Return voyage control method and device of underwater robot and underwater robot

Abstract

The invention discloses a return voyage control method and device of an underwater robot and the underwater robot, and the method comprises the steps: receiving a remote control instruction transmitted by a control device of the underwater robot, and judging whether the remote control instruction is a specified return voyage instruction or not; when the remote control instruction is the specified return flight instruction, performing reverse view angle switching on a camera of the underwater robot, and performing reverse mapping operation on a pre-constructed navigation control strategy to obtain a reverse control strategy; and controlling the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy. The homeward voyage control method and device of the underwater robot and the underwater robot can efficiently and quickly perform homeward voyage control on the underwater robot.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of system control technology, and in particular to a return control method and device for an underwater robot, an underwater robot, and a storage medium. Background Art

[0002] With the continuous development of artificial intelligence technology, unmanned technology (for example, unmanned aerial vehicles) has been gradually applied to various fields, greatly improving the efficiency of traditional operations. For example, underwater robots are used to help people explore the unknown underwater world, including but not limited to using underwater robots to photograph underwater environments and using underwater robots to replace humans in underwater operations such as exploration.

[0003] In the prior art, after an underwater robot performs underwater operations or encounters a malfunction, it needs to return. However, in the prior art, when an underwater robot returns underwater, the underwater robot is usually pulled back by a cable. When the underwater robot returns in a narrow environment such as a pipeline, this method is prone to malfunction and is not fast and efficient enough. Summary of the invention

[0004] The embodiments of the present application provide a return control method and device for an underwater robot, an underwater robot, and a computer-readable storage medium, which can efficiently and quickly control the return of the underwater robot.

[0005] The present application provides, on the one hand, a method for controlling the return of an underwater robot, comprising:

[0006] Receiving a remote control command sent by a control device of the underwater robot, and determining whether the remote control command is a designated return command;

[0007] When the remote control instruction is the designated return instruction, the camera of the underwater robot is switched to a reverse perspective, and a reverse mapping operation is performed on the pre-built navigation control strategy to obtain a reverse control strategy;

[0008] According to the remote control instruction and the reverse control strategy, the underwater robot is controlled to return in the target channel.

[0009] Optionally, before determining whether the remote control instruction is a designated return instruction, the method further includes:

[0010] Generate verification information according to the remote control instruction, and send the verification information to the remote control of the underwater robot;

[0011] Receiving a verification result fed back by a remote controller of the underwater robot;

[0012] When the verification result is that the verification is passed, the operation of determining whether the remote control instruction is a designated return instruction is performed.

[0013] Optionally, the camera includes a front camera and a rear camera, and the reverse viewing angle switching of the camera of the underwater robot includes:

[0014] In the interface of the application APP that interacts with the underwater robot, the front image captured by the front camera and the rear image captured by the rear camera are displayed interchangeably.

[0015] Optionally, the pre-built navigation control strategy includes forward control, backward control, left control and right control; the reverse mapping operation includes reverse mapping of forward control and backward control, and reverse mapping of left control and right control.

[0016] Optionally, controlling the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy includes:

[0017] Identifying a control direction included in the remote control command;

[0018] If the control direction is forward control, the underwater robot is controlled to move backward according to the reverse control strategy; if the control direction is backward control, the underwater robot is controlled to move forward; and

[0019] If the control direction is left control, the underwater robot is controlled to move to the right; if the control direction is right control, the underwater robot is controlled to move to the left.

[0020] Optionally, controlling the underwater robot to return in the target channel includes:

[0021] Acquire the traveled route and power change information sequence of the underwater robot, wherein the power change information sequence includes the power output direction and strength of the power equipment cluster of the underwater robot that changes over time, and the remote control instructions received by the underwater robot in each historical time period;

[0022] The underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence.

[0023] Optionally, controlling the underwater robot to return in the target channel according to the traveled route and the power change information sequence includes:

[0024] Using a curve posture recognition network in a pre-trained complex road section analysis model to extract features of the traveled route, to obtain a curve feature set;

[0025] Using the power change feature recognition network in the complex road section analysis model, feature extraction is performed on the power change information sequence to obtain a power change feature set;

[0026] Performing a corresponding moment feature splicing operation on the curve feature set and the dynamic change feature set to obtain a fused feature set, and performing a context attention enhancement operation on the fused feature set to obtain an enhanced feature set;

[0027] Scoring the enhanced feature set using the route complexity analysis network in the complex road segment analysis model to obtain a route complexity score sequence;

[0028] According to a preset complexity score threshold and the route complexity score sequence, the traveled route is segmented and output to obtain complex sections in the traveled route and non-complex sections in the traveled route;

[0029] When the underwater robot returns to the complex road section, the underwater robot is controlled to return along the original route according to the driving route corresponding to the complex road section;

[0030] When the underwater robot navigates to the non-complex section, the shortest distance analysis is performed on the traveled route corresponding to the non-complex section to obtain the shortest return route, and the underwater robot is controlled to return according to the shortest return route.

[0031] The present application also provides a return control device for an underwater robot, including:

[0032] A receiving module, used to receive a remote control instruction sent by a control device of the underwater robot, and determine whether the remote control instruction is a designated return instruction;

[0033] A switching module, used for, when the remote control instruction is the designated return instruction, switching the camera of the underwater robot to a reverse perspective, and performing a reverse mapping operation on the pre-built navigation control strategy to obtain a reverse control strategy;

[0034] The return module is used to control the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy.

[0035] On the one hand, an embodiment of the present application also provides an underwater robot, including: a memory and a processor; the memory stores executable program code; the processor coupled to the memory calls the executable program code stored in the memory to execute the return control method of the underwater robot provided in the above embodiment.

[0036] On one hand, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the return control method of the underwater robot provided in the above embodiment is implemented.

[0037] It can be seen from the above embodiments of the present application that by receiving the remote control command sent by the control device of the underwater robot, it is determined whether the remote control command is a designated return command. The remote control command sent by the pre-built underwater robot remote controller is received; the remote control command is determined whether the remote control command is a pre-built return command; when the remote control command is the designated return command, the pre-built camera of the underwater robot is switched to the front and back reverse perspectives, and the pre-built robot navigation control strategy is reversely mapped to obtain a reverse control strategy; according to the remote control command and the reverse control strategy, the power equipment cluster in the pre-built target channel of the underwater robot is controlled and returned. Since the camera of the underwater robot is reversely switched to the reverse perspective and the pre-built navigation control strategy is reversely mapped, the user does not need to consider the difference between the front and back correspondence and the left and right correspondence when the underwater robot returns and when it advances, and directly operates the front and back left and right control of the underwater robot when returning (i.e., retreating) in the same way as when the underwater robot advances, which greatly improves the convenience of underwater robot control, improves the efficiency of the underwater robot's return, and achieves the purpose of efficiently and quickly controlling the underwater robot to return. At the same time, it reduces the risk of collision caused by user misoperation and improves the safety of the underwater robot when returning. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 A flowchart of a method for controlling the return of an underwater robot according to an embodiment of the present application;

[0040] Figure 2 A schematic diagram of the structure of a return control device for an underwater robot provided in one embodiment of the present application;

[0041] Figure 3 A schematic diagram of the hardware structure of an underwater robot provided in one embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figure 1 , Figure 1 This is a flowchart of the implementation of the return control method of the underwater robot provided in one embodiment of the present application. The method described in this embodiment can be applied to the underwater robot. Figure 1 As shown, the method specifically includes:

[0044] Step S11, receiving a remote control instruction sent by the control device of the underwater robot, and determining whether the remote control instruction is a designated return instruction.

