Obstacle recognition method and apparatus, and control method for autonomous mobile robot

Abstract

Provided in the present disclosure are an obstacle recognition method and apparatus, and a control method for an autonomous mobile robot. The method comprises: obtaining feature information of an obstacle to be recognized; on the basis of the feature information, performing feature point matching on the obstacle to be recognized and a labeled obstacle to obtain a feature point matching rate; determining whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined on the basis of an avoidance weight of the labeled obstacle; and when the feature point matching rate is greater than or equal to the matching threshold, the obstacle to be recognized being categorized as the labeled obstacle. By means of the present disclosure, various types of obstacles, especially non-conventional or movable obstacles, can be effectively recognized.

Description

Obstacle identification method and device, and control method of self-moving robot

[0001] Cross-reference to related applications

[0002] The present disclosure is based on and claims priority from a Chinese patent application No. 202410904409.X filed on July 5, 2024, the content of which is hereby incorporated by reference in its entirety into the present disclosure. TECHNICAL FIELD

[0003] The present disclosure relates to the technical field of self-moving robots, in particular, to an obstacle identification method and device, and a control method of a self-moving robot. BACKGROUND

[0004] Self-moving robots need to cope with complex surrounding environment when moving. For example, when performing a cleaning task, the self-moving robot often needs to cope with various obstacles. According to different types of obstacles, the specific coping manner can be crossing or avoiding.

[0005] SUMMARY

[0006] Some embodiments of the present disclosure provide an obstacle identification method and device, and a control method of a self-moving robot, which can effectively identify various types of obstacles, especially some unconventional or movable obstacles.

[0007] An obstacle identification method is provided in an embodiment of the present disclosure, which includes: obtaining feature information of an obstacle to be identified; performing feature point matching between the obstacle to be identified and a labeled obstacle according to the feature information, to obtain a feature point matching rate; determining whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle; and when the feature point matching rate is greater than or equal to the matching threshold, the obstacle to be identified belongs to the labeled obstacle.

[0008] In some embodiments, further comprising: when the feature point matching rate is less than the matching threshold, sending a labeling request and setting the avoidance weight of the obstacle to be identified according to feedback information of the labeling request.

[0009] In some embodiments, after the step of sending a labeling request and setting the avoidance weight of the obstacle to be identified according to feedback information of the labeling request, further comprising: collecting feature information of the obstacle to be identified, and the collection amount of the feature information of the obstacle to be identified is positively correlated with the avoidance weight of the obstacle to be identified.

[0010] In some embodiments, the matching threshold is negatively correlated with the avoidance weight of the labeled obstacle.

[0011] In some embodiments, the avoidance weight is determined according to an avoidance type of the obstacle, and the more strict the avoidance type is, the higher the corresponding avoidance weight is.

[0012] In some embodiments, in the step of obtaining the feature information of the to-be-identified obstacle, the feature information of the to-be-identified obstacle includes image information of the to-be-identified obstacle.

[0013] In some embodiments, in the step of performing feature point matching between the to-be-identified obstacle and the labeled obstacle according to the feature information to obtain a feature point matching rate, the number of feature points of the labeled obstacle is positively correlated with the avoidance weight of the labeled obstacle.

[0014] In some embodiments, when the to-be-identified obstacle is a labeled obstacle, it is determined whether the position of the labeled obstacle changes; if the position changes, an electronic forbidden zone is set at the current position of the labeled obstacle, and an electronic forbidden zone at the original position of the labeled obstacle is deleted.

[0015] In some embodiments, the step of setting an electronic forbidden zone at the current position of the labeled obstacle includes: generating an electronic forbidden zone according to the avoidance weight corresponding to the labeled obstacle, wherein the range of the electronic forbidden zone is positively correlated with the avoidance weight.

[0016] Embodiments of the present disclosure provide a control method of a self-moving robot, including: obtaining feature information of a to-be-identified obstacle; performing feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information to obtain a feature point matching result; determining whether the position of the to-be-identified obstacle changes based on the matching result of the to-be-identified obstacle and the labeled obstacle; if the position changes, setting an electronic forbidden zone at the current position of the to-be-identified obstacle, and performing an obstacle avoidance operation on the newly set electronic forbidden zone.

[0017] In some embodiments, the control method further includes: deleting an electronic forbidden zone at the original position of the to-be-identified obstacle, and performing a cleaning operation on the deleted electronic forbidden zone.

