Method for handling a robot in a stuck situation and robot

Abstract

The application relates to the field of cleaning robots, and discloses a robot and a robot escape treatment method. The escape treatment method comprises the following steps: when a rolling brush is in a cleaning working state, if it is detected that the load current of the rolling brush is greater than a preset current threshold, a camera unit is controlled to collect a rolling brush image; the type of the winding object is determined according to the rolling brush image; if the winding object type is a non-adhesive tape winding object, the robot is controlled to enter a first escape mode; if the robot has not escaped after the first escape mode is executed for a first preset time, the robot is controlled to enter a second escape mode; if the robot has not escaped after the second escape mode is executed for a second preset time, it is determined that the robot has failed to escape, the robot is controlled to be urgently stopped, and an alarm is triggered. The escape treatment method and the robot provided by the application can effectively make the robot escape from the predicament of being stuck by foreign matter, thereby effectively improving the efficiency of automatic escape of the cleaning robot.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of cleaning robots, in particular to a robot and a robot out-of-trouble processing method. BACKGROUND

[0002] With the rapid development of robots, robots are becoming more and more important in life, learning and other scenarios. At present, in the working process of the robot, the robot may be trapped by obstacles, so that the robot cannot work normally. Therefore, how to solve the problem of the robot being trapped is very important.

[0003] At present, the cleaning robot mainly uses a roller brush to collect garbage when sweeping the environment, but it faces more and more complex working environments, such as nylon belts in office environments and large plastic bags in supermarket environments, which are easy to cause foreign matter entanglement of the roller brush. Without foreign matter entanglement processing mechanism, the frequency of human intervention will be greatly increased, which will affect the customer experience.

[0004] However, the current out-of-trouble research of the cleaning robot is mainly aimed at the operation processing of specific complex environments. For example, the existing technology generally adopts the following methods: 1. Increase the diameter of the driving wheel, but the larger the wheel, the smaller the torque, and the method of increasing the diameter of the driving wheel will result in poor power; 2. Increase the driving motor power, but increasing the driving motor power will upgrade the supporting circuit, reduce the battery usage time, and increase the cost; 3. Use a driving wheel obstacle climbing spring structure, but the driving wheel spring structure has poor obstacle climbing ability in actual operation. It basically does not involve reverse out-of-trouble of foreign matter entanglement. SUMMARY

[0005] The present application provides an out-of-trouble processing method and a robot to solve the problem that the cleaning robot in the prior art cannot effectively escape from the predicament of being entangled by foreign matter such as adhesive tape, branches or plastic bags.

[0006] To achieve the above object, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides an out-of-trouble processing method of a robot, the robot comprising a roller brush and a camera unit, comprising:

[0008] When the roller brush is in a cleaning working state, if it is detected that the load current of the roller brush is greater than a preset current threshold, the camera unit is controlled to collect a roller brush image;

[0009] According to the roller brush image, the type of the entangling object is determined;

[0010] If the type of the entangling object is a non-adhesive tape type entangling object, the robot is controlled to enter a first out-of-trouble mode, wherein the first out-of-trouble mode comprises controlling the roller brush to reverse at a first preset speed;

[0011] If the robot is not unstuck after the first preset time of the first unstuck mode, the robot is controlled to enter a second unstuck mode, where the second unstuck mode includes controlling the roller brush to reverse at a second preset speed, the second preset speed being greater than the first preset speed.

[0012] If the robot is not unstuck after the second preset time of the second unstuck mode, it is determined that the robot fails to be unstuck, and the robot is controlled to stop urgently and trigger an alarm.

[0013] In some embodiments, according to the roller brush image, the type of the entanglement is determined, including:

[0014] Based on the target detection model, the roller brush image is preprocessed to obtain a to-be-detected roller brush image.

[0015] If it is detected that there is an entanglement target in the to-be-detected roller brush image, the entanglement target is feature-extracted to determine the type of the entanglement target.

[0016] In some embodiments, the method further includes:

[0017] If it is detected that there is no entanglement target in the to-be-detected roller brush image, the robot is controlled to re-enter a cleaning working state, and the load current of the roller brush is re-detected.

[0018] In some embodiments, the robot further includes a drive wheel;

[0019] The first unstuck mode includes:

[0020] The roller brush is controlled to reverse at a first preset speed, and the drive wheel is controlled to retreat at a first auxiliary preset speed to assist in reversing, where the first preset speed is less than the maximum rotating speed of the roller brush, and the first auxiliary preset speed is less than the maximum rotating speed of the drive wheel.

[0021] The second unstuck mode includes:

[0022] The roller brush is controlled to reverse at a second preset speed, and the drive wheel is controlled to retreat at a second auxiliary preset speed to assist in reversing, where the first preset speed is equal to the maximum rotating speed of the roller brush, and the first auxiliary preset speed is equal to the maximum rotating speed of the drive wheel.

[0023] In some embodiments, the method further includes:

[0024] After the first preset time of the first unstuck mode, if it is detected that the load current of the roller brush is less than or equal to a preset current threshold, it is determined that the robot is successfully unstuck.

[0025] After the second preset time of the second unstuck mode, if it is detected that the load current of the roller brush is less than or equal to a preset current threshold, it is determined that the robot is successfully unstuck.

[0026] In some embodiments, the method further comprises:

[0027] If the winding type is a tape type winding, the robot is controlled to stop and an alarm is triggered.

