Dynamic full-coverage path planning method and device, cleaning device, and storage medium

Abstract

The application discloses a dynamic full-coverage path planning method and device, and the method comprises the following steps: dynamically updating a coverage state map corresponding to a target area according to obtained environment change information, wherein in the coverage state map, a cleaned area in the target area is marked as a covered area, and an uncleaned area is marked as an uncovered area; when a preset trigger condition occurs, selecting an area to be path planned from the uncovered area according to the coverage state map, and performing full-coverage path planning on the selected area to be path planned. According to the scheme, the cleaning path currently relied on can be dynamically adjusted according to the environment change, the change of the environment can be better adapted, the flexibility of the cleaning task execution is improved, the cleaning habit of people is better met, and the cleaning efficiency is ensured.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a dynamic full-coverage path planning method, a dynamic full-coverage path planning device, as well as a cleaning device and a storage medium. Background Technology

[0002] In recent years, with the development of autonomous driving technology in the cleaning field, we have increasingly seen driverless vehicles freely performing cleaning tasks in large squares, supermarkets, and underground parking garages, greatly reducing the burden on sanitation workers. For this cleaning method, achieving full coverage in a given scenario is a crucial technical challenge that needs to be overcome. In particular, in real-world scenarios, environmental conditions are often dynamically changing, and traditional full-coverage path planning methods often use a one-time generated cleaning path. When facing complex and dynamic environments, a one-time generated cleaning path often leads to missed areas, and may even render the entire cleaning path ineffective, making the cleaning task impossible. Summary of the Invention

[0003] This invention provides a dynamic full-coverage path planning scheme to solve the problem that existing technologies often use one-time generated cleaning paths that cannot adapt to environmental changes, resulting in unusable cleaning paths or low cleaning efficiency.

[0004] In a first aspect, embodiments of the present invention provide a dynamic full-coverage path planning method, which includes...

[0005] The coverage status map corresponding to the target area is dynamically updated based on the acquired environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas, and uncleaned areas are marked as uncovered areas.

[0006] When the preset triggering conditions occur, a region to be planned for the route is selected from the uncovered regions based on the coverage status map, and full coverage route planning is performed on the selected region to be planned for the route.

[0007] In a second aspect, embodiments of the present invention provide a dynamic full-coverage path planning device, comprising a memory for storing executable instructions; and

[0008] A processor for executing executable instructions stored in memory, which, when executed by the processor, cause the processor to perform the dynamic full-coverage path planning method as described in the first aspect.

[0009] Thirdly, embodiments of the present invention provide a dynamic full-coverage path planning device, comprising:

[0010] The status update module is used to dynamically update the coverage status map corresponding to the target area based on the acquired environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas, and uncleaned areas are marked as uncovered areas.

[0011] The dynamic scheduling module is used to detect trigger conditions in real time, and when the preset trigger conditions occur, it selects the area to be planned in the uncovered area according to the coverage status map and calls the path planning module to perform full coverage path planning.

[0012] The route planning module is used to perform full-coverage route planning for the selected area to be planned.

[0013] Fourthly, embodiments of the present invention provide a cleaning device, which includes:

[0014] fuselage; and

[0015] The aforementioned dynamic full-coverage path planning device is mounted on the machine body.

[0016] Fifthly, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0017] In a sixth aspect, the present invention provides a computer program product containing instructions that, when the computer program product is run on a computer, cause the computer to execute the above-described dynamic full-coverage path planning method.

[0018] The beneficial effects of the embodiments of the present invention are as follows: The method provided by the embodiments of the present invention can dynamically maintain the coverage status map according to the environmental change information, and dynamically trigger full coverage path planning according to the updated coverage status map, so that the current path planning can be dynamically adjusted according to the environmental changes, so as to better adapt to the environmental changes, improve the flexibility of cleaning task execution, and better conform to people's cleaning habits, thus ensuring cleaning efficiency. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a dynamic full-coverage path planning method according to an embodiment of the present invention;

[0021] Figure 2 A flowchart of another embodiment of the dynamic full-coverage path planning method of the present invention;

[0022] Figure 3 This is a flowchart of a full-coverage path planning method according to an embodiment of the present invention;

[0023] Figure 4 The diagram illustrates the effect of the outermost path before and after optimization.

[0024] Figure 5 The diagram schematically illustrates the effect of simulating the outer contour coverage of the cleaning equipment on the optimized outermost path;

[0025] Figure 6 The diagram illustrates the effects of the outer and inner contours before and after optimization.

[0026] Figure 7 This is a flowchart of the full-coverage path planning method according to Embodiment 1 of the present invention;

[0027] Figure 8 The diagram illustrates the effect of the transformation on the target area map before and after the transformation process. Figure 8 A shows a raster map with outlined boundaries. Figure 8 B shows the pair Figure 8 The distance transformation map obtained by performing distance transformation on the raster map shown in A;

[0028] Figure 9 A schematic diagram illustrating the resulting contour path formed by the method according to an embodiment of the present invention is shown, wherein, Figure 9 A is a cluster of contour lines generated based on the distances in the distance transformation map. Figure 9 B showed the pair Figure 9 The drivable contour path is obtained by processing the contour line cluster of A.

[0029] Figure 10 This is a method flow for connecting generated contour line paths according to one embodiment of the present invention;

[0030] Figure 11 This is a method flow for performing connectivity processing on the traversed current road segment according to one embodiment of the present invention.

[0031] Figure 12 The diagram illustrates the connectivity processing effect after performing one round of traversal.

[0032] Figure 13 The diagram illustrates the effect of a complete cleaning path obtained after connection.

[0033] Figure 14 This is a flowchart of a method for full-coverage path planning processing according to another embodiment of the present invention;

[0034] Figure 15 The diagram illustrates the effect of identifying contour lines with smaller areas as those that do not meet preset requirements and need to be deleted. Figure 15 This demonstration uses a raster map as an example.

[0035] Figure 16 The illustration shows the effect of filling an area with an arch-shaped pattern after removing contour lines. Figure 16 This demonstration uses a raster map as an example.

[0036] Figure 17 This is a flowchart of a method for full-coverage path planning processing according to another embodiment of the present invention;

[0037] Figure 18 The diagram schematically illustrates the relationship between the Vino skeleton line and the leaked coverage area. Figure 18 A shows the relationship between the uncovered area obtained after simulating the cover motion and the Vinno skeleton line. Figure 18 B shows the correspondence between the generated Vinno skeleton lines and the missing coverage areas on the map;

[0038] Figure 19 The diagram illustrates the coverage effect of the optimized Vino skeleton line on the map;

[0039] Figure 20 The image shows the effect of small, missed areas that were not swept during the actual cleaning process;

[0040] Figure 21 The diagram illustrates the effect of another complete cleaning path obtained after connection.

[0041] Figure 22 This is a schematic diagram of a dynamic full-coverage path planning device according to an embodiment of the present invention;

[0042] Figure 23 This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention;

[0043] Figure 24 This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention.

[0044] Figure 25 This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention.

[0045] Figure 26 This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention.

[0046] Figure 27This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention.

[0047] Figure 28 This is a schematic diagram of a dynamic full-coverage path planning device according to another embodiment of the present invention.

[0048] Figure 29 This is a schematic block diagram of a cleaning device according to an embodiment of the present invention;

[0049] Figure 30 This is a schematic diagram of the structure of an embodiment of the electronic device of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0052] This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, elements, data structures, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0053] In this invention, terms such as "module," "device," and "system" refer to relevant entities applied to a computer, such as hardware, combinations of hardware and software, software, or software in execution. More specifically, for example, an element can be, but is not limited to, a process running on a processor, a processor, an object, an executable element, an execution thread, a program, and / or a computer. Furthermore, an application program or script running on a server, and the server itself, can also be an element. One or more elements may be in an execution process and / or thread, and elements may be localized on a single computer and / or distributed across two or more computers, and may be run on various computer-readable media. Elements can also communicate via local and / or remote processes based on signals having one or more data packets, for example, signals from data interacting with another element in a local system, a distributed system, and / or interacting with other systems via signals over a network of the Internet.

[0054] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising" or "including" include not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0055] The dynamic full-coverage path planning method in this invention can be applied to dynamic full-coverage path planning devices, enabling users to dynamically execute path planning using these devices, thereby improving the flexibility and environmental adaptability of cleaning tasks. These dynamic full-coverage path planning devices include, but are not limited to, smartphones, tablets, personal PCs, computers, and cloud servers. Specifically, the dynamic full-coverage path planning method in this invention can also be applied to intelligent devices with cleaning functions (or mobile platforms providing automatic cleaning functions), such as unmanned cleaning vehicles, robotic vacuum cleaners, or autonomous cleaning vehicles; this invention does not limit its application to these applications.

[0056] Figure 1The illustration schematically depicts a dynamic full-coverage path planning method according to an embodiment of the present invention. This method is applicable to dynamic full-coverage path planning devices such as smartphones, personal computers, and cloud servers, enabling these devices to perform dynamic cleaning path analysis based on received environmental change information and plan cleaning paths that are dynamically updated according to environmental changes. It is also applicable to cleaning equipment performing cleaning tasks, such as unmanned sanitation vehicles, unmanned cleaning vehicles, unmanned sweeping vehicles, sweeping robots, and unmanned floor scrubbers, to dynamically execute path planning and cleaning tasks based on environmental change information. Figure 1 As shown, the method of this embodiment of the invention includes:

[0057] Step S10: Dynamically update the coverage status map corresponding to the target area based on the obtained environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas, and uncleaned areas are marked as uncovered areas.

[0058] Step S11: When the preset triggering condition occurs, select the area to be planned in the uncovered area according to the coverage status map, and perform full coverage path planning on the selected area to be planned.

[0059] This invention maintains a dynamically updated coverage status map to dynamically perform full-coverage path planning based on environmental changes, thereby dynamically generating and replanning the current cleaning path. This ensures the effectiveness of the real-time cleaning route even in complex dynamic environments, improves the applicability and scope of the full-coverage route planning algorithm, and avoids problems such as low efficiency, cleaning failures, and incomplete cleaning caused by the failure of a one-time generated cleaning path. It is more in line with human cleaning processing ideas, more intelligent, and more feasible.

