Patent Application
20240197135
This application claims priority to Chinese Patent Application No. 202211607717.3, filed in Chinese Patent Office on Dec. 14, 2022. The entire content of the above-referenced application is incorporated herein by reference in this application.
The present disclosure generally relates to a detection method for an autonomous mobile devices an autonomous mobile device, and a storage medium, and more specifically, to a method for detecting a doorsill on a work surface of an autonomous mobile device.
As the advance of technologies of autonomous mobile devices, autonomous mobile devices having multiple functions have been developed. An important category is cleaning robot used for indoor cleaning. For an indoor cleaning robot, an important capability is the obstacle clearance capability. However, due to the limitations of space and cost, the obstacle clearance capability is typically difficult to be well designed. Thus, when the obstacle is relatively high, the cleaning robot will have difficulty in negotiating the obstacle. A doorsill is a common design used at home. Because there is no fixed standard for the doorsill, there may be large differences in the heights and shapes for different doorsills. Some doorsills that are overly high and that have shapes not friendly to the chassis of the cleaning robot may cause the cleaning robot to be stuck or to have difficulty in negotiating the doorsills. Ultimately, the cleaning robot may generate an alarm or may miss some areas-to-be-cleaned.
In current technology, object recognition may be realized using cameras. Specifically, for example, the line profile of a door or a door frame may be recognized in images of the environment in front of, or above, the cleaning robot, thereby determining a corresponding location of the doorsill. In some existing technology, the location of the door or the doorsill may be determined based on information relating to a wall or room in the map. In some existing technology, the corresponding location of the doorsill may be determined based on the pose of the cleaning robot or a distance of the chassis of the cleaning robot from the floor. In addition, in theory, the detection of the doorsill may also be realized by 3D mapping of the work environment through a structured light sensor. In some other existing technology, the obstacle clearance capability of the cleaning robot may be increased by increasing the performance of the hardware components.
However, the above-described existing technology places a relatively high requirement on the hardware, software, or algorithm. For example, on one hand, the technical solution of object recognition through camera requires a relatively high computing power, which increases the requirement on the performance of the processor, as well as the cost of the cleaning robot. On the other hand, obstacle recognition usually uses deep learning methods. The rate of recognition is limited and there are occasions when the doorsills are mistakenly detected. This may interfere with the normal work/cleaning process of the cleaning robot, and lower the user experience. As another example, similar issues exist in the technical solutions using a structured light.
The object of the present disclosure is to overcome or at least mitigate the deficiencies of the existing technology, and to provide a doorsill detection method for an autonomous mobile device. The method utilizes limited hardware and computing power. A location of a doorsill may be obtained after the autonomous mobile device performs a limited movement in the environment, and the location of the doorsill may be saved in a map for subsequent operations.
According to a first aspect of the present disclosure, a detection method for an autonomous mobile device is provided. The method includes: obtaining an environmental map of an environment in which the autonomous mobile device is located, the environmental map being a grid map; obtaining grids-to-be-processed in the environmental map; clustering the grids-to-be-processed to obtain one or more groups of grids-to-be-processed; and for each group of grids-to-be-processed, when a number of grids-to-be-processed included in the group of grids-to-be-processed is within a predetermined range, determining whether the group of grids-to-be-processed corresponds to a doorsill.
According to a second aspect of the present disclosure, a detection method for an autonomous mobile device is provided. The method includes: obtaining an environmental map of an environment in which the autonomous mobile device is located, the environmental map being a grid map; obtaining grids-to-be-processed in the environmental map; clustering the grids-to-be-processed to obtain one or more groups of grids-to-be-processed; and for each group of grids-to-be-processed, determining whether the group of grids-to-be-processed corresponds to a doorsill based on environmental information.
In some embodiments, determining whether the group of grids-to-be-processed corresponds to a doorsill based on the environmental information includes: when a number of grids-to-be-processed included in the group of grids-to-be-processed is within a predetermined number range, determining whether the group of grids-to-be-processed corresponds to a doorsill based on the environmental information.
In some embodiments, the environmental information includes at least one of: obstacle information in a light detection and ranging (LiDAR) map of the environment; at least portion of video information of the environment; or structured light scanning result of the environment.