[0045] In this embodiment, the underwater robot can navigate in underwater environments such as pipelines and tunnels of a water plant.

[0046] In this embodiment, the control device of the underwater robot is a device for controlling the underwater robot. The control device of the underwater robot can control the underwater robot in one or more directions of up, down, front, back, left, and right.

[0047] For example, the control device is an application APP for controlling an underwater robot, or the control device is a control remote sensor.

[0048] The remote control command may be triggered by a user operation. For example, the user touches a target button to trigger the remote control command, and the target button is a button in the interface of the application APP that controls the underwater robot.

[0049] In this embodiment, the underwater robot includes a power equipment cluster and a signal transceiver. The power equipment cluster is used to maintain the stability of the underwater robot and move in all directions, and at least includes a plurality of power equipment; the signal transceiver is used to communicate with the control device of the underwater robot, for example, to receive instructions issued by the remote control device that controls the underwater robot.

[0050] In this embodiment, when the underwater robot is in forward navigation, the underwater robot is in forward navigation (i.e., in the direction of the head of the underwater robot) in the target pipeline. The designated return instruction is used to trigger the underwater robot to return backward, i.e., to navigate in the direction of the tail of the underwater robot. Specifically, the designated return instruction instructs the underwater robot not to turn around, but to directly switch the navigation direction of the underwater robot to return.

[0051] Specifically, after receiving the remote control command sent by the remote controller of the underwater robot, the remote control command is parsed to obtain information such as the command operation code and parameters, and the command type (for example, movement commands such as forward navigation or backward return) is determined based on this information.

[0052] Specifically, judging whether the remote control command is a designated return command includes matching the information contained in the remote control command with the designated return information to determine whether the remote control command is a designated return command. If so, the underwater robot is controlled to return directly without turning around.

[0053] Furthermore, in an optional embodiment of the present invention, before determining whether the remote control command is a designated return command, the method further includes:

[0054] Generate verification information according to the remote control instruction, and send the verification information to the remote control of the underwater robot;

[0055] Receiving a verification result fed back by a remote controller of the underwater robot;

[0056] When the verification result is that the verification is passed, the operation of determining whether the remote control instruction is a designated return instruction is performed.

[0057] Specifically, verification information can be generated according to the remote control command through a preset algorithm (such as a cyclic redundancy check algorithm, an asymmetric encryption algorithm, a hash-based message authentication code algorithm), and the verification information is sent to the remote control of the underwater robot for verification, and the verification result is sent to the underwater robot. When the verification is passed, the underwater robot returns that the verification is passed, and then performs the operation of determining whether the remote control command is a specified return command.

[0058] Step S12: when the remote control instruction is the designated return instruction, the camera of the underwater robot is switched to a reverse perspective, and a reverse mapping operation is performed on the pre-built navigation control strategy to obtain a reverse control strategy.

[0059] In this embodiment, the underwater robot further includes a camera, through which images of the environment in which the underwater robot is navigating can be collected.

[0060] In this embodiment, the camera of the underwater robot is switched to a reverse perspective, that is, the content displayed by the camera of the underwater robot is marked with a reverse perspective. The front camera of the underwater robot during the original navigation is marked as the rear camera of the current underwater robot when returning; or, the rear camera of the underwater robot during the original navigation is marked as the front camera of the current underwater robot when returning; or, the left camera of the underwater robot during the original navigation is marked as the right camera of the current underwater robot when returning; or, the right camera of the underwater robot during the original navigation is marked as the left camera of the current underwater robot when returning.

[0061] In specific implementation, according to the pre-built robot navigation control strategy, the power equipment cluster can output the power throttle according to the instructions issued by the application APP. According to the reverse control strategy, the power equipment cluster can output the power throttle in the opposite direction of the instructions issued by the application APP.

[0062] Furthermore, the pre-built navigation control strategy includes forward control, backward control, left control and right control; the reverse mapping operation includes reverse mapping between forward control and backward control, and reverse mapping between left control and right control.

[0063] Specifically, the forward control is used to control the underwater robot to move forward, for example, moving forward in an underwater pipe; similarly, the backward control is used to control the underwater robot to move backward; the left control is used to control the underwater robot to move left; and the right control is used to control the underwater robot to move right.

[0064] Specifically, the reverse mapping of the forward control and the backward control refers to swapping the forward control with the backward control, and the reverse mapping of the left control and the right control refers to swapping the left control with the right control.

[0065] Further, in an optional embodiment of the present invention, the camera includes a front camera and a rear camera, and the reverse viewing angle switching of the camera of the underwater robot includes:

[0066] In the interface of the application APP that interacts with the underwater robot, the front image captured by the front camera and the rear image captured by the rear camera are displayed interchangeably.

[0067] In this embodiment, the front camera is installed in front of the underwater robot, and the rear camera is installed at the rear of the underwater robot.

[0068] Specifically, swapping the front image captured by the front camera and the rear image captured by the rear camera means switching the front picture captured by the front camera to the rear picture captured by the rear camera, and switching the rear picture captured by the rear camera to the rear picture captured by the front camera.

[0069] That is, in the interface of the application APP that interacts with the underwater robot, the rear picture captured by the rear camera is displayed and the rear picture is marked as the front view; or, the front picture captured by the front camera is displayed and the front picture is marked as the rear view.

[0070] For example, in the interface of an application APP that interacts with an underwater robot, there are a forward view window and a rear view window. When the underwater robot is moving forward in a target channel, the forward view window displays the image captured by the front camera of the underwater robot, and the rear view window displays the image captured by the rear camera of the underwater robot. When the camera of the underwater robot is switched to a reverse perspective, the forward view window of the underwater robot displays the image captured by the rear camera, and the rear view window displays the image captured by the front camera.

[0071] In this embodiment, the front image captured by the front camera and the rear image captured by the rear camera are swapped and displayed, which can not only allow the user to see the image on the return channel in real time and intuitively, but also in this process, the user does not need to consider the corresponding relationship between forward and backward, and the user can operate the return (i.e., backward) of the underwater robot in the same way as the forward operation, which greatly improves the convenience of underwater robot control and improves the efficiency of the return of the underwater robot. In addition, it is beneficial for the user to carry out timely control and intervention when necessary, and improve the safety, reliability and stability of the underwater robot when returning.

[0072] Step S13, controlling the underwater robot to return to the target channel according to the remote control instruction and the reverse control strategy.

[0073] In this embodiment, due to the reverse mapping operation, a reverse control strategy is obtained. Therefore, after mapping, when the received instruction is a forward instruction, the underwater robot is controlled to fly backward; when the received instruction is a backward instruction, the underwater robot is controlled to fly forward; when the received instruction is a left instruction, the underwater robot is controlled to fly right; when the received instruction is a backward instruction, the underwater robot is controlled to fly left.

