Mobile robot jamming detection method and mobile robot detrapping method
By combining vision sensors and line laser sensors and body state judgments, the problem of misjudgment of existing mobile robot trap detection methods is solved, and the detection accuracy and accuracy are improved.
Patent Information
- Application Number
- CN202311407990.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-27
- Publication Date
- 2025-04-29
AI Technical Summary
The existing mobile robot trap detection methods have misjudgment problems, especially when the threshold setting is unreasonable, which can easily lead to misjudgment.
The method of combining vision sensors and line laser sensors is used for trap detection, and by judging the body state, it determines which sensor to use for detection. The specific steps include: determine whether the body is raised. If it is raised, use any sensor to determine the jamming. If it is not raised, both need to be judged to be stuck before confirmation.
Improve the accuracy of stuck-up detection, reduce misjudgment, and ensure that the mobile robot can accurately detect stuck-up status.
Smart Images
Figure CN119896422A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent mobile robots, and in particular to a method for detecting when a mobile robot is stuck and a method for escaping from a stuck state. Background Art
[0002] At present, more and more families are using intelligent mobile robots such as sweeping robots. In the actual user home environment, mobile robots may be stuck due to various complex environments. If they are stuck, many actions cannot be performed, which brings a very bad experience to users. The current common method of detecting stuck is to judge the changes in the pitch angle and roll angle of the gyroscope. This judgment method has defects. When the threshold is set unreasonably, it is easy to make a misjudgment. Summary of the invention
[0003] The present application provides a mobile robot stuck detection method and escape method, the specific technical solution is as follows: A mobile robot jam detection method specifically includes the following steps: the mobile robot determines whether the body is tilted, and if the body is not tilted, uses a visual sensor and a line laser sensor to determine whether a jam occurs, and when the mobile robot uses both the visual sensor and the line laser sensor to determine that a jam occurs, the mobile robot is considered to be jammed, otherwise the mobile robot is considered to be not jammed; if the body is tilted, the mobile robot uses the visual sensor and the line laser sensor to determine whether a jam occurs, and when the mobile robot uses either the visual sensor or the line laser sensor to determine that a jam occurs, the mobile robot is considered to be jammed, otherwise the mobile robot is considered to be not jammed.
[0004] Furthermore, the method for a mobile robot to use a visual sensor to determine whether it is stuck specifically includes: the mobile robot continuously collects images through the visual sensor, and then compares the similarity between adjacent images. If the similarity of a preset number of consecutive images exceeds a first threshold, it is considered that the mobile robot is stuck.
[0005] Furthermore, when the number of images whose similarity exceeds the first threshold value does not continuously increase to a preset number, the counted number is cleared and counting starts again.
[0006] Furthermore, the method for comparing the similarity between adjacent images specifically includes: after the mobile robot collects images through a visual sensor, the image is cropped; using a normalized cross-correlation matching algorithm to calculate the correlation of the cropped adjacent images to obtain a normalized cross-correlation coefficient; wherein, the larger the normalized cross-correlation coefficient, the more similar the adjacent images are.
[0007] Furthermore, the method for a mobile robot to use a line laser sensor to determine whether a jam occurs specifically includes: the mobile robot continuously collects point cloud data through the line laser sensor, and then determines whether adjacent frames of point cloud data are similar; if the accumulated number of frames of similar point cloud data reaches a preset number of frames, it is considered that the mobile robot is stuck.
[0008] Furthermore, the method for determining whether adjacent frames of point cloud data are similar specifically includes: a mobile robot collects point cloud data through a line laser sensor, and converts the coordinates of point cloud data with a height greater than or equal to a preset height into grid coordinates; compares the current frame of point cloud data with the previous frame of point cloud data, and at the same grid position, if the current frame of point cloud data and the previous frame of point cloud data exist at the same time, the height is subtracted and the absolute value is taken to obtain the height difference, and if the height difference is less than or equal to the preset difference, it is considered that the current frame of point cloud data and the previous frame of point cloud data at the grid position are similar points; counts the number of similar points between the current frame of point cloud data and the previous frame of point cloud data, and when the number of similar points is greater than or equal to the preset number, the adjacent frames of point cloud data are considered to be similar, otherwise the adjacent frames of point cloud data are considered to be dissimilar.
