A control optimization method for bionic underwater robot

Abstract

The present invention relates to a control optimization method for a bionic underwater robot, and relates to the field of robot control, including: collecting neighborhood space image information; performing species recognition on the neighborhood space image information to obtain a species recognition set, and when there are sensitive species, collecting underwater temperature information and underwater pressure information; optimizing the escape speed of sensitive species according to the underwater temperature information and underwater pressure information to obtain an escape speed threshold; taking the direction away from the sensitive species as the escape direction, initializing the motion controller at a speed greater than or equal to the escape speed threshold, and controlling the bionic underwater robot to perform interference avoidance movement. By identifying sensitive species and optimizing the escape speed threshold according to underwater environmental parameters, the bionic underwater robot is controlled to avoid in the direction away from the sensitive species, thereby achieving the technical effect of improving the task execution efficiency and adaptability of the bionic underwater robot while protecting the underwater sensitive species and the ecological environment.

Description

Technical Field

[0001] The invention relates to the field of robot control, and in particular to a control optimization method for a bionic underwater robot. Background Art

[0002] With the continuous development of science and technology, bionic underwater robots have been widely used in the fields of ocean exploration and underwater operations. However, existing bionic underwater robots often encounter various sensitive species, such as rare fish and coral reefs, when performing underwater tasks. These sensitive species are easily disturbed by bionic underwater robots, which not only affects the normal life of sensitive species, but also may have a negative impact on the entire underwater ecological environment. At the same time, frequent interference will also affect the task execution efficiency of bionic underwater robots and reduce their practicality. At present, some studies have proposed to reduce interference with sensitive species by presetting obstacle avoidance paths, but this method lacks real-time and flexibility and is difficult to cope with complex and changeable underwater environments; there are also studies that try to improve the environmental perception ability of bionic underwater robots by increasing the number and types of sensors, but this often leads to a significant increase in the cost and energy consumption of the robot, and its practicality is limited. Summary of the invention

[0003] The present invention aims to solve the technical problem in the prior art that when a bionic underwater robot performs underwater tasks, it is easy to cause interference when encountering sensitive species, which affects the efficiency of task execution and has a negative impact on the underwater ecological environment. A control optimization method for a bionic underwater robot is provided to solve the problem.

[0004] The technical solution of the present invention to solve the above technical problems is as follows:

[0005] In a first aspect, the present invention provides a control optimization method for a bionic underwater robot, comprising: collecting neighborhood space image information through a visual sensor; performing species identification on the neighborhood space image information to obtain a species identification set, and when the species identification set has sensitive species, collecting underwater temperature information and underwater pressure information through an underwater environment sensor; optimizing the escape speed of the sensitive species based on the underwater temperature information and the underwater pressure information to obtain an escape speed threshold; taking the direction away from the sensitive species as the escape direction, initializing the motion controller at a speed greater than or equal to the escape speed threshold, and controlling the bionic underwater robot to perform interference avoidance movement.

[0006] In a second aspect, the present application provides an electronic device, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is used to execute a bionic underwater robot control optimization method provided in the present application.

[0007] In a third aspect, the present application provides a computer-readable storage medium storing a computer program, and the computer program is used to execute a control optimization method for a bionic underwater robot provided by the present application.

[0008] The beneficial effects of the present invention are:

[0009] Through visual sensors, the image information of the neighborhood space is collected, and the image data of the environment around the bionic underwater robot is obtained to provide input for subsequent species identification; the species identification is performed on the neighborhood space image information to obtain a species identification set. When sensitive species exist in the species identification set, the underwater temperature information and underwater pressure information are collected through underwater environmental sensors. By judging whether there are sensitive species around that need special protection, and obtaining the current underwater temperature and pressure data through environmental sensors when there are sensitive species, a basis is provided for subsequent escape speed optimization; based on the obtained underwater temperature information and underwater pressure information, the escape speed of sensitive species is optimized to obtain The escape speed threshold is reached, which can effectively avoid interference with sensitive species and ensure the robot's movement efficiency; the direction away from sensitive species is taken as the escape direction, the motion controller is initialized at a speed greater than or equal to the escape speed threshold, and the bionic underwater robot is controlled to perform interference avoidance movement. The escape speed threshold and avoidance direction obtained by optimization are set to the motion controller, so that the bionic underwater robot can autonomously stay away from sensitive species, and the bionic underwater robot can minimize interference with underwater sensitive species and the ecological environment while performing tasks, thereby improving the robot's environmental adaptability and work efficiency, and balancing the needs of ecological protection and task completion. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of a flow chart of a control optimization method for a bionic underwater robot provided by the present invention;

[0011] Figure 2 A schematic diagram of the structure of an electronic device provided by the present invention;

[0012] Figure 3 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0013] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0014] Electronic device 100 , memory 110 , processor 120 , first computer program 111 , computer-readable storage medium 200 , second computer program 211 . DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0016] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0017] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention. Embodiment 1

[0018] like Figure 1 As shown, an embodiment of the present invention provides a control optimization method for a bionic underwater robot, which is applied to a bionic underwater robot. The bionic underwater robot includes a visual sensor, an underwater environment sensor, a central control unit and a motion controller.