[0074] Furthermore, controlling the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy includes:

[0075] Identifying a control direction included in the remote control command;

[0076] If the control direction is forward control, the underwater robot is controlled to move backward according to the reverse control strategy; if the control direction is backward control, the underwater robot is controlled to move forward; and

[0077] If the control direction is left control, the underwater robot is controlled to move to the right; if the control direction is right control, the underwater robot is controlled to move to the left.

[0078] Specifically, in actual applications, the remote control instruction may include a return instruction identifier and may also include direction control information when returning. And the remote control instruction may be an instruction sent continuously or an instruction sent at intervals, including a series of control commands. For example, the remote control instruction is used to control the underwater robot to return backwards, and then to control the underwater robot to move forward, left, right, up, down, etc. when returning backwards.

[0079] Specifically, controlling the underwater robot to return in the target channel includes power control of the power equipment cluster of the underwater robot. If the control direction is forward control, the power equipment cluster is controlled to generate a backward force; if the control direction is backward control, the power equipment cluster is controlled to generate a forward force; if the control direction is left control, the power equipment cluster is controlled to generate a right force; if the control direction is right control, the power equipment cluster is controlled to generate a left force.

[0080] Furthermore, in other optional embodiments, the automatic return function can be started or canceled by applying an APP. When the automatic return function is turned on, the automatic return can be performed without the user controlling the remote sensing. When the automatic return function is turned off, the user needs to control the return of the underwater robot by controlling the remote sensing or applying an APP or other control device.

[0081] Further, in an optional embodiment of the present invention, controlling the underwater robot to return in the target channel includes:

[0082] Acquire the traveled route and power change information sequence of the underwater robot, wherein the power change information sequence includes the power output direction and strength of the power equipment cluster of the underwater robot that changes over time, and the remote control instructions received by the underwater robot in each historical time period;

[0083] The underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence.

[0084] The route traveled by the underwater robot refers to the route traveled by the underwater robot in the current environment before receiving the current remote control command, for example, the route traveled by the underwater robot in the space of the target pipeline, specifically including the up, down, left, right, front, and back information of the underwater robot during navigation. Specifically, in the embodiment of the present invention, when the underwater robot is sailing forward, the route traveled by the underwater robot and the sequence of power change information can be obtained and saved through image analysis, gyroscope acquisition and other methods. When the specified return command is received, the saved route traveled is directly obtained.

[0085] In the embodiment of the present invention, the remote control instructions received by the underwater robot in each historical time period are used to determine the flight state of the underwater robot. Specifically, during the forward navigation process of the underwater robot, the underwater robot can cruise automatically or be controlled by an operator. The remote control instructions received by the underwater robot in each historical time period are used to instruct the underwater robot to navigate in a certain navigation type, and specifically, the navigation type includes but is not limited to driving forward close to the pipe wall, driving at a constant speed, and accelerating.

[0086] In this embodiment, the underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence. The underwater robot can return automatically without the need for operator operation, thereby improving the return efficiency.

[0087] Further, in an optional embodiment of the invention, the controlling the underwater robot to return in the target pipeline according to the traveled route and the power change information sequence includes:

[0088] Using a curve posture recognition network in a pre-trained complex road section analysis model to extract features of the traveled route, to obtain a curve feature set;

[0089] Using the power change feature recognition network in the complex road section analysis model, feature extraction is performed on the power change information sequence to obtain a power change feature set;

[0090] Performing a corresponding moment feature splicing operation on the curve feature set and the dynamic change feature set to obtain a fused feature set, and performing a context attention enhancement operation on the fused feature set to obtain an enhanced feature set;

[0091] Scoring the enhanced feature set using the route complexity analysis network in the complex road segment analysis model to obtain a route complexity score sequence;

[0092] According to a preset complexity score threshold and the route complexity score sequence, the traveled route is segmented and output to obtain complex sections in the traveled route and non-complex sections in the traveled route;

[0093] When the underwater robot returns to the complex road section, the underwater robot is controlled to return along the original route according to the driving route corresponding to the complex road section;

[0094] When the underwater robot navigates to the non-complex section, the shortest distance analysis is performed on the traveled route corresponding to the non-complex section to obtain the shortest return route, and the underwater robot is controlled to return according to the shortest return route.

[0095] In this embodiment, the complex section analysis model is obtained by pre-training. The complex route analysis model includes networks with different purposes, for example, a route complexity analysis network, a power change feature recognition network and a route complexity analysis network.

[0096] In this embodiment, a curve feature set and a dynamic change feature set are first obtained by feature extraction, and then the features of the curve feature set and the dynamic change feature set at the same time are spliced ​​to realize the construction of a fused feature set, thereby improving the relationship and correlation between different features, and then a contextual attention enhancement operation is performed through an attention enhancement algorithm to improve the before-and-after relationship and correlation of different features, thereby obtaining an enhanced feature set, and finally a route complexity analysis network (i.e., a neural network) is used to perform a fully connected classification judgment to score the enhanced feature set, and a route complexity score sequence of each section is obtained (specifically, each time in a different section may correspond to a different score), and the scores are summarized and output to obtain complex sections and non-complex sections.

[0097] Specifically, the complex road section analysis model can be used to analyze the position coordinates and power change information sequence of the traveled route at each time node. By analyzing the changes in the position coordinates and power change information sequence between adjacent time nodes, the curve characteristics and power change characteristics between adjacent time nodes can be identified. Then, according to the curve characteristics and power change characteristics, the road section complexity score is scored to determine the complexity of the road section between the adjacent time nodes. After the scores in each adjacent time period in the traveled route are calculated, the route complexity score sequence can be obtained, and then segmented according to a complexity judgment standard (such as a threshold score) to obtain complex sections and non-complex sections.

[0098] Specifically, after the scores of each adjacent time period in the traveled route are calculated, the scores of each adjacent time period can be arranged according to the time sequence to obtain a route complexity score sequence. Then, the complexity judgment standard (such as the threshold score) is compared with each score in the route complexity score sequence, and the time period with a score greater than the complexity judgment standard is extracted to obtain a complex time period. According to the time correspondence, the section of the traveled route corresponding to the complex time period is determined as a complex section, and the remaining sections of the traveled route are determined as non-complex sections.

[0099] In this embodiment, a complex section refers to a section with a greater number of bends than a preset number or a continuous bend angle greater than a preset angle. Alternatively, a complex section refers to a section with a greater number of obstacles in the target pipeline than a preset number (e.g., 2 obstacles). Alternatively, a complex route refers to a section with a less than preset number of obstacles in the target pipeline, and a size of the obstacle greater than a preset volume (e.g., one-third of the volume of the underwater robot), or a position of the obstacle is a target position (e.g., directly in front of the target pipeline).

[0100] Accordingly, a complex section refers to a section where the number of bends is not greater than a preset number or the continuous bend angle is less than or equal to a preset angle. Alternatively, a complex section refers to a section where the number of obstacles in the target pipeline is less than a preset number (for example, 2 obstacles). Alternatively, a complex route refers to a section where the number of obstacles in the target pipeline is less than a preset number, and the size of the obstacle is not greater than a preset volume (such as one-third of the volume of the underwater robot), or the location of the obstacle is not the target location (such as directly in front of the target pipeline).

[0101] Specifically, when a complex section or a non-complex section is obtained, the underwater robot is controlled by the power output by the power equipment cluster of the underwater robot to return to the original route or return to a non-original route (i.e., return quickly). Specifically, the power equipment cluster of the underwater robot includes multiple power devices, and each power device operates in a coordinated relationship, and each power device can output power in multiple directions.