[0028] In a second aspect, the embodiments of the present application provide a robot, comprising:

[0029] a main controller configured to perform part or all of the robot escape processing method as described in the first aspect;

[0030] a driver connected to the main controller and configured to execute the control instructions issued by the main controller;

[0031] a cleaning execution unit connected to the driver and configured to execute the control instructions issued by the driver to control the robot to escape, stop or alarm;

[0032] The cleaning execution unit comprises:

[0033] a rolling brush;

[0034] a rolling brush motor configured to control the movement of the rolling brush;

[0035] a driving wheel;

[0036] a walking motor configured to control the movement of the driving wheel.

[0037] In some embodiments, the driver comprises a winding detection unit, which comprises:

[0038] a load current detector integrated in the driver and configured to collect the load current of the rolling brush;

[0039] a camera unit configured to collect images of the rolling brush.

[0040] In some embodiments, the driver further comprises a processing unit and an escape execution unit, comprising:

[0041] the processing unit connected to the main controller, the winding detection unit and the escape execution unit, configured to upload the load current and the images of the rolling brush collected by the winding detection unit to the main controller, and to analyze the control instructions issued by the main controller to control the escape execution unit;

[0042] the escape execution unit connected to the cleaning execution unit and the winding detection unit, configured to receive the instructions sent by the winding detection unit to control the movement of the rolling brush and the movement of the driving wheel.

[0043] In some embodiments,

[0044] the escape execution unit comprises:

[0045] A forward and reverse rotation control circuit is configured to control forward or reverse rotation of the roller brush and forward or reverse movement of the drive wheel.

[0046] A rotation speed control circuit is configured to control rotation speed of the roller brush and / or the drive wheel.

[0047] The robot escape processing method provided by the embodiment of the present application has the beneficial effects that: different from the prior art, the robot escape processing method provided by the embodiment of the present application comprises the following steps: when the roller brush is in a cleaning working state, if it is detected that the load current of the roller brush is greater than a preset current threshold, a camera unit is controlled to collect a roller brush image; the type of the winding object is determined according to the roller brush image; if the type of the winding object is a non-adhesive tape winding object, the robot is controlled to enter a first escape mode; if the robot has not escaped after the first escape mode is executed for a first preset time, the robot is controlled to enter a second escape mode; if the robot still has not escaped after the second escape mode is executed for a second preset time, it is determined that the robot fails to escape, and the robot is controlled to be urgently stopped and an alarm is triggered.

[0048] The embodiment of the present application determines the type of the winding object and controls the robot to execute the corresponding escape mode by twice confirming whether the roller brush is wound, which can make the robot effectively escape from the predicament of being stuck by foreign matter, thereby effectively improving the automatic escape capability of the cleaning robot and further reducing the frequency of manual intervention. BRIEF DESCRIPTION OF DRAWINGS

[0049] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, which are schematic and not intended to be limiting of the embodiments, and in which like reference numerals designate similar items in the figures, in which: the drawings are not to scale.

[0050] Figure 1 is a schematic diagram of an overall structure of a cleaning robot provided by the embodiment of the present application;

[0051] Figure 2 is a schematic diagram of a flow of a robot escape processing method provided by the embodiment of the present application;

[0052] Figure 3 is a detailed flow of step S204 in Figure 2

[0053] Figure 4 is a schematic diagram of a structure of a cleaning robot provided by the embodiment of the present application;

[0054] Figure 5 is a schematic diagram of a structure of a main controller provided by the embodiment of the present application;

[0055] Figure 6 is a schematic diagram of a structure of a winding detection unit provided by the embodiment of the present application;​

[0056] Figure 7 is a structural schematic diagram of a cleaning execution unit provided by an embodiment of the present application;

[0057] Figure 8 is a structural schematic diagram of a cleaning execution unit provided by an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0059] In addition, the technical features involved in each of the embodiments of the present application described below can be combined with each other as long as there is no conflict.

[0060] Please refer to Figure 1 , Figure 1 is a structural schematic diagram of a cleaning robot provided by an embodiment of the present application;

[0061] As shown in Figure 1 , a cleaning robot provided by the present application generally can include a robot 100, the robot 100 is internally provided with a main controller device 101 and a driver device 102; the robot 100 is externally provided with a walking device 103 for controlling the movement of the robot and a rolling brush device 104 for cleaning.

[0062] The main controller device 101 is located above the side of the water tank inside the robot, the driver device 102 is located below the side of the water tank inside the robot, the walking device 103 and the rolling brush device 104 are located on the chassis of the robot for executing specific escape actions. The walking device 103 further includes a driving wheel and a driving motor; the rolling brush device 104 further includes a rolling brush and a rolling brush motor.

[0063] In the process of cleaning by the cleaning robot, the roller brush, as a structure directly contacting the ground, needs to carry the garbage to the internal garbage can by rolling. However, because the diameter is relatively small, it is easy to be entangled by the rope, plastic bag, adhesive tape and other hard and small strip objects in the environment, thereby affecting the normal operation of the roller brush device 104; and because the speed of the roller brush device 104 is not easy to monitor, therefore, the robot provided by the present application needs to detect whether the roller brush of the robot is blocked by the above similar objects to cause the roller brush device 104 to stop rotating, at this time, the driving wheel under the walking device 103 needs to cooperate to perform the retreat / advance operation to perform the escape operation.

