WO2023124621A1

WO2023124621A1 - Path planning method and system based on obstacle marker, and self-moving robot

Info

Publication number: WO2023124621A1
Application number: PCT/CN2022/132723
Authority: WO
Inventors: 盛蕴霞; 丘伟楠; 张聪
Original assignee: 追觅创新科技(苏州)有限公司
Priority date: 2021-12-31
Filing date: 2022-11-18
Publication date: 2023-07-06
Also published as: CN116414118A

Abstract

A path planning method and system based on an obstacle marker, and a self-moving robot. The method comprises the following steps: planning an obstacle-free navigation path for a self-moving robot from the current starting point to a target point (S10); if the obstacle-free navigation path cannot be obtained, identifying an obstacle path segment available for passing (S20); constructing a navigation passage from a starting point to a target point according to the identified obstacle path segment (S30); and controlling the self-moving robot to move to the target point according to the navigation passage (S40). Therefore, the path planning and navigation capabilities of a self-moving robot are improved, thereby improving the intelligence of the self-moving robot.

Description

A path planning method, system and self-mobile robot based on obstacle marking

This disclosure claims the priority of the following patent application: a Chinese patent submitted to the China Patent Office on December 31, 2021, with the application number 202111681899.4, and the title of the invention is "A Method, System and Self-Moving Robot for Path Planning Based on Obstacle Marking" application; the entire content of the aforementioned patent application is incorporated by reference into this disclosure.

technical field

The invention belongs to the technical field of robot path planning, and in particular relates to a path planning method and system based on obstacle marks and a self-moving robot.

Background technique

With the advancement of technology, cleaning robots have gradually entered people's daily lives. To complete an efficient and intelligent cleaning task in an indoor environment, a cleaning robot needs to plan a path in the working scene with a short path distance, high walking efficiency and good safety performance. At the same time, the robot is required to avoid all static and dynamic obstacles along the way, and while pursuing efficient cleaning, reduce the wear and tear of the robot as much as possible to prolong its service life. Therefore, the point-to-point (PTP) path planning technology of mobile robots has important application value and has become a research hotspot of researchers at home and abroad. Currently, many PTP path planning algorithms have been proposed for intelligent robots.

However, as people rely more and more on cleaning robots in their daily lives, they hope that they can meet the needs of cleaning the whole house to the greatest extent. At the same time, they hope to cope with the complex and changeable characteristics of the indoor environment. high demands.

Existing cleaning robots use the PTP path search algorithm for path planning and navigation. In one case, when the cleaning robot enters a room, if the door of the room is closed or there is an obstacle at the door, the cleaning robot uses the PTP path when leaving the room The search algorithm cannot find an unobstructed path out of the room, causing the search algorithm to fail. At this time, the cleaning robot will be trapped in the room and do not know how to walk.

It can be seen that the traditional PTP algorithm can no longer meet the needs of users for path planning and navigation of cleaning robots in various refined scenarios. Therefore, how to improve the path planning and navigation capabilities of cleaning robots in refined scenes has become one of the key technologies in the research of cleaning robots.

Contents of the invention

Therefore, the technical problem to be solved by the present invention is that the existing self-mobile robot cannot plan a navigation path by using the traditional path navigation algorithm, resulting in that the self-mobile walking device does not know how to walk to reach the target point, and needs manual troubleshooting, resulting in poor user experience.

In order to solve the above technical problems, the present invention provides a path planning method based on obstacle markings for self-mobile robots, including:

Planning an unobstructed navigation path from the current starting point to the target point of the self-mobile robot;

In the event that said navigation path without obstacles is unavailable, identifying path segments with obstacles to pass through;

constructing a navigation path from the starting point to the target point based on the identified obstacle path segment;

controlling the self-mobile robot to move to the target point according to the navigation path.

In one of the embodiments,

The identification of obstacle path segments with possible passage includes:

The obstacle path segment is located on the connecting path between the current area and the area where the target point is located.