[0019] Specifically, a disclosed control optimization method for a bionic underwater robot is applied to a bionic underwater robot, which includes several components, namely a visual sensor, an underwater environment sensor, a central control unit and a motion controller. Among them, the visual sensor is used to collect image information of the neighborhood space around the bionic underwater robot, providing necessary data for subsequent species identification. The underwater environment sensor is used to collect information such as the temperature and pressure of the underwater environment in which the bionic underwater robot is located, providing a reference basis for optimizing the escape speed when avoiding sensitive species. The central control unit is the core processing and decision-making unit of the bionic underwater robot, responsible for coordinating the work of each component and executing each step in a control optimization method for a bionic underwater robot. The motion controller receives the escape direction and escape speed threshold from the central control unit, and controls the actual movement of the bionic underwater robot accordingly.

[0020] Through the mutual cooperation of visual sensors, underwater environment sensors, central control units and motion controllers, the bionic underwater robot can effectively execute a control optimization method for bionic underwater robots, perform intelligent avoidance when encountering sensitive underwater species, and minimize interference with the underwater ecology while improving its own work efficiency.

[0021] The steps executed by the central control unit include:

[0022] S100: Collecting neighborhood space image information through visual sensors.

[0023] Specifically, the bionic underwater robot collects image information of the neighborhood space through its equipped visual sensors. Among them, the visual sensors include but are not limited to visible light cameras, infrared cameras, sonar imagers and other devices that can obtain image information of the neighborhood space. The type and parameters of the visual sensor are selected and configured according to the specific application scenarios and requirements of the bionic underwater robot.

[0024] The neighborhood space image information collected by the visual sensor is the visual image data within a certain range around the bionic underwater robot, and its content includes but is not limited to underwater terrain, underwater objects, underwater creatures, etc. These neighborhood space image information will serve as the data basis for subsequent species identification.

[0025] At the same time, in order to obtain clear and comprehensive neighborhood space image information, the layout position and acquisition frequency of the visual sensor and other parameters need to be reasonably set according to the specific application. At the same time, the collected neighborhood space image information also needs to undergo necessary preprocessing, such as denoising and enhancement, to improve the accuracy of subsequent species identification.

[0026] Through the collection of visual sensors, the bionic underwater robot obtains image information of the surrounding space, providing the necessary data basis for the control optimization of the bionic underwater robot.

[0027] S200: performing species identification on the neighborhood spatial image information to obtain a species identification set, and when the species identification set contains sensitive species, collecting underwater temperature information and underwater pressure information through an underwater environment sensor.

[0028] Specifically, after obtaining the neighborhood space image information, the neighborhood space image information is processed for species identification to obtain a species identification set. The species identification process uses an image recognition algorithm based on deep learning. By training a large amount of image data of labeled species, a neural network model that can map image features to species categories is established. When the neighborhood space image information to be identified is input, the model can automatically extract the key features in the image and compare them with the features of known species, thereby realizing the identification and classification of species in the image. The result of species identification is a species list, namely the species identification set, which contains all species identified in the neighborhood space image information and their corresponding confidence levels. The confidence level indicates the credibility of the identification result, and the higher the value, the more accurate the identification.

[0029] After obtaining the species identification set, it is further determined whether there are sensitive species in the species identification set. The so-called sensitive species refers to underwater organisms that are highly sensitive to the presence and activities of the bionic underwater robot and are easily disturbed or harmed. Among them, the determination of sensitive species is achieved by comparing with a pre-set sensitive species library. If there are sensitive species in the species identification set, the bionic underwater robot collects the temperature and pressure information of the current underwater environment through its equipped underwater environment sensor, providing the necessary parameter basis for the subsequent escape speed optimization.

[0030] By identifying species in the neighborhood spatial image information, it is determined whether there are sensitive species, and relevant underwater environmental parameters are collected when necessary, providing a decision-making basis for bionic underwater robots to avoid sensitive species.

[0031] S300: Optimizing the escape speed of the sensitive species according to the underwater temperature information and the underwater pressure information to obtain an escape speed threshold.