[0009] Furthermore, the method for determining whether adjacent frames of point cloud data are similar specifically includes: a mobile robot collects point cloud data through a line laser sensor, and converts the coordinates of point cloud data with a height greater than or equal to a preset height into grid coordinates; the current frame of point cloud data is compared with the previous frame of point cloud data, and at the same grid position, if the current frame of point cloud data and the previous frame of point cloud data exist at the same time, the height is subtracted and the absolute value is taken to obtain the height difference, and if the height difference is less than or equal to the preset difference, it is considered that the current frame of point cloud data and the previous frame of point cloud data at the grid position are similar points; the obtained height differences are accumulated to obtain a height difference accumulated value, and when the height difference accumulated value is less than or equal to the preset value and the number of similar points is greater than or equal to the preset number, the adjacent frames of point cloud data are considered to be similar, otherwise the adjacent frames of point cloud data are considered to be dissimilar.
[0010] Furthermore, after the mobile robot converts the coordinates of the point cloud data having a height greater than or equal to the preset height into grid coordinates, if there are multiple point cloud data at the same grid position, the point cloud data with the largest height is retained.
[0011] Furthermore, in the process of judging whether adjacent frame point cloud data are similar, when adjacent frame point cloud data are similar, the cumulative number of frames is increased by 1, and when adjacent frame point cloud data are not similar, the cumulative number of frames is decreased by 1. If the cumulative number of frames reaches the preset number of frames, it is considered that the mobile robot is stuck.
[0012] A method for a mobile robot to escape from a jam, the method comprising a mobile robot jam detection method, and the method further comprising the following steps: when the mobile robot body is not tilted and is stuck, the mobile robot controls the body to rotate, then moves in an arbitrary direction, and continuously executes the mobile robot jam detection method within a preset time; if it is detected that the mobile robot is not stuck, it is considered that the mobile robot has escaped successfully; if the mobile robot is still detected to be stuck after the preset time, it is considered that the mobile robot has failed to escape; when the mobile robot body is tilted and is stuck, the mobile robot moves in an arbitrary direction, and continuously executes the mobile robot jam detection method within a preset time; if it is detected that the mobile robot is not stuck, it is considered that the mobile robot has escaped successfully; if the mobile robot is still detected to be stuck after the preset time, the mobile robot controls the body to rotate and determines whether the body is tilted; if the body is no longer tilted, it is considered that the mobile robot has escaped successfully; if the body is still tilted, it is considered that the mobile robot has failed to escape.
[0013] Furthermore, when the mobile robot body is not tilted and is stuck, the method of the mobile robot controlling the body to rotate and then moving in any direction specifically includes: the mobile robot rotates the body by changing the speed of two wheels, and then detects whether the heading angle of the gyroscope changes; if the heading angle does not change, the speed of the two wheels is increased and then the next step is entered; if the heading angle changes, the next step is entered directly; wherein there is a speed difference between the two wheels; the mobile robot controls the body to move in a first preset direction, and then controls the body to move in a second preset direction; wherein the first preset direction is in a different direction from the second preset direction.
[0014] The method for detecting a mobile robot getting stuck described in the present application first determines the body state of the mobile robot, and then detects whether the mobile robot is stuck by combining a visual sensor and / or a line laser sensor according to different body states. If the body is not tilted, a visual sensor and a line laser sensor are used to determine whether a jam has occurred. When the judgment results of the visual sensor and the line laser sensor are both yes, the mobile robot is considered to be stuck, otherwise the mobile robot is considered not to be stuck. If the body is tilted, the mobile robot uses a visual sensor and a line laser sensor to determine whether a jam has occurred. When the judgment result of any one of the visual sensor and the line laser sensor is yes, the mobile robot is considered to be stuck, otherwise the mobile robot is considered not to be stuck. The method for detecting a mobile robot getting stuck described in the present application adds the auxiliary judgment of vision and / or line laser to improve the accuracy of jam detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 A flowchart of a mobile robot jam detection method according to an embodiment of the present application.