In some embodiments, the environmental map is obtained by the autonomous mobile device in an explore mode.
In some embodiments, the environmental map is created through at least one of: a light detection and ranging (LiDAR) device; a structured light sensor; or a visual sensor.
In some embodiments, obtaining grids-to-be-processed in the environmental map includes, for each grid of a plurality of grids traversed by the autonomous mobile device when moving in the environment: detecting the grid through a floor sensor mounted on the autonomous mobile device to obtain a floor sensor signal corresponding to the grid; and determining that the grid is a grid-to-be-processed based on the floor sensor signal.
In some embodiments, the floor sensor is an ultrasound sensor, the floor sensor signal is a reflected ultrasound wave. Detecting the grid through a floor sensor mounted on the autonomous mobile device to obtain a floor sensor signal corresponding to the grid includes: transmitting, through the ultrasound sensor, a plurality of ultrasound pulses toward the grid and receiving a plurality of reflected ultrasound wave; and determining that the grid is a grid-to-be-processed based on the floor sensor signal includes: counting, in the plurality of reflected ultrasound waves, a number of reflected ultrasound waves having a reflected wave intensity within a predetermined ultrasound intensity threshold range; and when the number of reflected ultrasound waves is within a count threshold range, determining that the grid is a grid-to-be-processed.
In some embodiments, the floor sensor is an infrared sensor, and the floor sensor signal is a reflected infrared light. Detecting the grid through a floor sensor mounted on the autonomous mobile device to obtain a floor sensor signal corresponding to the grid includes: transmitting, through the infrared sensor, a plurality of infrared pulses toward the grid and receiving a plurality of reflected infrared lights; and determining that the grid is a grid-to-be-processed based on the floor sensor signal includes: counting, in the plurality of reflected infrared lights, a number of reflected infrared lights having a light intensity within a predetermined light intensity threshold range; and when the number of reflected infrared lights is within a predetermined count threshold range, determining that the grid is a grid-to-be-processed.
In some embodiments, the floor sensor is a video sensor, and the floor sensor signal is a video frame. Detecting the grid through a floor sensor mounted on the autonomous mobile device to obtain a floor sensor signal corresponding to the grid includes: capturing, through the video sensor, a video frame of a floor portion corresponding to the grid; and determining that the grid is a grid-to-be-processed based on the floor sensor signal includes: determining that the grid is a grid-to-be-processed based on the video frame.
In some embodiments, the method also includes: when the number of grids-to-be-processed included in the group of grids-to-be-processed is greater than a maximum value of the predetermined number range, determining that the group of grids-to-be-processed corresponds to a carpet.
In some embodiments, the method also includes: when the number of grids-to-be-processed included in the group of grids-to-be-processed is lower than a minimum value of the predetermined number range, determining that the group of grids-to-be-processed corresponds to a lower base of an object in the environment.
In some embodiments, the environmental map is a LiDAR map; the environmental information includes obstacle information in the LiDAR map of the environment; and determining whether the group of grids-to-be-processed corresponds to a doorsill based on the environmental information includes: calculating a location of a center of the group of grids-to-be-processed in the environmental map; based on the location, obtaining a map portion that includes the location from the environmental map; and when the map portion includes obstacle information indicating a door or a hallway, determining that the group of grids-to-be-processed corresponds to the doorsill.
In some embodiments, after determining that the group of grids-to-be-processed corresponds to the doorsill, the method also includes: labelling the doorsill in the environmental map.
In some embodiments, after determining that the group of grids-to-be-processed corresponds to the doorsill, the method also includes: in response to a determination that skidding occurred to the autonomous mobile device and a distance from a current location of the autonomous mobile device to the doorsill is smaller than or equal to a predetermined distance value, controlling the autonomous mobile device to perform predetermined predicament avoidance actions.
In some embodiments, the autonomous mobile device is a cleaning robot operable in a wet-mopping mode; and after determining that the group of grids-to-be-processed corresponds to the doorsill, the method also includes: when the autonomous mobile device is in the wet-mopping mode, controlling the autonomous mobile device to move across the doorsill and to continue moving in the wet-mopping mode on the other side of the doorsill.
