Energy Management Method, Device and Equipment for Multiple Power-Consuming Modules of an Underwater Robot

Abstract

The present invention provides an energy management method, device and equipment for multiple power-consuming modules of an underwater robot, which relates to the technical field of equipment power supply management. The method includes: obtaining information on multiple power-consuming modules; constructing a Bayesian network, making inference decisions according to the Bayesian network to obtain an energy allocation strategy; obtaining an energy management cost function, optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy; constructing a longitudinal dynamics model of the underwater robot and a lithium-ion battery SOC prediction model; adjusting the energy allocation of the underwater robot according to the energy management strategy, obtaining the actual operating conditions of the underwater robot, and performing real-time feedback and adjustment on the energy management strategy according to the actual operating conditions. Through the present invention, the underwater robot can optimize energy use, extend the battery life, and ensure the safety and reliability of the robot while ensuring the task execution efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of device power supply management, and more particularly, to an energy management method, device and equipment for multiple power-consuming modules of an underwater robot. Background Art

[0002] The ocean contains rich resources. With the continuous growth of the resource demand of countries around the world, the ocean has gradually become a key development direction. As a commonly used deep-sea exploration tool, an autonomous underwater vehicle (AUV) can carry various sensors for ocean exploration and plays an increasingly important role in military, civilian and scientific research.

[0003] Currently, most of the research objects of energy management are hybrid electric vehicles and ships. For research on a single power source of a battery, it is only to control the motor to operate in the high-efficiency area to improve energy utilization rate. For an underwater robot equipped with multiple sensors, there are multiple power-consuming parts. Unreasonable power distribution will affect the performance of the underwater robot such as endurance and work to a certain extent. Summary of the Invention

[0004] The problem solved by the present invention is how to perform reasonable energy management on multiple power-consuming modules of an underwater robot to improve energy utilization rate.

[0005] To solve the above problems, the present invention provides an energy management method, device and equipment for multiple power-consuming modules of an underwater robot.

[0006] In a first aspect, the present invention provides an energy management method for multiple power-consuming modules of an underwater robot, including:

[0007] Obtaining information of multiple power-consuming modules, where the information of multiple power-consuming modules includes camera module information, sonar module information and thruster module information;

[0008] Constructing a Bayesian network, and making an inference decision on the information of multiple power-consuming modules according to the Bayesian network to obtain an energy distribution strategy;

[0009] Obtaining an energy management cost function, optimizing the energy distribution strategy according to the energy management cost function, and performing dynamic optimization to update the energy distribution strategy;

[0010] Adjusting the energy distribution of the underwater robot according to the energy distribution strategy, obtaining the actual operating condition of the underwater robot, and performing real-time feedback and adjustment on the energy distribution strategy according to the actual operating condition.

[0011] Optionally, the constructing a Bayesian network, making an inference decision on the information of multiple power-consuming modules according to the Bayesian network to obtain an energy distribution strategy includes:

[0012] Infer the multi-power-consuming module information through the Bayesian network;

[0013] Determine whether the camera module information and sonar module information are responsive to obtain a response result;

[0014] Determine the power response mode of the thruster module information to obtain a target response mode;

[0015] Obtain the energy allocation strategy according to the response result and the target response mode.

[0016] Optionally, the determining whether the camera module information and sonar module information are responsive to obtain a response result; determining the power response mode of the thruster module information to obtain a target response mode; obtaining the energy allocation strategy according to the response result and the target response mode includes:

[0017] Obtain sample data, train the Bayesian network according to the sample data, and adjust the conditional probability parameters between network nodes;

[0018] Obtain the battery node state corresponding to the multi-power-consuming module information, input the battery node state into the Bayesian network, and correct it through the conditional probability parameters;

[0019] When the conditional probability parameter corresponding to the battery node is greater than or equal to a preset probability threshold, the response result is responsive, and then obtain the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode;

[0020] When the conditional probability parameter corresponding to the battery node is less than the preset probability threshold, the response result is non-responsive, and then no energy is allocated to the camera module information and the sonar module information.