[0102] In specific implementation, when returning to the original route, the power matrix for maintaining the stability of the underwater robot and the forward power matrix for each complex section can be obtained according to the power change information sequence. Then the forward power matrix is ​​transposed to obtain the modified power matrix. And the modified power matrix is ​​used to control the direction distribution and power output of each power device in the power equipment cluster to achieve the original route return.

[0103] Specifically, in the embodiment of the present invention, the shortest distance analysis is performed on the traveled route corresponding to the non-complex road section, and the shortest return route is obtained, which means that when the tunnel of the target pipeline is very wide, very long, and has multiple directions, it is not necessary to control the underwater robot to navigate according to the original forward navigation position (for example, when the underwater robot is navigating according to the original forward navigation, it will move up, down, left, right, etc. at a certain position point), and when returning in these non-complex sections, it is sufficient to navigate within the limit where the underwater robot will not collide, and there is no need to move in different directions at a certain position point. For example, the underwater robot can turn close to the tunnel wall when returning, so as to reduce the return route and improve the return efficiency; in addition, when sailing to the surface after exiting the tunnel of the target pipeline, it can directly sail in a straight line to the recovery point (i.e., the horizontal plane when the underwater robot dives into the water), so as to achieve a rapid return after leaving the target pipeline.

[0104] Specifically, the historical travel coordinates of the traveled route corresponding to the non-complex road section may be obtained, various expected return routes may be calculated based on the historical travel coordinates, and the expected return route with the shortest distance may be determined as the shortest return route.

[0105] Alternatively, the shortest distance can be calculated based on the starting point and the end point of the non-complex road section to obtain the shortest return route, thereby controlling the power equipment cluster to return.

[0106] In specific implementation, when the underwater robot navigates to a non-complex section, since the return route of the underwater robot is different from the original return route, the power matrix required to keep the robot stable can be calculated in real time, and then an additional power matrix in the opposite direction of the most recent route can be applied to realize the movement of the underwater robot.

[0107] In this embodiment, the return efficiency of the underwater robot can be improved by classifying the traveled route into complex sections and non-complex sections and then selecting different return modes.

[0108] Furthermore, in an optional embodiment of the present invention, before extracting features of the traveled route using the curve posture recognition network in the complex road section analysis model, the method further includes:

[0109] Obtaining a pre-constructed complex road section analysis model and a pre-constructed underwater navigation experiment sample set, and allocating the underwater navigation experiment sample set into a training set and a test set according to a preset ratio;

[0110] Iteratively training the complex road section analysis model using the training set to obtain an updated complex road section analysis model, and obtaining the recognition accuracy of the updated complex road section analysis model according to the test set;

[0111] Determining whether the recognition accuracy is greater than a preset qualified threshold;

[0112] When the recognition accuracy is greater than the qualified threshold, the iterative training process is stopped to obtain a pre-trained complex road section analysis model.

[0113] In this embodiment, the pre-constructed complex road section analysis model may be constructed based on a binary classification model.

[0114] In this embodiment, an underwater navigation experiment sample set can be obtained by testing various actual underwater pipeline environments, and the underwater navigation experiment sample set can be grouped into a training set and a test set according to a preset ratio (such as 7:3).

[0115] Specifically, the complex road section analysis model can be iteratively trained using the training set according to the cross entropy loss algorithm and the gradient descent method. And the training progress of the model is controlled by the recognition accuracy. When the recognition accuracy of the complex road section analysis model is greater than the preset qualified threshold, such as 95%, it indicates that the model training effect is good, the iteration process is stopped, and the pre-trained complex road section analysis model is obtained. In the embodiment of the present application, by receiving the remote control command sent by the control device of the underwater robot, it is judged whether the remote control command is a designated return command. The remote control command sent by the pre-built underwater robot remote control is received; the remote control command is judged whether the remote control command is a pre-built return command; when the remote control command is the designated return command, the pre-built camera of the underwater robot is switched to the front and back reverse perspectives, and the pre-built robot navigation control strategy is reversely mapped to obtain a reverse control strategy; according to the remote control command and the reverse control strategy, the power equipment cluster controlling the underwater robot in the pre-built target channel is powered to return. Since the camera of the underwater robot is reversely switched to a reverse perspective and the pre-built navigation control strategy is reversely mapped, the user does not need to consider the difference between the front-to-back correspondence and the left-to-right correspondence when the underwater robot is returning and when it is advancing. The user can directly operate the front-to-back, left-to-right, and right-to-left control of the underwater robot when returning (i.e., retreating) in the same way as when the underwater robot is advancing, which greatly improves the convenience of underwater robot control and the efficiency of the underwater robot's return, and achieves the purpose of efficient and rapid control of the underwater robot's return. At the same time, it reduces the risk of collision caused by user misoperation and improves the safety of the underwater robot when returning.

[0116] like Figure 2 , which is a functional module diagram of a return control device for an underwater robot provided in one embodiment of the present invention.

[0117] The return control device 100 of the underwater robot of the present invention can be installed in the underwater robot. According to the functions to be implemented, the return control device 100 of the underwater robot includes: a receiving module 101, a switching module 102 and a return module 103. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by the underwater robot processor and can complete fixed functions, which are stored in the memory of the underwater robot.

[0118] In this embodiment, the functions of each module / unit are as follows:

[0119] The receiving module 101 receives a remote control instruction sent by a control device of the underwater robot, and determines whether the remote control instruction is a designated return instruction.

[0120] In this embodiment, the underwater robot can navigate in underwater environments such as pipelines and tunnels of a water plant.

[0121] In this embodiment, the control device of the underwater robot is a device for controlling the underwater robot. The control device of the underwater robot can control the underwater robot in one or more directions of up, down, front, back, left, and right.

[0122] For example, the control device is an application APP for controlling an underwater robot, or the control device is a control remote sensor.

[0123] The remote control command may be triggered by a user operation. For example, the user touches a target button to trigger the remote control command, and the target button is a button in the interface of the application APP that controls the underwater robot.

[0124] In this embodiment, the underwater robot includes a power equipment cluster and a signal transceiver. The power equipment cluster is used to maintain the stability of the underwater robot and move in all directions, and at least includes a plurality of power equipment; the signal transceiver is used to communicate with the control device of the underwater robot, for example, to receive instructions issued by the remote control device that controls the underwater robot.

[0125] In this embodiment, when the underwater robot is in forward navigation, the underwater robot is in forward navigation (i.e., in the direction of the head of the underwater robot) in the target pipeline. The designated return instruction is used to trigger the underwater robot to return backward, i.e., to navigate in the direction of the tail of the underwater robot. Specifically, the designated return instruction instructs the underwater robot not to turn around, but to directly switch the navigation direction of the underwater robot to return.

[0126] Specifically, after receiving the remote control command sent by the remote controller of the underwater robot, the remote control command is parsed to obtain information such as the command operation code and parameters, and the command type (for example, movement commands such as forward navigation or backward return) is determined based on this information.

[0127] Specifically, judging whether the remote control command is a designated return command includes matching the information contained in the remote control command with the designated return information to determine whether the remote control command is a designated return command. If so, the underwater robot is controlled to return directly without turning around.