In one of the embodiments, the constructing the navigation path from the starting point to the target point according to the identified obstacle path segment specifically includes:

Using a common local navigation mode, planning a first partial navigation path from the starting point to the proximal end of the obstacle path segment, and planning a second partial navigation path from the far end of the obstacle path segment to the target point;

The first partial navigation path, the obstacle path segment and the second partial navigation path are sequentially connected end-to-end to construct the navigation path.

In one of the embodiments, the controlling the self-mobile robot to move to the target point according to the navigation path includes:

In the normal local navigation mode, controlling the self-mobile robot to move to the proximal end of the obstacle path segment according to the first local navigation path;

When the self-mobile robot moves to the near end of the obstacle path segment, controlling the self-mobile robot to switch from the normal local navigation mode to the fine local navigation mode;

In the fine local navigation mode, acquiring current environment information, and judging whether the obstacle path segment is passable based on the environment information;

If the judgment result is passable, the self-mobile robot is controlled to move along the obstacle path segment to the far end of the obstacle path segment.

In one of the embodiments, after controlling the self-mobile robot to move along the obstacle path segment to the far end of the obstacle path segment, it includes:

controlling the self-mobile robot to switch from the fine local navigation mode to the common local navigation mode;

Controlling the self-mobile robot to move from the current position to the target point in a common local navigation mode.

In one of the embodiments, the identification of obstacle path segments with possible passage includes:

If the target point is located in an area that has been traveled during the execution of the current work task, obtain the trajectory map of the self-mobile robot performing the current work task, and determine the obstacle path segment based on the trajectory map; wherein, the obstacle path Segments are located on the trajectory map.

If the target point is located in an area that the current work task does not drive through, constructing the obstacle path segment with predetermined attributes; or,

Obtaining a historical track map saved during the historical cleaning process, and determining the obstacle path segment through the historical track map; wherein, the obstacle path segment is located on the historical track map.

In one of the embodiments, the construction of the obstacle path segment with predetermined attributes specifically includes:

Identify whether the area where the target point is located is connected to the current area;

In the case where the area where the identification target point is located is connected to the current area, the surrounding environment information is scanned, and the obstacle path segment with the possibility of passing is identified based on the surrounding environment information.

In one of the embodiments, the constructing the obstacle path segment with predetermined properties includes:

The obstacle path segment with predetermined attributes is constructed by using SLAM map and/or AI intelligent recognition technology.

In one of the embodiments, the construction of the obstacle path segment with predetermined attributes through SLAM map and/or AI intelligent recognition technology includes:

Use the SLAM map to identify whether the area where the target point is located is connected to the current area;

In the case that the area where the identification target point is located is connected to the current area, the surrounding environment information is scanned by AI intelligent identification technology, and the obstacle path segment with possible passage is identified based on the surrounding environment information.

Obtain the position coordinates of the obstacle whose attribute is marked;

The obstacle path segment is determined according to the location coordinates of the obstacle marked with attributes.

In addition, the present invention also provides a path planning system based on obstacle markings, including:

The navigation module is used to plan an obstacle-free navigation path from the current starting point to the target point of the self-mobile robot;

The obstacle path identification module is connected in communication with the navigation module, and is used to identify the obstacle path segment with possible passage when the obstacle-free navigation path cannot be obtained;

A navigation path planning module, communicatively connected with the obstacle path identification module, for constructing a navigation path from the starting point to the target point based on the obstacle path segment;

The control module is connected in communication with the navigation path planning module, and is used to control the self-mobile robot to move to the target point according to the navigation path.