[0032] Specifically, after collecting the underwater temperature information and underwater pressure information, the escape speed of the identified sensitive species is optimized to obtain the final escape speed threshold. First, according to the current underwater temperature information and underwater pressure information, the historical escape data of sensitive species are retrieved in the pre-established sensitive species database. These historical data record the speed values ​​of the prey of each sensitive species successfully escaping from hunting under similar underwater environmental conditions. Secondly, for the historical escape speed data of each sensitive species, the method of removing outliers is used to remove possible outliers or noise data, and a relatively concentrated escape speed record range is obtained. On this basis, the maximum value of the range is extracted as the escape speed threshold of the sensitive species under the current underwater environmental conditions. After that, after obtaining the escape speed thresholds of all sensitive species, the one with the largest value is selected as the final escape speed threshold, which represents the minimum escape speed that the bionic underwater robot needs to achieve in order to effectively avoid all sensitive species under the current underwater environmental conditions. The above escape speed optimization process fully considers the differences in escape ability of different sensitive species under different underwater environmental conditions. Through big data analysis and threshold selection, a highly applicable escape speed threshold is obtained, which provides an important reference for the subsequent motion control of the bionic underwater robot.

[0033] Driven by underwater environmental data, the optimal escape speed threshold for avoiding sensitive species is adaptively obtained, effectively balancing the mission execution efficiency of the bionic underwater robot and the needs of underwater biological protection.

[0034] S400: taking the direction away from the sensitive species as the escape direction, initializing the motion controller at a speed greater than or equal to the escape speed threshold, and controlling the bionic underwater robot to perform interference avoidance movement.

[0035] Specifically, after obtaining the escape speed threshold, the direction away from the sensitive species is used as the escape direction, and the motion controller is initialized at a speed greater than or equal to the escape speed threshold, thereby controlling the bionic underwater robot to perform interference avoidance motion.

[0036] First, the relative position of the sensitive species to itself is determined based on the species identification results and the neighborhood space image information. The opposite direction of the position is used as the escape direction, that is, the target direction of the interference avoidance movement. Secondly, the obtained escape speed threshold is set as the minimum movement speed of the motion controller. Subsequently, the bionic underwater robot performs interference avoidance movement along the specified escape direction and speed under the control of the motion controller according to the set escape direction and escape speed threshold, and stays away from the area where the sensitive species are located.

[0037] Through the setting of the motion controller, the bionic underwater robot has achieved the ability to autonomously avoid sensitive species when encountering them, thereby improving its mission execution efficiency and success rate while protecting the underwater ecology.

[0038] Furthermore, the embodiment of the present application also includes:

[0039] S410: obtaining the rated motion speed of the bionic underwater robot through the motion controller;

[0040] S420: when the escape speed threshold is greater than the rated speed of the bionic underwater robot, the direction away from the sensitive species is used as the escape direction, the non-inertial movement trajectory is started at the rated speed of the bionic underwater robot to initialize the motion controller, and the bionic underwater robot is controlled to perform interference avoidance movement;

[0041] S430: When the escape speed threshold is less than or equal to the rated speed of the bionic underwater robot, the motion controller is initialized at a speed greater than or equal to the escape speed threshold, with the direction away from the sensitive species as the escape direction, to control the bionic underwater robot to perform interference avoidance movement.

[0042] In a preferred embodiment, first, the motion rated speed of the robot itself, i.e., the bionic underwater robot motion rated speed, is obtained through the motion controller. The motion rated speed refers to the maximum safe motion speed of the bionic underwater robot under normal working conditions, which is the upper limit of its motion capacity. Then, the obtained escape speed threshold is compared with the obtained bionic underwater robot motion rated speed. If the escape speed threshold is greater than the bionic underwater robot motion rated speed, the speed parameter of the motion controller is set to the bionic underwater robot motion rated speed, and at the same time, the direction away from the sensitive species is used as the escape direction, and the non-inertial movement trajectory is started to initialize the motion controller, and the bionic underwater robot is controlled to perform interference avoidance movement. Among them, the non-inertial movement trajectory refers to an unpredictable motion trajectory that is significantly different from the conventional movement trajectory of the bionic underwater robot. Through the non-inertial movement trajectory, the unpredictability and adaptability of the evasive movement can be enhanced when the escape speed is insufficient, thereby improving the evasion effect. For example, if the bionic underwater robot usually moves in a straight line, the non-inertial movement trajectory can be a curve, a Z-shape, a spiral, etc.; if the bionic underwater robot usually moves in a plane, the non-inertial movement trajectory can be a three-dimensional, multi-level space curve, etc. The non-inertial movement trajectory can be generated by a random function or selected from a preset non-inertial trajectory library. If the escape speed threshold is less than or equal to the rated speed of the bionic underwater robot, the speed parameter of the motion controller is set to a value greater than or equal to the escape speed threshold, and the motion controller is initialized with the direction away from the sensitive species as the escape direction to control the bionic underwater robot to avoid interference. Among them, when the escape speed threshold exceeds the rated speed of the bionic underwater robot, in order to ensure the movement safety of the bionic underwater robot, the rated speed of the bionic underwater robot can only be used as the maximum escape speed, but at the same time, the non-inertial movement trajectory is started to enhance the avoidance effect; when the escape speed threshold does not exceed the rated speed of the movement, it can be directly used as the escape speed without starting the non-inertial movement trajectory.

