之间的空间距离小于预设距离阈值。
在一些实施例中,所述将至少两个所述遗漏区域合并为一个所述组合区域,包括:
确定至少两个遗漏区域中各遗漏区域的边界;
对各遗漏区域边界进行边界拟合,得到所述组合区域。
在一些实施例中,所述对各遗漏区域边界进行边界拟合,得到所述组合区域,包括:
沿不同方向分别对各遗漏区域边界作一个外接矩形;
根据几何特征确定各外接矩形中的最小外接矩形,其中所述几何特征包括面积、边长、长宽比;
将所述最小外接矩形所覆盖的区域作为所述组合区域。
在一些实施例中,所述根据几何属性选择各遗漏区域对应的最小外接矩形,包括:
将各外接矩形中面积最小的外接矩形作为所述最小外接矩形;或,
将各外接矩形中边长最小的外接矩形作为所述最小外接矩形;或,
将各外接矩形中长宽比最大的外接矩形作为所述最小外接矩形。
在一些实施例中,所述补偿路径包括若干直行路径和若干转弯路径,相邻的直行路径通过转弯路径相连。
在一些实施例中,所述规划覆盖所述组合区域的补偿路径,包括:
确定所述最小外接矩形的长边和短边,其中所述长边的长度大于短边的长度;
根据所述长边确定所述直行路径。
在一些实施例中,所述根据所述长边确定所述直行路径,包括:
确定所述长边的方向;
根据所述长边的方向确定所述直行路径的方向。
在一些实施例中,所述规划覆盖所述组合区域的补偿路径,包括:
确定所述最小外接矩形所覆盖的预设作业路径;
将所述预设作业路径在所述最小外接矩形内的直行部分作为所述直行路径。
在一些实施例中,所述规划覆盖所述组合区域的补偿路径,包括:
确定所述预设作业路径的直行部分的直行方向;
根据所述预设作业路径的直行部分的直行方向确定所述补偿路径的直行路径的方向。
在一些实施例中,所述预设作业路径引导所述自移动设备在草坪护理区域内移动;
还包括:确定所述外接矩形在所述草坪护理区域中的位置;
根据所述位置确定转弯路径的类型。
在一些实施例中,还包括:
选择所述遗漏区域的至少两种空间分布特征;
分别根据至少两种空间分布条件将至少两个所述遗漏区域合并得到至少两个所述组合区域;
规划覆盖各组合区域的补偿路径,并计算各所述补偿路径对应的补偿时间;
根据所述补偿时间选择所述组合区域。
在一些实施例中,所述根据所述补偿时间选择所述组合区域,包括
对各补偿时间进行排序,根据排序结果选择所述组合区域。
在一些实施例中,还包括:
判断至少两个所述遗漏区域之间是否满足阻隔条件;
当满足所述阻隔条件时,将至少两个所述遗漏区域划分到不同的组合区域。
在一些实施例中,所述阻隔条件包括以下条件中的一个或多个:
所述遗漏区域间的高度差大于预设高度阈值;
所述遗漏区域间存在不可穿越障碍物;
所述遗漏区域间的切割参数不同。
在一些实施例中,还包括:
确定所述遗漏区域的面积;
忽略面积小于面积阈值的遗漏区域。
在第二方面,本公开提供一种自移动设备,包括:
工作组件,被配置为执行草坪护理任务;
移动组件,被配置为带动所述工作组件运动,以使所述工作组件在移动中执行所述草坪护理任务;以及
处理器,被配置为与所述移动组件以及所述工作组件连接,以实现任意一项所述的方法。
在一些实施例中,所述工作组件包括切割组件;
所述处理器,被配置为根据草坪护理区域的地图,规划所述自移动设备在所述草坪护理区域中的作业路径,并向所述切割组件和所述移动组件输出第一指令,以控制所述自移动设备在所述草坪护理区域中沿所述作业路径进行自主遍历和一步到位的割草,其中,一步到位的割草通过在割草时同步地碎草或集草实现。
在第三方面,本公开提供一种计算机可读存储介质,所述存储介质存储有计算机程序,所述计算机程序在被处理器执行时,使所述处理器执行任意一项所述的方法。
本公开实施例提供了一种路径规划方法、自移动设备及存储介质。基于本公开提供的路径规划方法,自移动设备可以及时识别遗漏区域并执行草坪护理任务,以提高作业的完成度。此外,补偿路径可以基于遗漏区域的聚类结果构建,从而使各个单独的遗漏区域可以在一个草坪护理任务中同时被处理,相对于对各个遗漏区域单独执行草坪护理任务,减少了作业时间,以提高整体作业效率。
附图说明
为了更清楚地说明本发明实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,应当理解,以下附图仅示出了本发明的某些实施例,因此不应被看作是对范围的限定,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相关的附图。
图1是本公开一示例性实施例提供的自移动设备的应用场景图。
图2是本公开一示例性实施例提供的自移动设备的结构示意图。
图3是本公开一示例性实施例提供的路径规划方法的流程示意图。
图4是本公开一示例性实施例提供的执行预设算法的流程示意图。
图5是本公开一示例性实施例提供的确定遗漏区域集合的示意图。
图6是本公开一示例性实施例提供的栅格地图示意图;
图7是本公开一示例性实施例提供的补偿作业路径示意图。
具体实施方式
具体实施方式下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,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,都属于本公开保护的范围。自移动设备适用于在草坪护理区域执行至少一种草坪护理任务,通常包括机身以及与机身连接的移动组件、传感器组件、工作组件以及控制电路。控制电路与移动组件、传感器组件、工作组件耦合,根据传感器组件输出的传感数据控制移动组件和工作组件动作。传感器组件可以包括但不限于:视觉组件、定位装置、碰撞传感组件、超声传感组件等中的至少一种。移动组件通常包括驱动电机以及与驱动电机连接的驱动轮,可选的,还可以包括从动轮,以与驱动轮共同带动自移动设备移动。工作组件可以包括但不限于:割草用的刀盘、吹落叶的吹风工作头、扫雪用的扫雪工作头中的至少一种。
在一些实施例中,工作组件可以根据自移动设备110需要执行的草坪护理类型确定。例如,对于清洁任务,工作组件可以包括吹落叶工作头、吹雪工作头。再例如,对于切割任务,工作组件可以包括刀具。又例如,对于喷洒任务,工作组件可以包括喷洒组件。
在一些实施例中,移动组件可以根据自移动设备110的实际工作环境选取。例如,对于在地面移动的自移动设备,移动组件可以为轮胎、履带等能在路面运动的装置。再例如,对于在空中移动的自移动设备,移动组件可以为空气螺旋桨、喷气发动机等能在空中移动的装置。
自移动设备的草坪护理任务一般呈现为移动工作,即自移动设备需要在移动过程中进行作业。例如,对于无人机,其草坪护理任务可以为在目标区域内飞行并在飞行过程中喷洒农药或清水。再例如,对于割草机器人,其草坪护理任务为可以为在草坪移动并在移动过程中修剪草坪,可以为在草坪移动并在移动过程释放除虫剂、释放香氛,还可以在移动过程中吹散落叶,吸取草坪上散落的落叶或碎草等。
以割草机器人为例,上述能够自主工作的割草机器人目前主要应用在家用领域,在商用割草领域鲜有看到采用智能控制的割草机器人。
在商用割草领域,商业团队通常根据排好的订单,在一天内驾驶装载由割草机器人和其他园林工具的运输车前往不同的用户家庭进行草坪维护,包括割草、修边、修枝、吹落叶等多项工作。这些工作通常需由工作人员手动完成,导致团队的人力支出较高。并且,天气、地形、植被的分布和生长状况等因素都会影响团队的作业时间,作业时间越长,商业团队的人力支出也越多。除此之外,对于商用割草机器人而言,商业团队中不同的人的驾驶技术也各有差异,容易导致工作效率和切割质量的不稳定。
在一些特殊情况下,也有商业团队将商用割草机器人替换为多台小型家用割草机器人,以减少人力,实现自动控制割草。但这类方式通常需要配备多台智能割草机,成本仍较高,并且,家用的割草机器人通常移动速度较慢,工作效率较低,适用于草况稳定的每日维护性质的切割。而商用割草机器人通常需应用于草况各异的草坪(如工况较恶劣的高草和/或密草草坪),且每次切割需达到一定日期(例如1~2周)内无需再次进行维护的质量,也就是说,家用割草机器人工作频次通常为一天一次,而商用割草机器人的工作频次通常为一到两周一次。在商业团队的草坪维护过程中,为实现商用割草机器人对草坪护理任务中草坪护理区域的全覆盖,可以采用全覆盖路径规划算法预先规划商用割草机器人在草坪护理区域的预设作业路径。但在实际中,因为草坪护理区域的实际地形的影响,商用割草机器人在作业过程中很容易产生遗漏区域。例如,对于在草坪上移动的商用割草机器人,当草坪护理区域中存在斜坡、沟道或存在不规则边缘时,很可能形成遗漏区域。再例如,在执行预设作业路径时,受环境(如风、地形、障碍物)影响,商用割草机器人本身的运动可能不稳定从而产生遗漏区域。