[0021] Optionally, when the conditional probability parameter corresponding to the battery node is greater than or equal to a preset probability threshold, the response result is responsive, and then obtain the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode includes:

[0022] Obtain the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode, where P is the maximum available discharge power of the battery at present; Po is the sum of the powers allocated to the camera module information and the sonar module information; Pd is the power required by the thruster;

[0023] When P - Po ≥ Pd and the target response mode is full - power response, the energy allocation strategy is full - power response. With sufficient remaining allocable power, the thruster responds with full power;

[0024] When P - Po ≥ Pd and the target response mode is under - power response, the energy allocation strategy is under - power response. With sufficient remaining allocable power, the thruster responds with under - power.

[0025] When P - Po < Pd and the target response mode is full - power response, the energy allocation strategy is full - power response. With insufficient remaining allocable power, the thruster responds with under - power.

[0026] When P - Po < Pd and the target response mode is under - power response, the energy allocation strategy is under - power response. With insufficient remaining allocable power, the thruster responds with under - power.

[0027] Optionally, obtaining the energy management cost function, optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy includes:

[0028] Obtaining the energy management cost function, where the energy management cost function is as follows:

[0029] ;

[0030] where L is the energy management cost function, , reflecting the loss of the overall dynamic performance of the AUV, is the two - norm, , representing the lost battery power, is a constant, Pd is the required power of the thruster, is the power actually allocated to the thruster, and SOC is the remaining battery power;

[0031] Optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy.

[0032] Optionally, optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy includes:

[0033] Optimizing the energy allocation strategy according to the energy management cost function;

[0034] Obtaining the under - power response K value in the energy allocation strategy, performing dynamic optimization on the under - power response K value to obtain the target K value, and updating the energy allocation strategy according to the target K value.

[0035] Optionally, it further includes:

[0036] Construct the longitudinal dynamics model of the underwater robot, where the longitudinal dynamics model includes velocity, position, and direction vectors, as well as the transformation matrix between the body coordinate system and the world coordinate system;

[0037] Construct a lithium-ion battery SOC prediction model, where the lithium-ion battery SOC prediction model is used to predict the remaining power of the battery;

[0038] Through the longitudinal dynamics model and the lithium-ion battery SOC prediction model, simulate the thruster module information to obtain a simulation result;

[0039] Optimize the energy distribution strategy according to the simulation result.

[0040] In a second aspect, the present invention provides an energy management device for multiple power-consuming modules of an underwater robot, including:

[0041] An acquisition unit for acquiring information on multiple power-consuming modules, where the information on multiple power-consuming modules includes camera module information, sonar module information, and thruster module information;

[0042] An inference unit for constructing a Bayesian network and making an inference decision on the information on multiple power-consuming modules according to the Bayesian network to obtain an energy distribution strategy;

[0043] A correction unit for acquiring an energy management cost function, optimizing the energy distribution strategy according to the energy management cost function, and performing dynamic optimization to update the energy distribution strategy;

[0044] A processing unit for adjusting the energy distribution of the underwater robot according to the energy distribution strategy, obtaining the actual operating condition of the underwater robot, and performing real-time feedback and adjustment on the energy distribution strategy according to the actual operating condition.

[0045] In a third aspect, the present invention provides an electronic device, including a memory and a processor;

[0046] The memory is used to store a computer program;

[0047] The processor is used to implement the energy management method for multiple power-consuming modules of the underwater robot as described in the first aspect when executing the computer program.

[0048] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the energy management method for multiple power-consuming modules of the underwater robot as described in the first aspect is implemented.