According to a third aspect of the present disclosure, an autonomous mobile device is provided. The autonomous mobile device includes: a motion assembly configured to move the autonomous mobile device in an environment; a storage device configured to store computer-executable instructions; and a processor configured to retrieve and execute the computer-executable instructions to: obtain an environmental map of an environment in which the autonomous mobile device is located, the environmental map being a grid map; obtain grids-to-be-processed in the environmental map; cluster the grids-to-be-processed to obtain one or more groups of grids-to-be-processed; and for each group of grids-to-be-processed, when a number of grids-to-be-processed included in the group of grids-to-be-processed is within a predetermined range, determine whether the group of grids-to-be-processed corresponds to a doorsill.
According to a fourth aspect of the present disclosure, an autonomous mobile device is provided. The autonomous mobile device includes: a motion assembly configured to move the autonomous mobile device in an environment; a storage device configured to store computer-executable instructions; and a processor configured to retrieve and execute the computer-executable instructions to: obtain an environmental map of an environment in which the autonomous mobile device is located, the environmental map being a grid map; obtain grids-to-be-processed in the environmental map; cluster the grids-to-be-processed to obtain one or more groups of grids-to-be-processed; and for each group of grids-to-be-processed, determine whether the group of grids-to-be-processed corresponds to a doorsill based on environmental information.
In some embodiments, the processor is also configured to, when a number of grids-to-be-processed included in the group of grids-to-be-processed is within a predetermined range, determine whether the group of grids-to-be-processed corresponds to a doorsill based on environmental information.
In some embodiments, the environmental map is created through at least one of the following components mounted on the autonomous mobile device: a LiDAR device; a structured light sensor; or a visual sensor.
In some embodiments, the autonomous mobile device also includes a floor sensor mounted on the autonomous mobile device, for each grid of a plurality of grids traversed by the autonomous mobile device when moving in the environment: the floor sensor is configured to detect the grid to obtain a floor sensor signal corresponding to the grid; and the processor is configured to determine that the grid is a grid-to-be-processed based on the floor sensor signal.
In some embodiments, the floor sensor is an ultrasound sensor, the floor sensor signal is a reflected ultrasound wave. The ultrasound sensor is configured to transmit a plurality of ultrasound pulses toward the grid and receive a plurality of reflected ultrasound wave. The processor is configured to: count, in the plurality of reflected ultrasound waves, a number of reflected ultrasound waves having a reflected wave intensity within a predetermined ultrasound intensity threshold range; and when the number of reflected ultrasound waves is within a count threshold range, determine that the grid is a grid-to-be-processed.
In some embodiments, the floor sensor is an infrared sensor, and the floor sensor signal is a reflected infrared light. The infrared sensor is configured to transmit a plurality of infrared pulses toward the grid and receive a plurality of reflected infrared lights; and the processor is configured to: count, in the plurality of reflected infrared lights, a number of reflected infrared lights having a light intensity within a predetermined light intensity threshold range; and when the number of reflected infrared lights is within a predetermined count threshold range, determine that the grid is a grid-to-be-processed.
In some embodiments, the floor sensor is a video sensor, and the floor sensor signal is a video frame. The video sensor is configured to capture a video frame of a floor portion corresponding to the grid; and the processor is configured to: determine that the grid is a grid-to-be-processed based on the video frame.
In some embodiments, the processor is configured to: when the number of grids-to-be-processed included in the group of grids-to-be-processed is greater than a maximum value of the predetermined number range, determine that the group of grids-to-be-processed corresponds to a carpet.
In some embodiments, the processor is configured to: when the number of grids-to-be-processed included in the group of grids-to-be-processed is lower than a minimum value of the predetermined number range, determine that the group of grids-to-be-processed corresponds to a lower base of an object in the environment.
In some embodiments, the environmental map is a LiDAR map; the environmental information includes obstacle information in the LiDAR map of the environment; and the processor is configured to: calculate a location of a center of the group of grids-to-be-processed in the environmental map; based on the location, obtain a map portion that includes the location from the environmental map; and when the map portion includes obstacle information indicating a door or a hallway, determine that the group of grids-to-be-processed corresponds to the doorsill.
