Obstacle recognition method, system and cleaning robot

Abstract

The application relates to an obstacle recognition method, system and cleaning robot. The method comprises the following steps: acquiring a grid map of a region to be cleaned, detecting the state of all laser point clouds in the grid map; when a plurality of laser point clouds are detected to be accumulated to form a point cloud set, acquiring the state of the point cloud set; and when the shape of the point cloud set is detected to be a regular shape, judging that the point cloud set is an obstacle. The application can solve the problem that, in the prior art, due to the limitation of mapping technology, errors occur in the recognition of obstacles in the process of creating a cleaning map, and the accuracy of the cleaning map is affected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cleaning equipment, in particular to an obstacle recognition method and system and a cleaning robot. BACKGROUND

[0002] With the continuous development of automation technology and artificial intelligence technology, the application of cleaning robots such as floor washing machines and floor sweeping machines is becoming more and more widespread, bringing great convenience to people's lives. In traditional technology, in order to improve the cleaning efficiency of the cleaning robot, a cleaning map of the to-be-cleaned area needs to be established for the cleaning robot in the process of cleaning the to-be-cleaned area. However, due to the limitations of mapping technology, errors occur in the recognition of obstacles in the process of creating a cleaning map, which affects the accuracy of the cleaning map. SUMMARY

[0003] Therefore, the technical problem to be solved by the present application is that in traditional technology, due to the limitations of mapping technology, errors occur in the recognition of obstacles in the process of creating a cleaning map, which affects the accuracy of the cleaning map.

[0004] To solve the above technical problems, the present application provides an obstacle recognition method applied to a cleaning robot, which comprises:

[0005] obtaining a grid map of a to-be-cleaned area and detecting the state of all laser point clouds in the grid map;

[0006] when a plurality of laser point clouds are detected to accumulate to form a point cloud set, obtaining the state of the point cloud set;

[0007] when the shape of the point cloud set is detected to be a regular shape, judging that the point cloud set is an obstacle.

[0008] Optionally, the obtaining of the grid map of the to-be-cleaned area and the detection of the state of all laser point clouds in the grid map comprises:

[0009] detecting the to-be-cleaned area by a laser detection mechanism to obtain all laser point clouds of the to-be-cleaned area; wherein the laser point clouds comprise a plurality of laser detection points;

[0010] obtaining a grid map of the to-be-cleaned area according to the obtained all laser point clouds of the to-be-cleaned area;

[0011] obtaining the state of all laser point clouds in the grid map according to the obtained grid map of the to-be-cleaned area.

[0012] Optionally, the judging that the point cloud set is an obstacle when the shape of the point cloud set is detected to be a regular shape comprises:

[0013] obtaining a matching state of the normal vector of the point cloud set when it is detected that the shape of the point cloud set is a regular long strip shape;

[0014] judging that the point cloud set is a wall body when it is detected that the matching degree of the normal vector of the point cloud set is high.

[0015] Optionally, the obtaining a matching state of the normal vector of the point cloud set when it is detected that the shape of the point cloud set is a regular long strip shape comprises:

[0016] obtaining the normal vector of each laser point cloud in the point cloud set when it is detected that the shape of the point cloud set in the grid map is a regular long strip shape;

[0017] obtaining a matching state of the normal vector of the laser point cloud in the point cloud set according to the obtained normal vector of each laser point cloud.

[0018] Optionally, the obtaining the normal vector of each laser point cloud in the point cloud set comprises:

[0019] fitting all the laser point clouds in the point cloud set respectively to obtain corresponding point cloud surfaces according to the obtained point cloud set in the grid map;

[0020] obtaining the normal vector of the point cloud surface according to the obtained point cloud surface of the laser point cloud;

[0021] obtaining the normal vector of each laser point cloud according to the obtained normal vector of each point cloud surface.

[0022] Optionally, the obtaining a matching state of the normal vector of the laser point cloud in the point cloud set according to the obtained normal vector of the laser point cloud comprises:

[0023] detecting consistency of the normal vector direction of each laser point cloud in the point cloud set according to the obtained normal vector of the laser point cloud in the point cloud set;

[0024] judging that the matching degree of the normal vector of each laser point cloud in the point cloud set is high when it is detected that the consistency of the normal vector direction of each laser point cloud is high.

[0025] Optionally, after the detecting consistency of the normal vector direction of each laser point cloud in the point cloud set, the method further comprises:

[0026] judging that the matching degree of the normal vector of the laser point cloud in the point cloud set is low when it is detected that the consistency of the normal vector direction of part of the laser point cloud with the normal vector direction of other laser point clouds is low.