[0064] In the embodiment of the present application, the roller brush is used to collect garbage in the rotating process; the roller brush motor is connected with the roller brush for controlling the rotation of the roller brush; when the cleaning robot is stuck in the process of collecting garbage by using the roller brush device 104, the cleaning robot performs the corresponding escape operation, controls the roller brush device 104 to perform the reverse operation, and controls the driving wheel device 103 to perform the retreat / advance operation, for assisting the roller brush device 104 to push the entangled objects to the two sides or one side, so as to avoid the roller brush device 104 from being stuck, thereby not affecting the normal work of the cleaning robot.

[0065] Optionally, the installation positions of the walking device 103 and the roller brush device 104 are adjustable.

[0066] In a specific embodiment, as shown in Figure 1 the roller brush device 104 is arranged on the front side of the walking device 103, for assisting the roller brush device 104 to push the entangled objects to the two sides or one side in the retreat / advance process of the walking device 103 when the robot performs the escape operation.

[0067] In another specific embodiment, the roller brush device 104 is arranged on the rear side of the walking device 103, for assisting the roller brush device 104 to push the entangled objects to the two sides or one side in the retreat / advance process of the walking device 103 when the robot performs the escape operation.

[0068] It needs to be understood that the roller brush device 104 can be arranged on the front side of the walking device 103, or can be arranged on the rear side of the walking device 103; when the roller brush device 104 is arranged on the front side of the walking device 103, the escape process is generally retreat of the walking device 103, of course, it can also be a combination action of retreat and advance; when the roller brush device 104 is arranged on the rear side of the walking device 103, the escape process is generally advance of the walking device 103, of course, it can also be a combination action of advance and retreat.

[0069] Please refer to Figure 2 , Figure 2 a flowchart of an escape processing method of a robot provided in the embodiment of the present application.

[0070] The first aspect of this application proposes a method for robot obstacle avoidance, wherein the robot obstacle avoidance method is executed by a cleaning robot and is applied to, for example... Figure 1 The robot shown in the image has the following steps S201-S213 for escaping obstacles:

[0071] Step S201: When the roller brush is in the cleaning working state, obtain the load current of the roller brush;

[0072] In this embodiment of the application, the load current flowing through the roller brush is compared with a preset current threshold by automatic detection by the robot to determine whether the roller brush current is normal, and thus to determine the cleaning working state of the robot. The cleaning working state of the robot includes normal cleaning working state and abnormal cleaning working state.

[0073] Step S202: Determine whether the load current of the roller brush is greater than the preset current threshold.

[0074] Specifically, after detecting the instantaneous load current value of the robot's roller brush, this instantaneous load current value can be compared with a preset current threshold. Based on the comparison result, different instructions are executed to enable the robot to extricate itself from the entanglement. Specifically, the preset current threshold in this embodiment is a critical current threshold obtained through multiple experiments and statistical analysis when the robot is stuck by an entangled object. For example, if the critical current threshold is set to 0.6A, when the instantaneous load current value of the roller brush exceeds this critical current threshold, it indicates that one or more roller brushes of the robot may be stuck by an entangled object. In this case, the robot will execute other instructions to further confirm the type of entangled object.

[0075] In this embodiment of the application, when the robot uses a roller brush to collect garbage, the roller brush may get stuck with tangled objects. To address this, a preset condition can be set. When the robot meets this preset condition, it is assumed that the robot may be stuck with tangled objects during the cleaning task, and the type of tangled object can be further identified.

[0076] Specifically, when the robot passes through the load current detector configured on the robot at a preset time, it detects the instantaneous current value on the robot's roller brush, and after the detection is completed, it determines whether the load current of the roller brush is within a preset current threshold. Preferably, in practical application scenarios, the preset time setting can be set according to actual needs, and is not limited here.

[0077] Specifically, it determines whether the load current of the roller brush is greater than a preset current threshold in order to determine the robot's cleaning working status.

[0078] If the detected load current of the roller brush is less than or equal to the preset current threshold, it is determined that the robot is in normal cleaning operation state, and the process returns to step S201.

[0079] If the load current of the roller brush is detected to be greater than the preset current threshold, it is determined that the robot is in an abnormal cleaning working state, and the process proceeds to step S203.

[0080] Step S203: Determine that the robot is in an abnormal cleaning working state and control the robot to stop suddenly.

[0081] In this embodiment, during the process of the robot collecting garbage on the ground using a roller brush, the roller brush, as a structure in direct contact with the ground, needs to transport the garbage into the internal garbage bin by rolling. However, because of its relatively small diameter, it is easily entangled by strip-shaped objects with low hardness, such as ropes, plastic bags, and tape in the environment. When the roller brush is entangled, the resistance generated by the entanglement may cause changes in the current of the brush's drive motor. Therefore, the cleaning status of the robot can be confirmed first by detecting changes in the current, and then a secondary confirmation of whether it is entangled can be made through image detection.

[0082] Therefore, in this embodiment of the application, during the cleaning process of the robot, the load current value flowing through the robot's roller brush is detected in real time. When the instantaneous load current value of the robot's roller brush is detected to be greater than the preset current threshold, it indicates that the robot's cleaning working state is abnormal and it may be entangled by objects.

[0083] This application embodiment first determines the robot's cleaning working status by current detection, so as to further perform image detection on the roller brush to determine whether the roller brush is entangled by tangled objects, and then perform a freeing operation. By performing a second confirmation of entangled objects, the ineffective freeing caused by misjudgment is effectively reduced, and the problem of failure to free the robot when it is entangled by foreign objects with adhesive properties or other difficult-to-free features is reduced, which may even lead to robot malfunction. This ensures that the robot can get out of trouble smoothly and improves the robot's freeing efficiency.