In some embodiments, the processor is configured to: after determining that the group of grids-to-be-processed corresponds to the doorsill, label the doorsill in the environmental map.
In some embodiments, the processor is configured to: after determining that the group of grids-to-be-processed corresponds to the doorsill, and in response to a determination that skidding occurred to the autonomous mobile device and a distance from a current location of the autonomous mobile device to the doorsill is smaller than or equal to a predetermined distance value, control the autonomous mobile device to perform predetermined predicament avoidance actions.
In some embodiments, the autonomous mobile device is a cleaning robot operable in a wet-mopping mode; and the processor is configured to: after determining that the group of grids-to-be-processed corresponds to the doorsill, when the autonomous mobile device is in the wet-mopping mode, control the autonomous mobile device to move across the doorsill and to continue moving in the wet-mopping mode on the other side of the doorsill.
According to a fifth aspect of the present disclosure, a computer-readable storage medium is provided. The storage medium stores a computer program including computer-executable instructions. When the instructions are executed by a processor of an autonomous mobile device, the above detection method of the present disclosure is executed.
According to the technical solutions of the present disclosure, for example, without significantly increasing the cost, the location of the doorsill may be relatively accurately detected, and false detection of the doorsill may be avoided.
Next, the embodiments of the present disclosure will be explained in detail with reference to the following drawings. Other features and aspects of the present disclosure will become clearer.
The drawings, which are included in the specification as parts of the specification, together with the specification, illustrate the exemplary embodiments, features and aspects of the present disclosure, and are used to explain the principle of the present disclosure.
Various illustrative embodiments, features, and aspects of the present disclosure will be described in detail with reference to the drawings. The same labels in the drawings indicate elements having the same or similar functions. Although various aspects of the embodiments are shown in the drawings, the drawings are not to scale, unless otherwise noted.
The term “illustrative” as used herein means “as an example, an embodiment, or explanatory.” Here, any embodiment described using the term “illustrative” does not necessarily mean that the embodiment is more advantageous or better than other embodiments.
In addition, to better explain the present disclosure, various specific details are described in the following detailed implementations. A person having ordinary skills in the art would understand that without certain specific details, the present disclosure can still be implemented. In some embodiments, methods, means, elements, and electric circuits that are well-known to a person having ordinary skills in the art are not described in detail, such that the main principle of the present disclosure can be better illustrated.
The detection method of the present disclosure may be applied to or may be performed by the autonomous mobile device 10 shown in
In some embodiments, in step 210, the environmental map may be obtained by the autonomous mobile device 10 in an explore mode. A typical cleaning robot or other autonomous mobile device is usually equipped with an “explore mode,” in which the autonomous mobile device may create a map of the environment in which the autonomous mobile device is located before performing the cleaning tasks or other tasks. The environmental map of the indoor space may be stored (locally or remotely) for use in subsequent cleaning tasks or other tasks. The environmental map of the present disclosure may be obtained using various techniques. For example, in the embodiment shown in
In the present disclosure, the environmental map may be a grid map. That is, the environment in which the autonomous mobile device 10 is located may be divided into a series of grids, such that each point in the environment is included in a grid, thereby forming the grid map. Each grid has a characteristic value to indicate whether the grid is occupied, i.e., whether the location corresponding to the grid is a space that the autonomous mobile device can freely move through, or is occupied by an obstacle.
In step 220, in order to obtain the grids-to-be-processed in the environmental map, steps 222-224 may be executed for each grid in at least a portion of the grids of the environmental map. The at least a portion of the grids of the environmental map may be grids traversed by the autonomous mobile device 10 in the explore mode or when executing other tasks. In the embodiment shown in
In the present disclosure, in order to recognize a doorsill, it is needed to find the “grids-to-be-processed” in the environmental map that may correspond to a doorsill. In step 222, a grid may be detected through a floor sensor 108 mounted on the autonomous mobile device 10 to obtain a floor sensor signal corresponding to the grid.