为了方便说明,可以将草坪护理区域分为遗漏区域和未遗漏区域,遗漏区域为割草机器人偏移预设作业路径和/或所述工作组件未覆盖所形成,相反的,未遗漏区域为割草机器人沿着预设作业路径和所述工作组件覆盖所形成。通俗的讲,若割草机器人沿着按照预设作业路径移动时,草坪护理区域将不会出现未被工作组件切割的区域,然而当割草机器人由于偶发时间产生偏移,此时其真实路径与预设作业路径产生了偏移,那么将会在真实路径和预设作业路径之间产生遗漏区域,即遗漏区域的草坪未被工作组件切割。
上述遗漏区域会直接影响移动草坪护理任务的完成情况,当存在遗漏区域时,则说明存在部分草坪未被切割,遗漏区域的草坪高度相对于未遗漏区域的草坪高度更加突出,从而影响整体草坪的外观评价。具体地,由于家用割草机器人体积较小,单次切割量较小,因此其工作频次通常为每日切割,其遗漏区域与未遗漏区域之间的高度差较小,用户肉眼感受不明显,不会影响用户的观感。
但是每天都雇佣商业团队进行草坪维护价格十分昂贵,所以大多数用户会选择商业团队与商用割草机器人的搭配,相比于家用割草机器人,商用割草机器人可以提升单次切割量以降低雇佣商业团队的次数,具体为1-2周一次。在某些雨量充沛、光照强烈、温度适宜的季节下,草的生长速度较快,因此在出现遗漏区域时,商用割草机器人的遗漏区域与未遗漏区域之间的高度差较大,有时甚至可以达到五厘米以上,用户肉眼感受强烈,若不对漏草区域进行补割将会影响用户对于该商业团队的好感度,所以商用割草机器人对漏草区域进行补割是比较重要的。
在相关技术的实施例中,对上述遗漏区域的补偿作业一般有两个模式。一种为实时补偿,即当产生遗漏区域时立刻停止作业进行补偿;另一种为完成补偿,即当完成草坪护理任务后对遗漏区域逐一进行补偿。在一些实施例中,当自移动设备为割草机器人时,可以将对遗漏区域进行补割认为是一种补偿,当然,当自移动设备为自动灌溉机时,可以将对遗漏区域进行补水认为是一种补偿,当自移动设备为自动吹落叶机时,可以将对遗漏区域进行补吹落叶认为是一种补偿。
本公开首先简单叙述商用割草机器人如何执行草坪护理任务,以及如何进行补割:
在商业团队将商用割草机运输或控制其到达草坪护理区域后,首先商用割草机器人可以获取商业团队绘制好的地图,并通过地图去确定预设作业路径,该作业路径可以是商业团队预先设计好的,也可以是商用割草机器人即时生成的,还可以是草坪护理区域主人指定的,本公开对此不做限定,在获取到预设作业路径时,商用割草机器人将严格参照预设作业路径执行草坪护理任务,在商用割草机器人沿着预设作业路径执行草坪护理任务时,草坪将会被切割,通常,商用割草机器人的移动中心严格按照预设作业路径执行草坪护理任务时,将不会产生遗漏区域,但是草坪所处的室外环境十分复杂,很容易对商用割草机器人造成影响导致其移动中心不能严格按照预设作业路径执行草坪护理任务,此时将有可能产生遗漏区域。在一些实施例中,移动中心可以为商用割草机器人的几何中心,也可以是其定位模块的中心。此时,商用割草机器人将会去解决遗漏区域,但是现在的方式通常为出现遗漏区域就去解决这个遗漏区域,很容易出现商用割草机器人反复掉头回退的现象,因此效率较低。
所以商用割草机器人对遗漏区域的实时补偿至少存在以下技术问题:完成补偿一般基于各个遗漏区域的中心位置进行路径规划,从而实现对遗漏区域的逐一补偿,即使是对于相距比较近的遗漏区域也需要分别补偿,从而严重拖慢的补偿速度;
基于上述技术问题,本公开提供了一种路径规划方法、自移动设备及存储介质。本公开提供的路径规划方法可以确定已草坪护理区域中的遗漏区域,并对遗漏区域进行聚类,以使至少两个遗漏区域处理为一个组合区域,确定组合区域的补偿路径,通过补偿路径引导自移动设备同时组合区域中的至少两个遗漏区域执行草坪护理任务(补偿),在本公开中,处理手段包括聚合、合并或选取等。由此,基于本公开提供的路径规划方法,自移动设备可以同时识别至少两个遗漏区域,并对至少两个遗漏区域在一个连续的补偿路径中执行草坪护理任务,以提高自移动设备对草坪护理区域作业的作业覆盖度。此外,补偿路径可以基于遗漏区域的聚类结果构建,从而使至少两个遗漏区域可以在一个草坪护理任务中同时被处理,相对于对各个遗漏区域单独依次执行草坪护理任务,减少了作业时间,以提高整体作业效率。
如图1所示,本公开实施例中的控制方法应用于自移动设备110,自移动设备110被配置为在草坪护理区域200中行驶和/或工作。其中,自移动设备110与草坪护理区域200可以根据实际需要适应性调整。为方便描述,在本公开中以割草机器人进行说明,本领域技术人员可以根据需要适应性修改技术内容,以适应无人机、扫地机器人、播种机器人、灌溉机器人、吹落叶机器人、驱虫机器人等自移动设备。
在一些实施例中,自移动设备110可以为用于执行割草作业的割草机器人,则草坪护理区域200可以为待割草的草坪。自移动设备110可以根据设定的草坪护理任务在草坪护理区域200进行移动割草。在自移动设备110移动割草的过程中,草坪护理区域200可能出现遗漏区域(未割草区域)。基于本公开提供的路径规划方法,自移动设备110可以基于聚类算法生成补偿路径,从而对遗漏区域进行草坪护理任务(割草)。
在本公开的一种实施方式中,用户可以通过用户终端120与自移动设备110交互。如图1所示,自移动设备110可以通过网络130获取用户终端120发送的草坪护理任务,也可以通过网络130向用户终端120发送草坪护理任务的完成情况。草坪护理任务包括但不限于:作业类型和草坪护理区域200。其中,用户终端120为可以与自移动设备110建立通信连接以发送控制指令的终端设备,控制指令可以包括移动草坪护理任务。如图1所示,用户终端120可以是移动设备120-1、平板计算机120-2、膝上型计算机120-3、台式计算机120-4等其他具有输入和/或输出功能的设备中的一种或其任意组合。
在一些实施例中,考虑到预设作业路径规划过程可能涉及多方交互。例如,前述自移动设备110可以包括多个设备,并互相协同完成草坪护理任务。再例如,用户终端120也可能与路径规划过程有关(如设置作业边界)。则前述控制器210还可以被配置为路径规划过程中涉及的计算设备的集合。例如,控制器210可以包括用户终端120处设置路径规划区域的相关设备以及各个自移动设备110的相关控制器。
在本公开的另一种实施方式中,草坪护理区域200的地图可以预先下载到自移动设备100的存储介质220或控制器210中,自移动设备110基于草坪护理区域200的地图和设定的移动草坪护理任务在草坪护理区域200中执行移动作业。
其中,控制器210可以是执行本公开提供的路径规划方法的计算设备。例如,本公开提供的路径规划方法由自移动设备110执行,则控制器210可以包括设置在自移动设备110中并集成设置有路径规划算法的控制器(如嵌入式控制器)。再例如,本公开提供的路径规划方法由云端服务器执行,则控制器210可以包括云端服务器中的计算设备(如CPU、SOC等)。此时,云端服务器对本公开提供的路径规划方法的处理结果可以配置为自移动设备110中各个控制器的控制指令,并通过网络130发送到自移动设备110,以使自移动设备110响应于云端处理器的控制指令。
在本公开中,网络130可以是有线网络或无线网络中的任意一种或多种。例如,网络130可以包括电缆网络、光纤网络、电信网络、互联网、局域网络(LAN)、广域网络(WAN)、无线局域网络(WLAN)、城域网(MAN)、公共交换电话网络(PSTN)、蓝牙网络、紫蜂网络(ZigBee)、近场通信(NFC)、设备内总线、设备内线路、线缆连接等或其任意组合。各部分之间的网络连接可以是采用上述一种方式,也可以是采取多种方式。在一些实施例中,网络可以是点对点的、共享的、中心式的等各种拓扑结构或者多种拓扑结构的组合。在一些实施例中,网络130可以包括一个或以上网络接入点。该应用场景内的一个或多个组件可连接到网络130上以交换数据和/或信息。