In addition, the present invention also provides a self-moving robot for walking and working automatically in the working area, including:

body;

a controller, arranged on the body;

Wherein, the controller is used for:

Obtain an unobstructed navigation path from the current starting point to the target point of the self-mobile robot;

When the obstacle-free navigation path cannot be obtained, identifying obstacle path segments with possible passage;

The technical scheme provided by the invention has the following advantages:

The path planning method and system based on obstacle marks provided by the present invention, and the self-mobile robot, in the case that the traditional navigation algorithm cannot obtain an obstacle-free navigation path, by identifying the obstacle path segment with the possibility of passing, and according to the identified obstacle path Construct a navigation path from the current starting point to the target point, and control the self-mobile robot to move to the target point according to the navigation path, thereby improving the chance and possibility of the self-mobile robot to reach the target point, making the self-mobile robot more intelligent and improving the user experience , which reduces the embarrassment that the self-mobile robot is stuck in the current area and does not know how to walk when the transmission navigation algorithm cannot plan the navigation path.

Description of drawings

In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings that need to be used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description These are some implementations of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without creative work.

FIG. 1 is a schematic flowchart of a path planning method based on obstacle markers provided by an embodiment of the present invention;

Fig. 2 is a simple schematic diagram of a navigation path across two connected domains provided by an embodiment of the present invention;

Fig. 3 is a simple schematic diagram of a navigation path across two connected domains provided by another embodiment of the present invention;

FIG. 4 is a schematic diagram of a simple scene of a common local navigation mode in a single-connected domain provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of a simple scene of a local fine navigation mode of an obstacle path segment provided by an embodiment of the present invention;

Fig. 6 is a schematic diagram of a simple scene of a local fine navigation mode of an obstacle (step) path segment provided by an embodiment of the present invention;

Fig. 7 is a path planning system based on obstacle markers provided by an embodiment of the present invention.

Explanation of reference signs:

102-navigation module; 104-obstacle path recognition module; 106-navigation path planning module; 108-control module.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Apparently, the described embodiments are part of the embodiments of the present invention, but not all of them. Hereinafter, the present invention will be described in detail with reference to the drawings and examples. It should be noted that, in the case of no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

It should be noted that the terms "first" and "second" in the description and claims of the present invention and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence.

In the present invention, unless stated otherwise, the used orientation words such as "upper, lower, top, bottom" are usually for the directions shown in the drawings, or for the parts themselves in the vertical, In terms of vertical or gravitational direction; similarly, for the convenience of understanding and description, "inner and outer" refer to the inner and outer relative to the outline of each component itself, but the above orientation words are not used to limit the present invention.

This embodiment provides a path planning method based on obstacle markings for self-mobile robots.

The above-mentioned self-mobile robot is a robot that automatically performs work tasks in a work area. In a specific implementation scenario, the self-mobile robot is a cleaning robot, and correspondingly, the working area is the floor of the room to be cleaned. The cleaning robot automatically executes a cleaning plan covering the floor in the room. In another specific implementation scenario, the self-mobile robot is a mowing robot, and correspondingly, the working area is a lawn to be mowed. Currently, the self-mobile robot mentioned above may also include other types of robots, such as inspection robots, nanny robots, etc., which are not limited here.

At present, many PTP (Point-to-Point, abbreviated as PTP) path planning algorithms have been proposed for self-mobile robots, such as Probabilistic Roadmap (Probabilistic Roadmap, PRM), Rapidly-Exploring Random Tree Algorithm (Rapidly-Exploring Random Tree) , RRT), Artificial Potential Field (APF), A star algorithm and some other heuristic algorithms. Compared with these algorithms, A star algorithm is widely used because it combines the ideas of BFS ((Breadth First Search) algorithm and Dijkstra algorithm, and has the advantages of high search efficiency and short planning path. These algorithms are used in path planning , it is necessary to plan a path with short distance, high walking efficiency and good safety performance in the working scene. At the same time, it is required that the self-mobile robot can avoid all static and dynamic obstacles along the way, while pursuing efficient cleaning, as much as possible Reduce wear and tear on cleaning robots to extend their lifespan.