[0043] The motion controller initialization process is further optimized through the rated speed of the bionic underwater robot. While ensuring the avoidance effect, the bionic underwater robot's own motion safety is taken into account, improving practicality and reliability. At the same time, the startup mechanism of the non-inertial movement trajectory further enriches the bionic underwater robot's avoidance motion mode and enhances its adaptability to complex underwater environments.

[0044] Furthermore, the embodiment of the present application also includes:

[0045] S210: Obtain a sensitive species library, and extract matching species of the species identification set belonging to the sensitive species library;

[0046] S220: When the number of the matching species is equal to 0, continue to perform the underwater monitoring task;

[0047] S230: When the number of the matching species is not equal to 0, setting the matching species as the sensitive species;

[0048] The steps of constructing the sensitive species library include:

[0049] S211: Obtain bionic underwater robot bionic species;

[0050] S212: Adding underwater creatures that take the bionic underwater robot-bionic species as prey into the sensitive species library.

[0051] In a feasible implementation, after performing species identification on the neighborhood space image information and obtaining the species identification set, when determining whether there are sensitive species, first, obtain a pre-constructed sensitive species library. The sensitive species library stores a series of characteristic information of underwater organisms that are highly sensitive to the existence and activities of the bionic underwater robot. Then, the bionic underwater robot matches each species in the species identification set with the species in the sensitive species library, and extracts the matching species belonging to the sensitive species library. Subsequently, the number of matching species is determined. If the number is equal to 0, that is, the species identification set does not contain any sensitive species, the bionic underwater robot can continue to perform the original underwater monitoring task without starting the avoidance mechanism. If the number of matching species is not equal to 0, that is, the species identification set contains at least one sensitive species, the bionic underwater robot marks all matching species as sensitive species and starts the subsequent avoidance mechanism, including collecting underwater temperature and pressure information, calculating the escape speed threshold, etc.

[0052] To build a sensitive species library, first, we need to obtain the biological species that the bionic underwater robot imitates, that is, its biomimetic species. Among them, the biomimetic species are selected according to the design goals and application scenarios of the biomimetic underwater robot, representing the ecological role it plays in the underwater environment. Then, based on existing biological knowledge and practical experience, underwater organisms that take the biomimetic species of the biomimetic underwater robot as prey are identified as potential sensitive species and added to the sensitive species library. This is because, in natural ecosystems, predators are usually highly sensitive to the presence and activities of their prey.

[0053] By determining sensitive species and building a sensitive species library, bionic underwater robots can more accurately and efficiently identify sensitive species in underwater environments, providing a reliable basis for subsequent avoidance decisions.

[0054] Furthermore, the embodiment of the present application also includes:

[0055] S310: Retrieving first sensitive species abandonment hunting record data of the sensitive species using the underwater temperature information and the underwater pressure information, wherein the first sensitive species abandonment hunting record data includes a first prey escape speed record value;

[0056] S320: performing outlier analysis on the first prey escape speed record value to obtain a first prey escape speed concentrated record value, extracting a maximum value of the first prey escape speed concentrated record value, and setting the maximum value as a first sensitive species escape speed threshold;

[0057] S330: until the Nth sensitive species abandoning hunting record data of the sensitive species is retrieved by using the underwater temperature information and the underwater pressure information, wherein the Nth sensitive species abandoning hunting record data includes the Nth prey escape speed record value;

[0058] S340: performing outlier analysis on the Nth prey escape speed record value to obtain the Nth prey escape speed concentrated record value, extracting the maximum value of the Nth prey escape speed concentrated record value, and setting it as the Nth sensitive species escape speed threshold;

[0059] S350: Extract the maximum value of the escape speed threshold of the first sensitive species to the escape speed threshold of the Nth sensitive species, and set it as the escape speed threshold.