[0016] Figure 2 A flowchart of a method for a mobile robot to use a visual sensor to determine whether a jam has occurred, as described in one embodiment of the present application.
[0017] Figure 3 A flowchart of a method for a mobile robot to use a line laser sensor to determine whether a jam has occurred, as described in one embodiment of the present application. Implementation
[0018] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0019] It should be understood that when used in this application, the term "comprising" indicates the presence of the described features, integral bodies, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integral bodies, steps, operations, elements, components and / or their collections. It should also be understood that the term "and / or" used in this application refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.
[0020] As used in this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.
[0021] In addition, in the description of the present application, the terms "first", "second", "third", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance. The reference to "one embodiment" or "some embodiments" described in the specification of the present application means that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in the differences in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0022] At present, more and more families are using intelligent mobile robots such as sweeping robots. In the actual user home environment, mobile robots may be stuck due to various complex environments. If they are stuck, many actions cannot be performed, which brings a very bad experience to users. The current common method of detecting stuck is to judge the changes in the pitch angle and roll angle of the gyroscope. This judgment method has defects. When the threshold is set unreasonably, it is easy to make a misjudgment.
[0023] In order to solve the above technical problems, an embodiment of the present application provides a method for detecting a mobile robot being stuck. First, the body state of the mobile robot is determined. Then, based on different body states, a visual sensor and / or a line laser sensor is used to detect whether the mobile robot is stuck. The auxiliary judgment of vision and / or line laser is added to improve the accuracy of jam detection.
[0024] like Figure 1 As shown, the mobile robot stuck detection method specifically includes the following steps: The mobile robot determines whether the body is tilted. If the body is not tilted, the visual sensor and the line laser sensor are used to determine whether a jam occurs. When the mobile robot determines that a jam occurs using both the visual sensor and the line laser sensor, the mobile robot is considered to be stuck. Otherwise, the mobile robot is considered not to be stuck. If the body is tilted, the mobile robot uses the visual sensor and the line laser sensor to determine whether it is stuck. When the mobile robot uses either the visual sensor or the line laser sensor to determine that it is stuck, the mobile robot is considered to be stuck; otherwise, the mobile robot is considered not to be stuck.
[0025] Among them, regarding the judgment order of the visual sensor and the line laser sensor, the visual sensor can be used first to determine whether the mobile robot is stuck, and then the line laser sensor can be used to make a judgment, or vice versa. This application does not limit the judgment order. However, when the mobile robot body is tilted, if any sensor is used to detect that the mobile robot is stuck, it is no longer necessary to use another sensor to make a judgment, so as to improve the detection efficiency. Because when the body is tilted, it has been shown that the mobile robot has the possibility of being stuck, combined with the judgment that any sensor detects that the mobile robot is stuck, it is sufficient to draw this conclusion.
[0026] It should be noted that the mobile robot will continuously perform simultaneous localization and mapping (SLAM) through gyroscopes, visual sensors and / or line laser sensors during operation. This is a prior art and will not be described in detail here. On the other hand, the mobile robot can use a gyroscope to determine whether the body is tilted, which is also a prior art and will not be described in detail here.
[0027] As one of the implementation methods, Figure 2 As shown, the method for a mobile robot to use a visual sensor to determine whether it is stuck specifically includes: the mobile robot continuously collects images through the visual sensor, and then compares the similarity between adjacent images. If the similarity of a preset number of consecutive images exceeds a first threshold, it is considered that the mobile robot is stuck. If the similarity of multiple consecutive images is high, for example, the mobile robot collects an image every 1 second, and the similarity of 5 consecutive images is high, it indicates that the mobile robot is stuck in place and the images collected are at the same location. The method described in this embodiment has high accuracy and strong practicality.