[0060] Specifically, in step S10, environmental change information refers to real-time information indicating current environmental changes, which can be obtained from other modules or other devices or passively received. For example, it refers to a real-time environmental map, which can be obtained from a camera device that captures the current environmental map in real time. In other embodiments, it can also refer to the real-time driving trajectory of the equipment performing the current cleaning task or the driving trajectory within a specified time period, such as the real-time driving trajectory of a robotic vacuum cleaner or an unmanned sweeping vehicle, or the driving trajectory within a specified time period. In this case, the environmental change information can be obtained from the navigation module. In other embodiments, depending on the requirements, the environmental change information can also be input by the user through a user interface provided by a full-coverage path planning device or cleaning equipment to obtain the environmental change information. In this embodiment of the invention, the coverage status map is a map corresponding to the target cleaning area of ​​the current cleaning task, which marks covered areas, uncovered areas, and missed covered areas. This map can be a raster map, a real-time environmental map, or a distance-transformed map obtained by performing distance transformation on a raster map. This embodiment of the invention does not impose any restrictions on this, as long as it is possible to maintain a dynamically marked coverage status map so that dynamic path planning can be performed based on the dynamically changing coverage status marks on the coverage status map. Preferably, the coverage status map in this embodiment of the invention refers to a map obtained by marking the coverage status on the raster map corresponding to the target cleaning area. The coverage status marking specifically includes marking cleaned areas in the target area as covered areas and uncleaned areas as uncovered areas.

[0061] In step S11, the triggering condition is a preset condition used to trigger the determination of the area to be planned and the full-coverage path planning. This condition can be set according to requirements or as a system default, allowing dynamic path planning to be performed based on preset scenarios or user-defined needs, combined with a dynamic coverage map. The full-coverage path planning is performed on the selected area to be planned. After full-coverage path planning, the cleaning path used for the cleaning task can be dynamically replaced with a newly planned path that better reflects the actual environmental conditions, thus achieving dynamic updates to the cleaning path used for the task and enabling flexible execution of the cleaning task.

[0062] In practical applications, there are many situations where the cleaning path based on the current driving path needs to be dynamically updated. A typical example is when the state of obstacles changes. For instance, if a new obstacle suddenly appears in the target cleaning area, the previously planned offline cleaning path will become invalid due to the presence of the new obstacle; or if an existing obstacle in the target cleaning area is suddenly removed, if the cleaning task is performed according to the previously planned offline cleaning path, the removed area will not be cleaned, resulting in incomplete or unclean cleaning. Therefore, embodiments of the present invention can flexibly set trigger conditions according to needs. That is, based on needs and actual conditions, the situations and areas requiring full-coverage path planning can be determined, and full-coverage path planning can be performed accordingly based on preset trigger conditions and the selected area to be planned.

[0063] The following will use two different scenarios as specific examples to illustrate... Figure 1 The specific implementation of the embodiments shown will be described in detail. Of course, it will be easy for those skilled in the art to understand that the two specific embodiments listed below are not the only situations that need to be updated and changed in reality. Those skilled in the art can flexibly apply the concept of the present invention to other specific situations.

[0064] Taking the scenario where a new obstacle suddenly appears in the target cleaning area, potentially rendering the originally planned offline cleaning path ineffective, as an example, the environmental change information can be implemented as a real-time environmental map. In step S10, dynamically updating the coverage status map corresponding to the target area based on the environmental change information can specifically include: performing obstacle detection based on the real-time environmental map to determine if a new first obstacle exists, and adding the new first obstacle to the corresponding position on the coverage status map based on the detected real-time obstacle information. The obstacle detection based on the real-time environmental map can be implemented using existing technology, and will not be elaborated upon here. Furthermore, in scenarios where the coverage status map is a grid map, this embodiment of the invention will also convert the real-time environmental map into a grid map according to existing technology, and use the converted grid map to mark the new first obstacle and update the coverage status, thereby generating an updated coverage status map. Correspondingly, determining whether a new obstacle exists based on the acquired environmental change information can be based on comparing the obstacle detection results with the obstacle positions in the original, unupdated coverage status map.

[0065] In practical applications, the impact of newly added obstacles on the planned cleaning path and cleaning task usually occurs only when the newly added obstacle is located in an uncleaned area. Therefore, in the preferred embodiment, when a new first obstacle is added, the newly added first obstacle is added to the corresponding position in the coverage status map. This can also be implemented by further determining the position of the newly added first obstacle in the target area, and adding the newly added first obstacle to the corresponding position in the uncleaned area of ​​the coverage status map only when the newly added first obstacle is located in an uncovered area.

[0066] In other embodiments, the environmental change information may not be limited to the real-time environment map, but may directly be the information of the newly added first obstacle. In this case, in step S10, it is not necessary to perform obstacle detection processing on the real-time environment map. Instead, the newly added first obstacle can be directly added to the corresponding position in the coverage state map based on the environmental change information including the newly added first obstacle, or when it is determined that the newly added first obstacle is located in an uncovered area, the newly added first obstacle can be added to the corresponding position in the uncovered area of ​​the coverage state map.

[0067] In the case of a newly added first obstacle, the preset trigger conditions can be flexibly set according to the user's path planning needs. For example, in one embodiment of the present invention, the preset trigger conditions may include the presence of a newly added first obstacle in the acquired environmental change information, and the newly added first obstacle being located in an uncovered area. Correspondingly, the selected path planning area can be an uncleaned area of ​​the target area containing the first obstacle. In other embodiments, taking global path planning as an example, since global path planning involves overall path planning for the entire uncleaned area of ​​the target area, the addition of a first obstacle in an uncovered area of ​​the target area creates a need for path planning for the target area. Therefore, for example, the preset trigger conditions can be set to include the presence of a newly added first obstacle in the acquired environmental change information, and the newly added first obstacle being located in an uncovered area. In this case, the path planning area selected from the uncovered area according to the coverage status map can specifically be an uncleaned area of ​​the target area. In the example of local path planning, path planning is generally performed only for the current clean area. Therefore, a dynamic path planning requirement for the current clean area will only be generated if the newly added first obstacle falls into the unclean area of ​​the current clean area. Thus, for example, in this case, the preset triggering conditions include the presence of a newly added first obstacle in the obtained environmental change information and the newly added first obstacle being located in the unclean area of ​​the current clean sub-region. Accordingly, the area to be planned for path planning in the uncovered area according to the coverage status map is the unclean area of ​​the current clean sub-region. Of course, it is not difficult to understand that in other scenarios, there may be other needs and possibilities for triggering dynamic path planning. For example, when there are multiple unconnected passable areas formed by the division of impassable areas in the target area, in general, path planning and cleaning tasks are performed separately for each passable area (in this embodiment of the invention, each passable area is referred to as a clean sub-area in the target area). In this case, for local path planning, it is generally only concerned with the dynamic changes of the clean sub-area currently performing the cleaning task. The need for re-path planning will only arise when the newly added first obstacle is located in the uncleaned area of ​​the current clean sub-area. At this time, the preset triggering condition can also be set to include the existence of a newly added first obstacle in the obtained environmental change information and the newly added first obstacle being located in the uncleaned area of ​​the current clean sub-area. Correspondingly, at this time, the area to be planned in the uncovered area according to the coverage status map is the uncleaned area of ​​the current clean sub-area.In scenarios where cloud servers are used for full-coverage path planning, for both global and local path planning, if there are multiple cleaning sub-areas in the target area, and the newly added first obstacle is located in another cleaning sub-area that has not yet been cleaned, a re-path planning requirement will also arise to update the cleaning path stored in the cloud or sent to the cleaning equipment. In this case, the preset trigger condition can be set to include the presence of a newly added first obstacle in the acquired environmental change information and the newly added first obstacle being located in an uncovered area. Accordingly, in this case, the area to be planned in the uncovered area selected according to the coverage status map is the uncleaned area of ​​the cleaning sub-area where the first obstacle is located.

[0068] Taking the scenario where an obstacle is suddenly removed from the target cleaning area, reducing the cleaning efficiency of the originally planned offline cleaning path due to the disappearance of the original obstacle, as an example, step S10 further includes updating the coverage status map based on real-time obstacle information in the real-time environmental map. When it is determined that a second obstacle has been removed from the real-time environmental map, the area occupied by the removed second obstacle is updated to an uncovered area. Specifically, in step S10, this embodiment of the invention detects obstacles in the acquired real-time environmental map to determine real-time obstacle information. This real-time obstacle information is then compared with the obstacle information in the original, unupdated coverage status map, such as through obstacle grid status, position, or coordinates, to determine whether any obstacles have been removed from the currently acquired real-time environmental map. If so, the area where the removed second obstacle is located is marked as an uncovered area, thereby updating the coverage status map.

[0069] In other embodiments, the environmental change information may not be limited to a real-time environmental map, but may directly be information about the removed second obstacle. In this case, in step S10, it is not necessary to perform obstacle detection processing on the real-time environmental map, but can directly update the area occupied by the removed second obstacle to an uncovered area based on the environmental change information including the removed second obstacle.