[0060] In a preferred embodiment, each sensitive species of sensitive species is traversed, and one sensitive species is obtained each time as the first sensitive species. At the same time, the current underwater temperature information and underwater pressure information are obtained, combined with the first sensitive species, as the search condition, and the pre-constructed sensitive species database is accessed. Similar historical records are searched in the database to obtain the record data of the first sensitive species in the past hunting activities, which was forced to give up hunting due to the prey escaping too fast, that is, the first sensitive species gave up hunting record data, which includes the escape speed value of the prey at that time, that is, the first prey escape speed record value. By obtaining the escape speed data of the prey when the first sensitive species gave up hunting under similar underwater environmental conditions, a data basis is provided for subsequent threshold calculation. After obtaining the first prey escape speed record value, these data are subjected to outlier analysis. The so-called outlier refers to a data point in the data set that is significantly different from other data points and deviates from the normal range. The existence of outliers is usually caused by measurement errors, accidental factors, etc., which will interfere with subsequent statistical analysis. Outlier analysis is the process of identifying and eliminating these abnormal data points. Through this analysis, a relatively concentrated and stable range of first prey escape speed records was obtained, namely, the first prey escape speed concentrated record values. Then, the maximum value was selected from the first prey escape speed concentrated record values ​​and set as the first sensitive species escape speed threshold under the current underwater environmental conditions, indicating that when the escape speed of the first sensitive species' prey reaches or exceeds this value, the first sensitive species is likely to give up hunting, reflecting the upper limit of the first sensitive species' hunting ability.

[0061] Repeat the above process, but change the search object from the first sensitive species to the Nth sensitive species. Wherein, N is a variable, representing the total number of all sensitive species identified. By accessing the sensitive species database in sequence, the hunting abandonment record data of each sensitive species is searched, that is, the hunting abandonment record data of the first sensitive species to the hunting abandonment record data of the Nth sensitive species, including the respective prey escape speed record values, that is, the first prey escape speed record value to the Nth prey escape speed record value. Then, the prey escape speed record values ​​of each sensitive species are respectively subjected to outlier analysis to obtain the respective prey escape speed concentrated record values, that is, the first prey escape speed concentrated record value to the Nth prey escape speed concentrated record value, and the maximum value is selected from each prey escape speed concentrated record value, and is set as the escape speed threshold of each sensitive species under the current underwater environmental conditions, that is, the escape speed threshold of the first sensitive species to the escape speed threshold of the Nth sensitive species, as the upper limit of the hunting ability of each sensitive species, that is, the minimum escape speed that the prey of each sensitive species needs to reach in order to effectively avoid hunting.

[0062] Afterwards, the previously calculated escape speed thresholds of the first sensitive species to the Nth sensitive species are compared, and the maximum value is selected and set as the final escape speed threshold, which represents the minimum escape speed that the bionic underwater robot needs to achieve under the current underwater environmental conditions to effectively avoid being hunted by all sensitive species. The reason for selecting the maximum value is that only when the escape speed of the bionic underwater robot exceeds the upper limit of the hunting ability of all sensitive species can it be guaranteed to be foolproof. If a smaller threshold is selected, although some sensitive species can be avoided, it may still be captured by some sensitive species with stronger hunting abilities.

[0063] By first calculating the escape speed threshold of each sensitive species separately and then comprehensively comparing to obtain the final threshold, the differences in hunting capabilities of different sensitive species are fully taken into account, and a reliable escape speed threshold is obtained, which effectively improves the success rate of bionic underwater robots in avoiding sensitive species.

[0064] Furthermore, the embodiment of the present application also includes:

[0065] S421: configuring a conventional moving trajectory according to the bionic underwater robot biomimetic species;

[0066] S422: randomly generate a first moving trajectory;

[0067] S423: when the trajectory similarity between the regular movement trajectory and the first movement trajectory is less than or equal to a trajectory similarity threshold, setting the first movement trajectory as the non-inertial movement trajectory;

[0068] S424: Otherwise, update the first moving trajectory.

[0069] In a preferred embodiment, when configuring the non-inertial movement trajectory, first, according to the characteristics of the biomimetic species of the bionic underwater robot, a conventional movement trajectory is configured. Among them, the biomimetic species refers to the specific underwater organisms imitated by the biomimetic underwater robot during design, such as fish, shrimps, turtles, etc. Different biomimetic species usually have different typical motion modes in underwater environments, such as linear motion, curved motion, spiral motion, etc. The conventional movement trajectory is a reference trajectory set according to the typical motion mode of the biomimetic species, which reflects the conventional movement mode of the biomimetic underwater robot in a non-avoidance state, and provides a reference for the subsequent non-inertial trajectory generation.