[0049] The beneficial effects of the energy management method, device, electronic device and storage medium for multiple power-consuming modules of an underwater robot according to the present invention are as follows: The energy management method for multiple power-consuming modules of the underwater robot according to the present invention first collects the power consumption information of the camera module, sonar module and thruster module; uses a Bayesian network to perform inference and decision-making on these modules. Among them, the network nodes can include battery SOC, SOP, sonar module request, camera module request, battery power status, power status, thruster response, camera module response, sonar module response, etc. Through the Bayesian network of the present invention, it can be obtained whether each module should respond to its power consumption request, and whether the thruster should respond at full power or underpower; establish a cost function, which considers the loss of dynamic performance and SOC loss. Among them, the weight value in the cost function is adaptively variable, and it is based on the probability value of the inference and decision-making module making an underpower response decision for the thruster; optimize the energy allocation strategy according to the cost function to update the strategy, and perform real-time feedback and adjustment on the energy allocation strategy according to the actual operating conditions of the underwater robot. The method of the present invention solves the problem that for an underwater robot powered by a single battery and equipped with multiple sensors, unreasonable power distribution may affect the endurance and working performance of the underwater robot, optimizes power distribution, improves the performance and efficiency of the underwater robot, enables the underwater robot to optimize energy use while ensuring the task execution efficiency, extends the battery life, and ensures the safety and reliability of the robot, and is particularly suitable for application scenarios that need to perform tasks for a long time in a complex underwater environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic flow chart of an energy management method for multiple power-consuming modules of an underwater robot according to an embodiment of the present invention;

[0051] Figure 2 It is a schematic structural diagram of an energy management device for multiple power-consuming modules of an underwater robot according to an embodiment of the present invention;

[0052] Figure 3 It is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] To make the above objects, features and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention is provided in conjunction with the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0054] It should be understood that the various steps described in the method embodiments of the present invention can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0055] As used herein, the term "comprising" and its variations are open-ended, i.e., "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0056] It should be noted that the modifications of "one" and "plural" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0057] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0058] As Figure 1 shown, a method for energy management of multiple power-consuming modules of an underwater robot provided by an embodiment of the present invention includes:

[0059] Obtain multi-power-consuming module information, where the multi-power-consuming module information includes camera module information, sonar module information, and thruster module information.

[0060] Specifically, taking an underwater snake robot as an example of the underwater robot, in the method for energy management of multiple power-consuming modules of the underwater snake robot, first, it is necessary to obtain the information of each power-consuming module on the robot, which includes the camera module, the sonar module, and the thruster module. These modules are an indispensable part when the robot performs tasks, but at the same time, they will also impose a burden on the robot's energy system.

[0061] Construct a Bayesian network, and make an inference decision on the multi-power-consuming module information according to the Bayesian network to obtain an energy allocation strategy.

[0062] Specifically, Bayesian network is used for inference and decision-making. This network makes decisions based on the SOC and SOP of the battery and the request status of other modules. The Bayesian network can infer whether to respond to the power consumption requests of the camera and sonar modules, and whether the thruster should operate at full power or underpower, according to the current energy status and power status. This step is the core of the algorithm, which determines how the robot allocates limited energy under different circumstances.

[0063] Obtain the energy management cost function, optimize the energy allocation strategy according to the energy management cost function, and perform dynamic optimization to update the energy allocation strategy.

[0064] Specifically, the design of the energy management cost function is to balance the dynamic performance loss and SOC loss of the underwater snake robot. The cost function includes terms reflecting the overall dynamic performance loss of the AUV and terms representing the lost battery power, where the weight values are adaptively variable based on the results of the Bayesian network inference and decision-making. This step ensures that the energy allocation strategy minimizes energy consumption as much as possible while meeting the task requirements.

[0065] Adjust the energy allocation of the underwater robot according to the energy allocation strategy, obtain the actual operating conditions of the underwater robot, and perform real-time feedback and adjustment on the energy allocation strategy according to the actual operating conditions.

[0066] Specifically, according to the inference and decision-making of the Bayesian network and the optimization results of the cost function, the energy allocation of the underwater snake robot is adjusted in real time. This includes response decisions for the camera and sonar modules, and power allocation for the thruster. This step is the execution phase of the algorithm, which directly affects the energy use efficiency and task execution effect of the robot. At the same time, the algorithm needs to continuously optimize the energy allocation according to the latest operation data to adapt to possible environmental changes and robot state changes, ensuring the real-time and effectiveness of energy management.