[0084] Step S204: Acquire the image of the roller brush and determine the type of entanglement based on the image of the roller brush.

[0085] In this embodiment of the application, when the robot detects an abnormal current in the roller brush and determines that the robot is in an abnormal cleaning working state and performs an emergency stop operation, the camera is controlled to capture the roller brush image, and the type of entangled object is determined based on the roller brush image.

[0086] Specifically, in this embodiment of the application, the robot chassis includes a built-in close-range monitoring camera for real-time detection of the brush cleaning area. The camera captures images of the brush to obtain image data of the cleaning area. The image data is then preprocessed to obtain the brush image to be detected. If an entangled object is detected in the brush image to be detected, the entangled object is feature extracted to determine the type of entangled object. The entangled object type includes tape-type entangled objects and non-tape-type entangled objects.

[0087] This application embodiment issues corresponding instructions based on the type of entanglement. For example, when the roller brush is entangled by foreign objects with strong adhesiveness, such as transparent tape, directly executing the escape action may cause the robot to fail to escape or even malfunction and stop operating. Therefore, determining the type of entanglement based on image detection and then deciding whether to enter the escape mode can effectively reduce the problem of robot failure or even robot malfunction when the robot is entangled by foreign objects with adhesiveness that are difficult to escape. This ensures that the robot can escape smoothly and improves the robot's escape efficiency.

[0088] Step S205: Determine whether the type of wrapped material is tape-type wrapped material;

[0089] In this application embodiment, the type of entangled object is divided into tape-type entangled object and non-tape-type entangled object. It can be understood that by classifying the difficulty of entangled objects in terms of their entanglement difficulty, it is possible to determine whether to perform an entanglement action, which can effectively avoid the phenomenon of low entanglement efficiency caused by forced entanglement and even damage to the machine, thereby improving the robot's entanglement efficiency.

[0090] Specifically, in this embodiment, an internal camera mounted on a roller brush acquires image data, and the YOLOv7 target detection model is used to identify the images, determining the type of wrapped material and classifying the results into two categories: the first category is non-tape-type wrapped materials, such as nylon ropes and plastic bags; the second category is tape-type wrapped materials, such as transparent tape and waterproof tape. Different post-processing measures are applied based on the classification results. Specifically:

[0091] If the wrapped material is tape, proceed to step S213: control the robot to stop suddenly and trigger an alarm.

[0092] Specifically, the robot is controlled to stop immediately by cutting off the power to the escape execution unit by cutting off its control circuit, and the main controller is triggered to send an alarm command to the alarm device to trigger an alarm.

[0093] If the entangled object is not a tape-type entangled object, proceed to step S206: execute the first escape mode.

[0094] Step S206: Execute the first escape mode.

[0095] In this embodiment of the application, the escape mode is divided into two levels. The first level is the first escape mode. The first escape mode includes controlling the drive wheel to assist in reversing while controlling the winding roller to perform a reverse operation. By controlling the drive wheel to assist in reversing, the escape range of the roller can be increased, and the escape process of the roller can be made more flexible, thereby improving the robot's escape efficiency.

[0096] In this embodiment, the first escape mode controls the roller brush and drive wheel to perform reverse / backward movement at a preset speed less than the maximum rotation speed. This helps the roller brush avoid the problem of failing to escape due to excessively fast reverse rotation speed, or even getting tangled in the opposite direction, thereby improving the robot's escape efficiency.

[0097] Step S207: Control the roller brush to reverse at a first preset speed.

[0098] In this embodiment of the application, by setting a first preset speed, the winding roller is controlled to perform a reverse operation at the first preset speed, and at the same time, the drive wheel is controlled to perform a backward operation at the first preset auxiliary speed. It should be noted that the first preset speed is less than the maximum rotational speed of the roller, and the first auxiliary preset speed is less than the maximum rotational speed of the drive wheel.

[0099] Optionally, the first preset speed can be 0.5 times, 0.6 times, or 0.7 times the maximum rotational speed of the roller brush. The first preset auxiliary speed can be 0.5 times, 0.6 times, or 0.7 times the maximum rotational speed of the drive wheel. It is understood that the first preset speed and the first preset auxiliary speed can be set according to actual needs and are not limited here.

[0100] When a cleaning robot collects trash, its brush rotates at a relatively high speed. Therefore, when the brush encounters tangled debris, it may become entangled within a short time. If the reverse rotation speed of the brush is set too high, the brush may not be able to untangle itself, and the debris may be caught again by the reversing brush, resulting in reverse entanglement. Therefore, in this embodiment, by controlling the first preset speed of the brush reversal to be no greater than the maximum rotational speed of the brush during cleaning, and controlling the first preset auxiliary speed of the drive wheel reversing to be no greater than the maximum rotational speed of the drive wheel, reverse entanglement due to excessively high rotational speed during the robot's escape operation can be effectively avoided, thereby significantly improving the robot's escape efficiency.

[0101] Step S208: Determine whether the robot has successfully escaped the predicament.