[0070] Then, a new moving trajectory is randomly generated, which is called the first moving trajectory. Different from the conventional moving trajectory, the first moving trajectory is a randomly generated, uncertain trajectory, and its shape and direction have certain randomness and unpredictability. By introducing randomness, the first moving trajectory may present characteristics that are significantly different from the conventional moving trajectory, such as sudden turns, rapid acceleration and deceleration, etc. These characteristics help to enhance the unpredictability of the avoidance behavior of the bionic underwater robot and improve the avoidance effect. Subsequently, it is evaluated whether the non-inertial characteristics of the first moving trajectory are obvious enough. Specifically, the trajectory similarity between the first moving trajectory and the conventional moving trajectory is calculated, and it is compared with the preset trajectory similarity threshold. Among them, the trajectory similarity is an indicator that characterizes the similarity between the two trajectories in terms of shape, direction, etc. The smaller the value, the greater the difference between the two trajectories. If the trajectory similarity between the first moving trajectory and the conventional moving trajectory is less than or equal to the trajectory similarity threshold, it means that the first moving trajectory has sufficiently obvious non-inertial characteristics. At this time, it is set as the final non-inertial moving trajectory, thereby ensuring that the non-inertial moving trajectory is sufficiently different from the conventional moving trajectory, so that sensitive species can be effectively confused and avoided. If the trajectory similarity between the first moving trajectory and the conventional moving trajectory is greater than the trajectory similarity threshold, it means that the non-inertial feature of the first moving trajectory is not obvious enough. In this case, the first moving trajectory is not directly used, but updated and adjusted. The updating method can be to regenerate a completely different random trajectory, or to introduce some random disturbances on the basis of the first moving trajectory to change its shape and direction to a certain extent. Through one or more updates, a moving trajectory that meets the non-inertial requirements is finally obtained, which ensures the generation quality of the non-inertial moving trajectory and improves the effectiveness of the avoidance behavior.

[0071] By introducing the trajectory similarity as an evaluation indicator and through comparison with conventional movement trajectories and threshold judgment, the adaptive generation of non-inertial movement trajectories is achieved, which effectively enhances the task execution efficiency of bionic underwater robots in complex underwater environments.

[0072] Furthermore, the embodiment of the present application also includes:

[0073] S4231: deploying the conventional moving trajectory in a first space coordinate system to obtain a first trajectory curve;

[0074] S4232: deploying the first moving trajectory in the first space coordinate system to obtain a second trajectory curve;

[0075] S4233: Calculate the inverse of the area enclosed by the first trajectory curve and the second trajectory curve, and set it as the trajectory similarity.

[0076] In a preferred embodiment, first, a three-dimensional space coordinate system is established, referred to as the first space coordinate system. Then, the data of the conventional mobile trajectory is mapped into the coordinate system to obtain a three-dimensional curve representing the conventional mobile trajectory, referred to as the first trajectory curve. The establishment of the first space coordinate system enables the spatial position and shape characteristics of the conventional mobile trajectory to be accurately characterized in a mathematical way, laying the foundation for subsequent trajectory comparison. Subsequently, the same method is used to map the data of the first mobile trajectory into the first space coordinate system to obtain another three-dimensional curve, referred to as the second trajectory curve. The second trajectory curve characterizes the position and shape characteristics of the first mobile trajectory in space. By deploying the first mobile trajectory and the conventional mobile trajectory in the same coordinate system, the two trajectories are aligned in space, creating conditions for calculating the similarity between the two trajectories.

[0077] Subsequently, based on the first trajectory curve and the second trajectory curve, the trajectory similarity is calculated. Specifically, first, the enclosed area of ​​the first trajectory curve and the second trajectory curve in the first space coordinate system is calculated, that is, the area of ​​the area enclosed by the two curves; then, the reciprocal of the enclosed area is taken and defined as the trajectory similarity. The enclosed area reflects the degree of difference between the two trajectories in space. The larger the area, the greater the difference in shape and position of the two trajectories, and the lower the trajectory similarity. The reciprocal of the enclosed area is proportional to the similarity. The larger the reciprocal, the higher the trajectory similarity. By calculating the trajectory similarity, an indicator that quantitatively describes the similarity between the conventional moving trajectory and the first moving trajectory is obtained, which provides a basis for whether to set the first moving trajectory as a non-inertial moving trajectory.

[0078] By mapping the trajectory to the spatial coordinate system and using the inverse of the enclosed area to characterize the similarity, fast and accurate calculation of trajectory similarity is achieved, providing support for the adaptive generation of non-inertial motion trajectories.