[0070] In cases where a second obstacle has been removed, the preset trigger condition can be flexibly set according to the user's path planning needs. For example, in one embodiment of the invention, the preset trigger condition can be set to include the presence of a removed second obstacle in the acquired environmental change information; correspondingly, the selected path planning area can be an uncleaned area within the target area containing the second obstacle. In other embodiments of the invention, taking global path planning as an example, since global path planning involves overall path planning for the uncleaned areas of the entire target area, when a second obstacle is removed from the target area, the removed second obstacle area becomes a new uncovered area, thus creating a need for path planning for the target area. Therefore, for example, the preset trigger condition can be set to include the presence of a removed second obstacle in the acquired environmental change information. In this case, the path planning area selected from the uncovered area according to the coverage status map can specifically be an uncleaned area of ​​the target area. In the example of local path planning, path planning is generally only performed for the current clean area. In this case, there are two different situations: the removed second obstacle is located in a cleaned area of ​​the current clean area, or the removed second obstacle is located in an uncleaned area of ​​the current clean area. In both cases, for real-time cleaning tasks, a dynamic path planning requirement for the current cleaning area only arises if the removed second obstacle is located in an uncleaned area of ​​the current cleaning area. Therefore, for example, in this case, the preset triggering conditions include the presence of a removed second obstacle in the acquired environmental change information and the second obstacle being located in an uncleaned area of ​​the current cleaning sub-area. Correspondingly, the path planning area selected from the uncovered area according to the coverage status map is the uncleaned area of ​​the current cleaning sub-area. Of course, it is easy to understand that in other scenarios, there may be other needs and possibilities for triggering dynamic path planning. For example, in a scenario using a cloud server for full-coverage path planning, for global or local path planning, when the removed second obstacle is located in a cleaned area of ​​the current cleaning area, a new cleaning sub-area to be cleaned is formed at the location of the removed second obstacle. In this case, there may also be a need for pre-planning a path for this newly formed cleaning sub-area. In this case, the preset triggering conditions include the presence of a removed second obstacle in the acquired environmental change information, and correspondingly, the selected path planning area is the uncleaned area of ​​the cleaning sub-area where the second obstacle is located.In other embodiments, such as in the scenario of using a cloud server for full-coverage path planning, for global or local path planning, when there are multiple cleaning sub-areas in the target area, if the removed second obstacle is located in another cleaning sub-area that has not yet been cleaned, there may also be a need for pre-planned re-path planning to update the cleaning path stored in the cloud or sent to the cleaning equipment. In this case, the preset trigger condition can be set to include the presence of the removed second obstacle in the acquired environmental change information. Accordingly, in this case, the area to be planned in the uncovered area according to the coverage status map is the uncleaned area of ​​the cleaning sub-area where the second obstacle is located.

[0071] In practical applications, the execution of cleaning tasks may follow preset rules, such as being executed in a specified order (i.e., executing one cleaning sub-area in a preset order after completing another); or being executed according to a specified cycle or time (i.e., executing the cleaning task of the corresponding cleaning sub-area when the preset time or cycle arrives). Therefore, according to actual application needs, the preset triggering conditions in this embodiment of the invention can also be set to include the completion of the current cleaning sub-area and / or the arrival of the preset execution cycle and / or the arrival of the preset execution time. In this case, the selected path planning area can be an uncovered area in the coverage status map that conforms to the preset cleaning rules. The preset cleaning rules can be set by system default or user-defined. For example, the preset cleaning rules can be, for instance, executing cleaning tasks on cleaning sub-areas in a specified order. In this case, the selected path planning area can be the uncovered area corresponding to the next cleaning sub-area in the specified order. In other implementations, the preset cleaning rule can also be to perform cleaning tasks based on the proximity of the cleaning sub-regions. In this case, the selected path planning area can be the uncovered area corresponding to the next cleaning sub-region that is closest to the currently completed cleaning sub-region. In yet another embodiment, the preset cleaning rule can also be to perform cleaning tasks based on the size of the cleaning sub-regions in descending order of area. In this case, the selected path planning area can be the uncovered area corresponding to the next cleaning sub-region that is closest in area to the currently completed cleaning sub-region.

[0072] For example, considering that the new cleaning sub-area resulting from the removal of obstacles may be inconsistent with the current driving direction or trajectory, making it impossible to immediately and effectively move to that area for cleaning, this embodiment of the invention preferably performs path planning and cleaning tasks for the cleaning sub-area generated in this situation according to a preset cleaning execution time or a preset cleaning time cycle (e.g., set to one hour later). For example, taking a preset cleaning execution time as an example, when marking the uncovered area corresponding to the removal of the second obstacle area, an execution time for path planning and / or cleaning tasks to be performed can be set for the uncovered area, such as 5 PM on October 31, 2021. In this case, in step S11, path planning will be performed according to the set cleaning execution time, and then the cleaning task for the cleaning sub-area will be performed according to the set cleaning execution time.

[0073] Therefore, the embodiments of the present invention can update the uncleaned area in the coverage status map after the original obstacle has been detected and removed, and then clean the area again at an appropriate time, thereby achieving dynamic adaptation to environmental changes and ensuring cleaning efficiency and coverage.

[0074] As another embodiment, Figure 2 Taking environmental change information, including real-time driving trajectory, as an example, the paper demonstrates a method for implementing dynamic planning to execute cleaning paths based on real-time driving paths. Figure 2 As shown in the embodiments of the present invention, Figure 1 The step S10 shown is further implemented as including

[0075] Step S101: Determine the coverage status of the cleaning path based on the real-time driving trajectory, and update the coverage status map based on the determined coverage status.

[0076] In this embodiment of the invention, the coverage status map is dynamically updated based on the actual driving trajectory of the cleaning equipment, such as an unmanned sweeping vehicle, to plan and update the cleaning path in real time. Specifically, determining the coverage status of the cleaning path based on the real-time driving trajectory can be achieved by identifying the traversed cleaning path as covered and marking the traversed cleaning path as a covered area on the coverage status map. For example, a newly added cleaned area can be determined based on a preset cleaning width and the actual cleaning trajectory, and this newly added cleaned area can be marked as a covered area. For instance, the preset cleaning width and the length of the traversed path calculated from the actual cleaning trajectory can be multiplied to obtain the actual cleaning area, and the covered area can then be determined based on the actual cleaning area.

[0077] In embodiments where environmental change information is the actual cleaning trajectory, the preset triggering conditions can be set to include the completion of the current cleaning sub-area, and / or the arrival of the preset execution cycle, and / or the arrival of the preset execution time. Correspondingly, the selected area for path planning can be an uncovered area in the coverage status map that conforms to the preset cleaning rules. Taking real-time planning and updating of the cleaning path based on a preset cycle as an example, for instance, when setting the execution cycle for cleaning path planning, the real-time acquisition of the cleaning trajectory can be set to acquire the cleaning trajectory once after each of the cleaning paths is completed. Taking the loop path (i.e., the contour path) below as an example, the real-time cleaning trajectory can be acquired only after one loop is completed. Based on the real-time cleaning trajectory, the currently completed loop path is marked as a covered area on the coverage status map, and path planning is triggered. At this time, the selected area for path planning can be an uncovered area of ​​the current cleaning sub-area. That is, the triggering condition can be set to update the cleaning path based on the current driving once after each loop of cleaning is completed. In other implementation examples, the period for obtaining the real-time cleaning trajectory can be set to the completion of a cleaning sub-area, and the trigger condition can be set to the completion of the current cleaning sub-area, that is, when a certain subdivided cleaning sub-area is completed, the real-time cleaning trajectory is obtained to update the coverage status and perform dynamic planning of a cleaning path.

[0078] As an exemplary implementation, in the above embodiments, the specific implementation of selecting the area to be planned from the uncovered area based on the coverage status map can be pixel-based. For example, different pixel values ​​can be used to distinguish the uncovered areas of different clean sub-regions. For instance, a pixel value of 225 can be used to identify the uncovered area of ​​the current clean sub-region, a pixel value of 200-225 can be used to identify the uncovered area of ​​a clean sub-region formed due to the removal of a second obstacle, and a pixel value of 160 can be used to identify the uncovered area of ​​another uncleaned sub-region. Therefore, the required map area can be extracted as the area to be planned based on the pixel values, and full-coverage path planning processing can be performed on the extracted area to be planned.

[0079] The preferred embodiments of the full-coverage path planning described above will be described in detail below.

[0080] Taking full-coverage path planning as an example, which involves planning a looping path, Figure 7 This illustration schematically depicts a full-coverage path planning method according to one embodiment of the present invention. In this implementation example, a dynamically extracted area to be planned is used as the input target area to generate a clean path. The execution subject of this method embodiment and... Figure 1The execution entity shown can be the same, or it can be a dedicated processor of a full-coverage path planning device, such as... Figure 7 As shown, its implementation includes:

[0081] Step S80: Extract the contour boundary of the area to be planned and smooth the obtained contour boundary;

[0082] Step S81: Map the contour boundary to the raster map corresponding to the area to be planned, and obtain the distance transformation map of the raster map by performing distance transformation on the raster map mapped with the contour boundary.

[0083] Step S82: In the distance transformation map, using the contour boundary as a reference, generate contour path parallel to the contour boundary, and smooth each contour path respectively;

[0084] Step S83: Form a cleaning path corresponding to the area to be planned based on the smoothed contour path.

[0085] In step S80, the route planning area is preferably formed based on a grid map. The grid map identifies passable and impassable areas, for example, using white grids to represent passable areas and black grids to represent impassable areas. This embodiment of the invention performs full-coverage route planning based on an input grid map with passable and impassable area markers. Alternatively, passable and impassable areas in the grid map can be identified in other ways, such as filling grids with 0 to represent passable areas and filling them with 1 to represent impassable areas. Those skilled in the art can flexibly set these methods according to actual needs, and this application does not impose strict limitations. Specifically, in step S80, the grid map corresponding to the route planning area is first subjected to contour extraction. Contour extraction is achieved using the markers of impassable areas. For example, the markers of impassable areas can be identified, such as black grids, and the contour boundary is formed based on the range of the impassable area markers. That is, in this embodiment of the invention, the contour boundary refers to the contour boundary of the impassable area formed by the outer edge of the impassable area. It is easy to understand that, since the number of impassable areas in a target clean area is uncertain, the number of contour boundaries extracted based on impassable areas in this embodiment of the invention is also uncertain. For example, if there are two impassable areas, two contour boundaries can be extracted accordingly; if there are three impassable areas, three contour boundaries can be extracted accordingly. Figure 8As shown in example A, the target clean area includes three impassable regions (represented by black grids): region A, region B, and region C. Therefore, the resulting contour boundaries are contour boundary 1 corresponding to region A, contour boundary 2 corresponding to region B, and contour boundary 3 corresponding to region C. Contour boundary 1 is located at the outermost layer of the entire target clean area, defining its maximum extent; therefore, it is referred to as the outer contour boundary in this embodiment. Contour boundaries 2 and 3, being impassable regions within the target clean area, are referred to as inner contour boundaries in this embodiment. Since the initial grid map is typically generated by laser point clouds hitting obstacles, the resulting contour boundaries inevitably contain noise during contour extraction. To ensure contour smoothness, this embodiment requires smoothing the contour boundaries after extraction, resulting in a smoother, fully covered path and reducing unnecessary branches in the subsequently generated Vino skeleton. Smoothing processing improves the smoothness of the resulting contour boundaries, which in turn makes the contour clusters and connected paths constructed in subsequent steps smoother, thus creating a smooth, full-coverage cleaning path. For example, in this embodiment of the invention, smoothing processing can be performed using filtering or existing path smoothing algorithms, including but not limited to the Floyd path smoothing algorithm or multiple curve simulations. It should be noted that in this embodiment of the invention, the independently connected traversable areas formed by the segmentation of the target cleaning area by impassable areas are referred to as cleaning sub-regions. When actually performing cleaning tasks or path planning, a separate local path planning can be performed on a specific cleaning sub-region, or a global path planning can be performed on the entire target cleaning area. The triggering conditions for dynamic path planning and the corresponding extracted areas to be planned can be referred to the foregoing description and applied according to requirements; this embodiment of the invention does not impose any limitations on this.