However, for the navigation of complex work scenes, since the work area usually includes multiple different areas, and there are interconnected paths between multiple different areas, multiple different areas can also be called multiple different connected domains . For example, as shown in Figure 2 and Figure 3, when the starting point (English name start) and the target point (English name goal) of the self-mobile robot are in different connected domains, the self-mobile robot needs to be between two different connected domains. Find a passable path. However, the connected path between these two different connected domains is narrow and prone to closed situations. In this case, all possible passage paths from the starting point (Start) to the goal point (Goal) are blocked. , since the mobile robot cannot obtain an obstacle-free navigation path based on existing navigation algorithms.

In order to improve the intelligence of the self-mobile robot and increase the possibility of the self-mobile robot reaching the target point when the self-mobile robot cannot obtain an unobstructed navigation path. The present invention provides a method for path planning based on obstacle markings. When the method is specifically implemented, it may include the following steps:

S10. Planning an obstacle-free navigation path for the self-mobile robot from the current starting point to the target point;

S20. In the case where the obstacle-free navigation path cannot be obtained, identify an obstacle path segment that is possible to pass through;

S30. Construct a navigation path from the starting point to the target point according to the identified obstacle path segment;

S40. Control the self-mobile robot to move to the target point according to the navigation path.

In an implementation scenario, the aforementioned self-mobile robot is a cleaning robot. As people rely more and more on cleaning robots in their daily lives, they hope that they can meet the needs of cleaning the whole house to the greatest extent, and at the same time, they hope that they can cope with the complex and changeable characteristics of the indoor environment. Usually, for indoor work scenarios, the work area includes multiple different rooms, and each room has interconnected paths, so each room is equivalent to a connected domain. Exemplarily, the living room and the bedroom are two different connected domains, which are connected by opening the bedroom door. Therefore, the living room and the bedroom can be understood as two different connected domains. If the cleaning robot is currently located in the living room and the target point is located in the bedroom, the planned navigation path is a navigation path that crosses two connected domains. It is understandable that, compared with navigation within a connected domain, navigation across two connected domains needs to pass through a narrow connected path between two different connected domains, and the connected path is prone to be closed, which can easily lead to poor navigation path planning. Failure, causing the cleaning robot to stop in the current connected domain and do not know how to walk, it seems very clumsy and not intelligent enough. That is to say, when the starting point (starting point, Start) and the target point (end point, Goal) of the cleaning robot are in different connected domains (rooms), the cleaning robot needs to find a path between the two different connected domains. common path.

In the above step S10, the sampled navigation algorithm is the existing PTP route navigation algorithm. In one embodiment, the navigation algorithm adopts the A Star algorithm. Of course, other route navigation algorithms may also be used, which is not limited here. These navigation algorithms can plan a navigation path when there is a path from the current starting point to the target point. And when all the passing paths from the current starting point to the target point are blocked by obstacles, these conventional navigation algorithms will fail, and the above-mentioned navigation paths cannot be planned.

Among them, the starting point can be referred to simply as the starting point, which is called Start in English, and the target point can also be called the end point, which is called Goal in English.

In the event that an unobstructed navigation path is not available, path segments with obstacles are identified that are possible to pass through. The aforementioned "passable obstacle path segment" is an obstacle path segment that is currently identified as impassable by the navigation algorithm. At the same time, considering the attribute characteristics of the marked obstacles of the obstacle path segment, the identification still has the possibility of passing. Here, having "passability" should be understood as that the obstacle path segment is located on the connected path between the current area and the area where the target point is located, which can be the connected path indicated by the SLAM map or the connected path indicated by the track map. Can be a previously traveled path segment. Therefore, the self-mobile robot believes that the obstacle path segment still has the possibility to pass again. For example, the aforementioned obstacle path segment may be a currently closed door or a step marked as an obstacle.

A "navigation path" is actually a navigation path through obstacles. When all passages are blocked by obstacles, it can be considered that the self-mobile robot can still try to pass through the obstacle path segment with possible passage to reach the destination. In the implementation scenario of the cleaning robot, when someone is standing at the door, the path planning fails. By identifying the possible obstacle path segment, the self-mobile robot can still walk to the door through ordinary navigation. At this time, it is possible that the person has already If you leave the door, even if you haven't left, people will usually take the initiative to get out of the way and let the mobile robot pass. Therefore, for this situation, the construction of the above-mentioned navigation pathway obviously improves the intelligence and flexibility of the self-mobile robot, and has high practical value.