[0102] In this embodiment of the application, after the first preset time is executed in the first escape mode, it is determined whether the robot has successfully escaped. It should be noted that the first preset time can be set according to actual needs. In practical applications, the first preset time can be set according to empirical values ​​and / or experimental values. It can be to determine whether the robot has successfully escaped within 20 seconds or 30 seconds. If the load current of the roller brush is still greater than the preset current threshold after the first preset time is exceeded, it can be determined that the robot has failed to escape in the first escape mode. Otherwise, it is determined that the robot has successfully escaped.

[0103] Specifically, if the load current of the roller brush is detected to be less than or equal to the preset current threshold, then proceed to step S212: the robot has successfully escaped its predicament.

[0104] In this embodiment, the method for determining whether the robot has successfully escaped after executing the escape mode is to detect the instantaneous load current value on the robot's roller brush again, compare the instantaneous load current value with a preset current threshold, and determine whether the robot has completed the escape based on the comparison result. Specifically, the preset current threshold in this embodiment is a critical current threshold obtained by statistically analyzing multiple experiments when the robot is stuck by an entangled object. When the instantaneous load current value on the roller brush is greater than the critical current threshold, it indicates that the robot may have one or more roller brushes stuck by an entangled object.

[0105] If the load current of the roller brush is detected to be greater than the preset current threshold, it indicates that the robot has failed to escape from the obstacle, and then proceed to step S209: execute the second escape mode.

[0106] Step S209: Execute the second escape mode.

[0107] In this embodiment, the second escape mode is a second-level escape mode. When the robot fails to escape using the first-level escape mode, it is controlled to enter the second escape mode. The second escape mode includes simultaneously controlling the drive wheels to reverse at maximum speed while controlling the winding brush to reverse at maximum speed.

[0108] In this embodiment, the second escape mode controls the roller brush and drive wheel to perform reverse / backward motion at maximum speed, which can help the roller brush avoid the problem of slow escape due to slow reverse speed, thereby improving the robot's escape efficiency.

[0109] Step S210: Control the roller brush to reverse at a second preset speed.

[0110] In this embodiment of the application, by setting a second preset speed, the winding roller is controlled to perform a reverse operation at the second preset speed, and at the same time, the drive wheel is controlled to perform a backward operation at the second preset auxiliary speed. It should be noted that the above-mentioned second preset speed is equal to the maximum rotation speed of the roller, and the second auxiliary preset speed is equal to the maximum rotation speed of the drive wheel.

[0111] It is understandable that the maximum speed of the roller brush and the maximum speed of the drive wheel can be set according to actual needs, and no limitation is made here.

[0112] Step S211: Determine whether the robot has successfully escaped the predicament.

[0113] In this embodiment of the application, after the second escape mode is executed for a second preset time, it is determined whether the robot has successfully escaped. It should be noted that the second preset time can be set according to actual needs. In practical applications, the second preset time can be set according to empirical values ​​and / or experimental values. It can be to determine whether the robot has successfully escaped within 20 seconds or 30 seconds. If the load current of the roller brush is still greater than the preset current threshold after the second preset time is exceeded, it can be determined that the robot has failed to escape in the second escape mode. Otherwise, it is determined that the robot has successfully escaped.

[0114] Specifically, if the load current of the roller brush is detected to be less than or equal to the preset current threshold, then proceed to step S212: the robot has successfully escaped its predicament.

[0115] If the load current of the roller brush is detected to be greater than the preset current threshold, it indicates that the robot has failed to escape from the obstacle, and then proceed to step S213: control the robot to stop urgently and trigger an alarm.

[0116] Step S213: Control the robot to stop suddenly and trigger an alarm.

[0117] Specifically, the robot is controlled to stop immediately by cutting off the power to the escape execution unit by cutting off its control circuit, and the main controller is triggered to send an alarm command to the alarm device to trigger an alarm.

[0118] In this embodiment, when the cleaning robot is collecting garbage, the roller brush is easily entangled by nylon ropes, plastic bags, and other tangled objects. These entanglements are difficult to remove. If the roller brush's reversal speed is set too fast, the entangled object may be entangled again by the reversing roller brush, resulting in reverse entanglement. If the roller brush's reversal speed is set too slow, the roller brush may not be able to detach from the entanglement. Therefore, in this embodiment, two detachment modes are set. In the first detachment mode, the roller brush is controlled to rotate at a first preset speed. If the robot fails to escape after executing the first detachment mode, it enters the second detachment mode, controlling the roller brush to rotate at a second preset speed. If the robot fails to escape after executing the second detachment mode, it is controlled to stop suddenly and issue an alarm. This effectively solves the above problems and improves the robot's detachment efficiency.

[0119] Please refer to Figure 3 , Figure 3 for Figure 2 A detailed flowchart of step S204 in the process.

[0120] like Figure 3 As shown, step S204 includes the following steps:

[0121] Step S2041: Acquire images of the roller brush through the camera unit.

[0122] Specifically, an internal monitoring camera in the camera unit is used to capture images of the roller brush cleaning area, including the roller brush and debris.

[0123] Step S2042: Based on the target detection model, preprocess the roller brush image to obtain the roller brush image to be detected.

[0124] Specifically, deep learning is used to sample a large number of cleaning regions from the input brush image and adjust the region edges to more accurately obtain the ground-truth bounding boxes of the feature targets. Different models may use different region sampling methods.

[0125] In this embodiment, the target detection model includes a YOLOv7 model. The brush image is input into the YOLOv7 model, and the output is divided into two categories. The YOLOv7 network generally consists of three parts: an input network, a backbone network, and an output network. Specifically, the input network preprocesses the input brush image, for example, aligning it to a 640*640 pixel size to be detected. The to-be-detected brush image is in RGB format. Then, the to-be-detected brush image is input into the backbone network to further detect whether it contains an entangled target. For details, please refer to step S2043.