[0067] The energy management method for multiple power-consuming modules of the underwater robot in this embodiment first collects the power consumption information of the camera module, sonar module, and thruster module; uses a Bayesian network to perform inference and decision-making on these modules. Among them, the network nodes can include battery SOC, SOP, sonar module requests, camera module requests, battery power status, power status, thruster response, camera module response, sonar module response, etc. Through the Bayesian network of the present invention, it can be obtained whether each module should respond to its power consumption request, and whether the thruster should respond at full power or underpower. Establish a cost function that takes into account the loss of power performance and SOC loss. Among them, the weight value in the cost function is adaptively variable and is based on the probability value of the inference decision module making an underpower response decision for the thruster. Optimize the energy allocation strategy according to the cost function to update the strategy, and perform real-time feedback and adjustment on the energy allocation strategy according to the actual operating conditions of the underwater robot. The method of this embodiment solves the problem that for an underwater robot with a single battery power source and multiple sensors, unreasonable power allocation may affect the endurance and working performance of the underwater robot, optimizes power allocation, improves the performance and efficiency of the underwater robot, enables the underwater robot to optimize energy use while ensuring the task execution efficiency, extends the battery life, and ensures the safety and reliability of the robot, and is particularly suitable for application scenarios that need to perform tasks for a long time in complex underwater environments.

[0068] Optionally, the constructing of the Bayesian network, performing inference and decision-making on the multiple power-consuming module information according to the Bayesian network, and obtaining an energy allocation strategy includes:

[0069] Performing inference on the multiple power-consuming module information through the Bayesian network;

[0070] Determining whether the camera module information and sonar module information respond, and obtaining a response result;

[0071] Determining the power response mode of the thruster module information, and obtaining a target response mode;

[0072] Obtaining the energy allocation strategy according to the response result and the target response mode.

[0073] Specifically, the process of constructing a Bayesian network and making inference decisions is the core of the energy management method for multiple power-consuming modules of an underwater snake robot. First, the Bayesian network is used to infer the information of the camera module, sonar module, and thruster module, including the power consumption requests of the modules, the state of charge (SOC) and state of power (SOP) of the battery, etc. The node design of the Bayesian network takes into account multiple factors such as battery SOC, SOP, sonar module requests, camera module requests, battery power status, power status, thruster response, camera module response, sonar module response, etc. Then, based on the inference results of the Bayesian network, it is determined whether to respond to the power consumption requests of the camera and sonar modules, and whether the thruster responds at full power or underpower. In the preferred embodiment of the present invention, when the battery SOC and SOP are in a good state, the requests of the camera and sonar will be responded to, and the thruster will respond at full power; while when the energy and power status are poor, these module requests may not be responded to, and the thruster may also switch to underpower response. Finally, according to these response results and response modes, an energy allocation strategy is formed to determine the energy allocation of each module.

[0074] For the energy management method of multiple power-consuming modules of the underwater robot in this embodiment, the Bayesian network enables the energy management decision-making to be adaptive and can be flexibly adjusted according to different operating conditions, thereby extending the operation time of the underwater snake robot and improving its energy utilization efficiency. In addition, the model predictive control method is used to dynamically optimize the energy allocation strategy, further improving the accuracy and response speed of energy management, and ensuring that the robot can effectively perform tasks in a complex and changing underwater environment.

[0075] Optionally, determining whether to respond to the camera module information and sonar module information to obtain a response result; determining the dynamic response mode of the thruster module information to obtain a target response mode; and obtaining the energy allocation strategy according to the response result and the target response mode, including:

[0076] Obtain sample data, train the Bayesian network according to the sample data, and adjust the conditional probability parameters between network nodes;

[0077] Obtain the battery node state corresponding to the multiple power-consuming module information, input the battery node state into the Bayesian network, and correct it through the conditional probability parameters;

[0078] When the conditional probability parameter corresponding to the battery node is greater than or equal to a preset probability threshold, the response result is a response, and then the energy allocation strategy is obtained according to the camera module information, the sonar module information, and the target response mode;

[0079] When the conditional probability parameter corresponding to the battery node is less than the preset probability threshold, the response result is non-response, and no energy allocation is performed for the camera module information and the sonar module information.