[0027] When it is detected that the matching degree of the normal vector of the laser point cloud in the point cloud set is low, the laser point cloud with the low matching degree is deleted.

[0028] Optionally, after the point cloud set is determined as a wall, the method further comprises:

[0029] According to the normal vector of each laser point cloud of the obtained point cloud set, the category of the wall is determined.

[0030] When it is detected that each laser point cloud on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, the wall is determined as an interior wall.

[0031] When it is detected that each laser point cloud on one side of the point cloud set has a normal vector, and the matching degree of the normal vector is high, the wall is determined as an exterior wall.

[0032] In addition, the application further provides an obstacle recognition system applied to a cleaning robot, comprising:

[0033] A point cloud state detection module is configured to acquire a grid map of a to-be-cleaned area, and detect the state of all laser point clouds in the grid map.

[0034] A point cloud set detection module is in communication connection with the point cloud state detection module, and is configured to acquire the state of a point cloud set when it is detected that multiple laser point clouds are accumulated to form the point cloud set.

[0035] An obstacle determination module is in communication connection with the point cloud set detection module, and is configured to determine the point cloud set as an obstacle when it is detected that the shape of the point cloud set is a regular shape.

[0036] In addition, the application further provides a cleaning robot, comprising:

[0037] A robot body; and

[0038] A control processor is arranged on the robot body.

[0039] The control processor is configured to:

[0040] Acquire a grid map of a to-be-cleaned area, and detect the state of all laser point clouds in the grid map.

[0041] Acquire the state of a point cloud set when it is detected that multiple laser point clouds are accumulated to form the point cloud set.

[0042] Determine the point cloud set as an obstacle when it is detected that the shape of the point cloud set is a regular shape.

[0043] The technical solution provided by the application has the following advantages:

[0044] The obstacle recognition method provided by the present application can reduce the recognition error of the obstacle in the process of creating the cleaning map of the to-be-cleaned area, and can improve the accuracy of the cleaning map. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0046] Figure 1 The step flowchart of the obstacle recognition method described in the embodiments of the present application is shown in the figure.

[0047] Figure 2 The structure schematic diagram of the obstacle recognition system described in the embodiments of the present application is shown in the figure.

[0048] Figure 3 The structure schematic diagram of the cleaning robot described in the embodiments of the present application is shown in the figure.

[0049] Figure 4 The three-dimensional structure schematic diagram of the cleaning robot described in the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0050] The technical solutions of the present application will be described in detail below with reference to the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, not all. In the following, the present application will be described in detail with reference to the drawings and embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0051] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0052] In the present application, the orientation words such as "up", "down", "top", "bottom" are generally directed to the direction shown in the drawings, or to the vertical, vertical or gravity direction of the components themselves; similarly, for the convenience of understanding and description, "inner", "outer" refers to the inner and outer of the contour of each component itself, but the above orientation words are not used to limit the present application.

[0053] In the prior art, in order to improve the cleaning efficiency of the cleaning robot such as a sweeping robot, a cleaning map of a to-be-cleaned area needs to be established for the cleaning robot in advance during the cleaning of the to-be-cleaned area. However, due to the limitation of mapping technology, errors occur in the identification of obstacles during the creation of the cleaning map, which affects the accuracy of the cleaning map. In order to solve the above technical problems, the present application provides an obstacle identification method, system and cleaning robot.

[0054] The obstacle identification method and system provided by the present application can not only be applied to cleaning robots such as sweeping robots, but also be applied to other machine devices (such as vehicles, robots, etc. that automatically move in certain areas) that need to identify obstacles in a map. In the following embodiments, the present application is described by taking a cleaning robot such as a sweeping robot that needs to identify obstacles in a map as an example.

[0055] Embodiment 1

[0056] The present embodiment provides an obstacle identification method applied to a cleaning robot. As shown in the figure, the obstacle identification method can include the following steps: Figure 1 The obstacle identification method can include the following steps:

[0057] S100, acquiring a grid map of a to-be-cleaned area, detecting the state of all laser point clouds in the grid map;

[0058] S200, when a plurality of laser point clouds are detected to accumulate to form a point cloud set, acquiring the state of the point cloud set;

[0059] S300, when the shape of the point cloud set is detected to be a regular shape, judging that the point cloud set is an obstacle.

[0060] By acquiring the grid map of the to-be-cleaned area and detecting the state of the laser point cloud in the grid map, when a plurality of laser point clouds are detected to accumulate to form a point cloud set with a regular shape, it can be judged that the point cloud set is formed by an obstacle, that is, it can be judged that there is a corresponding obstacle at the point cloud set in the to-be-cleaned area. In this way, the identification error of the obstacle in the process of creating the cleaning map of the to-be-cleaned area can be reduced, and the accuracy of the cleaning map can be improved.