[0128] Furthermore, in an optional embodiment of the present invention, the receiving module 101 is further configured to:

[0129] Before determining whether the remote control instruction is a designated return instruction, generating verification information according to the remote control instruction, and sending the verification information to the remote control of the underwater robot;

[0130] Receiving a verification result fed back by a remote controller of the underwater robot;

[0131] When the verification result is passed, the operation of determining whether the remote control command is a designated return command is performed.

[0132] Specifically, verification information can be generated according to the remote control command through a preset algorithm (such as a cyclic redundancy check algorithm, an asymmetric encryption algorithm, a hash-based message authentication code algorithm), and the verification information is sent to the remote control of the underwater robot for verification, and the verification result is sent to the underwater robot. When the verification is passed, the underwater robot returns that the verification is passed, and then the operation of determining whether the remote control command is a specified return command is performed.

[0133] The switching module 102 is used to perform reverse viewing angle switching on the camera of the underwater robot and perform reverse mapping operation on the pre-built navigation control strategy to obtain a reverse control strategy when the remote control instruction is the designated return instruction.

[0134] In this embodiment, the underwater robot further includes a camera, through which images of the environment in which the underwater robot is navigating can be collected.

[0135] In this embodiment, the camera of the underwater robot is switched to a reverse perspective, that is, the content displayed by the camera of the underwater robot is marked with a reverse perspective. The front camera of the underwater robot during the original navigation is marked as the rear camera of the current underwater robot when it returns; or, the rear camera of the underwater robot during the original navigation is marked as the front camera of the current underwater robot when it returns; or, the left camera of the underwater robot during the original navigation is marked as the right camera of the current underwater robot when it returns; or, the right camera of the underwater robot during the original navigation is marked as the left camera of the current underwater robot when it returns.

[0136] In specific implementation, according to the pre-built robot navigation control strategy, the power equipment cluster can output the power throttle according to the instructions issued by the application APP. According to the reverse control strategy, the power equipment cluster can output the power throttle in the opposite direction of the instructions issued by the application APP.

[0137] Furthermore, the pre-built navigation control strategy includes forward control, backward control, left control and right control; the reverse mapping operation includes reverse mapping between forward control and backward control, and reverse mapping between left control and right control.

[0138] Specifically, the forward control is used to control the underwater robot to move forward, for example, moving forward in an underwater pipe; similarly, the backward control is used to control the underwater robot to move backward; the left control is used to control the underwater robot to move left; and the right control is used to control the underwater robot to move right.

[0139] Specifically, the reverse mapping of the forward control and the backward control refers to swapping the forward control with the backward control, and the reverse mapping of the left control and the right control refers to swapping the left control with the right control.

[0140] Further, in an optional embodiment of the present invention, the camera includes a front camera and a rear camera, and the reverse viewing angle switching of the camera of the underwater robot includes:

[0141] In the interface of the application APP that interacts with the underwater robot, the front image captured by the front camera and the rear image captured by the rear camera are displayed interchangeably.

[0142] In this embodiment, the front camera is installed in front of the underwater robot, and the rear camera is installed at the rear of the underwater robot.

[0143] Specifically, swapping the front image captured by the front camera and the rear image captured by the rear camera means switching the front picture captured by the front camera to the rear picture captured by the rear camera, and switching the rear picture captured by the rear camera to the rear picture captured by the front camera.

[0144] That is, in the interface of the application APP that interacts with the underwater robot, the rear picture captured by the rear camera is displayed and the rear picture is marked as the front view; or, the front picture captured by the front camera is displayed and the front picture is marked as the rear view.

[0145] For example, in the interface of an application APP that interacts with an underwater robot, there are a forward view window and a rear view window. When the underwater robot is moving forward in a target channel, the forward view window displays the image captured by the front camera of the underwater robot, and the rear view window displays the image captured by the rear camera of the underwater robot. When the camera of the underwater robot is switched to a reverse perspective, the forward view window of the underwater robot displays the image captured by the rear camera, and the rear view window displays the image captured by the front camera.

[0146] In this embodiment, the front image captured by the front camera and the rear image captured by the rear camera are swapped and displayed, which can not only allow the user to see the image on the return channel in real time and intuitively, but also in this process, the user does not need to consider the corresponding relationship between forward and backward, and the user can operate the return (i.e., backward) of the underwater robot in the same way as the forward operation, which greatly improves the convenience of underwater robot control and improves the efficiency of the return of the underwater robot. In addition, it is beneficial for the user to carry out timely control and intervention when necessary, and improve the safety, reliability and stability of the underwater robot when returning.

[0147] The return module 103 is used to control the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy.

[0148] In this embodiment, due to the reverse mapping operation, a reverse control strategy is obtained. Therefore, after mapping, when the received instruction is a forward instruction, the underwater robot is controlled to fly backward; when the received instruction is a backward instruction, the underwater robot is controlled to fly forward; when the received instruction is a left instruction, the underwater robot is controlled to fly right; when the received instruction is a backward instruction, the underwater robot is controlled to fly left.

[0149] Furthermore, the return module 103 is used to:

[0150] Identifying a control direction included in the remote control command;

[0151] If the control direction is forward control, the underwater robot is controlled to move backward according to the reverse control strategy; if the control direction is backward control, the underwater robot is controlled to move forward; and

[0152] If the control direction is left control, the underwater robot is controlled to move to the right; if the control direction is right control, the underwater robot is controlled to move to the left.

[0153] Specifically, in actual applications, the remote control instruction may include a return instruction identifier and may also include direction control information when returning. And the remote control instruction may be an instruction sent continuously or an instruction sent at intervals, including a series of control commands. For example, the remote control instruction is used to control the underwater robot to return backwards, and then to control the underwater robot to move forward, left, right, up, down, etc. when returning backwards.

[0154] Specifically, controlling the underwater robot to return in the target channel includes performing power control on a power equipment cluster of the underwater robot.

[0155] In detail, in an embodiment of the present invention, the power control of the pre-built power equipment cluster according to the reverse control strategy includes:

[0156] When receiving a navigation control instruction sent by the underwater robot remote controller, identifying a control direction included in the navigation control instruction;

[0157] According to the reverse control strategy, if the control direction is forward control, the power equipment cluster is controlled to generate a backward force; if the control direction is backward control, the power equipment cluster is controlled to generate a forward force; and

[0158] If the control direction is left-direction control, the power equipment cluster is controlled to generate right-direction force; if the control direction is right-direction control, the power equipment cluster is controlled to generate left-direction force.

[0159] Specifically, in this embodiment, the navigation control instruction is used to control the direction of the underwater robot when returning. The navigation control instruction of the underwater robot is reversely mapped, that is, the control direction of the navigation control instruction of the underwater robot is reversely mapped, specifically including: mapping the forward control to the backward control, mapping the backward control to the forward control, and mapping the left control to the right control, and mapping the right control to the left control.

[0160] Furthermore, in other optional embodiments, the automatic return function can be started or canceled by applying an APP. When the automatic return function is turned on, the automatic return can be performed without the user controlling the remote sensing. When the automatic return function is turned off, the user needs to control the return of the underwater robot by controlling the remote sensing or applying an APP or other control device.