[0079] Furthermore, the embodiment of the present application also includes:

[0080] S240: Obtaining a species template image set;

[0081] S250: traversing the species template image set, performing species identification on the neighborhood space image information, and obtaining the species identification set.

[0082] In a preferred embodiment, in order to perform species identification, first, a pre-constructed species template image set is obtained, which includes a large number of species template images of different underwater organisms. The acquisition of the species template image set provides a benchmark and reference for subsequent species identification, and is an important basis for image comparison and classification. Subsequently, the twin neural network is used to perform species identification on the neighborhood spatial image information in combination with the species template image set. The twin neural network is a model composed of two completely identical neural networks, namely the spatial image subnetwork and the template image subnetwork, and the two subnetworks share the same architecture and parameters. In species identification, the spatial image subnetwork is responsible for processing the current neighborhood spatial image information; while the template image subnetwork is responsible for processing the species template image, and then the species category to which the current neighborhood spatial image information belongs is determined through the measurement layer.

[0083] Specifically, first, the neighborhood space image information is segmented according to different species objects using image segmentation algorithms, such as semantic segmentation and instance segmentation, to obtain multiple independent neighborhood space sub-images, each of which corresponds to a specific species area in the original scene. The purpose of image segmentation is to isolate and extract image information of different species, reduce interference between species, and provide purer and more focused input data for subsequent twin neural network analysis. Among them, the image segmentation algorithm uses deep learning models, such as full convolutional networks, to achieve automatic image segmentation through end-to-end feature learning and pixel-level classification. These models are pre-trained on massive annotated data and can effectively capture the visual features and boundary information of different species. When applied to neighborhood space image information, the image segmentation algorithm can automatically detect and locate different species areas in the image and generate corresponding neighborhood space sub-images. After obtaining multiple neighborhood space sub-images, each neighborhood space sub-image is traversed and input into the twin neural network in turn, and compared and analyzed with the species template images in the species template image set and similarity calculated. The spatial image subnetwork of the twin neural network extracts the feature vector of the current neighborhood spatial subimage, and the template image subnetwork extracts the feature vector of each species template image. The similarity between the two feature vectors is then calculated through the measurement layer. The species template image with the greatest similarity to the current neighborhood spatial subimage is taken as the species category to which the corresponding current neighborhood spatial subimage belongs and is added to the species recognition set.

[0084] By utilizing species template image sets and similarity metrics, the twin neural network can efficiently and accurately identify underwater biological species in neighborhood space images, providing reliable recognition results for bionic underwater robots.

[0085] The embodiment of the present invention provides a control optimization method for a bionic underwater robot, which has at least the following technical effects:

[0086] Through visual sensors, the image information of the neighborhood space is collected, and the image data of the surrounding environment of the bionic underwater robot is obtained, which provides data support for judging whether there are sensitive species. The species identification is performed on the neighborhood space image information to obtain a species identification set. When the species identification set contains sensitive species, the underwater temperature information and underwater pressure information are collected through the underwater environment sensor. If sensitive species are identified, the temperature and pressure information of the current water area are collected through the underwater environment sensor to provide a basis for the subsequent optimization of the escape speed. According to the underwater temperature information and underwater pressure information, the escape speed of sensitive species is optimized to obtain the escape speed threshold. The speed threshold can ensure that the bionic underwater robot is away from sensitive species at a sufficient speed to reduce interference, and can avoid excessive speed affecting the robot's motion stability and energy consumption. The direction away from sensitive species is taken as the escape direction, and the motion controller is initialized at a speed greater than or equal to the escape speed threshold, and the bionic underwater robot is controlled to avoid interference movement, so as to ensure that the interference with sensitive species and their habitats is minimized during the detection or operation process, and the task execution efficiency and adaptability of the bionic underwater robot are improved while protecting underwater sensitive species and the ecological environment. Embodiment 2

[0087] See also Figure 2 , Figure 2 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, an electronic device 100 provided by an embodiment of the present invention includes a memory 110, a processor 120, and a first computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the first computer program 111, a control optimization method for a bionic underwater robot is implemented. Embodiment 3

[0088] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 200 on which a second computer program 211 is stored. When the second computer program 211 is executed by a processor, a control optimization method for a bionic underwater robot is implemented.