示例性自移动设备
为进一步说明自移动设备100在执行路径规划方法时的过程,下面结合图2对自移动设备的内部结构进行进一步说明。
如图2所示,自移动设备110可以包括但不限于工作组件111、移动组件112、传感器组件113、控制器210。
工作组件111、移动组件112、传感器组件113的具体组成情况可以与自移动设备110本身的功能匹配,并基于实际工作要求调整。以割草机器人为例,工作组件111可以包括割草用的刀盘。移动组件112通常包括驱动电机以及与驱动电机连接的运动装置(如轮胎、履带等)。传感器组件113可以包括视觉传感器、定位装置、碰撞传感器、超声传感器等中的至少一种。
前述工作组件111、移动组件112以及传感器组件113可以响应于控制指令变更工作状态。其中,工作组件111、移动组件112以及传感器组件113的控制方法可以根据实际情况调整。例如,工作组件111以及移动组件112可以由电机驱动,则控制工作组件111以及移动组件112时,需要生成对应的电机驱动信号(一般包括电机供电信号以及霍尔信号)。再例如,工作组件111、移动组件112以及传感器组件113由对应的控制器(如域控制器、域控制器、微控制器等)驱动,则可以通过生成对应的控制信号即可由对应的控制器实现控制。
在一些实施例中,控制器210可以包含一个或多个子处理设备(例如,单核处理设备或多核多芯处理设备)。仅作为示例,控制器210可以包括中央处理器(CPU)、专用集成电路(ASIC)、专用指令处理器(ASIP)、图形处理器(GPU)、物理处理器(PPU)、数字信号处理器(DSP)、现场可编程门阵列(FPGA)、可编辑逻辑电路(PLD)、控制器、微控制器单元、精简指令集电脑(RISC)、微处理器等或以上任意组合。
在本公开中,自移动设备110还可以包括集成在自移动设备上的存储介质220。存储介质220可以存储自移动设备110内的指令/数据。其中,存储介质220可以存储本公开路径规划方法的相关指令,以使相关指令被控制器210调用后,实现本公开实施例提供的路径规划方法。此外,存储介质220还可以缓存路径规划方法执行过程中的相关数据(如栅格地图、传感数据等),以保证路径规划方法的执行。
在本公开的一种实施方式中,控制器210可以集成设置在自移动设备110中。例如,自移动设备110可以设置有印刷电板(Printed Circuit Board,PCB)、系统级芯片(System on Chip,SOC)、电子控制器单元(Electronic Control Unit,ECU)等计算设备,其中,计算装置中的处理单元(如CPU)可以作为前述控制器210,存储单元(如RAM以及ROM)可以作为存储介质220。
在本公开中,控制器210可以为工作组件111、移动组件112、传感器组件113提供控制信号,以控制其工作状态。在本公开的一种实施方式中,当工作组件111、移动组件112、传感器组件113直接由电机驱动,则控制器210可以基于预设协议生成电机驱动信号,以控制工作组件111、移动组件112、传感器组件113中的至少一种。
在本公开的一种实施方式中,控制器210可以包括路径规划装置230,路径规划装置230为工作组件111、移动组件112、传感器组件113提供控制信号,以控制其工作状态。如图2所示,路径规划装置230可以包括未遗漏区域确定组件单元231、遗漏区域确定组件单元232、组合区域确定组件单元233以及路径规划组件单元。
未遗漏区域确定组件单元231可以用于控制自移动设备沿预设作业路径对目标草坪护理区域执行草坪护理任务,确定目标草坪护理区域中的未遗漏区域。
遗漏区域确定组件单元232可以用于确定草坪护理区域中的遗漏区域。
组合区域确定组件单元233可以用于按照预设算法将至少两个遗漏区域合并为一个组合区域。
路径规划组件单元234可以用于规划自移动设备的补偿路径,基于补偿路径对组合区域执行草坪护理任务。
在本公开中,路径规划装置230确定自移动设备的预设作业路径,并基于预设作业路径实时生成对应的控制信号,并发送到工作组件111以及移动组件112,以使自移动设备在移动组件112带动下沿预设作业路径移动并通过工作组件111进行移动作业。控制器210可以根据工作组件111以及移动组件112对预设作业路径的执行情况确定未遗漏区域和遗漏区域。
前述遗漏区域确定组件单元232的功能可以由传感器组件113、控制器210以及存储介质220实现。即传感器组件113可以实时感知自移动设备110的位置,并存储在存储介质220内。控制器210可以基于自移动设备110的实时位置确定遗漏区域(例如,基于栅格地图确定)。
前述组合区域确定组件单元233的功能可以由控制器210以及存储介质220实现。即控制器210可以基于存储介质220中的遗漏区域进行处理(如聚类),从而确定包含多个遗漏区域的组合区域。
前述路径规划组件单元234的功能可以由工作组件111、移动组件112以及控制器210实现。即可以由控制器210基于组合区域确定补偿路径,并基于补偿路径生成对应的控制信号,并发送到工作组件111以及移动组件112,以使自移动设备沿补偿路径移动并进行移动作业,以实现对遗漏区域的草坪护理任务。
需要说明的是,本公开实施例对功能单元的划分可以基于实际需要进行调整。即本公开实施例中对功能单元的划分是示意性的,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。例如,可以将两个或两个以上的功能单元集成在一个单元中。此外,上述集成的功能单元既可以采用硬件的形式实现,也可以采用软件功能模块的形式实现。
下面将基于前述应用场景以及自移动设备对本公开提供的基于遗漏区域的路径规划方法进行详细描述。
示例性路径规划方法
现有路径规划方法大多基于固定网格划分或实时感知遗漏区域进行路径回补,但未充分考虑遗漏区域的空间分布特征,往往采用逐个补偿的方式,导致自移动设备反复调头、路径重叠,作业时间延长,影响整体效率等问题。
如图3所示,在一些实施例中,控制器210可以独立控制或通过路径规划装置230控制自移动设备110执行以下步骤:
S310、确定遗漏区域,所述遗漏区域由所述自移动设备偏移所述预设作业路径和/或所述工作组件未覆盖所形成。在一些实施例中,S310可以由控制器210或遗漏区域确定单元232执行。
S320、按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域。在一些实施例中,S320可以由控制器210或组合区域确定组件单元233执行。
S330、规划覆盖所述组合区域的补偿路径,所述补偿路径被配置为引导所述自移动设备在所述组合区域中连续移动,所述工作组件在移动过程中执行所述草坪护理任务。在一些实施例中,S330可以由控制器210或路径规划单元件234执行。
草坪护理任务可以指自移动设备被设置需要执行的工作。在一些实施例中,草坪护理任务可以由用户通过用户终端设置。例如,用户可以通过用户终端向自移动设备输入草坪护理区域以及作业要求,从而确定需要自移动设备执行的草坪护理任务。示例性的,对于割草机器人,草坪护理任务可以包括在草坪护理区域的割草任务。其中,草坪护理区域指待割草区域,作业要求可以指割草后草坪的剩余高度。
特别要说明的是,所述补偿路径被配置为引导所述自移动设备在所述组合区域中连续移动中的“连续”可以表示自移动设备以补偿路径进行单次的移动,其在移动过程中工作组件覆盖过的区域,工作组件将不会再次覆盖,由此避免路径重叠,反复掉头。
草坪护理区域可以指自移动设备需要执行完成草坪护理任务所覆盖的区域。
预设作业路径可以是一种基于草坪护理区域的自移动设备路径。当自移动设备沿着预设作业路径的开始位置移动到预设作业路径的结束时,理论上自移动设备可以完成草坪护理区域的草坪护理任务。