[0018] Embodiments of the present disclosure provide an obstacle identification device, including: a feature information acquisition unit configured to obtain feature information of a to-be-identified obstacle; a feature point matching unit configured to perform feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information to obtain a feature point matching rate; a comparison unit configured to determine whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle; and an obstacle determination unit configured to determine that the to-be-identified obstacle belongs to the labeled obstacle when the feature point matching rate is greater than or equal to the matching threshold.

[0019] The self-moving robot provided by the embodiments of the present disclosure comprises a memory and a processor. The memory stores programs or instructions. The processor implements the obstacle identification method when executing the programs or instructions.

[0020] The non-transitory computer-readable storage medium provided by the embodiments of the present disclosure stores computer program instructions. The computer program instructions are called and executed by a processor to implement the method steps of any one of the above embodiments.

[0021] The computer program product provided by the embodiments of the present disclosure comprises computer program instructions stored in a computer-readable storage medium and adapted to be called and executed by a processor to make a computer device with the processor implement the method steps of any one of the above embodiments.

[0022] Compared with the prior art, the obstacle identification method, device and control method of the self-moving robot provided by the embodiments of the present disclosure can effectively identify various types of obstacles, especially some unconventional or movable obstacles. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure. It is clear that the accompanying drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those of ordinary skill in the art without creative effort. In the drawings:

[0024] FIG. 1 is a schematic diagram of the three-dimensional structure of a self-moving robot according to some embodiments of the present disclosure;

[0025] FIG. 2 is a schematic diagram of a self-moving robot and an obstacle according to some embodiments of the present disclosure;

[0026] FIG. 3 is a flowchart of an obstacle identification method according to some embodiments of the present disclosure;

[0027] FIG. 4 is a schematic diagram of an obstacle identification device according to some embodiments of the present disclosure;

[0028] FIGS. 5-7 are specific flowcharts of a cleaning task performed by a sweeping robot using the obstacle identification method according to the embodiments of the present disclosure;

[0029] FIG. 8 is a schematic diagram of a connection structure of a self-moving robot according to some embodiments of the present disclosure.

[0030] Explanation of reference signs:

[0031] The robot 10, the body 110, the front portion 111, the rear portion 112, the perception system 120, the camera 121, the bumper 122. DETAILED DESCRIPTION

[0032] In order to make the objects, technical solutions and advantages of the present disclosure clearer, the following further describes the present disclosure with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, and not all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present disclosure.

[0033] The terms used in the embodiments of the present disclosure are merely for the purpose of describing particular embodiments and are not intended to limit the present disclosure. The singular forms "a", "said" and "the" used in the embodiments of the present disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Plural" generally includes at least two.

[0034] It should be understood that the term "and / or" used herein is only to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0035] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that a product or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such product or device. Without more limitations, the element defined by the statement "including one" does not exclude the presence of additional identical elements in the product or device including the element.

[0036] The self-moving robot needs to cope with complex surrounding environment when traveling. For example, when performing a cleaning task, the self-moving robot often needs to cope with various obstacles. According to different types of obstacles, the specific coping method may be crossing or avoiding. If the self-moving robot cannot accurately identify and cope with the obstacles, the robot may be stuck by the obstacles or unintentionally push the obstacles during the cleaning process. Especially for unconventional obstacles with unique appearance and variable position, it is more difficult for the robot to correctly identify and bypass such obstacles, which seriously affects the efficiency and effect of the robot performing the task.

[0037] Therefore, how to effectively identify various types of obstacles has become a technical problem to be solved at present.

[0038] As mentioned above, the existing solutions for dealing with irregular or movable obstacles still need to be further improved.

[0039] For example, a common practice is that a user draws fixed electronic exclusion zones on the map of the robot, and the target obstacles are framed within the electronic exclusion zones, and the robot will automatically avoid when approaching these electronic exclusion zones. The disadvantage of this solution is that it cannot adapt to the changes in the position of the obstacles. If the obstacles are movable, the position of the obstacles marked in the fixed electronic exclusion zone may have changed at the next cleaning, resulting in poor avoidance effect and may also cause missed cleaning.

[0040] To solve the above problems, one aspect of an embodiment of the present application provides a method for identifying obstacles, comprising: obtaining feature information of a to-be-identified obstacle; performing feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching rate; and when the feature point matching rate is greater than or equal to a matching threshold, the to-be-identified obstacle belongs to the labeled obstacle, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle.