[0161] Specifically, the control direction included in the navigation control instruction refers to the direction control performed by the user in the control remote sensing or application APP, including left control, right control, forward control or backward control.

[0162] In this embodiment, when the underwater robot returns, if the user controls forward, since the forward control and backward control have been reversely mapped in the above operation, the underwater robot returns directly without turning around, and the operation method is consistent with that when moving forward. This allows the underwater robot to return quickly even if the environment in which it is located is narrow when navigating in a pipeline.

[0163] In this embodiment, when the underwater robot is returning, if the user controls to the left, since the left and right controls have been reversely mapped in the above operation, the user does not need to consider the difference between the left and right correspondence when the underwater robot is moving backward and when it is moving forward, and can directly operate the left and right control of the underwater robot when returning (i.e., moving backward) in the same way as when the underwater robot is moving forward, which greatly improves the convenience of underwater robot control and the efficiency of the underwater robot returning. At the same time, it avoids misoperation and improves the safety of the underwater robot when returning in the pipeline.

[0164] Furthermore, after switching the images displayed by the front camera and the rear camera in the application APP, the user can intuitively see the return content while performing extremely simple operations. In this process, the user does not need to consider the correspondence between forward and reverse. The user can operate the return (i.e., backward) of the underwater robot in the same way as operating forward, which greatly improves the convenience of controlling the underwater robot and improves the efficiency of the return of the underwater robot.

[0165] Further, in an optional embodiment of the present invention, controlling the underwater robot to return in the target channel includes:

[0166] Acquire the traveled route and power change information sequence of the underwater robot, wherein the power change information sequence includes the power output direction and strength of the power equipment cluster of the underwater robot that changes over time, and the remote control instructions received by the underwater robot in each historical time period;

[0167] The underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence.

[0168] In this embodiment, when the control direction is forward control, the underwater robot can be controlled to automatically return to the destination in the target pipeline according to the traveled route and the power change information sequence. Therefore, during the return travel, the efficiency can be further improved.

[0169] Among them, the route traveled by the underwater robot during forward navigation refers to the route traveled by the underwater robot in the current environment before receiving the current remote control command, for example, the route traveled by the underwater robot in the 3D space of the target pipeline, specifically including the up, down, left, right, front and back information of the underwater robot during navigation.

[0170] Specifically, in the embodiment of the present invention, when the underwater robot is sailing forward, the route traveled by the underwater robot and the power change information sequence can be obtained and saved by image analysis, gyroscope acquisition, etc. When a designated return instruction is received, the saved route traveled can be directly obtained.

[0171] In the embodiment of the present invention, the remote control instructions received by the underwater robot in each historical time period are used to determine the flight state of the underwater robot. Specifically, during the forward navigation process of the underwater robot, the underwater robot can cruise automatically or be controlled by an operator. The remote control instructions received by the underwater robot in each historical time period are used to instruct the underwater robot to navigate in a certain navigation type, and specifically, the navigation type includes but is not limited to driving forward close to the pipe wall, driving at a constant speed, and accelerating.

[0172] In this embodiment, the underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence. The underwater robot can return automatically without the need for operator operation, thereby improving the return efficiency.

[0173] Further, in an optional embodiment of the invention, the controlling the underwater robot to return in the target pipeline according to the traveled route and the power change information sequence includes:

[0174] Using a curve posture recognition network in a pre-trained complex road section analysis model to extract features of the traveled route, to obtain a curve feature set;

[0175] Using the power change feature recognition network in the complex road section analysis model, feature extraction is performed on the power change information sequence to obtain a power change feature set;

[0176] Performing a corresponding moment feature splicing operation on the curve feature set and the dynamic change feature set to obtain a fused feature set, and performing a context attention enhancement operation on the fused feature set to obtain an enhanced feature set;

[0177] Scoring the enhanced feature set using the route complexity analysis network in the complex road segment analysis model to obtain a route complexity score sequence;

[0178] According to a preset complexity score threshold and the route complexity score sequence, the traveled route is segmented and output to obtain complex sections in the traveled route and non-complex sections in the traveled route;

[0179] When the underwater robot returns to the complex road section, the underwater robot is controlled to return along the original route according to the driving route corresponding to the complex road section;

[0180] When the underwater robot navigates to the non-complex section, the shortest distance analysis is performed on the traveled route corresponding to the non-complex section to obtain the shortest return route, and the underwater robot is controlled to return according to the shortest return route.

[0181] In this embodiment, the complex section analysis model is obtained by pre-training. The complex route analysis model includes networks with different purposes, for example, a route complexity analysis network, a power change feature recognition network and a route complexity analysis network.

[0182] In this embodiment, a curve feature set and a dynamic change feature set are first obtained by feature extraction, and then the features of the curve feature set and the dynamic change feature set at the same time are spliced ​​to realize the construction of a fused feature set, thereby improving the relationship and correlation between different features, and then a contextual attention enhancement operation is performed through an attention enhancement algorithm to improve the before-and-after relationship and correlation of different features, thereby obtaining an enhanced feature set, and finally a route complexity analysis network (i.e., a neural network) is used to perform a fully connected classification judgment to score the enhanced feature set, and a route complexity score sequence of each section is obtained (specifically, each time in a different section may correspond to a different score), and the scores are summarized and output to obtain complex sections and non-complex sections.

[0183] Specifically, the complex road section analysis model can be used to analyze the position coordinates and power change information sequence of the traveled route at each time node. By analyzing the changes in the position coordinates and power change information sequence between adjacent time nodes, the curve characteristics and power change characteristics between adjacent time nodes can be identified. Then, according to the curve characteristics and power change characteristics, the road section complexity score is scored to determine the complexity of the road section between the adjacent time nodes. After the scores in each adjacent time period in the traveled route are calculated, the route complexity score sequence can be obtained, and then segmented according to a complexity judgment standard (such as a threshold score) to obtain complex sections and non-complex sections.

[0184] Specifically, after the scores of each adjacent time period in the traveled route are calculated, the scores of each adjacent time period can be arranged according to the time sequence to obtain a route complexity score sequence. Then, the complexity judgment standard (such as the threshold score) is compared with each score in the route complexity score sequence, and the time period with a score greater than the complexity judgment standard is extracted to obtain a complex time period. According to the time correspondence, the section of the traveled route corresponding to the complex time period is determined as a complex section, and the remaining sections of the traveled route are determined as non-complex sections.

[0185] In this embodiment, a complex section refers to a section with a greater number of bends than a preset number or a continuous bend angle greater than a preset angle. Alternatively, a complex section refers to a section with a greater number of obstacles in the target pipeline than a preset number (e.g., 2 obstacles). Alternatively, a complex route refers to a section with a less than preset number of obstacles in the target pipeline, and a size of the obstacle greater than a preset volume (e.g., one-third of the volume of the underwater robot), or a position of the obstacle is a target position (e.g., directly in front of the target pipeline).

[0186] Accordingly, a complex section refers to a section where the number of bends is not greater than a preset number or the continuous bend angle is less than or equal to a preset angle. Alternatively, a complex section refers to a section where the number of obstacles in the target pipeline is less than a preset number (for example, 2 obstacles). Alternatively, a complex route refers to a section where the number of obstacles in the target pipeline is less than a preset number, and the size of the obstacle is not greater than a preset volume (such as one-third of the volume of the underwater robot), or the location of the obstacle is not the target location (such as directly in front of the target pipeline).