[0089] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0090] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0092] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0094] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0095] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A control optimization method for a bionic underwater robot, characterized in that: Applied to a bionic underwater robot, the bionic underwater robot includes a visual sensor, an underwater environment sensor, a central control unit and a motion controller, and the central control unit performs the following steps: Collect neighborhood space image information through visual sensors; Performing species identification on the neighborhood spatial image information to obtain a species identification set, and when the species identification set contains sensitive species, collecting underwater temperature information and underwater pressure information through an underwater environment sensor; Optimizing the escape speed of the sensitive species according to the underwater temperature information and the underwater pressure information to obtain an escape speed threshold; Taking the direction away from the sensitive species as the escape direction, initializing the motion controller at a speed greater than or equal to the escape speed threshold, and controlling the bionic underwater robot to perform interference avoidance movement; The step of optimizing the escape speed of the sensitive species according to the underwater temperature information and the underwater pressure information to obtain an escape speed threshold includes: Retrieve the first sensitive species abandonment hunting record data of the sensitive species by using the underwater temperature information and the underwater pressure information, wherein the first sensitive species abandonment hunting record data includes the first prey escape speed record value; Performing outlier removal analysis on the first prey escape speed record value to obtain the first prey escape speed concentrated record value, extracting the maximum value of the first prey escape speed concentrated record value and setting it as the first sensitive species escape speed threshold; Until the Nth sensitive species abandoning hunting record data of the sensitive species is retrieved by using the underwater temperature information and the underwater pressure information, wherein the Nth sensitive species abandoning hunting record data includes the Nth prey escape speed record value; Performing outlier analysis on the Nth prey escape speed record value to obtain the Nth prey escape speed concentrated record value, extracting the maximum value of the Nth prey escape speed concentrated record value and setting it as the Nth sensitive species escape speed threshold; The maximum value of the escape speed threshold of the first sensitive species to the escape speed threshold of the Nth sensitive species is extracted and set as the escape speed threshold.

2. The method according to claim 1, characterized in that The motion controller is initialized at a speed greater than or equal to the escape speed threshold, with the direction away from the sensitive species as the direction, and the bionic underwater robot is controlled to perform interference avoidance movement, including: The rated speed of the bionic underwater robot is obtained by the motion controller; When the escape speed threshold is greater than the rated speed of the bionic underwater robot, the direction away from the sensitive species is used as the escape direction, and the non-inertial movement trajectory is started at the rated speed of the bionic underwater robot to initialize the motion controller, so as to control the bionic underwater robot to perform interference avoidance movement; When the escape speed threshold is less than or equal to the rated speed of the bionic underwater robot, the motion controller is initialized at a speed greater than or equal to the escape speed threshold, with the direction away from the sensitive species as the escape direction, to control the bionic underwater robot to perform interference avoidance movement.

3. The method according to claim 1, characterized in that Performing species identification on the neighborhood space image information to obtain a species identification set, and when the species identification set has sensitive species, collecting underwater temperature information and underwater pressure information through an underwater environment sensor, including: Obtain a sensitive species library, and extract matching species of the species identification set belonging to the sensitive species library; When the number of the matching species is equal to 0, continue to perform the underwater monitoring task; When the number of the matching species is not equal to 0, the matching species is set as the sensitive species; The steps of constructing the sensitive species library include: Obtain bionic underwater robot bionic species; The underwater creatures that take the bionic underwater robot-bionic species as prey are added into the sensitive species library.

4. The method according to claim 2, characterized in that The non-inertial movement trajectory configuration step comprises: Configure conventional movement trajectories according to the bionic underwater robot biomimetic species; Randomly generate a first moving trajectory; When the trajectory similarity between the conventional movement trajectory and the first movement trajectory is less than or equal to a trajectory similarity threshold, setting the first movement trajectory as the non-inertial movement trajectory; Otherwise, the first moving trajectory is updated.

5. The method according to claim 4, characterized in that When the trajectory similarity between the conventional movement trajectory and the first movement trajectory is less than or equal to a trajectory similarity threshold, setting the first movement trajectory as the non-inertial movement trajectory includes: Deploy the conventional moving trajectory in a first space coordinate system to obtain a first trajectory curve; Deploy the first moving trajectory in the first space coordinate system to obtain a second trajectory curve; The inverse of the area enclosed by the first trajectory curve and the second trajectory curve is calculated and set as the trajectory similarity.

6. The method according to claim 1, characterized in that Performing species identification on the neighborhood spatial image information to obtain a species identification set includes: Obtain a species template image collection; The species template image set is traversed, species recognition is performed on the neighborhood space image information, and the species recognition set is obtained.

7. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing a control optimization method for a bionic underwater robot as described in any one of claims 1-6.

8. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, which, when executed by a processor, implements a control optimization method for a bionic underwater robot as described in any one of claims 1 to 6.

Images