[0041] As can be seen, the obstacle identification method provided by the embodiment of the present application can effectively identify various types of obstacles, especially some irregular or movable obstacles.

[0042] The optional embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0043] The embodiment of the present application provides an obstacle identification method, which can be applied to various types of self-moving robots, such as a sweeping robot, a mopping robot, a dust collector, a weeding machine, etc. For ease of description, the control method will be described below taking the sweeping robot as an example.

[0044] Referring to FIGS. 1 and 2, the self-moving robot is a sweeping robot 10, which can further include a body 110, a perception system 120, a controller, a driving system, a cleaning system, an energy system, etc.

[0045] The body 110 includes a forward portion 111 and a rearward portion 112, wherein the forward portion 111 of the body 110 can be further provided with a bumper 122. The body 110 has a circular or approximately circular shape. In some embodiments, the body 110 is circular, and the center of the body 110 is the center of the circle. The circular design helps the sweeping robot 10 to be more flexible in obstacle avoidance when encountering obstacles.

[0046] The driving system includes a driving wheel assembly and a driven wheel assembly, which realizes the travel and steering of the sweeping robot 10.

[0047] The controller is configured to control the driving system, the cleaning system, etc., so that the robot cleaner 10 performs a cleaning operation.

[0048] The perception system 120 is configured to provide various position information and motion state information of the robot cleaner 10 to the controller. The perception system 120 can include various sensors disposed on or inside the body 110, such as image acquisition sensors, laser sensors, TOF sensors, collision sensors, cliff sensors, magnetometers, accelerometers, gyroscopes, odometers, etc.

[0049] The image acquisition sensor is configured to acquire image information around the body 110, and can provide visual information for the self-moving robot. Referring to FIG. 2, in some embodiments, the image acquisition sensor can be a camera 121 disposed at the front end of the body 110, which can acquire feature information of obstacles within a certain range in front of the body 110.

[0050] The laser sensor can be used to measure distance, and can determine the distance to an object by emitting a laser beam and detecting the reflected light. Common laser sensors include LDS laser distance sensors, laser radars, etc. The laser sensor can perform a large-angle environmental scan, such as a 360-degree environmental scan, so as to dynamically collect distance data of obstacles around the body 110.

[0051] The TOF sensor includes a type of light-based time-of-flight (TOF) distance measurement sensor, which calculates distance by emitting light pulses and measuring the time required for the pulses to reach a target and return.

[0052] Referring to FIG. 3, the present application provides a kind of obstacle identification method, suitable for various types of self-moving robot, which includes the following steps:

[0053] S110, obtaining feature information of an obstacle to be identified;

[0054] S120, according to the feature information, the feature point matching of the obstacle to be identified and the labeled obstacle is carried out, and the feature point matching rate is obtained;

[0055] S130, it is judged whether the feature point matching rate is greater than or equal to the matching threshold value, wherein the matching threshold value is determined according to the avoidance weight of the labeled obstacle;

[0056] S140, when the feature point matching rate is greater than or equal to the matching threshold value, the obstacle to be identified belongs to the labeled obstacle;

[0057] S150, when the feature point matching rate is less than a matching threshold, sending a labeling request and setting an avoidance weight of the to-be-identified obstacle according to feedback information of the labeling request.

[0058] In step S110, the feature information of the to-be-identified obstacle can be image information of the obstacle. The image information can describe the appearance characteristics of the obstacle, and the image information can further include visual-related information such as color, shape, texture, size, etc. of the obstacle.

[0059] The image information can be an image of the obstacle, which can be acquired by an image acquisition sensor (such as a camera) carried by the self-moving robot. The image information can also be one or more of color, shape, texture, size, etc. of the obstacle.

[0060] In step S120, the labeled obstacle refers to an obstacle that has been labeled. The labeling process can identify and classify the obstacle, and the data information of the labeled obstacle can be stored in the database of the self-moving robot.

[0061] In this step, the to-be-identified obstacle and the labeled obstacle are matched with feature points. Before feature point matching, the feature points of the to-be-identified obstacle need to be extracted from the feature information of the to-be-identified obstacle. The feature points or key feature points can represent and describe the image content of the obstacle, and the feature points can remain unchanged under different transformations of the image.

[0062] The feature points of the labeled obstacle have been pre-extracted. The extraction of the feature points of the labeled obstacle can be in the labeling process or in the image acquisition process after labeling. The extracted feature points of the labeled obstacle are stored in a designated database, such as a local database of the self-moving robot.