[0187] Specifically, when a complex section or a non-complex section is obtained, the underwater robot is controlled by the power output by the power equipment cluster of the underwater robot to return to the original route or return to a non-original route (i.e., return quickly). Specifically, the power equipment cluster of the underwater robot includes multiple power devices, and each power device operates in a coordinated relationship, and each power device can output power in multiple directions.

[0188] In specific implementation, when returning to the original route, the power matrix for maintaining the stability of the underwater robot and the forward power matrix for each complex section can be obtained according to the power change information sequence. Then the forward power matrix is ​​transposed to obtain the modified power matrix. And the modified power matrix is ​​used to control the direction distribution and power output of each power device in the power equipment cluster to achieve the original route return.

[0189] Specifically, in the embodiment of the present invention, the shortest distance analysis is performed on the traveled route corresponding to the non-complex road section, and the shortest return route is obtained, which means that when the tunnel of the target pipeline is very wide, very long, and has multiple directions, it is not necessary to control the underwater robot to navigate according to the original forward navigation position (for example, when the underwater robot is navigating according to the original forward navigation, it will move up, down, left, right, etc. at a certain position point), and when returning in these non-complex sections, it is sufficient to navigate within the limit where the underwater robot will not collide, and there is no need to move in different directions at a certain position point. For example, the underwater robot can turn close to the tunnel wall when returning, so as to reduce the return route and improve the return efficiency; in addition, when sailing to the surface after exiting the tunnel of the target pipeline, it can directly sail in a straight line to the recovery point (i.e., the horizontal plane when the underwater robot dives into the water), so as to achieve a rapid return after leaving the target pipeline.

[0190] Specifically, the historical travel coordinates of the traveled route corresponding to the non-complex road section may be obtained, various expected return routes may be calculated based on the historical travel coordinates, and the expected return route with the shortest distance may be determined as the shortest return route.

[0191] Alternatively, the shortest distance can be calculated based on the starting point and the end point of the non-complex road section to obtain the shortest return route, thereby controlling the power equipment cluster to return.

[0192] In specific implementation, when the underwater robot navigates to a non-complex section, since the return route of the underwater robot is different from the original return route, the power matrix needed to keep the robot stable can be calculated in real time, and then an additional power matrix in the opposite direction of the most recent route can be applied to realize the movement of the underwater robot.

[0193] In this embodiment, the return efficiency of the underwater robot can be improved by classifying the traveled route into complex sections and non-complex sections and then selecting different return modes.

[0194] Furthermore, in an optional embodiment of the present invention, before extracting features of the traveled route using the curve posture recognition network in the complex road section analysis model, the method further includes:

[0195] Obtaining a pre-constructed complex road section analysis model and a pre-constructed underwater navigation experiment sample set, and allocating the underwater navigation experiment sample set into a training set and a test set according to a preset ratio;

[0196] Iteratively training the complex road section analysis model using the training set to obtain an updated complex road section analysis model, and obtaining the recognition accuracy of the updated complex road section analysis model according to the test set;

[0197] Determining whether the recognition accuracy is greater than a preset qualified threshold;

[0198] When the recognition accuracy is greater than the qualified threshold, the iterative training process is stopped to obtain a pre-trained complex road section analysis model.

[0199] In this embodiment, the pre-constructed complex road section analysis model may be constructed based on a binary classification model.

[0200] In an embodiment of the present invention, an underwater navigation experiment sample set can be obtained by testing various actual underwater pipeline environments, and the underwater navigation experiment sample set can be grouped into a training set and a test set according to a preset ratio (such as 7:3).

[0201] Specifically, the present invention can use the training set to iteratively train the complex road section analysis model according to the cross entropy loss algorithm and the gradient descent method. And the training progress of the model is controlled by the recognition accuracy. When the recognition accuracy of the complex road section analysis model is greater than the preset qualified threshold, such as 95%, it indicates that the model training effect is good, and the iterative process is stopped to obtain the pre-trained complex road section analysis model.

[0202] In the embodiment of the present application, by receiving the remote control command sent by the control device of the underwater robot, it is judged whether the remote control command is a designated return command; receiving the remote control command sent by the pre-built underwater robot remote controller; judging whether the remote control command is a pre-built return command; when the remote control command is the designated return command, the pre-built camera of the underwater robot is switched to the front and back reverse perspective, and the pre-built robot navigation control strategy is reversely mapped to obtain a reverse control strategy; according to the remote control command and the reverse control strategy, the power equipment cluster in the pre-built target channel of the underwater robot is controlled and returned by power. Since the camera of the underwater robot is reversely switched to the reverse perspective, and the pre-built navigation control strategy is reversely mapped, the user does not need to consider the difference between the front and back correspondence and the left and right correspondence when the underwater robot returns and when it advances, and directly operates the front and back left and right control of the underwater robot when returning (i.e. retreating) in the same way as when the underwater robot advances, which greatly improves the convenience of underwater robot control, improves the efficiency of underwater robot return, and achieves the purpose of efficient and fast return control of the underwater robot. At the same time, it reduces the risk of collision caused by user misoperation and improves the safety of the underwater robot when returning.

[0203] See also Figure 3 , a schematic diagram of the hardware structure of an underwater robot provided in one embodiment of the present application.

[0204] The underwater robot 10 may include a control circuit, which may include a storage and processing circuit 30. The storage and processing circuit 30 may include a memory, such as a hard disk drive memory, a non-volatile memory (such as a flash memory or other electronically programmable memory with limited deletion for forming a solid-state drive, etc.), a volatile memory (such as a static or dynamic random access memory, etc.), etc., which is not limited by the embodiment of the present application. The processing circuit in the storage and processing circuit 30 can be used to control the operation of the underwater robot 10. The processing circuit can be implemented based on one or more microprocessors, microcontrollers, digital signal processors, baseband processors, power management units, audio codec chips, application-specific integrated circuits, display driver integrated circuits, etc.

[0205] The storage and processing circuit 30 can be used to run software in the underwater robot 10, such as Voice over Internet Protocol (VOIP) phone call applications, operating system functions, etc. These software can be used to perform some control operations, such as image acquisition based on a camera, ambient light measurement based on an ambient light sensor, proximity sensor measurement based on a proximity sensor, information display functions based on status indicators such as status indicators of light-emitting diodes, touch event detection based on a touch sensor, functions associated with displaying information on multiple (e.g., layered) displays, operations associated with performing wireless communication functions, operations associated with collecting and generating audio signals, control operations associated with collecting and processing button press event data, and other functions in the underwater robot 10, etc., which are not limited by the embodiments of the present application.

[0206] Furthermore, the memory stores an executable program code, and the processor coupled to the memory calls the executable program code stored in the memory to execute the above Figure 1 The return control method of the underwater robot described in the illustrated embodiment.

[0207] The executable program code includes the above Figure 2 The various modules in the return control device of the underwater robot described in the illustrated embodiment, for example: a receiving module, a switching module and a return module.