在一些实施例中,考虑到自移动设备基于预设作业路径需要完成目标草坪护理区域的草坪护理任务。则预设作业路径可以是一种基于工作组件的单次作业宽度的全覆盖路径。即基于单次作业宽度扩展预设作业路径后形成的平面可以完全覆盖草坪护理区域。仅作为一种示例,本公开中可以采用弓字型全覆盖路径规划算法进行全覆盖路径规划,以确定预设作业路径。在本公开中,当自移动设备为割草机器人,此时工作组件可以为切割组件,那么工作组件覆盖的区域指的是切割组件进行切割后所形成的区域。
在一些实施例中,前述S310可以基于自移动设备对预设作业路径的执行情况执行。即可以通过记录自移动设备的移动路径确定自移动设备对预设作业路径的执行情况,从而将自移动设备当前所在的草坪护理区域中已完成的预设作业路径对应的部分目标草坪护理区域作为已草坪护理区域。
在本公开中,遗漏区域可以指草坪护理区域中割草机器人偏移预设作业路径和为进行切割所形成的区域。例如,遗漏区域可以包括自移动设备受实际作业环境(如障碍物、斜坡等)干扰而切割到的区域,比如预设作业路径穿过一块斜坡,该预设作业路径处于穿过斜坡的中间部位,由于该斜坡的坡度较大,在割草机器人的侧向防滑无法保证其稳定的沿预设作业路径移动时,割草机器人将从斜坡的高处滑落到低处,此时割草机器人的真实路径与预设作业路径无法重合,无法重合的部分将会形成遗漏区域。需要说明的是,考虑到预设作业路径可以对目标草坪护理区域进行全覆盖,因此,通常情况下,草坪护理区域包括遗漏区域和未遗漏区域。
在一些实施例中,若遗漏区域的面积足够的小,甚至小于预设的面积阈值,那么可以认为这个遗漏区域不会引起用户注意,不会影响美观,所以可以将面积小于面积阈值的遗漏区域忽略,具体的遗漏面积阈值可以为0.5m2。
在一些实施例中,前述S310可以基于自移动设备的实际执行情况执行。例如,可以通过比较自移动设备的真实路径与预设作业路径的差异确定遗漏区域。具体来说,可以基于栅格地图通过记录自移动设备的覆盖范围从而确定遗漏区域。关于栅格地图的更多内容可以参见图6及其相关描述;此外,还可以通过比较沿着预设作业路径移动后,移动路径附近的草高确定遗漏区域,由于商用割草机单次切割量大,当某个区域相比于其他区域的草高高出很多时,可以认为这片区域为遗漏区域,具体的比较方式可以通过图片、轮廓等。
在一些实施例中,当对一个组合区域执行完成草坪护理任务时,遗漏区域确定组件232可以重新确定草坪护理区域中的遗漏区域,防止对组合区域执行草坪护理任务后,产生的新的遗漏区域。
预设算法可以是一种能对遗漏区域进行聚类或分类的算法。基于预设算法可以将至少两个遗漏区域处理为一个遗漏区域集合。
遗漏区域集合始终基于预设算法将多个相邻遗漏区域合并而成的逻辑集合。
组合区域可以指基于合并结果确定的需要进行草坪护理任务作业(补割)的区域。在一些实施例中,组合区域可以与遗漏区域集合一一对应,即,同一个遗漏区域集合中的遗漏区域位于同一个组合区域,同一个组合区域中的遗漏区域属于同一个遗漏区域集合。考虑到遗漏区域之间被未遗漏区域隔离,则组合区域可以包括遗漏区域集合以及联通遗漏区域集合中任意两个遗漏区域之间的未遗漏区域。
在一些实施例中,前述S320可以基于前述预设算法执行,即基于前述预设算法基于各个遗漏区域的分布情况确定反映聚类结果的遗漏区域集合,再基于遗漏区域集合对应生成组合区域。关于生成组合区域的更多内容可以参见图4、图5及其相关内容。
补偿路径可以指在组合区域范围内规划出的、可被自移动设备执行的覆盖性路径。即补偿路径用于引导自移动设备对组合区域执行草坪护理任务,使遗漏区域变为未遗漏区域。
在一些实施例中,前述S330可以基于路径规划算法执行。例如,可以对组合区域执行全覆盖路径规划算法确定组合区域的补偿路径;补偿路径引导自移动设备对组合区域执行全覆盖切割。此外,还可以对组合区域执行部分覆盖路径规划算法确定组合区域的补偿路径,补偿路径引导自移动设备对组合区域执行部分覆盖切割。本公开对于全覆盖和部分覆盖不做限定,凡是可以将组合区域中的遗漏区域进行草坪护理任务的方式均属于本公开的保护范围。
此外,还可以设计移动路径,移动路径引导自移动设备从当前位置移动到组合区域。在执行移动路径时,自移动设备110的工作组件可以不处于工作状态,例如,当自移动设备110为割草机器人时,当割草机器人沿移动路径移动时,工作组件可以不处于工作状态,即割草机器人不割草,而当割草机器人沿补偿路径移动时,割草机器人的工作组件处于工作状态,即割草机器人割草。
在一些实施例中,当存在多个组合区域时,可以基于自移动设备110的当前位置与多个组合区域之间的相对距离确定组合区域的执行顺序,并重复步骤S310至步骤S330以确定各个组合区域的补偿路径以及各个组合区域之间的移动路径。
在一些实施例中,所述按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域,包括:
确定所述遗漏区域之间的空间分布特征,其中所述空间分布特征包括空间距离、空间密度、空间方向;
选择空间分布特征满足预设空间分布条件的至少两个所述遗漏区域,作为一个遗漏区域集合;
根据所述遗漏区域集合得到所述组合区域,其中所述组合区域覆盖各遗漏区域。
以空间距离举例,可以确定两个遗漏区域的中心之间的距离,也可以确定两个遗漏区域的边界的最小距离。在使用空间距离时,对应的空间分布条件为两个遗漏区域之间的空间距离小于预设距离阈值。即两个遗漏区域之间的距离小于预设距离阈值时,可以作为一个遗漏区域集合。其中,预设距离阈值一般设置为单次作业宽度。本公开中单次作业宽度可以为45cm。需要说明的是,本公开并不限制距离的统计方法,具体的可以根据实际需要进行适应性修改。例如,距离可以为平面最近距离、平面最远距离,特定角度/任意角度投影的投影最近距离、投影最远距离等。
示例性地,基于图5所示的遗漏区域,可以先选取一个遗漏区域(如C),然后确定各个遗漏区域与遗漏区域C的距离,并将距离小于空间距离阈值的遗漏区域(如B)和遗漏区域C作为一个遗漏区域集合。再确定组合区域与各个遗漏区域的距离,重复上述步骤可知,基于空间分布条件,可以将B、C、D、E作为一个遗漏区域集合,遗漏区域A进行单独作业,并可以认为遗漏区域A作为另一个遗漏区域集合,此时遗漏区域A-E,得到两个遗漏区域集合。
此外,还可以通过预设算法确定各遗漏区域的空间密度,空间分布条件包括两个遗漏区域之间的空间密度小于预设密度阈值,当两个遗漏区域之间的空间密度小于预设密度阈值时,可以将这两个遗漏区域作为一个遗漏区域集合。
通过这种方式,可以考虑到遗漏区域的空间分布特征,使用不同的空间分布特征与对应的空间分布条件得到不同的遗漏区域集合。
在一些实施例中,为了提升对组合区域执行草坪护理任务的效率,还可以计算不同空间分布特征,得到的不同的遗漏区域集合,例如使用空间距离时,可以得到A作为一个遗漏区域集合,得到B-E作为另一个遗漏区域集合。距离来说,若使用空间密度,可以得到A-C作为一个遗漏区域集合,D-E作为另一个遗漏区域集合,此时,可以比较再这两个空间分布特征下,哪种对于A-E的补偿时间最短,将补偿时间最短的空间分布特征作为较佳的方式,然后通过选择得到的空间分布特征得到对应的组合区域。
具体的,控制器210选择不同的空间分布特征,得到至少两个遗漏区域集合;根据至少两个遗漏区域集合得到至少两个组合区域;规划各组合区域的补偿路径,并计算各所述补偿路径对应的补偿时间;根据所述补偿时间确定所述空间分布特征。进一步,比较不同空间分布特征下候选组合区域的补偿时间,按照补偿时间较短的空间分布特征选择对应的空间分布条件,选择满足空间分布条件的若干遗漏区域,作为遗漏区域集合。在本公开中,计算各补偿路径对应的补偿时间是指在模拟的条件下,基于补偿路径,计算自移动设备在补偿路径的时间,其中,根据至少两种预设空间分布条件确定至少两个遗漏区域集合包括但不限于:第一集合确定方法,第二集合确定方法。第一集合确定方法与第二集合确定方法可以使用两种不同的空间分布特征。例如,第一集合确定方法可以基于空间距离进行合并。第二集合确定方法可以基于空间密度进行合并。
为进一步描述该过程,图4反映了该过程(S330)的流程示意图。图5反映了两种集合确定方法确实确定遗漏区域集合时的示意图。下面将结合图4、5对该过程进行说明。
在一些实施例中,图4所示的前述S330可以包括如下步骤:
S410、基于第一集合确定方法确定第一补偿路径,确定基于第一补偿路径执行草坪护理任务所需要的第一补偿时间。
S420、基于第二集合确定方法确定第二补偿路径,确定基于第二补偿路径执行草坪护理任务所需要的第二补偿时间。
S430、比较第一补偿时间与第二补偿时间。
S440、当第一补偿时间小于第二补偿时间时,通过第一集合确定方法将至少两个遗漏区域处理为一个遗漏区域集合。
S450、当第一补偿时间大于第二补偿时间时,通过第二集合确定方法将至少两个遗漏区域处理为一个遗漏区域集合。
S460、当第一补偿时间等于第二补偿时间时,通过第一集合确定方法或第二集合确定方法将至少两个遗漏区域处理为一个遗漏区域集合。
第一补偿路径可以指基于第一集合确定方法确定的候选组合区域的作业路径。第一补偿时间可以指自移动设备完成第一补偿路径的时间长度。其中,第一补偿时间可以根据第一补偿路径的长度以及自移动设备的工作参数确定。可以理解的是,第二补偿路径可以指基于第二集合确定方法确定的候选组合区域的作业路径,第二补偿时间可以指自移动设备完成第二补偿路径的时间长度。其中,第二补偿时间可以根据第二补偿路径的长度以及自移动设备的工作参数确定。其中,自移动设备的工作参数可以包括但不限于:移动速度、转弯半径等。
在一些实施例中,在自移动设备的爬坡能力和防侧倾能力有限时,若不设计任何条件将满足预设空间分布条件的遗漏区域进行合并,将有可能导致处于同一遗漏区域集合的两个遗漏区域间高度相差过大,有可能影响自移动设备的通过性,因此在将空间分布特征满足预设空间分布条件的至少两个遗漏区域处理为一个遗漏区域集合之前,包括:判断至少两个所述遗漏区域之间是否满足阻隔条件;当满足所述阻隔条件时,将至少两个所述遗漏区域划分到不同的遗漏区域集合,当不满足所述阻隔条件时,可以将至少两个所述遗漏区域划分到同一遗漏区域集合。本公开中,阻隔条件包括以下条件中的一个或多个:遗漏区域间的高度差小于预设高度阈值;遗漏区域间存在不可穿越障碍物;遗漏区域间的切割参数不同。
举例说明,即便遗漏区域A和遗漏区域B的空间分布特征满足空间分布条件,若遗漏区域A和遗漏区域B之间设有障碍物,那么将极大的影响自移动设备的补偿效率,因此可以考虑将遗漏区域A和遗漏区域B放入不同的两个遗漏区域集合中;若遗漏区域A和遗漏区域B的切割参数不同,那么在沿着补偿路径进行补偿时将不断地调整切割参数,这样也会极大的影响自移动设备的补偿效率,因此可以考虑将遗漏区域A和遗漏区域B放入不同的两个遗漏区域集合中,本公开中的切割参数包括切割高度、单次切割量、切割速度等。若遗漏区域A和遗漏区域B的高度大于预设高度阈值,那么在沿着补偿路径进行补偿时将可能导致打滑,这样也会极大的影响自移动设备的补偿效率,因此可以考虑将遗漏区域A和遗漏区域B放入不同的两个遗漏区域集合中。在本公开中预设高度阈值为25cm。
在一些实施例中,如图4所示,S410还可以包括如下步骤:
S411、基于第一集合确定方法确定多个第一组合区域,确定每个第一组合区域的第一补偿路径。
S412、所述自移动设备沿多条第一补偿路径执行草坪护理所需的第一总时间,将第一总时间作为第一补偿时间。
在一些实施例中,如图4所示,S420还可以包括如下步骤:
S421、基于第二集合确定方法确定多个第二组合区域,确定每个第二组合区域的第二补偿路径。
S422、所述自移动设备沿多条第二补偿路径执行草坪护理所需的第二总时间,将第二总时间作为第二补偿时间。
为进一步说明第一集合确定方法与第二集合确定方法的具体确定过程,下面集合图5来进行描述。
通过这种方式可以计算对所有的遗漏区域进行草坪护理工作的总时间,将总时间最短的集合确定方法作为最终方法。
在一些实施例中,第一集合确定方法可以理解为基于矩形在已草坪护理区域对遗漏区域进行不重复框选,确定的每一个框选区域可以被配置为一个遗漏区域集合。其中,考虑到本公开采用弓字型全覆盖路径规划算法,则前述预设图形一般配置为宽度为自移动设备单次作业宽度的整数倍的矩形。
基于上述任意一种集合确定方法,确定完成遗漏区域集合后,可以基于遗漏区域集合确定组合区域。在一些实施例中,基于遗漏区域集合确定组合区域,包括:确定所述遗漏区域集合中各遗漏区域的边界;对各遗漏区域边界进行边界拟合,得到所述组合区域。
边界拟合可以指基于遗漏区域集合,确定包含该遗漏区域集合内所有遗漏区域的形状,并将该形状作为组合区域。例如,前述第一框选区域510与第二框选区域520可以直接作为矩形的拟合结果,并作为组合区域。
具体的,沿不同方向分别所述遗漏区域集合中的各遗漏区域边界作一个外接矩形,所述外接矩形覆盖所述遗漏区域集合中的所有遗漏区域;
根据几何特征确定各外接矩形中的最小外接矩形,其中所述几何特征包括面积、边长、长宽比;
将所述最小外接矩形所覆盖的区域作为所述组合区域。
在本公开中,可以选定一个初始方向,并以每10°为一个方向,依次得到18种不同的外接矩形。
在一些实施例中,为尽量减少组合区域的面积,本公开可以采用几何特征确定一个最小外接矩形。其中,最小外接矩形是指能够包含遗漏区域集合中的所有遗漏区域的最小矩形。在一些实施例中,最小外接矩形可以基于预设方向生成。在一些实施例中,最小外接矩形可以基于任意方向生成。在基于任意方向生成时,可以生成多种外接矩形,在本公开中,将多种外接矩形的面积进行排序,将面积最小的外接矩形作为最小外接矩形。同样的,还可以将多种外接矩形的周长进行排序,将周长最小的外接矩形作为最小外接矩形。
根据作最小外接矩形的方式,得到遗漏区域集合对应的组合区域。
在一些实施例中,遗漏区域集合的最小外接矩形可以基于最小外接矩形相关算法计算。例如,可以基于旋转卡壳算法(Rotating Calipers)、最小面积矩形算法(Minimum Area Rectangle)、随机增量法(Randomized Incremental Method)或相关算法确定最小外接矩形。
在得到一个较为恰当的组合区域后,需要规划组合区域的补偿路径,该补偿路径可以覆盖遗漏区域,当自移动设备沿着补偿路径移动时,可以将组合区域中所有的遗漏区域进行补偿。
由此,基于本公开实施例提供的基于遗漏区域的路径规划方法。自移动设备可以及时识别遗漏区域并执行草坪护理任务,以提高作业的完成度。此外,补偿路径可以基于遗漏区域的聚类结果构建,从而使各个单独的遗漏区域可以在一个草坪护理任务中被处理,相对于对各个遗漏区域单独执行草坪护理任务,减少了作业时间,以提高整体作业效率。
具体的,如图7所示的补偿作业路径示意图,具体的,可以通过任意方向生成多个外接矩形,外接矩形71和外接矩形72均覆盖遗漏区域73、74,然后确定各外接矩形的面积,将面积最小的外接矩形认定为最小外接矩形,在图7中最小外接矩形为外接矩形72,当然,还可以确定各外接矩形的周长,将周长最小的外接矩形认定为最小外接矩形,本公开对此不做限定。
然而上述方式的转向次数较多,可能导致自移动设备磨损草坪。因此,在得到一个较为恰当的组合区域后,需要规划组合区域的补偿路径,该补偿路径可以覆盖遗漏区域,当自移动设备沿着补偿路径移动时,可以将组合区域中所有的遗漏区域进行补偿。
具体的,本公开中,在将外接矩形72确定为最小外接矩形后,可以确定外接矩形的长边721和短边722,在本公开中,长边721的长度大于短边722。在确定完成最小外接矩形72后需要确定对组合区域执行草坪护理任务的补偿路径,在本公开中,补偿路径引导自移动设备对组合区域执行草坪护理任务。此外,还可以设计一条移动路径,移动路径为从自移动设备当前位置到组合区域的最短直线,这样可以保证自移动设备最快速的进入组合区域。本公开着重描述补偿路径。为了让补偿路径可以更加合理,保证自移动设备沿补偿路径可以更快的执行草坪护理任务,本公开考虑到了自移动设备沿直线行走的速度大于转弯速度,为了方便说明,本公开中的补偿路径包括两部分,第一部分为若干直行路径75,若干直行路径75的间距可以固定,第二部分为转弯路径76,该转弯路径76连接两条相邻的直行路径75,本公开将与长边平行的方向作为直行路径75的方向,本公开对于转弯路径76不做限定,凡是可以将直行路径75相互连接,且可以保证短边附近的遗漏区域73或遗漏区域74可以被清除的路径均可以作为转弯路径76。