[0028] Based on the above embodiment, the method for comparing the similarity between adjacent images specifically includes: after the mobile robot collects the image through the visual sensor, the image is cropped; the correlation of the cropped adjacent images is calculated using the normalized cross-correlation matching algorithm to obtain the normalized cross-correlation coefficient; wherein, the larger the normalized cross-correlation coefficient, the more similar the adjacent images are. Wherein, the cropping is an edge cropping process, that is, the pixel values of the edges of an image are cropped by a certain threshold, for example, the size of a 320*240 image is changed to 300*220 after cropping, and the middle area after cropping is taken, which can avoid the overlapping area of the edges of adjacent images from bringing too high correlation estimation. After cropping, the NCC algorithm (Normalized Cross Correlation, normalized cross-correlation matching algorithm) based on the matching method of image grayscale information calculates the mean and standard deviation of the two image areas, and normalizes the pixel values. Then, the normalized cross-correlation coefficient is obtained by calculating the correlation between the pixel values of the two normalized image areas. The value range of the normalized cross-correlation coefficient is in the range of -1 to 1, and the larger the value, the more similar the two image areas are. In one embodiment, when the similarity of five consecutive images exceeds the first threshold, the mobile robot is considered to be in a fixed position and is in a stuck state. Preferably, the first threshold is 0.9. It should be noted that when the number of images whose similarity exceeds the first threshold does not increase to five images in a row, the counted number is cleared and restarted. For example, when the mobile robot detects three images whose similarity exceeds the first threshold in a row, and the similarity of the next image to arrive does not exceed the first threshold with the previous image, then the next time the similarity of the two images is detected to exceed the first threshold, it is counted from 1 instead of 4. This can avoid the situation where the mobile robot mistakenly thinks that it is stuck due to the similarity of the surrounding environment. On the other hand, even if the mobile robot is indeed stuck, it may be possible to free itself in a short time. Such a setting is more reasonable.
[0029] As one of the implementation methods, Figure 3As shown, the method for a mobile robot to use a line laser sensor to determine whether a jam occurs specifically includes: the mobile robot continuously collects point cloud data through a line laser sensor, and then determines whether adjacent frames of point cloud data are similar. If the cumulative number of frames of similar point cloud data reaches a preset number of frames, it is considered that the mobile robot is stuck. In one embodiment, the mobile robot uses a single-line laser sensor to determine whether a jam occurs. The single-line laser sensor can obtain 130 point cloud data, and the point cloud data includes angle information, height information, and distance information. When the mobile robot moves, the point cloud data also changes. When multiple frames of point cloud data are almost unchanged, for example, 20 frames of point cloud data remain unchanged, it indicates that the mobile robot is stuck in place and the point cloud data collected is at the same position. The method described in this embodiment has high accuracy and strong practicality.
[0030] Based on the above embodiment, the method for judging whether adjacent frame point cloud data are similar specifically includes: the mobile robot collects point cloud data through a line laser sensor, and converts the coordinates of the point cloud data whose height is greater than or equal to the preset height into grid coordinates; compares the current frame point cloud data with the previous frame point cloud data, and at the same grid position, if the current frame point cloud data and the previous frame point cloud data exist at the same time, the height is subtracted and the absolute value is taken to obtain the height difference, and if the height difference is less than or equal to the preset difference, it is considered that the current frame point cloud data and the previous frame point cloud data at the grid position are similar points; count the number of similar points between the current frame point cloud data and the previous frame point cloud data, and when the number of similar points is greater than or equal to the preset number, it is considered that the adjacent frame point cloud data are similar, otherwise it is considered that the adjacent frame point cloud data are not similar. Among them, converting the coordinates of the point cloud data from the line laser coordinate system to the grid map coordinate system is a prior art and will not be repeated here. After obtaining the grid coordinates, the grid position of the point cloud data can be known, and the height information of the point cloud data is retained for subsequent judgment.
[0031] In another embodiment, in the process of counting the number of similar points between the current frame point cloud data and the previous frame point cloud data, the obtained height differences are accumulated to obtain a height difference accumulated value. When the height difference accumulated value is less than or equal to a preset value and the number of similar points is greater than or equal to a preset number, the adjacent frames of point cloud data are considered to be similar; otherwise, the adjacent frames of point cloud data are considered to be dissimilar. The method described in this embodiment improves the accuracy of judging whether two frames of point cloud data are similar.