[0061] Specifically, in step S100, the grid map of the to-be-cleaned area is acquired, and the state of all laser point clouds in the grid map is detected, which can specifically include the following steps:

[0062] S110, detecting the to-be-cleaned area by a laser detection mechanism to acquire all laser point clouds of the to-be-cleaned area; wherein the laser point cloud can include a plurality of laser detection points;

[0063] In the process of creating the cleaning map of the cleaning area, the laser detection mechanism provided on the cleaning robot can be used to detect the cleaning area, and the laser point cloud information of the cleaning area can be obtained. Moreover, the laser point cloud information is obtained by the laser detection mechanism detecting the cleaning area by laser, and mainly includes the obstacle information in the cleaning area. Moreover, each laser point cloud is formed by a plurality of laser detection points, and the laser detection points are formed by the laser beams emitted by the laser detection mechanism and irradiated on the obstacles around the cleaning robot.

[0064] S120, obtaining the grid map of the cleaning area according to all the laser point clouds of the cleaning area.

[0065] According to the laser point cloud information obtained by detecting the obstacles in the cleaning area by laser and the ground grid information of the cleaning area, the grid map of the cleaning area can be obtained.

[0066] S130, obtaining the state of all the laser point clouds in the grid map of the cleaning area according to the obtained grid map of the cleaning area.

[0067] After obtaining the grid map formed by the laser point clouds, the state of the laser point clouds in the grid map can be detected. Specifically, the distribution and shape of the laser point clouds in the grid map can be detected, and the obstacle information represented by the laser point clouds can be determined according to the state of the laser point clouds.

[0068] Moreover, the plurality of laser point clouds formed by the laser detection mechanism detecting the cleaning area, some of the laser point clouds are accumulated to form a point cloud set and present the shape of the obstacle, and some of the laser point clouds are in a scattered state and cannot be detected to have a specific shape. Therefore, in step S200, when it is detected that the plurality of laser point clouds are accumulated to form a point cloud set, the state of the point cloud set is obtained, so that whether it is an obstacle can be detected according to the state of the point cloud set.

[0069] Moreover, in step S300, when it is detected that the shape of the point cloud set is a regular shape, the point cloud set is determined to be an obstacle, which can specifically include the following steps:

[0070] S310, when it is detected that the shape of the point cloud set is a regular long strip, the matching state of the normal vector of the point cloud set is obtained.

[0071] S320, when it is detected that the matching degree of the normal vector of the point cloud set is high, the point cloud set is determined to be a wall.

[0072] When it is detected that the plurality of laser point clouds are accumulated to form a point cloud set, it can be preliminarily determined that the point cloud set represents that there is an obstacle at the position. Moreover, by analyzing the shape and state of the point cloud set, the type of the obstacle represented by the point cloud set can be further detected.

[0073] Specifically, when it is detected that the point cloud set is distributed in a long strip shape, it can be preliminarily judged that the point cloud set represents a wall. When it is detected that the normal vectors of the point cloud set have a high matching degree, it can be determined that the point cloud set is indeed a wall.

[0074] Further, in step S310, when it is detected that the shape of the point cloud set is a regular long strip shape, the matching state of the normal vector of the point cloud set is obtained, which can specifically include the following steps:

[0075] S312, when it is detected that the shape of the point cloud set in the grid map is a regular long strip shape, the normal vector of each laser point cloud in the point cloud set is obtained;

[0076] S314, according to the obtained normal vector of each laser point cloud, the matching state of the normal vector of the laser point cloud in the point cloud set is obtained.

[0077] Since the point cloud set is formed by a plurality of laser point clouds, detecting the matching degree of the normal vector of the point cloud set is to detect the matching degree of the normal vector of the plurality of laser point clouds. If the matching degree of the normal vector of the plurality of laser point clouds is high, it proves that the matching degree of the normal vector of the point cloud set is high.

[0078] Further, in step S312, the normal vector of each laser point cloud in the point cloud set is obtained, which can specifically include the following steps:

[0079] S3122, according to the obtained point cloud set in the grid map, fitting is performed on all laser point clouds in the point cloud set to obtain corresponding point cloud surfaces;

[0080] According to the obtained point cloud set in the grid map, fitting can be performed on the laser detection points of all laser point clouds in the point cloud set to obtain corresponding point cloud surfaces. Specifically, a plurality of laser detection points on each laser point cloud can be connected together to form a corresponding point cloud surface.