[0126] Step S2043: Determine whether there is an entangled target in the image of the roller to be detected.

[0127] Specifically, the image of the roller brush to be detected is input into the model parameters of the backbone network for training, feature information is extracted, and it is determined whether the feature information is an entangled target.

[0128] If so, proceed to step S2045: extract features from the image of the roller brush to be detected.

[0129] If not, proceed to step S2044: Control the robot to re-enter the cleaning work state.

[0130] Specifically, the robot is controlled to re-enter the cleaning operation state and the load current of the roller brush is re-detected.

[0131] Step S2045: Extract features from the image of the roller brush to be detected.

[0132] In this embodiment of the application, specifically, the image of the roller brush to be detected is input into the model parameters of the backbone network for training, and feature information is extracted. Then, the feature information is input into the RepVGG Block model in the convolutional layer through the output network for training to obtain feature results. The feature results include classification information, front and back background classification information, and bounding box information.

[0133] Step S2046: Determine whether it is a tape-like wrapped object.

[0134] Specifically, the feature results are used to determine whether the target of the wrapping is a type of tape. Specifically, the feature results are compared with the local dataset to confirm whether the target of the wrapping is a type of tape with strong adhesion, such as transparent tape.

[0135] If so, proceed to step S2047: Control the roller brush to stop suddenly and issue an alarm.

[0136] If not, proceed to step S2048: execute the preset escape action.

[0137] Specifically, the preset escape actions include a preset first escape mode or a preset second escape mode.

[0138] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a cleaning robot provided in an embodiment of this application.

[0139] like Figure 4 As shown, the cleaning robot 400 includes:

[0140] The system includes a main controller 410, a driver 420, and a cleaning execution unit 430. The driver 420 is communicatively connected to both the main controller 410 and the cleaning execution unit 430. For example, the driver 420 can be connected to the main controller 410 and the cleaning execution unit 430 via cable or wirelessly, such as through 4G, 5G, local area network, or Bluetooth.

[0141] The main controller 410 is used to execute some or all of the robot's obstacle avoidance methods as described in the embodiments of this application. The main controller 410 includes a decision-making system based on a dual-ARM processor architecture. The main ARM processor loads a non-real-time operating system for handling business scheduling and mechanism decisions. The slave ARM processor loads a lightweight real-time operating system for implementing functions such as task scheduling, message queue management, and software timers.

[0142] Specifically, in this embodiment of the application, when the robot makes a decision on the action after being entangled by a foreign object, the main controller 410 analyzes the sensing data collected by the front end to confirm whether the load current is greater than the preset current threshold, or whether it is a sticky foreign object that is difficult to get out of, such as transparent tape. If it is confirmed, the robot will directly execute a shutdown alarm to introduce manual intervention; otherwise, the robot will execute a preset escape action. The preset escape action is generated based on a test, stored in FLASH, and read from the ARM processor when executed.

[0143] In this embodiment of the application, the preset escape actions are stored in the program area of ​​the FLASH embedded in the ARM processor, used for task scheduling, and sent to the CPU unit of the driver 420.

[0144] Please refer to the following: Figure 5 , Figure 5 This is a schematic diagram of the structure of a main controller provided in an embodiment of this application;

[0145] like Figure 5 As shown, the main controller 410 includes one or more first processors 411, second processors 412, and a memory 413. Among them, Figure 5 Take a processor 411 as an example.

[0146] The first processor 411, the second processor 412, and the memory 413 can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0147] The first processor 411 is used to load a non-real-time operating system and to implement business scheduling and mechanism decision-making.

[0148] The second processor 412 is used to load a lightweight real-time operating system to implement functions such as task scheduling, message queue management, and software timers.

[0149] The memory 413, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the robot obstacle avoidance method in the embodiments of this application. The first processor 411 and the second processor 412 execute various functional applications and data processing by running the non-volatile software programs, instructions, and modules stored in the memory 413, thereby realizing the functions of the robot obstacle avoidance method provided in the above method embodiments and the various modules or units in the above device embodiments.

[0150] Memory 413 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 413 may optionally include memory remotely located relative to the first processor 411 and the second processor 412, and these remote memories may be connected to the first processor 411 and the second processor 412 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0151] The program instructions / modules are stored in the memory 413. When executed by one or more first processors 411 and second processors 412, they execute the robot's escape method in any of the above method embodiments.

[0152] The main controller 410 in this application embodiment exists in various forms, and executes the above-described... Figure 2 During the various steps shown, the main controller 410 mentioned above includes, but is not limited to, controllers such as microcontroller units (MCUs) and circuit board controllers.

[0153] The driver 420 is used to convert the upper computer protocol stack instructions into specific escape actions issued by the lower computer, and to upload the sensing data from the front-end sensors to the main controller 410. It is also used for secondary confirmation of whether foreign object entanglement has occurred and for identifying and classifying entangled objects, thereby controlling the forward and reverse rotation of the roller brush and the backward operation of the drive wheel. The driver 420 includes a processing unit 421, an entanglement detection unit 422, and an escape execution unit 423.

[0154] The processing unit 421 is connected to the winding detection unit 422 and the escape execution unit 423. It is used to upload the load current and image of the roller brush collected by the winding detection unit 422 to the main controller 410, and to parse the task instructions issued by the main controller 410 to forward the task instructions to the escape execution unit 423.