In some embodiments, the floor sensor may be an ultrasound sensor. As the autonomous mobile device 10 moves along a route, the ultrasound sensor may transmit an ultrasound wave toward each grid and may receive a reflected ultrasound wave. In some embodiments, for each grid, the ultrasound sensor may transmit a plurality of ultrasound pulses and receive a plurality of reflected ultrasound waves. In some embodiments, the floor sensor may be an infrared sensor. As the autonomous mobile device 10 moves along the route, the infrared sensor may transmit a plurality of infrared pulses toward each grid and receive a plurality of reflected infrared lights. In some embodiments, the floor sensor may be a video sensor. As the autonomous mobile device 10 moves along the route, the video sensor may capture one or more video frames of each grid along the route.
Step 224, determining whether the grid is a grid-to-be-processed based on the floor sensor signal.
In practice, when the autonomous mobile device passes a doorsill, for an ultrasound sensor or any other sensor for detecting a floor based on similar principle, the doorsill may show characteristics similar to some kind of floor material such as a carpet. Grids having such characteristics may be recognized as the grids-to-be-processed based on the floor sensor signal. In embodiments that use an ultrasound sensor, the reflected wave that is reflected back by the carpet may have a low intensity due to the characteristics of the material of the carpet Similarly, because the autonomous mobile device 10 may tilt when moving across a doorsill, due to the tilt angle, at least a portion of the reflected wave reflected by the floor may not be received, therefore, the reflected wave may have a low intensity. Correspondingly, a flat hard floor may provide a reflected wave having a high intensity. Therefore, for each grid, multiple ultrasound pulses may be transmitted and the intensity of multiple reflected ultrasound waves may be measured. A number of reflected ultrasound waves having a reflected wave intensity within a predetermined ultrasound intensity threshold range may be counted. If the number of reflected ultrasound waves is within a count threshold range, the grid may be determined as a grid-to-be-processed. For example, for each grid, the ultrasound sensor may transmit 50 ultrasound pulses, and may receive corresponding reflected ultrasound waves. The ultrasound intensity threshold range may be set as 10%-40% of the intensity of the transmitted ultrasound pulses, and a count threshold value may be set as 30, for example. If within the 50 received reflected ultrasound waves, there are 35 reflected ultrasound waves having an intensity that is within 10%-40% of the intensity of the transmitted ultrasound pulses, it may be determined that a floor portion corresponding to the grid has provided a reflected ultrasound wave that is relatively weak, thus, the floor portion may be a carpet or a doorsill, and the grid may be determined as a grid-to-be-processed. If within the 50 receives reflected ultrasound waves, there are 20 reflected ultrasound waves having an intensity that is within 10%-40% of the intensity of the transmitted ultrasound pulses, and the remaining 30 reflected ultrasound waves have an intensity that is greater than 40%, then it may be determined that a floor portion corresponding to the grid has provided a reflected ultrasound that is relatively strong, thus, the floor portion may be a floor that is harder than a carpet or flatter than a doorsill, such as a wood, cement, granite, or porcelain floor. The grid is not a grid-to-be-processed. The number of ultrasound pulses transmitted, the ultrasound intensity threshold range, and the count threshold range may be set as other values, which are not limited by the present disclosure.
In some embodiments, each grid may be assigned an initial value. The initial value may be increased or decreased based on the intensity of the reflected ultrasound wave, and the final changed value may be used to determine whether the grid is a grid-to-be-processed. For example, each grid may be assigned an initial value of 500. 50 ultrasound pulses may be transmitted toward the grid and the intensity of the reflected ultrasound waves may be measured. The ultrasound intensity threshold range may be set as 10%-40% of the intensity of the transmitted ultrasound pulses. After receiving each reflected ultrasound wave, if the intensity of the reflected ultrasound wave is within the ultrasound intensity threshold range, the value of the grid may be increased by 10. If the intensity of the reflected ultrasound wave is not within the ultrasound intensity threshold range, the value of the grid may be decreased by 10. After receiving all reflected ultrasound waves, if the value of the grid is greater than or equal to 600, the grid may be determined as a grid-to-be-processed. The number of ultrasound pulses transmitted, the ultrasound intensity threshold range, the value to be increased or decreased each time, and the value for the final determination may be set as other values, which are not limited by the present disclosure.