[0081] S3124, according to the obtained point cloud surface of the laser point cloud, the normal vector of the point cloud surface is obtained;

[0082] After obtaining the point cloud surface formed by fitting a plurality of laser detection points on each laser point cloud, the normal line of the point cloud surface can be obtained, that is, the normal vector of the point cloud surface can be obtained.

[0083] S3126, according to the obtained normal vector of each point cloud surface, the normal vector of each laser point cloud is obtained. The normal vector of each point cloud surface is the normal vector of the corresponding laser point cloud.

[0084] Further, in step S314, according to the obtained normal vector of the laser point cloud, the matching state of the normal vector of the laser point cloud in the point cloud set is obtained, which can specifically include the following steps:

[0085] S3142, detecting the consistency of the normal vector direction of each laser point cloud in the point cloud set according to the normal vector of the laser point cloud in the obtained point cloud set;

[0086] After obtaining the normal vector of all laser point clouds in the point cloud set, it can be detected whether the direction of the normal vector of each laser point cloud is the same or approximately the same as the direction of the normal vector of other laser point clouds, that is, the consistency of the normal vector direction of each laser point cloud in the point cloud set can be obtained.

[0087] S3144, when the consistency of the normal vector direction of each laser point cloud is detected to be high, it is judged that the matching degree of the normal vector of each laser point cloud in the point cloud set is high.

[0088] That is, when it is detected that the directions of the normal vectors of all laser point clouds in the point cloud set are mostly consistent (that is, the same or approximately the same), it can be judged that the matching degree of the normal vector of each laser point cloud in the point cloud set is high. For example, when the directions of the normal vectors of more than 70% (or more than 80%, the specific proportion can be set according to the actual situation) of the laser point clouds in the point cloud set are the same, it can be judged that the matching degree of the normal vector of each laser point cloud in the point cloud set is high.

[0089] Moreover, after detecting the consistency of the normal vector direction of each laser point cloud in the point cloud set in step S3142, the following steps can also be included:

[0090] S3146, when it is detected that the consistency of the normal vector direction of part of the laser point clouds with the normal vector direction of other laser point clouds is low, it is judged that the matching degree of the normal vector of the laser point cloud in the point cloud set is low;

[0091] That is, when it is detected that the directions of the normal vectors of all laser point clouds in the point cloud set are mostly inconsistent, it can be judged that the matching degree of the normal vector of each laser point cloud in the point cloud set is low. For example, when the directions of the normal vectors of less than 60% (or more than 50%, the specific proportion can be set according to the actual situation) of the laser point clouds in the point cloud set are the same, it can be judged that the matching degree of the normal vector of each laser point cloud in the point cloud set is low.

[0092] S3148, when it is detected that the matching degree of the normal vector of the laser point cloud in the point cloud set is low, the laser point cloud with low matching degree is deleted.

[0093] When it is detected that the directions of the normal vectors of all laser point clouds in the point cloud set are mostly inconsistent, it is proved that the laser point clouds in the point cloud set may have errors or errors, and cannot be used as a basis for judging obstacles, and can be ignored.

[0094] In addition, after judging that the point cloud set is a wall in step S300, the following steps can also be included:

[0095] S400, judging the type of the wall according to the normal vectors of each laser point cloud of the obtained point cloud set;

[0096] When the point cloud set is detected as a wall obstacle, it can be further judged whether the wall belongs to an interior wall or an exterior wall. It can be known that the wall can be divided into an exterior wall and an interior wall according to the position and direction in the building. The wall arranged along the periphery of the building can be referred to as an exterior wall, and the wall surrounded by the exterior wall can be referred to as an interior wall. The interior wall refers to the wall that plays a role in separating space in the room and does not directly contact with outdoor air.

[0097] S500, when it is detected that each laser point cloud on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, the wall is determined to be an interior wall;

[0098] When the wall is an interior wall, the two sides of the interior wall can be detected by the laser detection mechanism, and a plurality of laser point clouds can be formed on the two sides of the interior wall. Therefore, when the corresponding point cloud surface is obtained on both sides of the point cloud set formed by the plurality of laser point clouds of the wall, and the matching degree of the normal vectors of the plurality of laser point clouds on each side is high, it can be preliminarily judged that the wall is an interior wall. Further, when it is detected that the consistency of the normal vectors of the plurality of laser point clouds on one side of the point cloud set and the normal vectors of the plurality of laser point clouds on the other side is high, it can be determined that the wall is an interior wall (the normal vector directions of the two sides of the interior wall are basically consistent).