[0086] In step S81, this embodiment of the invention first maps the contour boundaries extracted in step S80 back to the grid map. Then, it performs a distance transformation on the grid map with the mapped contour boundaries to obtain a distance transformation map of the entire grid map. This allows subsequent path planning analysis to be based on the entire grid map, focusing on the entire target area for cleaning strategy analysis. Compared to strategy analysis focusing on local areas after region division, this effectively ensures the smoothness and coverage of the planned cleaning path. The specific implementation of performing distance transformation on the grid map with mapped contour boundaries to obtain the distance transformation map can refer to existing distance transformation methods, such as the brushfire algorithm, which will not be elaborated here, thus obtaining the distance transformation map. Figure 8 The diagram schematically illustrates the effects of transforming a raster map with mapped outlines before and after processing, such as... Figure 8 A shows a raster map with outline boundaries. The target clean area includes three impassable areas (areas represented by black grids), namely area A, area B, and area C. Therefore, the outline boundaries formed are outline boundary 1 corresponding to area A, outline boundary 2 corresponding to area B, and outline boundary 3 corresponding to area C. Figure 8 B shows the pair Figure 8 The distance transformation map obtained after performing distance transformation on the raster map shown in Figure A is as follows: Figure 8 As shown in B, the distance transformation map obtained after the transformation clearly presents the skeleton of the raster map mapped with contour boundaries.

[0087] In step S82, this embodiment of the invention preferably generates contour paths in the obtained distance transformation map, so that a clean path can be formed based on the generated contour paths. In this embodiment, a contour path refers to a set of line clusters including at least one contour line, wherein a contour line is a closed curve formed by connecting adjacent points with equal elevation on the distance transformation map. Equal elevation in this embodiment means that the distance to the contour boundary is the same. In the distance transformation map of this embodiment, the gray value of each point in the passable area of ​​the distance transformation map is preferably used to represent the distance of the point from the contour boundary of the inaccessible area, and points with the same gray value are combined to form a contour line. Preferably, in this embodiment, each contour line in the contour line cluster has a set distance interval from the contour boundary, that is, this embodiment finds contour lines and forms contour line clusters according to the set distance interval and gray value. As a preferred implementation, in step S82, the present invention generates contour line clusters based on the contour boundary. Exemplarily, this can be achieved by generating contour line clusters parallel to the contour boundary with the contour boundary as the initial position and a first preset width as the interval distance. Specifically, the method for generating contour line clusters in this embodiment of the invention can be as follows: taking the outer contour boundary or inner contour boundary of the region shown in the distance transformation diagram as the initial position, and generating contour lines parallel to the outer contour boundary or inner contour boundary respectively. The interval distance between two adjacent contour lines is set to be equal, such as being equal to the first preset width. Thus, starting from the selected contour boundary, multiple contour lines can be gradually generated into the passable area through iteration, thereby forming a contour line cluster. For example, taking the outer contour boundary as the starting point, the interval between two adjacent contour lines is set to the width d of a cleaning device (e.g., the cleaning width of the cleaning device). Multiple contour lines are generated iteratively towards the passable area: starting from the outer contour boundary, the first contour line is obtained by indenting it inward by a distance d; then, using the first contour line as the boundary, the second contour line is obtained by indenting it inward by another distance d; and so on, thus forming a contour line cluster. Therefore, the contour lines in these generated contour line clusters are essentially curves parallel to the inner or outer contour, and the spacing between adjacent clusters is a first preset width. This allows the establishment of a basic full-coverage cluster (also referred to as a loop path in this embodiment), achieving full coverage of the cluster in the entire area shown in the distance transformation diagram. Preferably, the first preset width is set to the width of a vehicle, ensuring that the generated contour lines are exactly one vehicle width apart, so that there are no gaps between cleaning lines during the cleaning process, guaranteeing the coverage of the cleaning path. Figure 9 The diagram schematically illustrates the result of contour line clusters formed by the method according to an embodiment of the present invention, such as... Figure 9 A shows the pair Figure 8 The contour line clusters obtained after processing the distance transformation map in step S82 are shown. In a preferred implementation, path points can be further generated based on these contour line clusters, and these path points can be smoothed to reduce the path curvature at turns, generating a contour path that the autonomous vehicle can drive, such as... Figure 9 B shows the correct answer. Figure 9 The image shows the effect of a drivable contour path (also referred to as a loop path in this embodiment of the invention) obtained after generating path points and smoothing the contour line clusters of A. Figure 9 As shown, the looping path generated based on the distance transformation map can achieve an area coverage of over 90%. After processing in step S82, a contour path covering the entire area is formed. However, the contour lines are independent of each other, meaning they are not connected. Therefore, to obtain a complete cleaning path, in step S83, the contour paths generated in step S82 are connected. Specifically, step S83 connects the contour paths as follows: starting from a point on the outermost contour path (i.e., the outermost ring contour path), find the nearest point on the next path as the starting point. Then, starting from this nearest starting point, select two locations two car widths away. Select a point on this selected location as the ending point. Connect the starting and ending points using a Bezier path, where the directions of the starting and ending points are the tangent directions at those locations. This process is repeated for each of the generated contour paths to obtain a complete path. Preferably, forward filtering refers to using the driving direction of the driving path as the forward direction, and moving along that forward direction to filter out the positions of the required points.

[0088] Specifically, Figure 10 One embodiment of the method for connecting generated contour paths is shown, such as Figure 10 As shown, the method includes:

[0089] Step S131: Record all contour lines in the contour line path separately, and record each contour line as a road segment;

[0090] Step S132: Using the outermost contour boundary as the starting segment for traversal, traverse all segments corresponding to the contour path, and perform connectivity processing on the current traversed segment.

[0091] In step S131, for example, each contour line in the contour path can be treated as a road segment and identified and recorded, thereby forming L1, L2...L nThere are n road segments in total, where n is the number of contour lines in the contour path, and L1 refers to the road segment corresponding to the outermost contour boundary. The road segments are marked and recorded sequentially from the outer contour boundary inwards. That is, the road segment corresponding to the next outermost contour line adjacent to the outer contour boundary is L2, and so on. It should be noted that this step is not mandatory in the specific implementation. Marking the road segments on the contour path is only to facilitate the traversal processing in step S132 and to facilitate the recording and storage of the contour path. In other implementation examples, this step can be omitted, and step S132 can be directly performed on the contour path to connect the contour path, that is, to connect all the resulting road segments.

[0092] In step S132, the outer contour boundary can be used as the current road segment, and all road segments corresponding to the contour line path can be traversed sequentially to connect all contour line paths. For example, the nearest road segment L1 to the current road segment L1 can be found starting from the outer contour boundary. k In section L k Find the point closest to the end of the current road segment L1 (e.g., the end of the current road segment L1 is located at the end of road segment L1). k (Projection point on the map), and then using that point as the starting point, filter the distance to the second preset width in the preset direction, and take the position that is farthest from the starting point by the second preset width as the road segment L. k The starting position of road segment L1 and the ending position of road segment L2 are determined. k The starting and ending points of the current road segment are connected to form a continuous cleaning path. This process continues until a complete, interconnected cleaning path is achieved. It's important to note that the ending point of the current road segment refers to the final location reached during cleaning, while the starting point is the initial position. By connecting the ending point of the current road segment to the starting point of the next segment, the cleaning process can smoothly transition to the next segment, ensuring a smooth workflow and high efficiency. Preferably, to achieve higher coverage, the starting and ending points of the same road segment (i.e., the same contour line) can be set to the same location. The starting and ending points of road segment L1 (the outer boundary) can be set based on specific road conditions or experience. Figure 11 The specific details of the connectivity processing performed on the traversed current road segment in step S132 are shown, such as... Figure 11 As shown, it specifically includes:

[0093] Step S132A: Among the remaining road segments that have not yet been traversed, select the road segment closest to the current road segment as the next current road segment;

[0094] Step S132B: Select the end position on the current road segment, and determine the point on the next current road segment that is closest to the selected end position as the starting point;

[0095] Step S132C: Using the starting point as a reference, filter and determine the starting point position in the preset direction of the current road segment;

[0096] Step S132D: Connect the selected end position on the current road segment with the determined start position on the next current road segment.

[0097] In step S132A, the remaining road segments refer to those that have not yet been connected to the cleaning path, i.e., those that have not yet been traversed, and are relative to the current road segment. Since the next road segment to be executed (i.e., connected) is searched for against the current road segment during a single traversal, all remaining road segments are not yet connected to the current road segment. In this embodiment, the next current road segment refers to the next road segment to be executed, i.e., the next road segment to be visited after the current road segment has completed its cleaning process. Selecting the road segment closest to the current road segment in step S132A is actually implemented by selecting the contour line closest to the current contour line from the contour line paths that have not yet been connected.

[0098] In step S132B, selecting the end position on the current road segment can be exemplarily implemented by: determining whether the current road segment is segment L1 (i.e., the outer contour boundary), and selecting the end position according to a pre-set or default rule; otherwise, using the starting position of the road segment determined in the previous traversal (i.e., the traversal of the previous road segment before the current road segment, where the current road segment in this round is the next current road segment in the previous traversal) as its end position. After determining the end position of the current road segment, selecting the point closest to the end position from the next current road segment determined in step S132A as the starting point. The distance can be determined using existing methods such as coordinate ranging, and this embodiment of the invention does not impose any limitations on this.