[0155] The entanglement detection unit 422 is connected to the processing unit 421 and the escape execution unit 423. It is used to collect front-end sensing data to realize a motor load current detection. When an abnormality is detected, it gives an emergency stop instruction to the escape execution unit 423 and reports to the processing unit 421 of the driver 410 for post-processing analysis.

[0156] The escaping execution unit 423 is connected to the processing unit 421 and the entanglement detection unit 422, and is used to issue corresponding escaping operations. Specifically, the escaping execution unit 423 includes a forward and reverse rotation control circuit and a speed control circuit. The forward and reverse rotation control circuit is used to control the forward or reverse rotation of the roller brush and the forward or backward movement of the drive wheel, and the speed control circuit is used to control the speed of the roller brush and / or the drive wheel.

[0157] The cleaning execution unit 430 is connected to the escape execution unit 423 and is used to execute the task instructions issued by the escape execution unit 423 to control the robot to escape, stop suddenly, or issue an alarm. The cleaning execution unit 430 includes: a roller brush / roller brush motor for controlling the movement of the roller brush; and a drive wheel / walking motor for controlling the movement of the drive wheel.

[0158] The processing unit 421 is connected to the main controller 410, the entanglement detection unit 422 and the escape execution unit 423, and the escape execution unit 423 is connected to the processing unit 421, the entanglement detection unit 422 and the cleaning execution unit 430.

[0159] In this embodiment, the first processor 411 makes a decision based on the load current data collected by the load current detector. When the load current of the roller brush is detected to be greater than the preset current threshold through automatic detection by the robot, it is confirmed that the robot's cleaning operation is in an abnormal state. At this time, the second processor 412 executes task scheduling and sends a stop command to the processing unit 421 of the driver 420. The processing unit 421 transmits the protocol stack instructions of the main controller 410 to the escape execution unit 423. The escape execution unit 423 then transmits the instructions to the cleaning execution unit 430. The cleaning execution unit 430 is used to complete the forward and reverse rotation of the roller brush and the backward operation of the drive wheel to complete the escape action.

[0160] It should be noted that the structural schematic diagram of the cleaning robot 400 provided in this application embodiment has the same specific implementation and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the structural embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0161] Please refer to Figure 6 , Figure 6 This is a schematic diagram of the structure of an entanglement detection unit for a cleaning robot provided in an embodiment of this application.

[0162] like Figure 5 As shown, the winding detection unit 422 includes:

[0163] The load current detector 4221, integrated into the driver, is used to collect the load current of the roller brush to detect whether the robot is entangled with foreign objects.

[0164] The camera unit 4222 includes an internal monitoring camera, which is installed on the robot's roller brush to collect images of the roller brush, enabling secondary confirmation of entanglement anomalies and identification and classification of entangled objects.

[0165] Please refer to Figure 6 , Figure 6 This is a schematic diagram of the structure of an obstacle-avoidance execution unit for a cleaning robot provided in an embodiment of this application;

[0166] like Figure 6 As shown, the escape execution unit 423 includes:

[0167] The forward / reverse control circuit 4231 is specifically electrically connected to both the roller brush device and the walking device. In this embodiment, the forward / reverse control circuit 4231 also includes a lithium battery electrically connected to it. The forward / reverse control circuit 4231 transmits and distributes power to the roller brush device and the walking device, thereby controlling the forward or reverse rotation of the robot roller brush and the forward or backward movement of the drive wheels. In this embodiment, the forward / reverse control circuit 4231 includes circuits such as a DC motor forward / reverse control circuit and a three-phase asynchronous motor forward / reverse control circuit.

[0168] The speed control circuit 4232 is specifically electrically connected to both the roller brush device and the walking device. In this embodiment, the speed control circuit 4232 also includes a lithium battery electrically connected to it. The speed control circuit 4232 transmits and distributes power to the roller brush device and the walking device, thereby controlling the speed of the robot's roller brush and / or drive wheels. In this embodiment, the speed control circuit 4232 includes a stepper motor speed control circuit, a PWM DC motor speed controller circuit, and other circuits.

[0169] Please refer to Figure 7 , Figure 7 This is a schematic diagram of the cleaning execution unit of a cleaning robot provided in an embodiment of this application.

[0170] like Figure 7 As shown, the cleaning execution unit 430 includes:

[0171] The roller brush 431 and the roller brush motor 432 are used to control the movement of the roller brush 431.

[0172] In some embodiments, the robot has a built-in roller brush device, including a roller brush, a roller brush motor, a sleeve seat, and a bracket. The bracket is provided with two parallel mounting end plates. The two ends of the roller brush are rotatably engaged with the two mounting end plates respectively. The sleeve seat is fixed on the mounting end plates and is located inside the roller brush. The roller brush motor is installed inside the sleeve seat, and the rotating shaft of the roller brush motor is engaged with the roller brush. The roller brush motor is used to drive the roller brush to rotate.

[0173] In this embodiment, the cleaning execution unit 430 is provided with a roller brush 431 and a roller brush motor 432. The roller brush motor 432 is installed inside the roller brush 431 through the aforementioned sleeve seat and other components, thereby driving the roller brush 431 to rotate from inside the roller brush 431. This avoids the power and transmission components from being exposed, ensuring that the roller brush 431 rotates at a more stable speed, with a higher safety factor, and preventing jamming or other phenomena.