[0099] S600, when it is detected that each laser point cloud on one side of the point cloud set has a normal vector, and the matching degree of the normal vector is high, the wall is determined to be an exterior wall.

[0100] When the wall is an exterior wall, only the inner side of the exterior wall can be detected by the laser detection mechanism, so that only the inner side of the exterior wall will correspond to a plurality of laser point clouds. When the corresponding point cloud surface is obtained on the inner side of the point cloud set formed by the plurality of laser point clouds of the wall, and the matching degree of the normal vectors of the plurality of laser point clouds of the point cloud set is high, it is judged that the wall is an exterior wall.

[0101] Embodiment 2

[0102] The embodiment provides an obstacle recognition system applied to a cleaning robot. As shown in the figure, the obstacle recognition system 100 can include: Figure 2

[0103] A point cloud state detection module 102 is configured to acquire a grid map of a cleaning area, and detect the state of all laser point clouds in the grid map.

[0104] A point cloud set detection module 104 is in communication connection with the point cloud state detection module 102, and is configured to acquire the state of a point cloud set when a plurality of laser point clouds are detected to be accumulated to form the point cloud set.​

[0105] The obstacle determination module 106, in communication with the point cloud set detection module 104, is configured to determine that the point cloud set is an obstacle when it is detected that the shape of the point cloud set is a regular shape.

[0106] Specifically, the point cloud state detection module 102, when used for obtaining a grid map of the to-be-cleaned area and detecting the state of all laser point clouds in the grid map, can be specifically used for:

[0107] detecting the to-be-cleaned area by the laser detection mechanism to obtain all laser point clouds of the to-be-cleaned area; wherein the laser point clouds can include a plurality of laser detection points;

[0108] obtaining a grid map of the to-be-cleaned area according to the obtained all laser point clouds of the to-be-cleaned area;

[0109] obtaining the state of all laser point clouds in the grid map according to the obtained grid map of the to-be-cleaned area.

[0110] Furthermore, the obstacle determination module 106, when used for determining that the point cloud set is an obstacle when it is detected that the shape of the point cloud set is a regular shape, can be specifically used for:

[0111] obtaining the matching state of the normal vector of the point cloud set when it is detected that the shape of the point cloud set is a regular long strip shape;

[0112] determining that the point cloud set is a wall when it is detected that the matching degree of the normal vector of the point cloud set is high.

[0113] Further, the obstacle determination module 106, when used for obtaining the matching state of the normal vector of the point cloud set when it is detected that the shape of the point cloud set is a regular long strip shape, can be specifically used for:

[0114] obtaining the normal vector of each laser point cloud in the point cloud set when it is detected that the shape of the point cloud set in the grid map is a regular long strip shape;

[0115] obtaining the matching state of the normal vector of the laser point cloud in the point cloud set according to the obtained normal vector of each laser point cloud.

[0116] Furthermore, the obstacle determination module 106, when used for obtaining the normal vector of each laser point cloud in the point cloud set, can be specifically used for:

[0117] fitting all laser point clouds in the point cloud set to obtain the corresponding point cloud surface according to the obtained point cloud set in the grid map;

[0118] obtaining the normal vector of the point cloud surface of the laser point cloud according to the obtained point cloud surface of the laser point cloud;

[0119] According to the normal vectors of the obtained point cloud surfaces, the normal vectors of the laser point clouds are obtained. The normal vector of each point cloud surface is the normal vector of the corresponding laser point cloud.

[0120] Further, the obstacle determination module 106, when used to obtain the matching state of the normal vectors of the laser point clouds in the point cloud set according to the obtained normal vectors of the laser point clouds, can be specifically used for:

[0121] According to the normal vectors of the laser point clouds in the obtained point cloud set, the consistency of the normal vector directions of the laser point clouds in the point cloud set is detected.

[0122] When it is detected that the consistency of the normal vector directions of the laser point clouds is high, it is determined that the matching degree of the normal vectors of the laser point clouds in the point cloud set is high.

[0123] Further, after the obstacle determination module 106 is used to detect the consistency of the normal vector directions of the laser point clouds in the point cloud set, the obstacle determination module 106 can be further used for:

[0124] When it is detected that the consistency of the normal vector directions of some laser point clouds and other laser point clouds is low, it is determined that the matching degree of the normal vectors of the laser point clouds in the point cloud set is low.

[0125] When it is detected that the matching degree of the normal vectors of the laser point clouds in the point cloud set is low, the laser point clouds with the low matching degree are deleted.