[0099] In step S132C, the starting point position is determined on the next current road segment based on the starting point determined in step S132B. For example, the starting point can be used as the reference to select the starting point position in a preset direction. As a preferred implementation, to facilitate turning operations, the preset direction can be determined as the travel direction of the autonomous cleaning equipment, such as an unmanned vehicle or robot. That is, based on the starting point, the starting point position is selected from the next current road segment along the travel direction of the unmanned cleaning equipment. For example, selecting the starting point position on the next current road segment involves using the position of a point located at a distance of a second preset width from the starting point in the preset direction as the starting point position. The second preset width can be determined based on the size, volume, and turning capability of the cleaning equipment. For example, for a larger autonomous cleaning equipment, the second preset width can be set to the distance of two vehicle widths. Figure 12 The diagram illustrates the connected component processing result after performing one round of traversal, as follows: Figure 12 As shown, taking the current road segment as L1 as an example, it displays the end position on road segment L1 and the determined next current road segment L. k In L k The starting point and starting point positions are determined above, and the current road segment L1 and the next current road segment L2 are also displayed. k The connectivity relationship means that the current road segment and the next current road segment are connected by the end position of the current road segment and the starting position of the next current road segment. The connected path is not a straight turning point, but has a smooth transition, which makes it convenient for cleaning equipment to turn.

[0100] In other preferred embodiments, a Bézier curve can be used to connect the current position of the cleaning equipment, such as the unmanned sweeping vehicle, to the contour path formed above, in order to form a more complete cleaning path. Specifically, the current position of the cleaning equipment is used as the starting point of the Bézier curve, and the point on the contour path closest to the current position is selected two vehicle widths away as the ending point of the Bézier curve, and the two are connected. Figure 13 The diagram illustrating the complete cleaning path thus generated is shown, such as... Figure 13 As shown in the splicing effect diagram, all contour line paths and the contour line paths are connected to the current position of the cleaning equipment to form a complete path.

[0101] The resulting complete cleaning path not only covers the entire map area, achieving a high coverage rate, but also, because the contour lines are connected by defined starting and ending points, there is a smooth transition between the starting and ending points, such as the distance of two vehicle widths. Therefore, the transition between two road segments can be achieved without sharp turns. The entire cleaning path has smooth continuity, which greatly reduces the number of times the cleaning mobile platform turns in place, improving cleaning efficiency and the smoothness of the cleaning process.

[0102] As another preferred implementation example, such as Figure 14 As shown, before smoothing each contour path individually, the method of the present invention... Figure 7 The method shown can be further expanded to include:

[0103] Step S12A: Delete contour clusters that do not meet the preset requirements based on the size of the area enclosed by the generated contour clusters;

[0104] Step S12C: Fill the area surrounded by the deleted contour line clusters in a "bow" shape to form a "bow" shaped clean path.

[0105] In this embodiment of the invention, in addition to the contour line cleaning strategy, a "bow"-shaped cleaning strategy is also combined to optimize the generated cleaning path. Specifically, before connecting the contour line clusters, this embodiment of the invention also determines the area of ​​the region enclosed by each contour line. Since contour lines are closed curves, in the cleaning strategy that forms contour line clusters, each contour line encloses a corresponding area. In step S12A, this embodiment of the invention calculates the area of ​​all generated contour line clusters, calculates the area of ​​the region enclosed by each contour line, and determines the area. For a given contour line, the area enclosed by it refers to the area of ​​the entire region within the closed area enclosed by that contour line. It can be understood that the area enclosed by a contour line actually includes the area enclosed by other inner contour lines within the closed area enclosed by that contour line. For example, the area enclosed by the generated contour lines can be calculated by coloring the inner region enclosed by the contour lines with a specific color, and then counting the number of points of that specific color. Since each point in the image is a pixel, and each pixel has a corresponding area value based on its size, the area enclosed by the contour lines can be obtained by multiplying the calculated number of pixels of the specific color by the corresponding area value of each pixel. The method for calculating the area of ​​all generated contour line clusters can also be implemented using other existing technologies, which will not be elaborated here. Preferably, this embodiment of the invention sets a preset standard. Judging the area can be achieved by comparing the calculated area enclosed by each contour line with the preset standard to determine the contour line deletion strategy. For example, the deletion strategy can be to delete contour lines whose area is less than a preset area threshold from the contour line cluster. Because cleaning equipment has characteristics such as volume, size, and turning angle, when the contour line area is small, the excessive curvature of the path can lead to poor tracking performance, making cleaning or turning difficult. In such cases, a minimum area value for the cleaning area can be set as a preset area threshold based on the characteristics of the cleaning equipment performing the cleaning task. The areas of each contour line area are compared with this threshold, and contour lines with areas smaller than the preset area threshold are deleted. In step S12C, to ensure cleaning coverage of the areas where contour lines have been deleted, a "bow"-shaped cleaning strategy can be used. This involves filling these smaller areas with "bow" shapes, which avoids missed areas and overcomes the difficulty of cleaning small areas in the contour line coverage path. Furthermore, since the "bow"-shaped filling is performed in smaller areas, the cleaning path within these areas is relatively short, thus minimizing the impact on cleaning efficiency.It should be noted that for a contour line whose area is smaller than a preset area threshold and which is surrounded by other contour lines, since the area enclosed by the contour line is larger than the area enclosed by the other contour lines in its inner circle, after deleting these contour lines, it is only necessary to fill the blank area after deleting these contour lines, that is, the area enclosed by the outermost contour line whose area is smaller than the preset area threshold, with a "bow" shape to cover all the smaller areas after the contour lines are deleted. Therefore, it is easy for those skilled in the art to understand that in the case of deleting multiple nested contour lines, it is not necessary to fill the area enclosed by each nested contour line with a separate "bow" shape, for example, . Figure 15 The image, set against a raster map background, illustrates the effect of identifying contour lines with smaller areas as those that do not meet preset requirements and are to be deleted. Figure 15 As can be seen, contour lines 1, 2, and 3 in the figure are identified as contour lines that need to be deleted because the areas they enclose are relatively small, and therefore they will be deleted. Figure 16 The diagram illustrates the effect of filling an area with an arc shape after contour lines have been removed, against a background of a raster map. Figure 16 As can be seen, deleting contour lines 1, 2, and 3 in the diagram resulted in two bow-shaped paths 4 formed by filling the areas enclosed by contour line 1 and contour line 3 respectively. Since contour line 2 is located within the inner circle of contour line 1, and the area it encloses is included within the area enclosed by contour line 1, there is no need to repeat the bow-shaped filling within the area enclosed by contour line 2. Figure 15 and 16 As can be seen, because these areas are relatively small, the resulting bow-shaped path is short, and there are few (or even none) turning points, therefore it does not have a significant impact on cleaning efficiency. This is understandable. Figure 15 and Figure 16 Alternatively, a distance transformation graph can be used as a background to illustrate the corresponding effects. In order to make it easier to see the effects clearly and intuitively and to provide a clearer display, the embodiments of the present invention use a grid map to illustrate the effects, but this is not considered a limitation on the method of the embodiments of the present invention.

[0106] In practical applications, when both contour path planning and "bow"-shaped filling strategies are employed, the cleaning strategy for the target cleaning area can be determined based on specific application requirements. For example, the cleaning path obtained through contour path planning and the cleaning path obtained through "bow"-shaped filling can be stored separately, and the cleaning paths generated by the two strategies can be executed according to requirements. Alternatively, in other implementations, the cleaning path obtained through contour path planning and the cleaning path obtained through "bow"-shaped filling can be connected, that is, the cleaning path obtained through "bow"-shaped filling can also be connected to the cleaning path obtained through contour path planning, so that the connected cleaning path serves as the complete cleaning path for the target cleaning area, and the cleaning strategy is executed based on this complete cleaning path. This embodiment of the invention does not impose limitations on this. Specifically, connecting the cleaning path obtained through the "bow"-shaped filling to the cleaning path planned using the contour lines can be achieved by: determining the intersection point between the bow-shaped cleaning path obtained through the bow-shaped filling and the original contour lines enclosing the filling area; and connecting the intersection point to the end of the cleaning path formed by the contour line cluster or to the end position of the contour line closest to the current intersection point. The original contour lines enclosing the filling area refer to the contour lines that have not yet been deleted and enclose the area used for bow-shaped filling. That is, before deletion, the area enclosed by contour lines with an area smaller than a preset area threshold is filled, and the intersection point between the filling route and the contour line is recorded. This intersection point is then connected to the cleaning path of the contour line cluster. Preferably, the selected intersection point for connecting to the cleaning path of the contour line cluster can be the intersection point closest to the end of the cleaning path formed by the contour line cluster or the intersection point closest to the end position of the contour line closest to the contour line enclosing the filling area. In an embodiment that connects the cleaning paths obtained from the two cleaning strategies, to make the distinction clearer, the complete cleaning path formed by connecting the contour line clusters is referred to as the first cleaning path, and the bow-shaped cleaning path obtained by filling in the "bow" shape is referred to as the second cleaning path.

[0107] Figure 3 This invention illustrates another embodiment of a dynamic full-coverage path planning method, taking a loop path planning method as an example. Figure 3 As shown, the method of this embodiment of the invention is in Figure 7 Based on the method shown, it also includes

[0108] Step S84: Perform the first optimization process on the outermost path in the cleaning path corresponding to the area to be planned, and

[0109] Step S85: Perform a second optimization process on the outermost loop of the cleaning path corresponding to the area to be planned.

[0110] In this embodiment of the invention, the cleaning path generated through full-coverage planning is a ring path (i.e., a contour path). The drawback of directly using distance transformation to generate a ring path is that it does not consider the vehicle's outer contour. For complex maps with many obstacles, the outermost ring path does not have good edge-fitting properties, easily leading to missed areas and a higher probability of collisions. Furthermore, if the inner ring path is generated based on this outermost ring path, the inner ring path will also produce some unnecessary back-and-forth head-swaying phenomena, affecting cleaning efficiency. In addition, when using distance transformation to generate a path inward based on an uneven outer ring path, fragmented areas are easily generated. Therefore, after generating the cleaning path corresponding to the area to be planned, this embodiment of the invention also optimizes the outermost and inner ring paths of the generated cleaning path to improve the coverage efficiency of the cleaning path.