这样相比于不设计补偿路径,本公开的补偿路径的转向次数降低。由此,基于本公开实施例提供的基于遗漏区域的路径规划方法。自移动设备可以及时识别遗漏区域并执行草坪护理任务,以提高作业的完成度。此外,补偿路径可以基于遗漏区域的聚类结果构建,从而使各个单独的遗漏区域可以在一个草坪护理任务中被处理,相对于对各个遗漏区域单独执行草坪护理任务,减少了作业时间,降低对草坪的磨损程度,以提高整体作业效率。
本公开给出一种确定转弯路径的方法,包括:确定最小外接矩形72在草坪护理区域7中的位置,本公开中,最小外接矩形72可能处于草坪护理区域7的中间部位,还可以处于草坪护理区域7的边缘部分,若草坪护理区域7的外侧为自移动设备无法通行的区域,则最小外接矩形72在草坪护理区域7中的位置将会影响转弯路径76的类型,具体的,若最小外接矩形72处于草坪护理区域7的中间部位,则转弯路径可以超出最小外接矩形72,且无需担心自移动设备与已草坪护理区域7外侧的无法通行区域发生碰撞,若最小外接矩形72处于草坪护理区域7的边缘部位,则转弯路径无法超出最小外接矩形72,且需要避免自移动设备与草坪护理区域7外侧的无法通行区域发生碰撞。
通过这种方式,可以保证自移动设备沿补偿路径的工作效率最高。
然而在一些情况下,若用户对于草坪护理区域的美观要求较高时,通过上述方式将导致补偿路径与预设作业路径产生交叉,通俗来说,自移动设备沿补偿路径在草坪护理区域产生的花纹与沿预设作业路径在草坪护理区域产生的花纹不一致。
此外,还可以设计一种补偿路径,该补偿路径与上述内容相同,区别在于直行路径的方向需要与预设作业路径完全重合,即规划覆盖所述组合区域的补偿路径,包括:
确定所述最小外接矩形所覆盖的预设作业路径;将所述预设作业路径在所述最小外接矩形内的直行部分作为所述直行路径。由于本公开中假设预设作业路径为弓字型路径,因此预设作业路径必然有直行部分和转向部分,为了保证花纹完全一致,本公开可以控制补偿路径中的直行路径与预设作业路径的直行部分完全重合,此时自移动设备沿补偿路径在草坪护理区域产生的花纹与沿预设作业路径在草坪护理区域产生的花纹一致。
通过这种方式可以提升美观性。
此外,还可以设计一种补偿路径,该补偿路径与上述内容相同,区别在于直行路径的方向需要与预设作业路径部分重合,即规划覆盖所述组合区域的补偿路径,包括:
确定所述预设作业路径的直行部分的直行方向;根据所述预设作业路径的直行部分的直行方向确定所述补偿路径的直行路径的方向。由于本公开中假设预设作业路径为弓字型路径,因此预设作业路径必然有直行部分和转向部分,为了保证花纹的方向一致,本公开可以控制补偿路径中的直行路径与预设作业路径的直行部分完全重合,此时自移动设备沿补偿路径在草坪护理区域产生的花纹与沿预设作业路径在草坪护理区域产生的花纹方向一致。
示例性栅格地图
在一些实施例中,自移动设备可以基于栅格地图确定草坪护理区域中的遗漏区域和未遗漏区域。其中,栅格地图(Raster Map),也称为栅格图像或栅格数据,是一种基于像素的地图表示方法。栅格地图将地理空间划分为规则的网格单元,每个网格单元被称为像素,每个像素包含一个特定的值或属性。在本公开中,栅格单元的状态包括未草坪护理状态和已草坪护理状态,其中,未草坪护理状态表征所述工作组件未覆盖,已草坪护理状态表征所述工作组件已覆盖。
在一些实施例中,在基于栅格地图确定遗漏区域的过程中,可以先确定草坪护理区域的栅格地图。在目标草坪护理任务中,将自移动设备执行草坪护理任务时工作组件覆盖的多个栅格单元的状态由未草坪护理状态更新为已草坪护理状态,并将多个已草坪护理状态的栅格单元组成的区域作为目标草坪护理区域中的已草坪护理区域。
图6是本公开一示例性实施例提供的栅格地图示意图。
如图6所示,随自移动设备移动,自移动设备会根据自身的定位信息将工作组件覆盖的栅格单元的状态更新为已草坪护理状态。其中,在图6中,已草坪护理状态被配置为斜条纹样式,未草坪护理状态可以配置为空白样式(即像素内无图案)。
已草坪护理区域可以基于互相连通的已草坪护理状态下的栅格单元的轮廓而确定。如图6所示,图6中的已草坪护理区域由处于已草坪护理状态下的栅格单元而形成,在草坪护理区域610还包括多个处于未草坪护理状态的栅格单元。
在一些实施例中,在草坪护理区域610中,基于多个栅格单元的状态,确定多个遗漏区域。即在图6所示的草坪护理区域610中包括区域620以及区域630存在处于未草坪护理状态的栅格单元,则可以将区域620以及区域630分别配置为遗漏区域。
在一些实施例中,当遗漏区域包括多个栅格时,可以认为每个遗漏区域对应的多个栅格单元处于未草坪护理状态,且遗漏区域内的各个栅格单元相互连通。即前述区域620包括三个连续的于未草坪护理状态的栅格单元。
为了方便说明,在本公开中,将自移动设备沿补偿路径执行草坪护理任务称为补割。
在一些实施例中,在确定遗漏区域之前,控制器210可以基于栅格地图将草坪护理区域拆分为多个第一子区域,以使自移动设备依次执行各个第一子区域的草坪护理任务。在本公开中,自移动设备110所在的第一子区域为需要进行草坪护理任务的区域,控制器210控制自移动设备110执行所在的第一子区域的草坪护理任务,当自移动设备110完成当前的第一子区域的草坪护理任务后,自移动设备对当前的第一子区域中的遗漏区域进行补割,当完成补割后,控制器210控制自移动设备110对下一个第一子区域执行草坪护理任务,然后再对下一第一子区域中的遗漏区域进行补割,重复上述步骤,直至完成整个草坪护理区域的草坪护理任务和对草坪护理区域中组合区域的补割。上述方法,通过对草坪护理区域分区,对每个第一子区域依次执行草坪护理任务和对组合区域的补割,可以减少对草坪护理区域中组合区域执行补割时距离跨度大的问题。
在一个实施例中,前述第一子区域划分还可以基于孤岛区域进行。具体地,可以沿着孤岛区域的两条外切线,将草坪护理区域划分为多个第一子区域。其中,孤岛区域的两条外切线与所述草坪护理区域中部分预设作业路径的方向平行,所述孤岛区域被配置为禁止所述自移动设备进入。
孤岛区域可以是栅格地图中用于反映障碍物的栅格单元集合。孤岛区域对应的栅格单元被配置为黑色纯色完全填充。在栅格地图中,还可以将栅格单元进一步分为反映障碍物的前景单元以及反映非障碍物的背景单元。可以通过连通组件分析等算法,可以将前景像素组成的连通区域提取出来,这些连通区域就是孤岛区域,以代表了障碍物的位置和形状。
外切线可以指与给定图形(如圆、椭圆、多边形等)相切且在切点处与图形曲线相切的直线。外切线与图形的曲线仅在一个点处相切,不穿过图形。孤岛区域的外切线可以指沿孤岛区域边缘延伸的各个线段。
上述分区过程也可以呈现在图6中。如图6所示,草坪护理区域可以包括孤岛区域640,沿图中垂直方向可以作孤岛区域640的两个外切线AA与BB,从而形成三个第一子区域:第一子区域650、第一子区域660,以及由AA与BB围成的第一子区域670,其中,第一子区域650位于AA的左侧,第二子区域660位于BB的右侧。
在一些实施例中,第一子区域还可以基于栅格单元的高度差划分。即可以先基于草坪护理区域的栅格地图,确定每两个栅格之间的高度差。当高度差大于预设差值时,将两个格栅划分到不同的第一子区域中。在一些实施例中,该高度差可以自移动设备在高度方向上的调整能力确定。
在一些实施例中,还可以对第一子区域进行多级分区,从而保证遗漏区域的准确识别。其中,可以确定草坪护理区域中的多个第一子区域,按照预设路径宽度将第一子区域分为至少两个第二子区域,控制所述自移动设备沿预设作业路径在所述第二子区域内移动。在确定完成第二子区域后,再确定第二子区域中的遗漏区域,例如,当完成当前的一个第二子区域的草坪护理任务时,确定第二子区域中的遗漏区域,以避免补偿过程的运动路径长度过长。
在一些实施例中,第二子区域可以基于预设路径宽度确定。例如,在第一子区域的宽度等于预设路径宽度的整数倍时,将第一子区域分为多个第二子区域,每个第二子区域的区域宽度等于预设路径宽度。再例如,在第一子区域的区域宽度不等于预设路径宽度的整数倍时,将第一子区域分为至少一个区域宽度等于预设路径宽度的第二子区域,以及一个区域宽度小于预设路径宽度的第二子区域。其中,预设路径宽度可以是自移动设备的单次作业宽度的整数倍。