[0126] In addition, after the obstacle determination module 106 is used to determine that the point cloud set is a wall, the obstacle determination module 106 can be further used for:

[0127] According to the normal vectors of the laser point clouds in the obtained point cloud set, the category of the wall is determined.

[0128] When it is detected that each laser point cloud on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, it is determined that the wall is an interior wall.

[0129] When it is detected that each laser point cloud on one side of the point cloud set has a normal vector, and the matching degree of the normal vector is high, it is determined that the wall is an exterior wall.

[0130] The obstacle recognition system 100 described in the embodiment corresponds to the obstacle recognition method described above. The functions of each module in the obstacle recognition system 100 are described in detail in the corresponding method embodiment, and will not be described here.

[0131] Embodiment 3

[0132] The embodiment provides a cleaning robot, such as Figure 3 and Figure 4As shown, the cleaning robot 10 can include a robot body 12, and a control processor 14 arranged on the robot body 12. The robot body 12 can be controlled by the control processor 14 to clean a to-be-cleaned area, and analyze and identify obstacles in a grid map corresponding to the to-be-cleaned area.

[0133] Further, the cleaning robot 10 can include a laser detection mechanism 16 arranged on the robot body 12, and the laser detection mechanism 16 is in communication connection with the control processor 14. The control processor 14 can control the laser detection mechanism 16 to perform laser detection on the to-be-cleaned area to obtain laser point cloud information of the to-be-cleaned area.

[0134] Specifically, the control processor 14 can be configured to: obtain a grid map of a to-be-cleaned area, detect states of all laser point clouds in the grid map; when a plurality of laser point clouds are detected to be accumulated to form a point cloud set, obtain a state of the point cloud set; and when a shape of the point cloud set is detected to be a regular shape, determine that the point cloud set is an obstacle.

[0135] Further, when the control processor is configured to obtain a grid map of a to-be-cleaned area, and detect states of all laser point clouds in the grid map, the control processor can be specifically configured to:

[0136] detect the to-be-cleaned area by the laser detection mechanism to obtain all laser point clouds of the to-be-cleaned area; wherein the laser point clouds can include a plurality of laser detection points;

[0137] obtain the grid map of the to-be-cleaned area according to the obtained all laser point clouds of the to-be-cleaned area;

[0138] obtain states of all laser point clouds in the grid map according to the obtained grid map of the to-be-cleaned area.

[0139] Further, when the control processor is configured to determine that the point cloud set is an obstacle when a shape of the point cloud set is detected to be a regular shape, the control processor can be specifically configured to:

[0140] when the shape of the point cloud set is detected to be a regular long strip shape, obtain a matching state of a normal vector of the point cloud set;

[0141] when the matching degree of the normal vector of the point cloud set is high, determine that the point cloud set is a wall.

[0142] Further, when the control processor is configured to obtain a matching state of a normal vector of a point cloud set when a shape of the point cloud set is detected to be a regular long strip shape, the control processor can be specifically configured to:

[0143] when the shape of the point cloud set in the grid map is detected to be a regular long strip shape, obtain a normal vector of each laser point cloud in the point cloud set;

[0144] According to the normal vectors of the obtained laser point clouds, a matching state of the normal vectors of the laser point clouds in the point cloud set is obtained.

[0145] Moreover, the control processor is configured to, in the process of obtaining the normal vectors of the laser point clouds in the point cloud set, specifically configured to:

[0146] According to the obtained point cloud set in the grid map, all laser point clouds in the point cloud set are fitted respectively to obtain corresponding point cloud surfaces;

[0147] According to the obtained point cloud surfaces of the laser point clouds, normal vectors of the point cloud surfaces are obtained.

[0148] According to the normal vectors of the obtained point cloud surfaces, the normal vectors of the laser point clouds are obtained. The normal vectors of the point cloud surfaces are the normal vectors of the corresponding laser point clouds.

[0149] Moreover, the control processor is configured to, in the process of obtaining the matching state of the normal vectors of the laser point clouds in the point cloud set according to the obtained normal vectors of the laser point clouds, specifically configured to:

[0150] According to the obtained normal vectors of the laser point clouds in the point cloud set, consistency of the normal vector directions of the laser point clouds in the point cloud set is detected.

[0151] When the consistency of the normal vector directions of the laser point clouds is high, it is determined that the matching degree of the normal vectors of the laser point clouds in the point cloud set is high.

[0152] Moreover, the control processor is configured to, after the process of detecting the consistency of the normal vector directions of the laser point clouds in the point cloud set, further configured to:

[0153] When the consistency of the normal vector directions of the part of the laser point clouds and the other laser point clouds is low, it is determined that the matching degree of the normal vectors of the laser point clouds in the point cloud set is low.