[0174] The cleaning execution unit 430 also includes a drive wheel 433 and a walking motor 434, wherein the walking motor 434 is used to control the movement of the drive wheel 433.

[0175] In some embodiments, the robot has a built-in walking device, including a drive wheel, a drive motor, a reducer, a drive ring, and wheels. Typically, the drive motor includes a motor housing, a drive assembly, and a shaft. The shaft of the drive motor is coupled with the drive wheel, and the drive motor is used to control the rotation of the drive wheel.

[0176] In this embodiment, the cleaning execution unit 430 is equipped with a drive wheel 433 and a walking motor 434. The walking motor 434 drives the drive wheel 433 to rotate via a reducer, thus achieving the purpose of rotating the drive wheel 433. The walking motor 434 is housed within the drive wheel 433, resulting in a more rational internal structure and reduced space occupation. This avoids the problem of the drive wheel being difficult to suspend on the robot due to the large size of the motor and reducer. The externally mounted motor drive wheel is simple, reliable, and compact. Furthermore, the use of a reducer makes the power output more reliable and provides greater torque.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them; under the concept of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of this application as described above. For the sake of brevity, they are not provided in detail; although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for robot extrication from difficult situations, characterized in that, The robot includes a roller brush and a camera unit, and the method includes: When the roller brush is in the cleaning working state, if the load current of the roller brush is detected to be greater than a preset current threshold, the camera unit is controlled to acquire the image of the roller brush. Based on the roller image, determine the type of entanglement; If the type of entanglement is a non-tape type, then the robot is controlled to enter a first escape mode, wherein the first escape mode includes controlling the roller brush to reverse at a first preset speed; If the robot fails to escape the obstacle after the first escape mode has been executed for a first preset time, the robot is controlled to enter the second escape mode. The second escape mode includes controlling the roller brush to reverse at a second preset speed, where the second preset speed is greater than the first preset speed. If the robot fails to escape the obstacle after the second preset time has elapsed in the second escape mode, it is determined that the robot has failed to escape the obstacle, and the robot is controlled to stop abruptly and an alarm is triggered.

2. The method according to claim 1, characterized in that, The step of determining the type of entanglement based on the roller image includes: Based on the target detection model, the roller brush image is preprocessed to obtain the roller brush image to be detected; If an entangled object is detected in the image of the roller to be detected, feature extraction is performed on the entangled object to determine the type of entangled object.

3. The method according to claim 2, characterized in that, The method further includes: If no entangled target is detected in the image of the roller to be detected, the robot is controlled to re-enter the cleaning working state and the load current of the roller is detected again.

4. The method according to any one of claims 1-3, characterized in that, The robot also includes drive wheels; The first escape mode includes: The roller brush is controlled to reverse at a first preset speed, and the drive wheel is controlled to reverse backward at a first auxiliary preset speed, wherein the first preset speed is less than the maximum rotational speed of the roller brush, and the first auxiliary preset speed is less than the maximum rotational speed of the drive wheel. The second escape mode includes: The roller brush is controlled to reverse at a second preset speed, and the drive wheel is controlled to reverse backward at a second auxiliary preset speed, wherein the second preset speed is equal to the maximum rotational speed of the roller brush, and the second auxiliary preset speed is equal to the maximum rotational speed of the drive wheel.

5. The method according to any one of claims 1-3, characterized in that, The method further includes: After the first preset time is executed in the first escape mode, if the load current of the roller brush is detected to be less than or equal to the preset current threshold, it is determined that the robot has successfully escaped the obstacle. If, after the second preset time has elapsed in the second escape mode, the load current of the roller brush is detected to be less than or equal to the preset current threshold, then the robot is determined to have successfully escaped the obstacle.

6. The method according to any one of claims 1-3, characterized in that, The method further includes: If the type of wrapped material is tape, then control the robot to stop abruptly and trigger an alarm.

7. A robot, characterized in that, include: The main controller is used to execute the robot's obstacle avoidance method as described in any one of claims 1-6; The driver is connected to the main controller and is used to execute control commands issued by the main controller; The cleaning execution unit is connected to the driver and is used to execute the control commands issued by the driver to control the robot to escape from trouble, stop suddenly, or issue an alarm. The cleaning execution unit includes: Roller brush; A roller brush motor is used to control the movement of the roller brush; Drive wheel; A walking motor is used to control the movement of the drive wheels.

8. The robot according to claim 7, characterized in that, The driver includes a winding detection unit, the winding detection unit comprising: A load current detector, integrated into the driver, is used to acquire the load current of the brush. The camera unit is used to capture images of the roller brush.

9. The robot according to claim 8, characterized in that, The driver also includes a processing unit and an escape execution unit, including: The processing unit is connected to the main controller, the winding detection unit, and the escaping execution unit. It is used to upload the load current of the roller and the roller image collected by the winding detection unit to the main controller, and to parse the control commands issued by the main controller to control the escaping execution unit. The obstacle-avoidance execution unit is connected to the cleaning execution unit and the entanglement detection unit, and is used to receive instructions sent by the entanglement detection unit to control the movement of the roller brush and the drive wheel.

10. The robot according to claim 9, characterized in that, The escape execution unit includes: A forward / reverse control circuit is used to control the forward or reverse rotation of the roller brush, and to control the forward or backward movement of the drive wheel; A speed control circuit is used to control the speed of the roller brush and / or drive wheel.

Images