In the embodiment shown in
In embodiments where an infrared sensor is used, similarly, the carpet and the doorsill may provide a reflection that is relatively weak, and a flat hard floor may provide a reflection that is relatively strong, which may be determined based on infrared lights transmitted by the infrared sensor. For each grid, multiple infrared pulses may be transmitted and the intensity of multiple reflected infrared lights may be measured. A number of reflected infrared lights having an intensity that is within a light intensity threshold range may be counted, e.g., by a processor of the autonomous mobile device. If the number counted is within a count threshold value range, the grid may be determined, e.g., by the processor, as a grid-to-be-processed. For example, for each grid, the infrared sensor may transmit 50 infrared pulses, and receive corresponding reflected infrared lights. The predetermined light intensity threshold range may be set as 10%-40% of the light intensity of the transmitted infrared pulses, and a predetermined count threshold value may be set as 30. If in the 50 reflected infrared lights, there are 35 reflected infrared lights that have a light intensity that is 10%-40% of the light intensity of the transmitted infrared lights, it may be determined that the floor portion where the grid is located has provided a relatively weak reflected infrared light, which may be a carpet or a doorsill, and the grid may be determined as a grid-to-be-processed. If in the 50 reflected infrared lights, there are 20 reflected infrared lights that have a light intensity that is 10%-40% of the light intensity of the transmitted infrared lights, and the remaining 30 reflected infrared lights have a light intensity that is greater than 40% of the light intensity of the transmitted infrared lights, it may be determined that the floor portion where the grid is located has provided a relatively strong reflected infrared light, which may be a wood, cement, granite, or porcelain floor, and the grid is not a grid-to-be-processed.
When a video sensor is used for the detection, the doorsill may display characteristics similar to slender floor tiles. Grids having such characteristics may be recognized based on the floor sensor signal, and may be determined as grids-to-be-processed. Video frames of the floor portion corresponding to each grid may be captured by the video sensor, and a determination may be made for each grid as to whether the grid is a grid-to-be-processed based on the video frames. For example, grayscale processing may be performed on the video frames, and if a “stripe” like image is recognized in the processed video frame, it may be determined that the floor portion may be a doorsill or a slender floor tile, and the grid may be determined as a grid-to-be-processed. If the grayscale in the processed video frame is relatively uniform, it may be determined that the floor portion may be a uniform cement floor, and the grid may not be a grid-to-be-processed.
Step 230, clustering the grids-to-be-processed to obtain one or more groups of grids-to-be-processed. Because a doorsill is an integral structure, it should include multiple grids-to-be-processed will be grouped. Therefore, clustering the grids-to-be-processed into groups of grids-to-be-processed would benefit the recognition of the doorsill.
Step 250, for each group of grids-to-be-processed, determining whether the group of grids-to-be-processed corresponds to a doorsill based on environmental information.
In some embodiments, the environmental information may be obstacle information in a LiDAR map, which serves as an environmental map. For example, the obstacle information in the environmental map may indicate a wall, and a determination may be made as to whether the group of grids-to-be-processed corresponds to a doorsill based on a location of the wall. For example, when the wall includes an opening that has a size similar to a width of a door, the floor location corresponding to the opening may likely be a doorsill. In some embodiments, a location of a center of the group of grids-to-be-processed in the environmental map may be calculated, the location may be represented by, for example, (x, y) coordinates in a coordinate system of the environmental map. A map portion of the environmental map that includes the location may be obtained based on the location coordinates. As such, the obtained map portion may include the group of grids-to-be-processed that potentially corresponds to a doorsill and may include surrounding obstacle information that may indicate a wall. For example, coordinates of a center of the group of grids-to-be-processed may be used as a center of the map portion, and a rectangular map portion having a width “a” and a height “b” may be obtained. It should be understood that the obtained map portion may have other size or shape, as long as the map portion includes the obstacle information, which is not limited by the present disclosure.
In some embodiments, the environmental information may include video information of a location corresponding to the group of grids-to-be-processed. For example, video frames of a ceiling over a location corresponding to the group of grids-to-be-processed may be obtained by a camera configured to point upwardly. If an obvious reduction in the height of the ceiling is detected in the video frames or if a door frame structure is recognized, it may be determined that the group of grids-to-be-processed corresponds to a doorsill.