[0154] When the matching degree of the normal vectors of the laser point clouds in the point cloud set is low, the laser point clouds with the low matching degree are deleted.

[0155] In addition, the control processor is configured to, after the process of determining that the point cloud set is a wall, further configured to:

[0156] According to the obtained normal vectors of the laser point clouds in the point cloud set, the category of the wall is determined.

[0157] When it is detected that each laser point cloud on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, it is determined that the wall is an interior wall.

[0158] When it is detected that each laser point cloud on one side of the point cloud set has a normal vector, and the matching degree of the normal vector is high, it is determined that the wall is an exterior wall.

[0159] Similarly, in the present embodiment, the control processor can be configured to control the cleaning robot to implement each step of the above-described obstacle recognition method, and the specific implementation manner can refer to the specific content of the above-described obstacle recognition method, which will not be described here again.

[0160] Furthermore, in the present embodiment, the cleaning robot can be configured as a sweeping robot. In addition, the cleaning robot can also be configured as a floor washing machine, an unmanned floor washing machine, a dust collector, etc. with a laser detection mechanism.

[0161] In addition, the present application also provides a computer readable storage medium, the computer readable storage medium stores computer execution instructions, and the computer execution instructions are used to implement all method steps or part of method steps of the above-described obstacle recognition method when executed by a processor.

[0162] The present application implements all or part of the above-described method, which can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium, and the computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0163] Based on the same inventive concept, the present application also provides an electronic device, which includes a memory and a processor, the memory stores a computer program running on the processor, and the processor implements all method steps or part of method steps of the above-described obstacle recognition method when executing the computer program.

[0164] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The processor is the control center of the computer device, and connects all parts of the computer device through various interfaces and lines.

[0165] The memory can be used to store computer programs and / or models. The processor realizes various functions of the computer device by running or executing the computer programs and / or models stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system and at least one application program required by a function (such as a sound playing function, an image playing function, etc.). The data storage area can store data created according to the use of the mobile phone (such as audio data, video data, etc.). In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0166] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, a server or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage, etc.) containing computer-usable program code.

[0167] The present application is described in reference to the appended drawings figures and / or block diagrams of methods, apparatus (systems), servers, and computer program products according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 one or more flowcharts and / or blocks

[0168] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 one or more flowcharts and / or blocks

[0169] The computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 one or more flowcharts and / or blocks

[0170] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method of obstacle recognition applied to a cleaning robot, the method comprising: The method comprises: acquiring a grid map of a to-be-cleaned area, detecting the state of all laser point clouds in the grid map; when a plurality of the laser point clouds are detected to be accumulated to form a point cloud set, acquiring the state of the point cloud set; when the shape of the point cloud set is detected to be a regular shape, judging that the point cloud set is an obstacle, wherein the judging that the point cloud set is an obstacle when the shape of the point cloud set is detected to be a regular shape comprises: when the shape of the point cloud set is detected to be a regular long strip shape, acquiring the matching state of the normal vector of the point cloud set; when the matching degree of the normal vector of the point cloud set is detected to be high, judging that the point cloud set is a wall; the acquiring the matching state of the normal vector of the point cloud set when the shape of the point cloud set is detected to be a regular long strip shape comprises: when the shape of the point cloud set in the grid map is detected to be a regular long strip shape, acquiring the normal vector of each of the laser point clouds in the point cloud set; according to the obtained normal vector of each of the laser point clouds, acquiring the matching state of the normal vector of the laser point clouds in the point cloud set; after the judging that the point cloud set is a wall, further comprising: according to the obtained normal vector of each of the laser point clouds in the point cloud set, judging the category of the wall; when each of the laser point clouds on both sides of the point cloud set is detected to have a normal vector, and the matching degree of the normal vectors on both sides is high, determining that the wall is an interior wall; when each of the laser point clouds on one side of the point cloud set is detected to have a normal vector, and the matching degree of the normal vector is high, determining that the wall is an exterior wall.

2. The obstacle recognition method according to claim 1, characterized by, The acquiring a grid map of a to-be-cleaned area, detecting the state of all laser point clouds in the grid map comprises: detecting the to-be-cleaned area by a laser detection mechanism to acquire all laser point clouds of the to-be-cleaned area; wherein the laser point clouds comprise a plurality of laser detection points; according to the obtained all laser point clouds of the to-be-cleaned area, obtaining a grid map of the to-be-cleaned area; according to the obtained grid map of the to-be-cleaned area, acquiring the state of all laser point clouds in the grid map.