[0208] The underwater robot 10 may further include an input-output circuit 42. The input-output circuit 42 may be used to enable the underwater robot 10 to input and output data, that is, to allow the underwater robot 10 to receive data from an external device and also to allow the underwater robot 10 to output data from the underwater robot 10 to an external device. The input-output circuit 42 may further include a sensor 32. The sensor 32 may include an ambient light sensor, a proximity sensor based on light and capacitance, a touch sensor (e.g., based on a light touch sensor and / or a capacitive touch sensor, wherein the touch sensor may be part of a touch display screen or may be used independently as a touch sensor structure), an acceleration sensor, and other sensors.

[0209] The input-output circuit 42 may also include one or more displays, such as the display 14. The display 14 may include a liquid crystal display, an organic light-emitting diode display, an electronic ink display, a plasma display, a display using other display technologies, or a combination of several of them. The display 14 may include a touch sensor array (i.e., the display 14 may be a touch-controlled display screen). The touch sensor may be a capacitive touch sensor formed by an array of transparent touch sensor electrodes (e.g., indium tin oxide (ITO) electrodes), or may be a touch sensor formed using other touch technologies, such as sonic touch, pressure-sensitive touch, resistive touch, optical touch, etc., which is not limited in the embodiments of the present application.

[0210] The underwater robot 10 may also include an audio component 36. The audio component 36 may be used to provide audio input and output functions for the underwater robot 10. The audio component 36 in the underwater robot 10 may include a speaker, a microphone, a buzzer, a tone generator, and other components for generating and detecting sound.

[0211] The communication circuit 38 can be used to provide the underwater robot 10 with the ability to communicate with external devices. The communication circuit 38 may include analog and digital input / output interface circuits, and wireless communication circuits based on radio frequency signals and / or optical signals. The wireless communication circuit in the communication circuit 38 may include a radio frequency transceiver circuit, a power amplifier circuit, a low noise amplifier, a switch, a filter, and an antenna. For example, the wireless communication circuit in the communication circuit 38 may include a circuit for supporting near field communication (NFC) by transmitting and receiving near field coupled electromagnetic signals. For example, the communication circuit 38 may include a near field communication antenna and a near field communication transceiver. The communication circuit 38 may also include a cellular phone transceiver and antenna, a wireless local area network transceiver circuit and antenna, etc.

[0212] The underwater robot 10 may further include a battery, a power management circuit and other input / output units 40. The input / output units 40 may include buttons, a touch pad, a camera, a light emitting diode and other status indicators.

[0213] The user can input commands through the input-output circuit 42 to control the operation of the underwater robot 10 , and can use the output data of the input-output circuit 42 to receive status information and other outputs from the underwater robot 10 .

[0214] Furthermore, an embodiment of the present invention further provides a computer-readable storage medium, which may be provided in the underwater robot in each of the above embodiments, and the computer-readable storage medium may be the memory in the storage and processing circuit 30 in the above embodiments. The computer-readable storage medium stores a computer program, which, when executed by the processor, implements the above Figure 1 The return control method of the underwater robot described in the illustrated embodiment. Further, the computer storage medium can also be a U disk, a mobile hard disk, a read-only memory (ROM), a RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0215] It should be noted that, for the convenience of description, the aforementioned method embodiments are all described as a series of action combinations, but those skilled in the art should be aware that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0216] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0217] The above is a description of the return control method, device, underwater robot and computer-readable storage medium of the underwater robot provided by the present invention. For a person skilled in the art, according to the ideas of the embodiments of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for controlling the return of an underwater robot, characterized in that: The method comprises: Receiving a remote control command sent by a control device of the underwater robot, and determining whether the remote control command is a designated return command; When the remote control instruction is the designated return instruction, the camera of the underwater robot is switched to a reverse perspective, and a reverse mapping operation is performed on the pre-built navigation control strategy to obtain a reverse control strategy; According to the remote control instruction and the reverse control strategy, the underwater robot is controlled to return in the target channel.

2. The method according to claim 1, characterized in that Before determining whether the remote control instruction is a designated return instruction, the method further includes: Generating verification information according to the remote control instruction, and sending the verification information to the remote control of the underwater robot; Receiving a verification result fed back by a remote controller of the underwater robot; When the verification result is that the verification is passed, the operation of determining whether the remote control instruction is a designated return instruction is performed.

3. The method according to claim 1, characterized in that The camera includes a front camera and a rear camera, and the reverse viewing angle switching of the camera of the underwater robot includes: In the interface of the application APP that interacts with the underwater robot, the front image captured by the front camera and the rear image captured by the rear camera are displayed interchangeably.

4. The method according to claim 1, characterized in that: The pre-built navigation control strategy includes forward control, backward control, left control and right control; the reverse mapping operation includes reverse mapping between forward control and backward control, and reverse mapping between left control and right control.

5. The method according to claim 4, characterized in that The step of controlling the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy includes: Identifying a control direction included in the remote control command; If the control direction is forward control, the underwater robot is controlled to move backward according to the reverse control strategy; if the control direction is backward control, the underwater robot is controlled to move forward; and If the control direction is left control, the underwater robot is controlled to move to the right; if the control direction is right control, the underwater robot is controlled to move to the left.

6. The method according to claim 1, characterized in that The controlling the underwater robot to return in the target channel comprises: Acquire the traveled route and power change information sequence of the underwater robot, wherein the power change information sequence includes the power output direction and strength of the power equipment cluster of the underwater robot that changes over time, and the remote control instructions received by the underwater robot in each historical time period; The underwater robot is controlled to return in the target channel according to the traveled route and the power change information sequence.

7. The method according to claim 6, characterized in that The method of controlling the underwater robot to return in the target channel according to the traveled route and the power change information sequence includes: Using a curve posture recognition network in a pre-trained complex road section analysis model to extract features of the traveled route, to obtain a curve feature set; Using the power change feature recognition network in the complex road section analysis model, feature extraction is performed on the power change information sequence to obtain a power change feature set; Performing a corresponding moment feature splicing operation on the curve feature set and the dynamic change feature set to obtain a fused feature set, and performing a context attention enhancement operation on the fused feature set to obtain an enhanced feature set; Scoring the enhanced feature set using the route complexity analysis network in the complex road segment analysis model to obtain a route complexity score sequence; According to a preset complexity score threshold and the route complexity score sequence, the traveled route is segmented and output to obtain complex sections in the traveled route and non-complex sections in the traveled route; When the underwater robot returns to the complex road section, the underwater robot is controlled to return along the original route according to the driving route corresponding to the complex road section; When the underwater robot navigates to the non-complex section, the shortest distance analysis is performed on the traveled route corresponding to the non-complex section to obtain the shortest return route, and the underwater robot is controlled to return according to the shortest return route.

8. A return control device for an underwater robot, characterized in that: The device comprises: A receiving module, used to receive a remote control instruction sent by a control device of the underwater robot, and determine whether the remote control instruction is a designated return instruction; A switching module, used for, when the remote control instruction is the designated return instruction, switching the camera of the underwater robot to a reverse perspective, and performing a reverse mapping operation on the pre-built navigation control strategy to obtain a reverse control strategy; The return module is used to control the underwater robot to return in the target channel according to the remote control instruction and the reverse control strategy.

9. An underwater robot, characterized in that: The underwater robot comprises: Memory and processor; The memory stores executable program code; The processor coupled to the memory calls the executable program code stored in the memory to execute the return control method of the underwater robot according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the return control method of the underwater robot according to any one of claims 1 to 7 is implemented.

Images