[0063] The matching process involves comparing the feature points of the to-be-identified obstacle with the feature points of the labeled obstacle in the database, and different matching algorithms can be used in the matching process. The comparison result obtained by the matching process is the feature point matching rate. The feature point matching rate can represent the similarity between the to-be-identified obstacle and a certain labeled obstacle in the database. The higher the feature point matching rate, the higher the similarity between the to-be-identified obstacle and the labeled obstacle.

[0064] In step S130, the avoidance weight is a quantitative value, which can be determined according to the avoidance type of the labeled obstacle. In some embodiments, the labeled obstacle can be further labeled as different avoidance types. The avoidance type can be set by the user, and is usually related to the potential risk, avoidance demand, avoidance degree, etc. of the obstacle.

[0065] As an example, the child's toy is labeled as a conditional avoidance type, which is suitable for obstacles that do not pose a serious threat to the self-moving robot, such as toys, clothes, etc. For obstacles of the conditional avoidance type, the self-moving robot can approach moderately and does not need to maintain a large safety distance.

[0066] As an example, the child's toy is labeled as a conditional avoidance type, which is suitable for obstacles that do not pose a serious threat to the self-moving robot, such as toys, clothes, etc. For obstacles of the conditional avoidance type, the self-moving robot can approach moderately and does not need to maintain a large safety distance.

[0067] It can be understood that the number and types of avoidance types can be determined according to the actual application environment and the hardware performance of the self-moving robot. By setting multiple levels of avoidance types, more flexible and accurate responses to different obstacles can be achieved.

[0068] In some embodiments, a corresponding avoidance weight is further generated according to the avoidance type, which is a further quantification of each avoidance type. As an example, this can be achieved by assigning a numerical value to each avoidance type, and the weight value of the avoidance weight can reflect the urgency and importance of avoidance. For example, for obstacles of the strict avoidance type, a higher weight value (such as 100) can be assigned, indicating that the self-moving robot must avoid contact in any case. For obstacles of the conditional avoidance type, the weight can be lower (such as 50), indicating that the self-moving robot needs to approach cautiously, but does not have to maintain a large safety distance.

[0069] The setting of the matching threshold is based on the avoidance weight. In some embodiments, the matching threshold is negatively correlated with the avoidance weight, that is, the matching threshold corresponding to the obstacle with a high avoidance weight is set lower, and the matching threshold corresponding to the obstacle with a low avoidance weight is set higher.

[0070] The present application associates the matching threshold with the avoidance weight, and can dynamically adjust the matching threshold of the obstacle according to the avoidance weight, so that the higher the priority of the obstacle, the lower the threshold of the matching identification, and the higher the success rate of obstacle identification, thereby maximizing the safety of the operation.

[0071] In some embodiments, the number of feature points of the labeled obstacle is also positively correlated with the avoidance weight of the labeled obstacle. During the process of collecting feature information of the obstacle, the number of collected feature information is determined based on the avoidance weight of the obstacle. That is, for the obstacle with high avoidance weight and strict avoidance requirement, the self-moving robot will record more feature points and image details, and for the obstacle with low avoidance requirement, the self-moving robot will collect relatively less feature information.

[0072] In step S140, if the matching rate of the to-be-identified obstacle is greater than or equal to the set matching threshold, it can be determined that the to-be-identified obstacle belongs to the labeled obstacle in the database.

[0073] Further, after the to-be-identified obstacle is determined to belong to the labeled obstacle, it can be further determined whether the position of the labeled obstacle changes. If the position of the labeled obstacle changes, an electronic forbidden zone is set at the current position of the labeled obstacle; if the position of the labeled obstacle changes, the electronic forbidden zone at the original position of the labeled obstacle is deleted.

[0074] In some embodiments, the electronic forbidden zone can be generated according to the avoidance weight corresponding to the labeled obstacle, wherein the range of the electronic forbidden zone is positively correlated with the avoidance weight. In step S150, when the feature point matching rate is less than the matching threshold, it indicates that the current to-be-identified obstacle cannot be matched with the obstacle stored in the database of the self-moving robot, and further measures need to be taken to identify the obstacle.

[0075] When the matching rate does not reach the threshold, the self-moving robot initiates a labeling request, which can be sent to the user through the human-computer interaction system of the self-moving robot, or can be sent to the client or remote service through the wireless network.