[0111] In step S84, the first optimization process specifically includes: using the outermost path before optimization in the cleaning path as a reference line, performing dynamic planning sampling on the outermost path, and stitching together all the sampling results to form the optimized outermost path. Specifically, the dynamic planning sampling of the outermost path can be performed using dynamic planning sampling of a Bezier curve, sampling once at small intervals. The small intervals can be set according to requirements, such as the size and volume of the equipment performing the cleaning task. The sampling interval is set to be positively correlated with the equipment size; that is, when the equipment size is relatively small, the sampling interval is also set to a smaller value, which makes the sampled path more flexible and better able to fit the edge; while when the equipment size is large, the sampling interval is also set to a larger value, which makes the sampled path smoother and more suitable for the turning operation of large vehicles. For example, the sampling interval for smaller equipment can be set to 0.3-0.5m, while the sampling interval for larger equipment can be set to 0.5-0.8m. More preferably, the first optimization process also includes: during the sampling process, performing collision detection on the sampling results, such as performing collision detection on each sampled Bézier curve and filtering out the sampled curves that collide, so as to obtain the sampling results without collisions. The sampling results obtained after collision detection and filtering are used as the basis for stitching together to form the optimized outermost path. Because the collision detection takes into account the outer contour of the vehicle, the final result has a better edge-fitting effect and is less prone to collisions. Figure 4 The illustration shows the effect of the outermost path before and after optimization, such as... Figure 4 As shown, the green curve a represents the outermost edge before optimization, and the blue curve b represents the result after optimization. Figure 5 The illustration schematically shows the coverage effect of the simulated cleaning equipment's outer contour along the optimized outermost path, from... Figure 5As can be seen, on the boundary with obstacles, the blue curve b tends to avoid the obstacles in advance, and collisions are almost non-existent; while on the boundary without obstacles, the fit is basically consistent before and after optimization.

[0112] In step S85, the second optimization process is implemented using smoothing optimization. For example, it is implemented using a constrained quadratic programming method, specifically including: First, calculating the curvature of the outermost path; then, determining the concavity / convexity of the outermost path based on the sign of its curvature, where concave contours are generally near obstacles, and convex contours are generally free of obstacles. Therefore, in this embodiment, points with positive curvature are defined as convex points, and points with negative curvature are defined as concave points; finally, smoothing optimization is performed on the innermost path by setting the size of the quadratic programming constraints, so that the offset optimization of the innermost path at the concave points of the outermost path is minimized, while the offset optimization at the convex points of the outermost path is increased. Specifically, the smooth optimization of the innermost loop path by setting the constraint size of quadratic programming can be achieved by: first, using the distance of shrinking the outermost loop path inward by half a car width to about one car width as the reference line for the innermost loop path; then, setting the cost terms of quadratic programming, preferably setting the cost terms of quadratic programming to include three cost terms: minimizing the curvature of the optimized result, minimizing the path length of the optimized result, and minimizing the deviation of the optimized result from the reference line; finally, using the maximum offset value of each discrete point on the outermost loop path as the constraint condition for that discrete point, and using a quadratic function to optimize the innermost loop path. When setting offset constraints for each discrete point, for example, since the curvature of the outermost path is negative, indicating a concave shape, and the outer edge of a concave point is generally caused by obstacles, the maximum offset constraint for a point at a concave point can be set to about half the width of a vehicle to ensure that it will not encounter obstacles. Conversely, the curvature of the outermost path is positive, indicating a convex shape, and the outer edge of a convex point is generally free of obstacles. Setting the maximum offset constraint for a point at this point to a small value (such as 0-0.2m) can allow the optimization result to fill the convex hull, while simplifying and smoothing the inner contour. Common smoothing methods used in the prior art (such as mean filtering and Gaussian filtering) only have the function of reducing waves, while the second optimization process used in this embodiment of the invention not only achieves offset optimization for the innermost path, but also uses the concavity and convexity of the outermost path contour as a constraint condition, thus taking into account the simplification of the outermost path contour, while further reducing the fragmented areas generated when generating the inward path. Figure 6 The diagram schematically shows the effects of the first optimized outer contour c (i.e., the outermost path) and the second optimized inner contour d (i.e., the second outermost path), as shown below. Figure 6As shown, through the second optimization process, the distance between the outer contour c and the inner contour d is less than one vehicle width. Therefore, when performing the cleaning task, the execution of these two circular paths can achieve good overlap and coverage, improving coverage efficiency and avoiding the generation of uncovered debris areas. Since in practical applications, cleaning tasks are executed sequentially in circles, as a preferred implementation, this embodiment of the invention can also perform dynamic path planning based on the execution cycle of the cleaning task. That is, this embodiment of the invention can set the trigger condition to be set when the cleaning of the second outermost circle path in the cleaning path is completed. At this time, the environmental change information obtained in step S10 can be the corresponding real-time cleaning trajectory obtained when the second outermost circle path in the cleaning path is completed. Thus, the real-time cleaning trajectory obtained in this embodiment of the invention actually includes the driving trajectory of the outermost circle path and the second outermost circle path; that is, the driving trajectory obtained at this time covers the outermost circle and the second outermost circle path once in succession. As a preferred embodiment, this embodiment of the invention optimizes the outermost circle and the second outermost circle path during path planning. Because the distance between the outermost circle path and the second outermost circle path is less than one vehicle width after optimization, they can overlap well. Therefore, after these two rounds of coverage are completed, because the coverage status map is dynamically updated, when the area to be planned for the next coverage route is extracted from the coverage status map, the optimization results of the previous round can be reflected in the planned route of the next round, realizing true dynamic path planning and updating based on environmental changes, and improving cleaning efficiency and coverage.

[0113] It should be noted that, in specific applications, the embodiments of the present invention are not limited to performing the above-mentioned optimization processes on both the outermost path and the second outermost path at the same time. They may also perform only the first optimization process or only the second optimization process as needed. The embodiments of the present invention do not impose any restrictions on this.

[0114] As a better implementation example, Figure 17 Another embodiment of the present invention, a full-coverage path planning processing method, is also shown, such as... Figure 17 As shown, it is in Figure 7 In addition to the method shown, it also includes:

[0115] Step S86: Simulate the coverage movement of the cleaning path corresponding to the area to be planned to determine the missed areas, and generate the Vino line coverage path of the grid map based on the determined missed areas, and add the generated Vino line coverage path to the cleaning path of the area to be planned.

[0116] During the research and development process, the inventors discovered that although the looping path could achieve a path coverage rate of over 90%, based on practical application experience, this path planning method still inevitably resulted in some areas where missed scans were unavoidable. The inventors summarized the main reasons for these missed scans, identifying the following:

[0117] (1) Because the loop path will generate a driving path with a large curvature in a narrow space, although it is not necessary to turn in place after the path is smoothed, making it more convenient for the unmanned vehicle to drive, it is easy to miss the scan at the turning point.

[0118] (2) The innermost circle of the loop path is often a small area, and the generated small loop path is not convenient for autonomous vehicles to drive.

[0119] (3) For more complex maps, the distance between contour lines generated by distance transformation is not strictly a distance of one car width.

[0120] Based on this discovery, the inventors devised a solution to address this deficiency by using a rectangular frame to approximate the coverage area of ​​the cleaning vehicle. Specifically, the rectangular frame moves along the coverage path to simulate the cleaning vehicle's coverage movement on the map corresponding to the target area. The inventors identified un-colored white areas by blacking out the area covered by the rectangular frame. The white areas revealed after the coverage movement is complete are the missed areas (i.e., unswept areas). Figure 18 As shown in Figure A. Through analysis, the inventor unexpectedly discovered that the missed areas were primarily distributed along the Vino skeleton of the map (i.e., the skeleton lines shown in the distance transformation diagram). Therefore, the inventor conceived of generating only the Vino skeleton lines of the map as supplementary cleaning paths (i.e., Vino line coverage paths) to cover the missed areas. The generated Vino skeleton lines are shown in Figure A. Figure 18As shown in B. It should be noted that the Voronoi skeleton line refers to a line on the Voronoi diagram, i.e., the edge of a Thiessen polygon. It consists of a set of continuous polygons formed by perpendicular bisectors of lines connecting two adjacent points. The characteristics of the Voronoi skeleton line are that it perpendicularly bisects the area at that location, and it also has the following features: each V-polygon has a generator; the distance from a point within a V-polygon to that generator is shorter than its distance to other generators; points on the polygon boundary are equidistant from the generator that generated that boundary; and the Voronoi polygon boundaries of adjacent graphs are subsets of the original adjacent boundaries. In specific implementations, the above-mentioned Voronoi skeleton lines can be generated using existing methods such as divide-and-conquer, scan-line algorithms, and Delaunay triangulation algorithms. In particular, this embodiment of the invention uses existing open-source algorithms to generate the Voronoi skeleton lines of the raster map of the target area. Based on this, by forming a cleaning path including the Voronoi skeleton lines, 100% coverage of missed areas can be achieved, improving the coverage rate and cleaning efficiency of the looping path. As a better implementation example, the generated Vino skeleton lines can be further optimized in the specific implementation to remove unnecessary Vino skeleton lines, thus forming the final Vino line coverage path. Specifically, optimizing the Vino skeleton lines can be implemented as follows: mapping the Vino skeleton lines onto a map showing missed coverage areas through simulated coverage movement, determining whether the Vino skeleton line can correspond to and match the missed coverage areas shown in the map, and removing Vino skeleton lines that do not correspond to missed coverage areas. For example, a map of missed coverage areas can be constructed offline based on the generated cleaning path, and then the coordinates of the points on the generated Vino skeleton lines can be obtained, converted, and overlaid onto the pixel coordinate system of the map of missed coverage areas. The degree of correspondence between the line segment and the missed coverage area can be determined by judging how many points on the Vino skeleton line segment fall on the missed coverage area of ​​the map. Line segments with a matching degree exceeding a certain threshold are determined to be matched with the missed coverage area, and line segments with a matching degree not meeting a certain threshold are determined to be line segments that need to be removed. Figure 19 The optimized Vino skeleton line is schematically shown, such as... Figure 19 As shown, the optimized Vino skeleton lines can effectively cover the missing coverage areas. The Vino skeleton lines used to generate the map can be obtained by extracting the outline of the map skeleton based on a distance transformation map, which can be implemented with reference to existing technologies; this embodiment of the invention will not elaborate further.