The path planning method based on obstacle markers provided in this embodiment, in the case that the traditional navigation algorithm cannot obtain an obstacle-free navigation path, identifies the obstacle path segments that have the possibility of passing, and constructs The navigation path from the starting point to the target point controls the self-mobile robot to move to the target point according to the navigation path, thereby improving the chance and possibility of the self-mobile robot to reach the target point, making the self-mobile robot more intelligent, improving the user experience, and reducing the self-moving The robot is trapped in the current area and does not know how to walk when the navigation path cannot be planned by the transmission navigation algorithm.

In one embodiment, step S30 is also the step of "constructing a navigation path from the starting point to the target point according to the identified obstacle path segment", which specifically includes:

S31. Using the common local navigation mode, plan a first local navigation path from the starting point to the near end of the obstacle path segment, and plan a second local navigation path from the far end of the obstacle path segment to the target point ;

S32. Connect the first partial navigation path, the obstacle path segment, and the second partial navigation path sequentially end-to-end to construct the navigation path.

The above common local navigation mode is understood as the segmented navigation of the cleaning robot in the same connected domain, the corresponding navigation algorithm adopted is the common navigation algorithm, and the corresponding navigation mode is the common local navigation mode. As shown in Figure 4, the current location is used as the starting point, and the end of the obstacle path segment in the current connected domain is used as the intermediate target point to perform ordinary navigation. Since the current starting point and the intermediate target point are located in the same connected domain, current navigation can be performed. Ordinary navigation in the connected domain obtains the first partial navigation path. Correspondingly, the far end of the obstacle path segment and the target point are located in the connected domain of the target, which is also the same connected domain, and ordinary navigation in the target connected domain can be performed, thereby obtaining the second partial navigation path.

According to the first partial navigation path, the obstacle path segment and the first part of the second partial navigation path are connected to form a navigation path. It can be understood that the above-mentioned navigation path is not a path that enables the self-mobile robot to pass without obstacles in an absolute sense, but a navigation path constructed in segments, and the obstacle path segment is located in the middle part of the navigation path.

In one embodiment, step S40 is the step of "controlling the self-mobile robot to move to the target point according to the navigation path", which specifically includes:

S41. In the normal local navigation mode, control the self-mobile robot to move to the near end of the obstacle path segment according to the first local navigation path;

S42. When the self-mobile robot moves to the near end of the obstacle path segment, control the self-mobile robot to switch from the normal local navigation mode to the fine local navigation mode;

S43. In the fine local navigation mode, acquire current environment information, and judge whether the obstacle path segment is passable based on the environment information;

S44. If the judgment result is passable, control the self-mobile robot to move along the obstacle path segment to the far end of the obstacle path segment.

That is to say, the above steps first control the self-mobile robot to walk to the near end of the obstacle path segment through ordinary local navigation, and complete the walking of the first local navigation path. After reaching the obstacle path segment, switch to the fine local navigation mode to cope with the fine navigation of the obstacle path segment, and improve the possibility of successfully passing the obstacle path segment navigation.

Figures 5 and 6 illustrate fine local navigation for two different scenarios. Identify whether the current obstacle path segment is passable by scanning the surrounding environment information in fine local navigation mode. In Fig. 5, the obstacle path segment has been completely closed, and the judgment result is that the current obstacle path segment is impassable, and the self-controlled mobile robot abandons the navigation of the obstacle path segment. The obstacle path section shown in FIG. 6 is a step obstacle, and the self-mobile robot can be controlled to pass through the obstacle path section according to the judgment result of the environment information in S43. Specifically, in order to ensure passing efficiency, an accelerated passing method is adopted, that is, accelerated obstacle navigation.