应理解,本文中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表示前后关联对象是一种“或”的关系。
应理解,在本公开的各种实施例中,上述各过程的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本公开实施例的实施过程构成任何限定。
在本公开所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本公开方案的目的。
另外,在本公开各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
在上述实施例中,可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时,可以全部或部分地以计算机程序产品的形式实现。计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时,全部或部分地产生按照本公开实施例的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、网络设备、用户设备、核心网设备、操作维护管理(operationadministration and maintenance,OAM)或者其他可编程装置。计算机指令可以存储在计算机可读存储介质中,或者从一个计算机可读存储介质向另一个计算机可读存储介质传输,例如,计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber Line,DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。计算机可读存储介质可以是计算机能够读取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。可用介质可以是磁性介质,(例如,软盘、硬盘、磁带)、光介质(例如,数字通用光盘(digital video disc,DVD))或者半导体介质(例如,固态硬盘(solid state disk,SSD))等。该计算机可读存储介质可以是易失性或非易失性存储介质,或可包括易失性和非易失性两种类型的存储介质。
以上,仅为本公开的具体实施方式,但本公开的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本公开揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本公开的保护范围之内。因此,本公开的保护范围应以权利要求的保护范围为准。

Claims (20)

  1. 一种路径规划方法,其特征在于,应用于具有工作组件的自移动设备,所述自移动设备被配置为沿预设作业路径移动,并在移动过程中通过所述工作组件执行草坪护理任务,所述方法包括:
    确定遗漏区域,所述遗漏区域由所述自移动设备偏移所述预设作业路径和/或所述工作组件未覆盖所形成;
    按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域;
    规划覆盖所述组合区域的补偿路径,所述补偿路径被配置为引导所述自移动设备在所述组合区域中连续移动,所述工作组件在移动过程中执行所述草坪护理任务。
  2. 根据权利要求1所述的方法,其特征在于,所述按照空间分布特征将至少两个所述遗漏区域合并为一个组合区域,包括:
    确定所述遗漏区域之间的空间分布特征,其中所述空间分布特征包括空间距离、空间密度、空间方向中的一种或多种;
    选择空间分布特征满足预设空间分布条件的至少两个所述遗漏区域,作为一个遗漏区域集合;
    根据所述遗漏区域集合得到所述组合区域,其中所述组合区域覆盖所述至少两个遗漏区域。
  3. 根据权利要求以上任一所述的方法,其特征在于,所述空间分布条件包括:
    两个遗漏区域之间的空间距离小于预设距离阈值。
  4. 根据权利要求以上任一所述的方法,其特征在于,所述根据所述遗漏区域集合得到所述组合区域,包括:
    确定所述遗漏区域集合中各遗漏区域的边界;
    对各遗漏区域边界进行边界拟合,得到所述组合区域。
  5. 根据权利要求以上任一所述的方法,其特征在于,所述对各遗漏区域边界进行边界拟合,得到所述组合区域,包括:
    沿不同方向分别所述遗漏区域集合中的各遗漏区域边界作一个外接矩形,所述外接矩形覆盖所述遗漏区域集合中的所有遗漏区域;
    根据几何特征确定各外接矩形中的最小外接矩形,其中所述几何特征包括面积、边长、长宽比;
    将所述最小外接矩形所覆盖的区域作为所述组合区域。
  6. 根据权利要求以上任一所述的方法,其特征在于,所述根据几何特征确定各外接矩形中的最小外接矩形,包括:
    将各外接矩形中面积最小的外接矩形作为所述最小外接矩形;或,
    将各外接矩形中边长最小的外接矩形作为所述最小外接矩形;或,
    将各外接矩形中长宽比最大的外接矩形作为所述最小外接矩形。
  7. 根据权利要求以上任一所述的方法,其特征在于,所述补偿路径包括若干直行路径和若干转弯路径,相邻的直行路径通过转弯路径相连。
  8. 根据权利要求以上任一所述的方法,其特征在于,所述规划覆盖所述组合区域的补偿路径,包括:
    确定所述最小外接矩形的长边和短边,其中所述长边的长度大于短边的长度;
    根据所述长边确定所述直行路径。
  9. 根据权利要求以上任一所述的方法,其特征在于,所述根据所述长边确定所述直行路径,包括:
    确定所述长边的方向;
    根据所述长边的方向确定所述直行路径的方向。
  10. 根据权利要求以上任一所述的方法,其特征在于,所述规划覆盖所述组合区域的补偿路径,包括:
    确定所述最小外接矩形所覆盖的预设作业路径;
    将所述预设作业路径在所述最小外接矩形内的直行部分作为所述直行路径。
  11. 根据权利要求以上任一所述的方法,其特征在于,所述规划覆盖所述组合区域的补偿路径,包括:
    确定所述预设作业路径的直行部分的直行方向;
    根据所述预设作业路径的直行部分的直行方向确定所述补偿路径的直行路径的方向。
  12. 根据权利要求以上任一所述的方法,其特征在于,所述预设作业路径引导所述自移动设备在草坪护理区域内移动;
    还包括:确定所述外接矩形在所述草坪护理区域中的位置;
    根据所述位置确定转弯路径的类型。
  13. 根据权利要求以上任一所述的方法,其特征在于,还包括:
    选择不同的空间分布特征,得到至少两个遗漏区域集合;
    根据至少两个遗漏区域集合得到至少两个组合区域;
    规划各组合区域的补偿路径,并计算各所述补偿路径对应的补偿时间;
    根据所述补偿时间确定所述空间分布特征。
  14. 根据权利要求以上任一所述的方法,其特征在于,所述根据所述补偿时间确定所述空间分布特征,包括:
    对各补偿时间进行排序,根据排序结果确定所述空间分布特征。
  15. 根据权利要求以上任一所述的方法,其特征在于,所述选择不同的空间分布特征,得到至少两个遗漏区域集合,包括:
    判断至少两个所述遗漏区域之间是否满足阻隔条件;
    当满足所述阻隔条件时,将至少两个所述遗漏区域划分到不同的遗漏区域集合。
  16. 根据权利要求以上任一所述的方法,其特征在于,所述阻隔条件包括以下条件中的一个或多个:
    所述遗漏区域间的高度差大于预设高度阈值;
    所述遗漏区域间存在不可穿越障碍物;
    所述遗漏区域间的切割参数不同。
  17. 根据权利要求以上任一所述的方法,其特征在于,还包括:
    确定所述遗漏区域的面积;
    忽略面积小于面积阈值的遗漏区域。
  18. 一种自移动设备,其特征在于,包括:
    工作组件,被配置为执行草坪护理任务;
    移动组件,被配置为带动所述工作组件运动,以使所述工作组件在移动中执行所述草坪护理任务;以及
    处理器,被配置为与所述移动组件以及所述工作组件连接,以实现上述权利要求1至16中任意一项所述的方法。