[0121] In practical applications, the inventors further discovered that for some large uncovered areas, the above-mentioned full-coverage path planning method can generate a looping path. Combined with supplementary sweeping along the Vino line coverage path, a coverage rate of over 90% can be achieved. However, due to dynamic environmental changes, the actual cleaning path, due to positioning and control errors, is not precisely executed according to the planned cleaning path. Therefore, some small, fragmented uncovered areas are unavoidable, such as... Figure 20 The diagram illustrates the effect of cleaning small, missed areas (E) during the actual cleaning process. To further address this issue, as another preferred implementation, this embodiment of the invention further includes, based on the above method: determining whether the area of ​​each cleaning sub-region in the path planning area is less than or equal to a preset area threshold; if so, generating the minimum area bounding rectangle of each cleaning sub-region, determining the straight road segments generated based on each bounding rectangle as the cleaning path for each cleaning sub-region, and connecting the cleaning paths of each cleaning sub-region according to a preset connectivity strategy to form the cleaning path corresponding to the path planning area. The preset connectivity strategy can be a proximity-based connectivity strategy. Specifically, this can be implemented by determining whether there are clean sub-regions (i.e., fragmented regions) with an area less than or equal to a threshold area in the area to be planned. If so, a minimum bounding rectangle is generated for each clean sub-region based on its area. Then, based on the length and width of this rectangle, it is determined whether it can be covered by a straight line segment. If so, a covering line segment corresponding to the clean sub-region is generated as a line segment covering path. Then, the line segment covering paths of each clean sub-region are spliced ​​into the generated loop path in the nearest order using Bézier curves, thereby forming a complete clean path with higher coverage. The splicing effect of the generated more complete clean path including the line segment covering path is as follows: Figure 21 As shown, it schematically illustrates the defined debris area E, the outer rectangle F used for coverage, and the defined line segment coverage path G. In a specific implementation, the possible uncovered debris areas can be determined by performing simulated coverage motion on the generated looping path, or the possible uncovered debris areas can be directly marked on the target clean area map based on experience. The specific implementation method is not limited in this embodiment of the invention.

[0122] Figure 22 The diagram schematically illustrates a dynamic full-coverage path planning device according to an embodiment of the present invention, such as... Figure 22 As shown, the device includes:

[0123] Memory 60 is used to store executable instructions; and

[0124] Processor 61 is used to execute executable instructions stored in memory.

[0125] In one preferred embodiment, the executable instructions stored in memory 60, when executed by the processor, cause the processor to perform the dynamic full-coverage path planning method described in any embodiment of the present invention.

[0126] Figure 23 The diagram schematically illustrates a dynamic full-coverage path planning device according to an embodiment of the present invention, such as... Figure 23 As shown, the device includes:

[0127] The status update module 100 is used to dynamically update the coverage status map corresponding to the target area based on the acquired environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas and uncleaned areas are marked as uncovered areas.

[0128] The dynamic scheduling module 200 is used to detect triggering conditions in real time, and when the preset triggering conditions occur, it selects the area to be planned in the uncovered area according to the coverage status map and calls the path planning module to perform full coverage path planning.

[0129] The route planning module 300 is used to perform full-coverage route planning for the selected area to be planned.

[0130] In a preferred embodiment of the present invention, the acquired environmental change information includes a newly added first obstacle. The state update module can be specifically used to add the first obstacle to the corresponding position in the uncovered area when the first obstacle is located in an uncovered area.

[0131] In this embodiment, for example, the preset triggering conditions may include the presence of a newly added first obstacle in the acquired environmental change information and the newly added first obstacle being located in an uncovered area; the path planning area selected in the dynamic scheduling module is either an uncleaned area of ​​the clean sub-region where the first obstacle is located or an uncleaned area of ​​the target region.

[0132] In this embodiment, for example, the preset triggering condition may also include the presence of a newly added first obstacle in the acquired environmental change information and the newly added first obstacle being located in an uncleaned area of ​​the current clean sub-region; the path planning area selected in the dynamic scheduling module is the uncleaned area of ​​the current clean sub-region.

[0133] In this embodiment, for example, the preset triggering condition may also include the presence of a newly added first obstacle in the acquired environmental change information and the newly added first obstacle being located in an uncovered area. Accordingly, the selected path planning area may be an uncleaned area of ​​the target area containing the first obstacle.

[0134] In another preferred embodiment of the present invention, the acquired environmental change information includes the removal of a second obstacle, and the state update module can be specifically used to update the area occupied by the second obstacle to an uncovered area.

[0135] In this embodiment, for example, the preset triggering condition may include the presence of a removed second obstacle in the acquired environmental change information; the selected path planning area is an uncleaned area of ​​the clean sub-region where the second obstacle is located, or an uncleaned area in the target area.

[0136] In this embodiment, for example, the preset triggering condition may also include the presence of a removed second obstacle in the acquired environmental change information and the second obstacle being located in an unclean area of ​​the current clean sub-region; the selected path planning area is an unclean area in the current clean sub-region or an unclean area in the target area.

[0137] In this embodiment, for example, the preset triggering condition may also be set to include the presence of a removed second obstacle in the acquired environmental change information; correspondingly, the selected path planning area may be an uncleaned area of ​​the target area containing the second obstacle.

[0138] In another preferred embodiment of the present invention, the acquired environmental change information may include the actual cleaning trajectory, and the status update module may also be specifically used to determine the newly cleaned area based on the preset cleaning width and the actual cleaning trajectory, and mark the newly cleaned area as the covered area.

[0139] In one embodiment of the present invention, the preset triggering conditions further include the completion of the current cleaning sub-area, and / or the arrival of the preset execution cycle, and / or the arrival of the preset execution time; the selected path planning area is an uncovered area in the coverage status map that conforms to the preset cleaning rules.

[0140] Figure 24 The diagram schematically illustrates a dynamic full-coverage path planning device according to an embodiment of the present invention, such as... Figure 24 As shown, the path planning module 300 in this device specifically includes:

[0141] Boundary extraction unit 301 is used to extract the contour boundary of the area to be planned and to smooth the contour boundary.

[0142] The distance transformation unit 302 is used to map the contour boundary to the grid map corresponding to the area to be planned, and to obtain the distance transformation map of the grid map by performing distance transformation on the grid map mapped with the contour boundary.

[0143] Contour line generation unit 303 is used to generate contour line paths in the distance transformation map, with the contour boundary as a reference and parallel to the contour boundary;

[0144] The smoothing unit 304 is used to smooth each contour path separately.

[0145] The cleaning path generation unit 305 is used to generate a cleaning path corresponding to the area to be planned based on the smoothed contour path.

[0146] For example, the cleaning path generation unit 305 can be implemented to specifically traverse all road segments corresponding to the contour line cluster, using the outermost contour boundary as the starting road segment, and perform the following processing on the traversed current road segment:

[0147] Among the remaining road segments that have not yet been traversed, select the road segment closest to the current road segment as the next current road segment;

[0148] Select the end point on the current road segment, and determine the point on the next current road segment that is closest to the selected end point as the starting point;

[0149] The starting point is determined by filtering and selecting the starting position in the preset direction of the current road segment based on the starting point;

[0150] The selected end position on the current road segment will be connected to the starting position determined on the next current road segment.

[0151] Figure 25 The diagram schematically illustrates a dynamic full-coverage path planning device according to another embodiment of the present invention, such as... Figure 25 As shown, in the device of this embodiment, the path planning module 300 in Figure 24 The device shown is further implemented by including:

[0152] The contour line filtering unit 306 is used to delete contour line paths that do not meet the preset requirements based on the size of the area enclosed by the generated contour line paths.

[0153] The second cleaning path generation unit 307 is used to fill the area surrounded by the deleted contour path in a "bow" shape to form a "bow" shaped cleaning path.

[0154] In this embodiment, before the clean path generation unit connects the contour paths or before the smoothing unit performs smoothing processing, that is, after the contour generation unit generates the contour paths or before the smoothing unit performs smoothing processing on the contour paths, the contour filtering unit is first called to delete contour lines in the contour paths that do not meet the preset conditions. After the deletion operation, the second clean path generation unit is called to fill the corresponding areas in a "bow" shape to form a "bow" shaped clean path.

[0155] In specific applications, those skilled in the art can execute the cleaning paths generated by the cleaning path generation unit and the second cleaning path generation unit according to predetermined strategies as needed. For example, the cleaning paths generated by the cleaning path generation unit and the second cleaning path generation unit can be executed at preset times respectively. Alternatively, an integration unit can be set up to integrate and connect the cleaning paths generated by the cleaning path generation unit and the second cleaning path generation unit. The specific implementation process can be found in the previous description and will not be repeated here.

[0156] Figure 26 The diagram schematically illustrates a dynamic full-coverage path planning device according to another embodiment of the present invention, such as... Figure 26 As shown, in the device of this embodiment, the path planning module 300 in Figure 24 The device shown is further implemented by including:

[0157] The first optimization processing unit 308 is configured to perform a first optimization process on the outermost path in the cleaning path corresponding to the path planning area; and / or

[0158] The second optimization processing unit 309 is used to perform a second optimization process on the outermost loop path in the cleaning path corresponding to the path planning area.

[0159] Figure 27 The diagram schematically illustrates a dynamic full-coverage path planning device according to another embodiment of the present invention, such as... Figure 27 As shown, in the device of this embodiment, the path planning module is... Figure 24 The device shown is further implemented by including:

[0160] The path correction unit 310 is used to simulate the coverage movement of the cleaning path corresponding to the area to be planned to determine the missed area, generate the Vino line coverage path of the grid map according to the determined missed area, and add the generated Vino line coverage path to the cleaning path of the area to be planned.

[0161] Figure 28 The diagram schematically illustrates a dynamic full-coverage path planning device according to another embodiment of the present invention, such as... Figure 28As shown, in the apparatus of this embodiment of the invention, in Figure 23 The device shown is further implemented by including:

[0162] The second path planning module 400 is used to determine whether the area of ​​each cleaning sub-region in the path planning area is less than or equal to a preset area threshold; if so, it generates the minimum area bounding rectangle of each cleaning sub-region, determines the straight road segment generated based on each bounding rectangle as the cleaning path of each cleaning sub-region, and connects the cleaning paths of each cleaning sub-region according to a preset connection strategy to form the cleaning path corresponding to the path planning area.