[0076] After sending the labeling request, the self-moving robot receives feedback information according to the labeling request, and sets the avoidance weight of the to-be-identified obstacle according to the feedback information.

[0077] In some embodiments, step S150 can further include:

[0078] S160, collecting feature information of the to-be-identified obstacle, and the amount of collected feature information of the to-be-identified obstacle is positively correlated with the avoidance weight of the to-be-identified obstacle.

[0079] In step S160, the self-moving robot further collects feature information of the to-be-identified obstacle, and determines the quantity of the collected feature information based on the avoidance weight of the obstacle. That is, for the obstacle with high avoidance weight and strict avoidance requirement, the self-moving robot records more feature points and image details, and for the obstacle with low avoidance requirement, the self-moving robot collects relatively less feature information.

[0080] The application associates the feature information collection quantity with the avoidance weight, and dynamically adjusts the feature information collection quantity of the obstacle according to the avoidance weight, so that the higher the priority of the obstacle is, the higher the feature information collection quantity is, thereby saving system resources and improving processing speed under the premise of ensuring operation safety.

[0081] Referring to FIG. 4, as a specific implementation of the self-moving robot control method, the application embodiment provides an obstacle identification device 300, which comprises:

[0082] A feature information acquisition unit 310 is configured to obtain feature information of a to-be-identified obstacle;

[0083] A feature point matching unit 320 is configured to perform feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching rate.

[0084] A comparison unit 330 is configured to determine whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle.

[0085] An obstacle determination unit 340 is configured to determine that the to-be-identified obstacle belongs to the labeled obstacle when the feature point matching rate is greater than or equal to the matching threshold.

[0086] The obstacle identification device 300 provided by the application embodiment can implement the processes of the obstacle identification method provided by the foregoing embodiments, and thus the details are not repeated here.

[0087] Compared with the prior art, the obstacle identification method and device provided by the application embodiment can effectively identify various types of obstacles, especially some unconventional or movable obstacles.

[0088] Referring to FIGS. 5-7, as a specific example, the application further provides a specific flow of a sweeping robot applying the obstacle identification method provided by the foregoing embodiments to perform a cleaning task.

[0089] As shown in FIG. 5, the robot vacuum cleaner establishes a cleaning map of the user's home environment based on data from its own laser sensor and / or TOF sensor, in combination with SLAM (Simultaneous Localization and Mapping) positioning technology. The cleaning map displays the expected cleaning trajectory as well as obstacle information, including unidentified obstacles.

[0090] As shown in FIG. 6, because the unidentified obstacle is not included in the robot vacuum cleaner's existing identification model library, the unidentified obstacle cannot be effectively avoided during the robot vacuum cleaner's cleaning process. At this time, a labeling request can be sent to the user, allowing the user to dynamically label the unidentified obstacle on the map and define the avoidance type. For example, for furniture that is easy to get stuck, the user can set it to "strict avoidance", meaning that the robot vacuum cleaner should completely avoid approaching these areas; for smaller movable objects such as a child's toys, it can be set to "conditional avoidance", under which setting the robot vacuum cleaner can approach the obstacle but needs to take avoidance measures before contact.

[0091] The robot vacuum cleaner sets the avoidance weight of the to-be-identified obstacle according to the information fed back by the user. Specifically, according to the avoidance type set by the user, the robot vacuum cleaner automatically adjusts the avoidance weight of each obstacle. The avoidance weight reflects the urgency and importance of avoidance, and can also be used for subsequent feature point collection. Obstacles with high avoidance weights will trigger the robot vacuum cleaner to collect more images and more feature points at different angles when approaching.

[0092] After obtaining the labeled obstacle position information, the robot vacuum cleaner will actively travel to the coordinate position of the just-labeled obstacle according to the current map information. Upon reaching the vicinity of the obstacle, the robot vacuum cleaner will collect images of the obstacle from different angles. These images can be captured by the robot vacuum cleaner's own camera system. Further, the robot vacuum cleaner uses image processing algorithms to select key feature points from the collected images. Key feature points can describe the visual characteristics of the obstacle in detail, such as edges, corner points, and specific texture patterns.

[0093] During the collection process, the robot vacuum cleaner will intelligently select the amount of feature point collection information according to the previously set obstacle avoidance weight value. For obstacles with high avoidance weights, the robot vacuum cleaner will collect images from multiple angles and record as many feature points as possible. For obstacles with low avoidance weights, the robot vacuum cleaner will reduce the number and angle of image collection.