In some embodiments, the environmental information may include a structured light scanning result. Three-dimensional information of the environment may be obtained through structured light scanning (i.e., scanning the environment using a structured light). If it is detected that the height of the floor corresponding to the group of grids-to-be-process increases abruptly or the height of the ceiling has an obvious reduction, it may be determined that the group of grids-to-be-processed corresponds to a doorsill.
Through the above detection method 200, a location of a doorsill may be accurately detected without increasing the cost, and false detection of the doorsill may be avoided.
In some embodiments, after the doorsill is detected, the doorsill may be labelled in the environmental map for subsequent use. For example, if skidding occurs to the autonomous mobile device 10 during operations, a determination may be made as to whether a distance from a location where the skidding occurred to a location of the doorsill is smaller than or equal to a distance threshold. If the distance from the location where the skidding occurred to the location of the doorsill is smaller than or equal to the distance threshold, it may be determined that the autonomous mobile device 10 is jammed by the doorsill, therefore, the autonomous mobile device 10 may be controlled to perform corresponding predetermined predicament avoidance actions. The predetermined predicament avoidance actions may include, for example, first moving backwardly and then rotating, or repeatedly rotating multiple times and then moving backwardly, or alternately repeating backward movement and rotation actions.
In some embodiments, the autonomous mobile device 10 may be a cleaning robot operable in a wet-mopping mode. When the cleaning robot moves to the location of the doorsill, although signals obtained through an ultrasound sensor may indicate that this location is a carpet that should not be wet-mopped, based on the location of the doorsill in the environmental map, the autonomous mobile device 10 may determine that this location actually does not correspond to a carpet, and hence, may move across the doorsill to continue the wet-mopping of the floor on the other side of the doorsill. As described above, this method can avoid false detection of the doorsill, and benefit the completion of the cleaning task.
It should be understood that the four groups of grids-to-be-processed shown in
For example, after obtaining the grids-to-be-processed using the ultrasound sensor or the infrared sensor, the grids-to-be-processed may indicate a carpet, or may indicate a doorsill, or may indicate a lower base of an object. After the grids-to-be-processed are clustered into groups of grids-to-be-processed, if the number of grids-to-be-processed in a group of grids-to-be-processed is overly large, it may imply that the floor structure corresponding to the group of grids-to-be-processed is relatively large, which may be a carpet and may unlikely be a doorsill. As shown in
If the number of grids-to-be-processed in the group of grids-to-be-processed is overly small, it may imply that the floor structure corresponding to the group of grids-to-be-processed is relatively small, which may be a lower base of an object, such as an edge of a lower base of a floor lamp, and may unlikely be a doorsill. As shown in
In the embodiment of
In some embodiments, if the number of grids-to-be-processed included in a group of grids-to-be-processed is within the predetermined number range, then the determination based on the environmental information in step 250 may not be needed, and the group of grids-to-be-processed may be directly determined as corresponding to a doorsill in step 260. As such, the computing power of the processor may be saved, and in the meantime, the location of the doorsill may be accurately recognized.
In some embodiments, the present disclosure also provides a non-transitory computer-readable storage medium or a computer program product. The computer-readable storage medium or the program product may include instructions, which may be executable by a processor to perform the above methods related to the operations of the autonomous mobile device. The processor may include, but not be limited to, application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gated array (FPGA), controller, micro-controller, micro-processor, or other electronic element.
For the autonomous mobile device in the above embodiments, the detailed implementations of the operations of each component have been described in detail in the embodiments of the related methods, which are not repeated herein.
The above described are various specific embodiments of the present disclosure. The above descriptions are illustrative, and are not exhaustive. The present disclosure is not limited to the described embodiments. Without deviating from the scope and principle of the various embodiments described above in the present disclosure, many modifications and substitutions are easy to be conceived by a person having ordinary skills in the art. The selection of the terms in the present disclosure is for the purpose of best describing the principle, actual application of the various embodiments, or technical improvement in the market, or for the purpose of letting other persons of ordinary skills in the art better understand the various embodiments.
| Number | Date | Country | Kind |
|---|---|---|---|
| 202211607717.3 | Dec 2022 | CN | national |