[0032] As one of the implementation methods, after the mobile robot converts the coordinates of the point cloud data whose height is greater than or equal to the preset height into grid coordinates, if there are multiple point cloud data at the same grid position, the point cloud data with the largest height is retained. The method described in the embodiment of the present application is for the same frame of point cloud data. When there are two or more point cloud data on the same grid, only the one with the largest height value is retained for comparison with the next frame of point cloud data.
[0033] As one of the implementation methods, in the process of judging whether adjacent frame point cloud data are similar, when adjacent frame point cloud data are similar, the cumulative number of frames is increased by 1, and when adjacent frame point cloud data are not similar, the cumulative number of frames is decreased by 1. If the cumulative number of frames reaches the preset number of frames, it is considered that the mobile robot is stuck. It should be noted that in the process of using a visual sensor to judge whether the mobile robot is stuck, the similarity of a preset number of images exceeds the first threshold, and the mobile robot is considered to be stuck. In the process of using a line laser sensor to judge whether the mobile robot is stuck, the cumulative number of frames is counted. Even if it is interrupted in the middle, as long as the cumulative number of frames reaches the preset number of frames, it can be considered that the mobile robot is stuck. This is because the information of a frame of line laser is less than that of a frame of image, and the line laser data is more affected by the jitter of the robot. A slight jitter may result in dissimilar results. Even if the visual sensor jitters, it has a greater impact on the edge information of the image. The above-mentioned cropping of the edge information of the image can greatly reduce this impact.
[0034] The embodiment of the present application provides a mobile robot escape method, the mobile robot escape method includes the mobile robot stuck detection method, and the mobile robot escape method also includes the following steps: When the mobile robot body is not tilted and is stuck, the mobile robot controls the body to rotate and then moves forward in any direction, and the mobile robot stuck detection method is continuously executed within a preset time. If it is detected that the mobile robot is not stuck, it is considered that the mobile robot is successfully freed; if it is still detected that the mobile robot is stuck after the preset time, it is considered that the mobile robot fails to be freed; When the body of the mobile robot is tilted and stuck, the mobile robot moves forward in any direction and continuously executes the mobile robot jam detection method within a preset time. If it is detected that the mobile robot is not stuck, it is considered that the mobile robot has escaped successfully. If the mobile robot is still detected to be stuck after the preset time, the mobile robot controls the body to rotate and determines whether the body is tilted. If the body is no longer tilted, it is considered that the mobile robot has escaped successfully. If the body is still tilted, it is considered that the mobile robot has failed to escape.
[0035] The mobile robot escape method described in the embodiment of the present application adopts different escape actions according to the different body states of the mobile robot, thereby improving the success rate of escape. Among them, when the body of the mobile robot is not tilted, it is equivalent to being stuck on a plane. A more common situation is that the mobile robot is trapped in a carpet and cannot move in a straight line, and cannot move normally whether it is moving backward or forward. Therefore, when being stuck on a plane, the mobile robot first controls the body to rotate, with the purpose of trying to change the direction to get the mobile robot out of trouble. When the body of the mobile robot is tilted, it is usually lifted by the fan base or the door sill, which is equivalent to being stuck on a slope. At this time, the mobile robot first moves in any direction, such as first retreating and then rushing forward (equivalent to a rushing action), and usually does not need to rotate to get out of trouble. If the escape is not successful, the mobile robot is controlled to rotate in order to get rid of the jam, or after rotating, it rushes again to get out of trouble.
[0036] As one of the implementation modes, when the mobile robot body is not tilted and is stuck, the mobile robot controls the body to rotate, and then moves forward in any direction. Specifically, the method includes: the mobile robot rotates the body by changing the speed of the two wheels, and then detects whether the heading angle of the gyroscope changes. If the heading angle does not change, the speed of the two wheels is increased and then the next step is entered. If the heading angle changes, the next step is entered directly; wherein, there is a speed difference between the two wheels; the mobile robot controls the body to move in a first preset direction, and then controls the body to move in a second preset direction; wherein, the first preset direction is in a different direction from the second preset direction. The method described in this embodiment first rotates gently, and when the mobile robot cannot move, the rotation force is increased so that the mobile robot can change direction, and then tries to move in different directions to increase the success rate of the mobile robot getting out of trouble.