3. The obstacle recognition method according to claim 1, characterized by, The acquiring the normal vector of each of the laser point clouds in the point cloud set comprises: according to the obtained point cloud set in the grid map, fitting all the laser point clouds in the point cloud set respectively to obtain corresponding point cloud surfaces; according to the obtained point cloud surface of the laser point cloud, acquiring the normal vector of the point cloud surface; according to the obtained normal vector of each of the point cloud surfaces, acquiring the normal vector of each of the laser point clouds.

4. The obstacle recognition method according to claim 1, characterized by, The acquiring the matching state of the normal vector of the laser point clouds in the point cloud set according to the obtained normal vector of the laser point clouds comprises: according to the obtained normal vector of the laser point clouds in the point cloud set, detecting the consistency of the direction of the normal vector of each of the laser point clouds in the point cloud set; when the consistency of the direction of the normal vector of each of the laser point clouds is detected to be high, judging that the matching degree of the normal vector of each of the laser point clouds in the point cloud set is high.

5. The obstacle recognition method according to claim 4, characterized by, After detecting the consistency of the normal vector directions of each laser point cloud in the point cloud set, the method further includes: When it is detected that the normal vector direction of some of the laser point clouds has low consistency with the normal vector direction of other laser point clouds, it is determined that the matching degree of the normal vectors of the laser point clouds in the point cloud set is low. When it is detected that the matching degree of the normal vector of the laser point cloud in the point cloud set is low, the laser point cloud with low matching degree is deleted.

6. An obstacle recognition system applied to a cleaning robot, characterized by, include: The point cloud state detection module is used to acquire a grid map of the area to be cleaned and detect the state of all laser point clouds in the grid map. The point cloud cluster detection module is communicatively connected to the point cloud state detection module and is used to obtain the state of the point cloud cluster when multiple laser point clouds are detected to have aggregated to form a point cloud cluster. The obstacle determination module is communicatively connected to the point cloud detection module and is used to determine that the point cloud is an obstacle when the detected point cloud has a regular shape. When the point cloud is detected to have a regular shape, it is determined to be an obstacle, including: When the point cloud is detected to have a regular elongated shape, the matching status of the normal vector of the point cloud is obtained; When a high degree of matching of the normal vectors of the point cloud is detected, the point cloud is determined to be a wall. When the point cloud is detected to have a regular elongated shape, the matching state of the normal vector of the point cloud is obtained, including: When the shape of the point cloud set in the grid map is detected to be a regular strip, the normal vector of each laser point cloud in the point cloud set is obtained; Based on the obtained normal vectors of each laser point cloud, the matching state of the normal vectors of the laser point clouds in the point cloud set is obtained; After determining that the point cloud is a wall, the process further includes: The type of the wall is determined based on the normal vectors of each laser point cloud in the obtained point cloud set; When it is detected that each of the laser point clouds on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, the wall is determined to be an inner wall. When it is detected that each of the laser point clouds on one side of the point cloud set has a normal vector and the matching degree of the normal vector is high, the wall is determined to be an external wall.

7. A cleaning robot, characterized in that, include: The robot itself; as well as, The control processor is located on the robot body; The control processor is used for: Obtain a grid map of the area to be cleaned, and detect the state of all laser point clouds in the grid map; When multiple laser point clouds are detected to have aggregated to form a point cloud cluster, the state of the point cloud cluster is acquired; When the shape of the point cloud is detected to be a regular shape, the point cloud is determined to be an obstacle; The step of determining that the point cloud is an obstacle when its shape is detected to be a regular shape includes: When the point cloud is detected to have a regular elongated shape, the matching status of the normal vector of the point cloud is obtained; When a high degree of matching of the normal vectors of the point cloud is detected, the point cloud is determined to be a wall. When the point cloud is detected to have a regular elongated shape, the matching state of the normal vector of the point cloud is obtained, including: When it is detected that the shape of the point cloud set in the grid map is a regular long strip, a normal vector of each laser point cloud in the point cloud set is obtained; According to the obtained normal vector of each laser point cloud, a matching state of the normal vector of the laser point cloud in the point cloud set is obtained; After the point cloud set is determined as a wall, the method further comprises: According to the obtained normal vector of each laser point cloud of the point cloud set, a category of the wall is determined; When it is detected that each laser point cloud on both sides of the point cloud set has a normal vector, and the matching degree of the normal vectors on both sides is high, the wall is determined as an interior wall; When it is detected that each laser point cloud on one side of the point cloud set has a normal vector, and the matching degree of the normal vector is high, the wall is determined as an exterior wall.

Images