That is to say, in the fine local navigation mode, the self-mobile robot obtains the current environment information, and judges whether the obstacle path segment is passable based on the environment information. If it is not passable, the navigation of the obstacle path segment is directly abandoned. For example, if a closed door is detected and it is impossible to pass through, it will be given up directly. When the environment information analysis shows that the result is passable, the self-mobile robot is directly controlled to move along the obstacle path segment to the far end of the obstacle path segment.

In one embodiment, after the self-mobile robot moves to the near end of the obstacle path segment according to the first local navigation path, the self-mobile robot uses the current position as the starting point and uses the far end of the obstacle path segment as the terminal to perform path planning and navigation, and obtains A navigation path passing through the obstacle path segment, the self-mobile robot is controlled to walk through the obstacle path segment along the navigation path, and reach the target connected domain.

In one embodiment, the self-mobile robot analyzes the type of the obstacle based on the current environment information, and when the type of the obstacle is surmountable, such as a low step, the self-mobile robot is controlled to try to pass the path segment of the obstacle.

In one embodiment, after step S44, that is, after the step of "controlling the self-mobile robot to move along the obstacle path segment to the far end of the obstacle path segment", it also includes:

S441. Control the self-mobile robot to switch from the fine local navigation mode to the common local navigation mode;

S442. Control the self-mobile robot to move from the current position to the target point in a common local navigation mode.

That is, when the self-mobile robot has reached the target connected domain after passing through the obstacle path segment, the self-mobile robot automatically switches the local fine navigation mode to the normal local navigation mode, and uses the normal local navigation mode to complete the walking of the second local navigation path.

The above-mentioned "passable obstacle path segment" includes various types. In different work scenarios, the identification of obstacle path segments also differs due to the different types of obstacles. In one scenario, the target point is located in an area that has been traveled during the execution of the current work task, that is, the area of the target point has been reached before. In this case, the obstacle path segment can be identified based on the historical trajectory map.

Therefore, in one embodiment, "identifying obstacle path segments with possible passage" in step S20 specifically includes:

If the area where the target point is located has not been reached during the execution of the current work task, but has been reached during the execution of the previous work task. For example, in the cleaning robot scene, the cleaning robot has not reached the master bedroom during this cleaning work, but based on the previous historical trajectory map, the master bedroom has been reached. If the target point is located in the master bedroom, it can be based on the historical trajectory The map identifies obstacle path segments with traversable potential.

Therefore, in another embodiment, in step S20, "identifying obstacle path segments with possible passage" specifically includes:

The above-mentioned "historical trajectory map" is also the trajectory map saved in the past work process.

In another embodiment, in step S20, "identifying obstacle path segments with possible passage" specifically includes:

If the target point is located in an area not passed by the current work task, the obstacle path segment with predetermined attributes is constructed through SLAM map and AI intelligent recognition technology.

Specifically, in one embodiment, the above step of "constructing the obstacle path segment with predetermined attributes through SLAM map and/or AI intelligent recognition technology" includes:

Specifically, the SLAM map shows that the area where the target point is located is connected to the current area, and further uses AI intelligent recognition technology to identify the attributes of the obstacle path segment, for example, the area recognized by AI as a "door" will be recognized as a "door" area Constructed as obstacle path segments. AI can also identify many other obstacles with predetermined attributes, which are not completely impassable, such as sliding doors, movable tables and chairs, etc. In this embodiment, through the AI recognition technology, the attributes of obstacles can be analyzed. When the preset predetermined attributes are met, the path where the obstacle is located can be constructed as an obstacle path segment, thereby establishing a navigation path and improving the self-moving robot. work efficiency.

In the embodiment of the cleaning robot, the cleaning robot will set up a virtual wall by marking the position of some special areas encountered during the working process, such as the carpet area, and adopt a special cleaning method during normal cleaning , avoid entering these marked areas. In another case, before the cleaning robot completes the cleaning of the current area, the cleaning robot is controlled not to climb over these special areas, such as door steps and sliding door slide rails. After the cleaning of the current area is completed, the cleaning robot is controlled to climb over the special obstacle. area.