[0037] As one of the implementation methods, when the mobile robot body is tilted and stuck, the mobile robot uses the method of rushing (first retreating and then rushing) to get out of the jam. If it cannot move during the retreat process, it will try to twist and retreat to increase the force, and then rush forward. If the mobile robot is still stuck after rushing, it will try to get out of the jam by rotating. During the rotation process, if the pitch angle of the gyroscope drops below a preset angle, such as below 2 degrees, it is considered that the mobile robot has successfully escaped.
[0038] Based on the above embodiment, when the mobile robot cannot escape successfully, the user is notified of the relevant situation through the application layer. For example, a pop-up window prompt is displayed through the APP on the smart terminal that is compatible with the mobile robot, so that the user can help the mobile robot escape in time.
[0039] The embodiment of the present application also provides a chip, the chip stores a computer program, and when the computer program is executed, the mobile robot jam detection method or the mobile robot escape method is implemented. The chip described in the embodiment of the present application adopts different jam detection methods or escape methods according to different body states of the mobile robot, thereby improving the jam detection accuracy or escape success rate.
[0040] The embodiment of the present application also provides a mobile robot, the mobile robot is equipped with the chip, and the mobile robot also includes a gyroscope, a visual sensor and a line laser sensor. Among them, the gyroscope is used to detect the state of the mobile robot body, and judge whether the body is tilted by the pitch angle, the visual sensor is used to collect environmental images, and the line laser sensor is used to collect point cloud data. The data of the above three sensors are all transmitted to the chip for processing, so that the mobile robot can detect stuck or get out of trouble.
[0041] It will be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the references to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory ROM, programmable memory PROM, electrically programmable memory DPROM, electrically erasable programmable memory DDPROM or flash memory. Volatile memory may include random access memory RAM or external cache memory.
[0042] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0043] The above embodiments only express several embodiments of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.
Claims
1. A mobile robot stuck detection method, characterized in that: The mobile robot stuck detection method specifically comprises the following steps: The mobile robot determines whether the body is tilted. If the body is not tilted, the visual sensor and the line laser sensor are used to determine whether a jam occurs. When the mobile robot determines that a jam occurs using both the visual sensor and the line laser sensor, the mobile robot is considered to be stuck. Otherwise, the mobile robot is considered not to be stuck. If the body is tilted, the mobile robot uses the visual sensor and the line laser sensor to determine whether it is stuck. When the mobile robot uses either the visual sensor or the line laser sensor to determine that it is stuck, the mobile robot is considered to be stuck; otherwise, the mobile robot is considered not to be stuck.
2. A mobile robot stuck detection method according to claim 1, characterized in that: The method for a mobile robot to use a visual sensor to determine whether a jam has occurred specifically includes: The mobile robot continuously collects images through a visual sensor, and then compares the similarity between adjacent images. If the similarity of a preset number of consecutive images exceeds a first threshold, it is considered that the mobile robot is stuck.
3. A mobile robot stuck detection method according to claim 2, characterized in that: When the number of images whose similarity exceeds the first threshold value does not continuously increase to a preset number, the counted number is cleared and counting starts again.
4. A mobile robot stuck detection method according to claim 2, characterized in that: The method for comparing the similarity between adjacent images specifically includes: After the mobile robot collects images through the visual sensor, it crops the images; The normalized cross-correlation matching algorithm is used to calculate the correlation of the cropped adjacent images to obtain the normalized cross-correlation coefficient; Among them, the larger the normalized cross-correlation coefficient is, the more similar the adjacent images are.
5. A mobile robot stuck detection method according to claim 1, characterized in that: The method for a mobile robot to use a line laser sensor to determine whether a jam occurs specifically includes: The mobile robot continuously collects point cloud data through a line laser sensor, and then determines whether adjacent frames of point cloud data are similar. If the cumulative number of frames of similar point cloud data reaches a preset number of frames, the mobile robot is considered to be stuck.