  19. 根据权利要求18所述的自移动设备,其特征在于,
    所述工作组件包括切割组件;
    所述处理器,被配置为根据草坪护理区域的地图,规划所述自移动设备在所述草坪护理区域中的作业路径,并向所述切割组件和所述移动组件输出第一指令,以控制所述自移动设备在所述草坪护理区域中沿所述作业路径进行自主遍历和一步到位的割草,其中,一步到位的割草通过在割草时同步地碎草或集草实现。
  20. 一种计算机可读存储介质,其特征在于,所述存储介质存储有计算机程序,所述计算机程序在被处理器执行时,使所述处理器执行权利要求1-17中任意一项所述的方法。
PCT/CN2025/086261 2024-03-29 2025-03-31 一种路径规划方法、自移动设备及存储介质 Pending WO2025201559A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410382121 2024-03-29
CN202410382121.0 2024-03-29

Publications (1)

Publication Number Publication Date
WO2025201559A1 true WO2025201559A1 (zh) 2025-10-02

Family

ID=97216318

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/086261 Pending WO2025201559A1 (zh) 2024-03-29 2025-03-31 一种路径规划方法、自移动设备及存储介质

Country Status (1)

Country Link
WO (1) WO2025201559A1 (zh)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107843262A (zh) * 2017-10-30 2018-03-27 洛阳中科龙网创新科技有限公司 一种农用机械全覆盖运动路径规划的方法
CN111643011A (zh) * 2020-05-26 2020-09-11 深圳市杉川机器人有限公司 清洁机器人控制方法、装置、清洁机器人和存储介质
CN112168066A (zh) * 2020-09-30 2021-01-05 深圳市银星智能科技股份有限公司 清洁机器人的控制方法、装置、清洁机器人及存储介质
CN112567958A (zh) * 2019-09-12 2021-03-30 南京德朔实业有限公司 自行走割草系统及其漏割区域的补充作业的方法
CN113219985A (zh) * 2021-05-27 2021-08-06 九天创新(广东)智能科技有限公司 一种清扫车的路径规划方法、装置、清扫车
US20220091613A1 (en) * 2020-09-18 2022-03-24 Scythe Robotics, Inc. Coverage planner
CN114815814A (zh) * 2022-03-31 2022-07-29 深圳市正浩创新科技股份有限公司 自移动设备的作业方法、计算机设备及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107843262A (zh) * 2017-10-30 2018-03-27 洛阳中科龙网创新科技有限公司 一种农用机械全覆盖运动路径规划的方法
CN112567958A (zh) * 2019-09-12 2021-03-30 南京德朔实业有限公司 自行走割草系统及其漏割区域的补充作业的方法
CN111643011A (zh) * 2020-05-26 2020-09-11 深圳市杉川机器人有限公司 清洁机器人控制方法、装置、清洁机器人和存储介质
US20220091613A1 (en) * 2020-09-18 2022-03-24 Scythe Robotics, Inc. Coverage planner
CN112168066A (zh) * 2020-09-30 2021-01-05 深圳市银星智能科技股份有限公司 清洁机器人的控制方法、装置、清洁机器人及存储介质
CN113219985A (zh) * 2021-05-27 2021-08-06 九天创新(广东)智能科技有限公司 一种清扫车的路径规划方法、装置、清扫车
CN114815814A (zh) * 2022-03-31 2022-07-29 深圳市正浩创新科技股份有限公司 自移动设备的作业方法、计算机设备及存储介质

Similar Documents

Publication Publication Date Title
US11297755B2 (en) Method for controlling a soil working means based on image processing and related system
WO2021139683A1 (zh) 自移动设备
CN112486157B (zh) 自动工作系统及其转向方法、自移动设备
US12171159B1 (en) Autonomous vehicle navigation
CN114342640A (zh) 数据处理方法、自动园艺设备及计算机程序产品
Cuevas-Velasquez et al. Real-time stereo visual servoing for rose pruning with robotic arm
CN113156929A (zh) 自移动设备
US12572898B2 (en) Site maintenance utilizing autonomous vehicles
WO2021139397A1 (zh) 自移动设备的控制方法
WO2021031441A1 (zh) 路径规划方法、系统,机器人及可读存储介质
CN118444687B (zh) 割草机器人的自学习避障控制方法
WO2024045357A1 (zh) 自移动设备的控制方法、装置和自移动设备
CN108490953A (zh) 一种基于激光雷达和iBeacon的割草机装置及方法
US20250255213A1 (en) Autonomous working apparatus and system, and working method
CN118362125B (zh) 割草机导航方法、装置、设备及存储介质
US20260036999A1 (en) Control method and control apparatus
US20260123572A1 (en) Object Classification Based Dynamic Machine Positioning
CN121311843A (zh) 控制方法、装置、存储介质及电子设备
WO2023175580A1 (en) Working method for controlling an unmanned autonomous vehicle based on a soil type
US20260104705A1 (en) Robotic lawn mower
US20260123573A1 (en) Mowing Path Planning Including Skipped Passes
WO2025140341A1 (zh) 割草的控制方法、装置、割草机及存储介质
WO2025247384A1 (zh) 自主工作系统的控制方法以及自主工作系统
WO2025246750A1 (zh) 一种割草路径生成方法及相关装置
CN207867343U (zh) 一种智能家居割草机器人

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: 25776934

Country of ref document: EP

Kind code of ref document: A1