[0094] After completing the image collection and feature point selection steps, the robot vacuum cleaner continues its cleaning task and uses the collected image information to search and identify the target obstacle in the environment in real time.

[0095] For different priority obstacles, the robot adopts different feature point matching strategies. For obstacles with high avoidance weight, even if there is only a small feature point matching rate, it will be identified as a successful identification. For obstacles with low avoidance weight, such as objects that can be approached but need to be carefully avoided, a higher feature point matching rate is required for successful identification. The above strategy can ensure that the robot is more lenient in key obstacle identification, ensuring the safety of operation, while being more strict and accurate in less critical obstacle identification, avoiding unnecessary avoidance behavior and improving cleaning efficiency.

[0096] As shown in FIG. 7, after completing the identification of the obstacle, the robot will further process the coordinate point position information of the obstacle to ensure that its navigation and avoidance strategy is always based on the latest environmental data. The robot will constantly update and adjust the forbidden area on the map to adapt to changes in the environment.

[0097] Specifically, after the robot identifies and locates the obstacle through its sensing system, it will compare the latest coordinate points with the previously recorded positions on the map. If it finds that the position of the obstacle has changed (for example, the position has moved or a new obstacle has been added), the robot will update the above information to the map, thereby ensuring the accuracy of the map. Based on the updated coordinate information, if the obstacle is located outside the original forbidden area, the robot will mark a new forbidden area on the map. The new forbidden area marks the new position of the obstacle, and the robot will avoid the new area in subsequent cleaning. As can be seen, according to the obstacle identification method provided in the embodiment, the robot can flexibly respond to changes in the position of the obstacle in the environment, ensuring the smooth progress of the cleaning work.

[0098] The application also provides a self-moving robot, which comprises:

[0099] a body 110;

[0100] a memory disposed in the body 110, the memory storing programs or instructions; and

[0101] a processor disposed in the body 110, the processor executing the programs or instructions to implement the obstacle identification method of the foregoing embodiments.

[0102] Some embodiments of the application provide a non-transitory computer-readable storage medium storing computer program instructions, which when called and executed by a processor, implement the method steps of any of the above embodiments.

[0103] Some embodiments of the present application provide a computer program product, which comprises computer program instructions stored in a computer readable storage medium and adapted to be invoked and executed by a processor to cause a computer device having the processor to perform the steps of the method of any one of the above embodiments.

[0104] As shown in FIG. 8, the self-moving robot can include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 1101, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1102 or loaded from a storage device 1108 into a random access memory (RAM) 1103. Various programs and data required for base station operation are also stored in the RAM 1103. The processing device 1101, the ROM 1102, and the RAM 1103 are connected to each other through a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0105] In general, the following devices can be connected to the I / O interface 1105: input devices 1106 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 1107 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1108 including, for example, a hard disk, etc.; and communication devices 1109. The communication devices 1109 can allow the base station to communicate wirelessly or wiredly with other devices to exchange data.

[0106] The embodiments of the present disclosure further provide a control method of a self-moving robot, comprising:

[0107] S210, obtaining feature information of a to-be-identified obstacle;

[0108] S220, performing feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching result;

[0109] In this step, the feature point matching result obtained by the feature point matching can represent the similarity between the to-be-identified obstacle and the labeled obstacle in the database, and can be represented by a specific numerical value, i.e., a matching rate. The higher the feature point matching rate, the higher the similarity between the to-be-identified obstacle and the labeled obstacle.

[0110] S230, determining whether the position of the to-be-identified obstacle changes based on the matching result of the to-be-identified obstacle and the labeled obstacle;

[0111] In this step, the matching result can be compared with a specific numerical value, for example, whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold can be determined according to the avoidance weight of the labeled obstacle.

[0112] In S240, if the position changes, an electronic exclusion zone is set at the current position of the to-be-identified obstacle, and an obstacle avoidance operation is performed on the newly set electronic exclusion zone.

[0113] In this step, if the position of the to-be-identified obstacle changes, an electronic exclusion zone can be set at the current position of the to-be-identified obstacle, and the self-moving robot can be instructed to perform an obstacle avoidance operation in the electronic exclusion zone.

[0114] In some embodiments, in addition to setting an electronic exclusion zone, the electronic exclusion zone at the original position of the to-be-identified obstacle can be further selected and deleted. After the electronic exclusion zone is deleted, the self-moving robot can perform a cleaning operation on the deleted electronic exclusion zone.