In order to deal with the above specially marked obstacle path segment situation. In one embodiment, "identifying obstacle path segments with possible passage" in step S20 specifically includes:

Obtain the position coordinates of the obstacle whose attribute is marked; wherein, the obstacle includes one or more of the sliding door guide rail area, the carpet area, and the door step;

The above-mentioned "obstacles with marked attributes" are "obstacles" that can be passed by the self-mobile robot. These obstacles are marked with attributes during normal work. Therefore, in the above method steps, by obtaining the Position coordinates, obstacles with these marked attributes can be used to construct obstacle path segments.

The path planning method based on obstacle marks provided by the present invention is oriented to the design of PTP path planning and navigation schemes in refined scenarios. Existing PTP path planning and navigation technologies, when faced with cleaning tasks in complex scenes, usually show increased navigation collisions, poor arrival and escape capabilities in small spaces, etc. The present invention improves the arrival rate of the self-moving robot to narrow spaces by designing a path planning scheme that crosses connected domains and can support refined navigation, and at the same time, enhances the ability of the self-mobile robot to get out of trouble in a narrow space and its ability in fine navigation. The passability of navigation in the scene.

Referring to FIG. 7, the present invention also provides a path planning system 100 based on obstacle markings, including:

The navigation module 102 is used to plan an unobstructed navigation path from the current starting point to the target point from the mobile robot;

Obstacle path identification module 104, communicated with the navigation module 102, used to identify obstacle path segments with possible passage when an unobstructed navigation path cannot be obtained;

The navigation path planning module 106 is communicatively connected with the obstacle path identification module 104, and is used to construct the navigation path from the starting point to the target point based on the obstacle path segment;

The control module 108 is in communication connection with the navigation path planning module 106, and is used for controlling the self-mobile robot to move to the target point according to the navigation path.

The path planning system 100 based on obstacle marks described in this embodiment corresponds to the path planning method based on obstacle marks described above. In this embodiment, the functions of each module in the path planning system 100 based on obstacle marks in the corresponding method embodiment described in detail and will not be repeated here.

The present invention also provides a self-moving robot, which is used for walking and working automatically in a working area, comprising: a body; and a controller arranged on the body.

Among them, the controller is used for:

Obtain an unobstructed navigation path from the mobile robot from the current starting point to the target point;

Similarly, the function of the controller is to implement the above-mentioned path planning method based on obstacle markings. For details, please refer to the description of the above-mentioned cleaning control method, which will not be repeated here.

Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, servers or computer program products. Accordingly, the present invention can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), servers and computer program products according to embodiments of the invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

Apparently, the above-described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, those skilled in the art may make other changes or changes in different forms without creative work, which shall fall within the protection scope of the present invention.

Claims

A path planning method based on obstacle markings, characterized in that it is used for a self-moving robot, the method comprising:

Planning an unobstructed navigation path from the current starting point to the target point of the self-mobile robot;

In the event that said navigation path without obstacles is unavailable, identifying path segments with obstacles to pass through;

constructing a navigation path from the starting point to the target point based on the identified obstacle path segment;

controlling the self-mobile robot to move to the target point according to the navigation path.
The path planning method based on obstacle markers according to claim 1, wherein the identification of obstacle path segments with possible passage comprises:

The obstacle path segment is located on the connecting path between the current area and the area where the target point is located.
The path planning method based on obstacle markers according to claim 1, wherein said constructing a navigation path from said starting point to said target point according to said identified obstacle path segment specifically comprises:

Using a common local navigation mode, planning a first partial navigation path from the starting point to the proximal end of the obstacle path segment, and planning a second partial navigation path from the far end of the obstacle path segment to the target point;

The first partial navigation path, the obstacle path segment and the second partial navigation path are sequentially connected end-to-end to construct the navigation path.
The path planning method based on obstacle markings according to claim 3, wherein the controlling the self-mobile robot to move to the target point according to the navigation path comprises:

In the normal local navigation mode, controlling the self-mobile robot to move to the proximal end of the obstacle path segment according to the first local navigation path;

When the self-mobile robot moves to the near end of the obstacle path segment, control the self-mobile robot to switch from the common local navigation mode to the fine local navigation mode;

In the fine local navigation mode, acquiring current environment information, and judging whether the obstacle path segment is passable based on the environment information;

If the judgment result is passable, the self-mobile robot is controlled to move along the obstacle path segment to the far end of the obstacle path segment.
The path planning method based on obstacle markings according to claim 4, wherein the controlling the self-mobile robot to move along the obstacle path segment to the far end of the obstacle path segment includes:

controlling the self-mobile robot to switch from the fine local navigation mode to the common local navigation mode;

Controlling the self-mobile robot to move from the current position to the target point in a common local navigation mode.
The path planning method based on obstacle markers according to any one of claims 1-5, wherein the identification of obstacle path segments with possible passage includes:

If the target point is located in an area that has been traveled during the execution of the current work task, obtain the trajectory map of the self-mobile robot performing the current work task, and determine the obstacle path segment based on the trajectory map; wherein, the obstacle path Segments are located on the trajectory map.
The path planning method based on obstacle markers according to any one of claims 1-5, wherein the identification of obstacle path segments with possible passage includes:

If the target point is located in an area that the current work task does not drive through, constructing the obstacle path segment with predetermined attributes; or,

Obtaining a historical track map saved during the historical cleaning process, and determining the obstacle path segment through the historical track map; wherein, the obstacle path segment is located on the historical track map.
The path planning method based on obstacle markings according to claim 7, wherein said constructing said obstacle path segment of predetermined attribute specifically comprises:

Identify whether the area where the target point is located is connected to the current area;

In a case where the area where the identified target point is located is connected to the current area, the surrounding environment information is scanned, and the obstacle path segment with possible passage is identified based on the surrounding environment information.
The path planning method based on obstacle markers according to claim 8, wherein said constructing said obstacle path segment with predetermined attributes comprises:

The obstacle path segment with predetermined attributes is constructed by using SLAM map and/or AI intelligent recognition technology.
The path planning method based on obstacle markings according to claim 9, wherein said constructing said obstacle path segment with predetermined attributes through SLAM map and/or AI intelligent recognition technology comprises:

Use the SLAM map to identify whether the area where the target point is located is connected to the current area;

In the case that the area where the identification target point is located is connected to the current area, the surrounding environment information is scanned by AI intelligent identification technology, and the obstacle path segment with possible passage is identified based on the surrounding environment information.
The path planning method based on obstacle markers according to any one of claims 1-5, wherein the identification of obstacle path segments with possible passage includes:

Obtain the position coordinates of the obstacle whose attribute is marked;

The obstacle path segment is determined according to the location coordinates of the obstacle marked with attributes.
A path planning system based on obstacle marking, characterized in that it includes:

The navigation module is used to plan an obstacle-free navigation path from the current starting point to the target point of the self-mobile robot;

The obstacle path identification module is connected in communication with the navigation module, and is used to identify the obstacle path segment with possible passage when the obstacle-free navigation path cannot be obtained;

A navigation path planning module, communicatively connected with the obstacle path identification module, for constructing a navigation path from the starting point to the target point based on the obstacle path segment;

The control module is connected in communication with the navigation path planning module, and is used to control the self-mobile robot to move to the target point according to the navigation path.
A self-moving robot for autonomous walking and working in a working area, characterized in that it includes:

body;

a controller, arranged on the body;

Wherein, the controller is used for:

Obtain an unobstructed navigation path from the current starting point to the target point of the self-mobile robot;

When the obstacle-free navigation path cannot be obtained, identifying obstacle path segments with possible passage;

constructing a navigation path from the starting point to the target point based on the identified obstacle path segment;

controlling the self-mobile robot to move to the target point according to the navigation path.