[0163] It should be noted that the specific implementation process and implementation principle of each module and unit of the dynamic full-coverage path planning device in this embodiment of the invention can be found in the corresponding descriptions of the above method embodiments. For example, the corresponding descriptions of contour boundaries, distance transformation, and the generation and connectivity of contour line clusters in the method embodiments, as well as the corresponding descriptions of the first optimization process, the second optimization process, and the specific dynamic update process, etc., will not be repeated here. Exemplarily, the dynamic full-coverage path planning device in this embodiment of the invention can be any intelligent device with a processor, including but not limited to computers, smartphones, personal computers, robots, cloud servers, etc.

[0164] Figure 29 The illustration schematically shows a cleaning device according to one embodiment of the present invention, such as... Figure 29 As shown, the cleaning equipment includes:

[0165] fuselage 70;

[0166] A dynamic full-coverage path planning device 71 is mounted on the body 70.

[0167] The dynamic full-coverage path planning device 71 can be any of the devices described in the above embodiments. The specific implementation process and principle of the dynamic full-coverage path planning device can be found in the corresponding descriptions of the above embodiments, and will not be repeated here. It should be noted that the cleaning equipment in the embodiments of the present invention can be an unmanned cleaning vehicle, an unmanned sweeper, a sweeping robot, etc., with automatic cleaning functions.

[0168] In some embodiments, the present invention provides a non-volatile computer-readable storage medium storing one or more programs including execution instructions, which can be read and executed by electronic devices (including but not limited to computers, servers, or network devices) to perform the dynamic full-coverage path planning method of any of the above embodiments of the present invention.

[0169] In some embodiments, the present invention also provides a computer program product, the computer program product including a computer program stored on a non-volatile computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the dynamic full-coverage path planning method of any of the above embodiments.

[0170] In some embodiments, the present invention also provides an electronic device comprising: at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the dynamic full-coverage path planning method of any of the above embodiments.

[0171] In some embodiments, the present invention also provides a storage medium storing a computer program, characterized in that the program, when executed by a processor, implements the dynamic full-coverage path planning method of any of the above embodiments.

[0172] Figure 30 This is a schematic diagram of the hardware structure of an electronic device for performing a dynamic full-coverage path planning method according to another embodiment of this application, as shown below. Figure 30 As shown, the device includes:

[0173] One or more processors 610 and memory 620, Figure 30 Take the 610 processor as an example.

[0174] The device for performing the dynamic full-coverage path planning method may also include an input device 630 and an output device 640.

[0175] The processor 610, memory 620, input device 630, and output device 640 can be connected via a bus or other means. Figure 30 Taking the example of a connection between China and Israel via a bus.

[0176] The memory 620, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the dynamic full-coverage path planning method in the embodiments of this application. The processor 610 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 620, thereby implementing the dynamic full-coverage path planning method of the above-described method embodiments.

[0177] The memory 620 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the dynamic full-coverage path planning method, etc. Furthermore, the memory 620 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 620 may optionally include memory remotely located relative to the processor 610, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0178] Input device 630 can receive input digital or character information, and generate signals related to user settings and function control of the dynamic full-coverage path planning device. Output device 640 may include display devices such as a display screen.

[0179] The one or more modules are stored in the memory 620, and when executed by the one or more processors 610, they execute the dynamic full-coverage path planning method in any of the above method embodiments.

[0180] The above-described product can perform the methods provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects for performing the methods. Technical details not described in detail in this embodiment can be found in the methods provided in the embodiments of this application.

[0181] The electronic devices in this application embodiments exist in various forms, including but not limited to:

[0182] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and primarily aim to provide voice and data communication. These terminals include: smartphones (e.g., iPhones), multimedia phones, feature phones, and low-end phones, etc.

[0183] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, possessing computing and processing capabilities, and generally also have mobile internet access features. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0184] (3) Portable entertainment devices: These devices can display and play multimedia content. This category includes: audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.

[0185] (4) Server: A device that provides computing services. The components of a server include a processor, hard disk, memory, system bus, etc. Servers are similar to general computer architectures, but because they need to provide highly reliable services, they have higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0186] (5) Other electronic devices with data interaction functions.

[0187] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0188] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0189] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A dynamic full-coverage path planning method, characterized in that, The method includes: The coverage status map corresponding to the target area is dynamically updated based on the acquired environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas, and uncleaned areas are marked as uncovered areas. When the preset triggering condition occurs, a region to be planned is selected from the uncovered region according to the coverage status map, and full coverage path planning is performed on the selected region to be planned to generate a cleaning path corresponding to the region to be planned. The generated cleaning path includes a contour path. Perform a first optimization process on the outermost ring path of the cleaning path corresponding to the area to be planned; and / or perform a second optimization process on the second outermost ring path of the cleaning path corresponding to the area to be planned. The first optimization process includes using the outermost path before optimization in the clean path as a reference line, performing dynamic planning sampling on the outermost path, performing collision detection on the sampling results during the sampling process, filtering out the sampling curves that have collisions to obtain the sampling results that do not have collisions, and splicing the sampling results obtained after collision detection and filtering to form the optimized outermost path. The second optimization process includes calculating the curvature of the outermost path, determining the concavity or convexity of the outermost path based on the sign of its curvature, and performing smooth optimization on the innermost path by setting the size of the quadratic programming constraint. This minimizes the offset optimization of the innermost path at the concave points of the outermost path and maximizes the offset optimization at the convex points of the outermost path.

2. The method according to claim 1, characterized in that, The environmental change information includes newly added first obstacles, and the dynamic updating of the coverage status map corresponding to the target area based on the acquired environmental change information includes: If the first obstacle is located in an uncovered area, then the first obstacle is added to the corresponding position in the uncovered area; The preset triggering conditions include the presence of a newly added first obstacle in the acquired environmental change information and the newly added first obstacle being located in an uncovered area; the selected path planning area is an uncleaned area of ​​the target area containing the first obstacle.

3. The method according to claim 1, characterized in that, Environmental change information includes the removal of a second obstacle, and the dynamic updating of the coverage status map corresponding to the target area based on the acquired environmental change information includes: The area occupied by the second obstacle will be updated to an uncovered area. The preset triggering conditions include the presence of a removed second obstacle in the acquired environmental change information; the selected path planning area is an uncleaned area of ​​the target area containing the second obstacle.

4. The method according to claim 1, characterized in that, Environmental change information includes actual cleaning trajectories, and the dynamic updating of the coverage status map corresponding to the target area based on the acquired environmental change information includes: The newly added cleaned area is determined based on the preset cleaning width and the actual cleaning trajectory, and the newly added cleaned area is marked as the covered area.

5. The method according to claim 1, characterized in that, The preset triggering conditions also include the completion of the current cleaning sub-area, and / or the arrival of the preset execution cycle, and / or the arrival of the preset execution time; the selected path planning area is an uncovered area in the coverage status map that meets the preset cleaning rules.

6. The method according to any one of claims 1 to 5, characterized in that, The process of performing full-coverage path planning for the selected area to be planned specifically includes: Extract the contour boundary of the area to be planned and smooth the contour boundary; The contour boundary is mapped onto the raster map corresponding to the area to be planned, and a distance transformation map of the raster map is obtained by performing a distance transformation on the raster map mapped with the contour boundary. In the distance transformation diagram, contour line paths are generated parallel to the contour boundary, using the contour boundary as a reference. Smooth the paths of each contour line separately; A clean path is formed for the area to be planned based on the smoothed contour lines.

7. The method according to claim 6, characterized in that, Before smoothing each contour path individually, the method further includes: Based on the size of the area enclosed by the generated contour paths, delete contour paths that do not meet the preset requirements; The area enclosed by the deleted contour lines is filled with a "bow" shape to form a clean "bow" shaped path.

8. The method according to claim 6, characterized in that, The method further includes: The cleaning path corresponding to the area to be planned is simulated to determine the missed areas, and the Vino Line coverage path of the grid map is generated based on the determined missed areas. The generated Vino Line coverage path is then added to the cleaning path of the area to be planned.

9. A dynamic full-coverage path planning device, characterized in that, include Memory, used to store executable instructions; and A processor for executing executable instructions stored in memory, which, when executed by the processor, cause the processor to perform the dynamic full-coverage path planning method as described in any one of claims 1 to 8.

10. A dynamic full-coverage path planning device, characterized in that, The device includes: The status update module is used to dynamically update the coverage status map corresponding to the target area based on the acquired environmental change information. In the coverage status map, cleaned areas in the target area are marked as covered areas, and uncleaned areas are marked as uncovered areas. The dynamic scheduling module is used to detect trigger conditions in real time, and when the preset trigger conditions occur, it selects the area to be planned in the uncovered area according to the coverage status map and calls the path planning module to perform full coverage path planning. The path planning module is used to perform full-coverage path planning for the selected area to be planned, and generate a cleaning path corresponding to the area to be planned, wherein the generated cleaning path includes a contour path. The path planning module includes a first optimization processing unit and / or a second optimization processing unit, wherein the first optimization processing unit is used to perform a first optimization processing on the outermost ring path in the cleaning path corresponding to the area to be planned, and the second optimization processing unit is used to perform a second optimization processing on the second outermost ring path in the cleaning path corresponding to the area to be planned. The first optimization process includes using the outermost path before optimization in the clean path as a reference line, performing dynamic planning sampling on the outermost path, performing collision detection on the sampling results during the sampling process, filtering out the sampling curves that have collisions to obtain the sampling results that do not have collisions, and splicing the sampling results obtained after collision detection and filtering to form the optimized outermost path. The second optimization process includes calculating the curvature of the outermost path, determining the concavity or convexity of the outermost path based on the sign of its curvature, and performing smooth optimization on the innermost path by setting the size of the quadratic programming constraint. This minimizes the offset optimization of the innermost path at the concave points of the outermost path and maximizes the offset optimization at the convex points of the outermost path.

11. A cleaning device, characterized in that, include: body; and The dynamic full-coverage path planning device according to claim 9 or 10 is disposed on the body of the device.

12. A storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1 to 8.

13. A computer program product containing instructions, characterized in that, When the computer program product is run on a computer, the computer performs the dynamic full-coverage path planning method as described in any one of claims 1 to 8.

Images