[0115] It can be seen that the control method of the self-moving robot provided in this embodiment not only can effectively identify various types of obstacles, especially some unconventional or movable obstacles, but also can further determine whether the position of the obstacle changes according to the obstacle identification result, update the action path of the self-moving robot, improve the avoidance effect, and avoid missing.

[0116] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions, and operations of the systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a part of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than those noted in the drawings. For example, two blocks represented in succession can actually be executed substantially in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0117] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0118] The above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit it. Although this disclosure 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 disclosure.

Claims

1. An obstacle recognition method characterized by, The method comprises: obtaining feature information of a to-be-identified obstacle; performing feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching rate; determining whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle; when the feature point matching rate is greater than or equal to the matching threshold, the to-be-identified obstacle belongs to the labeled obstacle.

2. The obstacle recognition method according to claim 1, characterized by, The method further comprises: when the feature point matching rate is less than the matching threshold, sending a labeling request and setting the avoidance weight of the to-be-identified obstacle according to feedback information of the labeling request.

3. The obstacle recognition method according to claim 2, characterized by, The method further comprises: collecting the feature information of the to-be-identified obstacle, and the amount of the collected feature information of the to-be-identified obstacle is positively correlated with the avoidance weight of the to-be-identified obstacle.

4. The obstacle recognition method according to any one of claims 1 to 3, characterized in that, The matching threshold is negatively correlated with the avoidance weight of the labeled obstacle.

5. The obstacle recognition method according to any one of claims 1 to 4, characterized in that, The avoidance weight is determined according to an avoidance type of the obstacle, and the more strict the avoidance type is, the higher the corresponding avoidance weight is.

6. The obstacle recognition method according to any one of claims 1 to 5, characterized in that, The feature information of the to-be-identified obstacle comprises image information of the to-be-identified obstacle.

7. The obstacle recognition method according to any one of claims 1 to 6, characterized by, The number of feature points of the labeled obstacle is positively correlated with the avoidance weight of the labeled obstacle.

8. The obstacle recognition method according to any one of claims 1 to 7, characterized in that, The method further comprises: when the to-be-identified obstacle is a labeled obstacle, determining whether a position of the labeled obstacle changes; if the position changes, setting an electronic exclusion zone at a current position of the labeled obstacle and deleting an electronic exclusion zone at an original position of the labeled obstacle.

9. The obstacle recognition method according to claim 8, characterized by, The step of setting the electronic exclusion zone at the current position of the labeled obstacle comprises: generating the electronic exclusion zone according to the avoidance weight corresponding to the labeled obstacle, wherein the range of the electronic exclusion zone is positively correlated with the avoidance weight.

10. A control method of a self-moving robot, characterized by, The method comprises: obtaining feature information of a to-be-identified obstacle; performing feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching result; determining whether a position of the to-be-identified obstacle changes based on the feature point matching result; if the position changes, setting an electronic exclusion zone at a current position of the to-be-identified obstacle and performing an obstacle avoidance operation on the newly set electronic exclusion zone.

11. The control method according to claim 10, characterized by, The method further comprises: deleting an electronic exclusion zone at an original position of the to-be-identified obstacle and performing a cleaning operation on the deleted electronic exclusion zone.

12. An obstacle recognition device, characterized by The method comprises: a feature information acquisition unit configured to obtain feature information of a to-be-identified obstacle; a feature point matching unit configured to perform feature point matching between the to-be-identified obstacle and a labeled obstacle according to the feature information, to obtain a feature point matching rate; a comparison unit configured to determine whether the feature point matching rate is greater than or equal to a matching threshold, wherein the matching threshold is determined according to an avoidance weight of the labeled obstacle; an obstacle determination unit configured to determine that the to-be-identified obstacle belongs to the labeled obstacle when the feature point matching rate is greater than or equal to the matching threshold.

13. A self-moving robot, characterized in that, The method comprises: a memory storing a program or instructions; and a processor configured to implement the obstacle identification method according to any one of claims 1 to 11 when executing the program or instructions.

14. A computer program product, characterised in that, The computer program product comprises computer program instructions stored in a computer readable storage medium and adapted to be invoked and executed by a processor to cause a computer device having the processor to perform the steps of the method according to any one of claims 1 to 11.

15. A computer readable storage medium characterized by: The computer program instructions stored in the computer readable storage medium implement the steps of the method according to any one of claims 1 to 11 when invoked and executed by a processor.

Images