6. A mobile robot stuck detection method according to claim 5, characterized in that: The method for determining whether adjacent frame point cloud data are similar specifically includes: The mobile robot collects point cloud data through a line laser sensor and converts the coordinates of the point cloud data with a height greater than or equal to a preset height into grid coordinates; Compare the current frame point cloud data with the previous frame point cloud data. At the same grid position, if the current frame point cloud data and the previous frame point cloud data exist at the same time, then subtract the height and take the absolute value to get the height difference. If the height difference is less than or equal to the preset difference, then the current frame point cloud data and the previous frame point cloud data at the grid position are considered to be similar points. The number of similar points between the current frame point cloud data and the previous frame point cloud data is counted. When the number of similar points is greater than or equal to the preset number, the adjacent frame point cloud data are considered to be similar, otherwise the adjacent frame point cloud data are considered to be dissimilar.
7. A mobile robot stuck detection method according to claim 5, characterized in that: The method for determining whether adjacent frame point cloud data are similar specifically includes: The mobile robot collects point cloud data through a line laser sensor and converts the coordinates of the point cloud data with a height greater than or equal to a preset height into grid coordinates; Compare the current frame point cloud data with the previous frame point cloud data. At the same grid position, if the current frame point cloud data and the previous frame point cloud data exist at the same time, then subtract the height and take the absolute value to get the height difference. If the height difference is less than or equal to the preset difference, then the current frame point cloud data and the previous frame point cloud data at the grid position are considered to be similar points. The obtained height differences are accumulated to obtain a height difference accumulation value. When the height difference accumulation value is less than or equal to a preset value and the number of similar points is greater than or equal to a preset number, the adjacent frame point cloud data are considered to be similar, otherwise the adjacent frame point cloud data are considered to be dissimilar.
8. A mobile robot stuck detection method according to claim 6, characterized in that: After the mobile robot converts the coordinates of the point cloud data whose height is greater than or equal to the preset height into grid coordinates, if there are multiple point cloud data at the same grid position, the point cloud data with the largest height is retained.
9. A mobile robot stuck detection method according to claim 5, characterized in that: In the process of judging whether adjacent frame point cloud data are similar, when adjacent frame point cloud data are similar, the cumulative number of frames is increased by 1, and when adjacent frame point cloud data are not similar, the cumulative number of frames is decreased by 1. If the cumulative number of frames reaches the preset number of frames, it is considered that the mobile robot is stuck.
10. A method for escaping a mobile robot, characterized in that: The mobile robot escape method comprises the mobile robot stuck detection method according to any one of claims 1 to 9, and the mobile robot escape method further comprises the following steps: When the mobile robot body is not tilted and is stuck, the mobile robot controls the body to rotate and then moves forward in any direction, and the mobile robot stuck detection method is continuously executed within a preset time. If it is detected that the mobile robot is not stuck, it is considered that the mobile robot is successfully freed; if it is still detected that the mobile robot is stuck after the preset time, it is considered that the mobile robot fails to be freed; When the body of the mobile robot is tilted and stuck, the mobile robot moves forward in any direction and continuously executes the mobile robot jam detection method within a preset time. If it is detected that the mobile robot is not stuck, it is considered that the mobile robot has escaped successfully. If the mobile robot is still detected to be stuck after the preset time, the mobile robot controls the body to rotate and determines whether the body is tilted. If the body is no longer tilted, it is considered that the mobile robot has escaped successfully. If the body is still tilted, it is considered that the mobile robot has failed to escape.
11. A method for escaping a mobile robot according to claim 10, characterized in that: When the mobile robot body is not tilted and is stuck, the method for the mobile robot to control the body to rotate and then move forward in any direction specifically includes: The mobile robot rotates the body by changing the speed of the two wheels, and then detects whether the heading angle of the gyroscope changes. If the heading angle does not change, the speed of the two wheels is increased and then the next step is entered. If the heading angle changes, the next step is entered directly. There is a speed difference between the two wheels. The mobile robot controls the body to move forward in a first preset direction, and then controls the body to move forward in a second preset direction; wherein the first preset direction and the second preset direction are different directions.