[0080] Specifically, first, the Bayesian network is trained by collecting and analyzing sample data, and the conditional probability parameters between the nodes in the network are adjusted to ensure that the network can accurately reflect the relationship between the battery state and the module response. In the embodiment, for example, when the SOC and SOP states of the battery are judged to be good, the Bayesian network inference decision will tend to respond to the power consumption requests of the camera and sonar modules and allocate full power to the thruster module to support the robot in performing complex or energy-consuming tasks. On the contrary, when the battery state is poor or average, the network will decide not to respond to the power consumption requests of these modules, or adjust the thruster module to an underpower response mode to save energy and extend the operation time of the robot. This energy allocation strategy based on Bayesian network inference can be dynamically adjusted according to the real-time state of the battery, optimizing the energy usage efficiency and ensuring that the robot can effectively perform tasks under different energy states.

[0081] In the multi-power-consuming module energy management method of the underwater robot in this embodiment, through the dynamic inference decision of the Bayesian network, the robot can flexibly adjust the working modes of each power-consuming module according to the current energy status, thereby saving energy to the greatest extent while ensuring the task execution efficiency. Secondly, this method enhances the adaptability of the robot to environmental changes, enabling it to maintain the normal operation of key functions through intelligent decision-making under energy constraints.

[0082] Optionally, when the conditional probability parameter corresponding to the battery node is greater than or equal to the preset probability threshold, the response result is response, and the energy allocation strategy is obtained according to the camera module information, the sonar module information, and the target response mode, including:

[0083] The energy allocation strategy is obtained according to the camera module information, the sonar module information, and the target response mode, where P is the maximum available discharge power of the battery at present; Po is the sum of the power allocated to the camera module information and the sonar module information; Pd is the thruster demand power;

[0084] When P - Po ≥ Pd and the target response mode is full-power response, the energy allocation strategy is full-power response, the remaining available power is sufficient, and the thruster responds with full power;

[0085] When P - Po ≥ Pd and the target response mode is underpower response, the energy allocation strategy is underpower response, the remaining available power is sufficient, and the thruster responds with underpower.

[0086] When P - Po < Pd and the target response mode is full - power response, the energy allocation strategy is full - power response. The remaining available power for allocation is insufficient, and the thruster responds with under - power.

[0087] When P - Po < Pd and the target response mode is under - power response, the energy allocation strategy is under - power response. The remaining available power for allocation is insufficient, and the thruster responds with under - power.

[0088] Specifically, the determination of the energy allocation strategy is based on the result of Bayesian network inference and decision - making, the current maximum available discharge power of the battery (P), the total power requirements of the camera and sonar modules (Po), and the required power of the thruster (Pd). Among them, the total power requirements of the above - mentioned camera and sonar modules are obtained from the camera module information and sonar module information. Specifically, when the remaining available power of the battery (P - Po) is greater than or equal to the required power of the thruster (Pd), that is, P - Po ≥ Pd, the strategy selects full - power response, which means that the thruster will obtain all the required power to perform tasks, ensuring the propulsion efficiency and power performance of the robot. On the contrary, if P - Po < Pd, that is, the remaining available power is insufficient to meet the full - power requirements of the thruster, the strategy selects under - power response. At this time, the thruster will only obtain part of the required power, which may affect the propulsion efficiency of the robot but helps to extend the battery usage time and ensure that the robot can continue to operate for a longer time. In the embodiment, according to the SOC and SOP states of the battery, combined with the inference result of the Bayesian network, the power response mode of the thruster is dynamically adjusted to adapt to different task requirements and energy conditions.

[0089] The multi - power - consuming module energy management method of the underwater robot in this embodiment flexibly adjusts the energy allocation strategy of the underwater snake - shaped robot according to the real - time energy condition and task requirements, optimizes the energy usage efficiency, not only improves the power performance and efficiency of the robot when performing tasks, but also effectively extends the operation time of the robot and the service life of the battery by adopting under - power response under limited energy conditions.

[0090] Optionally, obtaining the energy management cost function, optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy includes:

[0091] Obtaining the energy management cost function, where the energy management cost function is as follows:

[0092] ;

[0093] where L is the energy management cost function, , reflecting the loss of the overall power performance of the AUV, is the second norm, , representing the lost battery power, is a constant, Pd is the required power of the thruster, is the power actually allocated to the thruster, and SOC is the remaining battery power;

[0094] Optimize the energy allocation strategy according to the energy management cost function and perform dynamic optimization to update the energy allocation strategy.

[0095] Specifically, the acquisition and application of the energy management cost function are the key links to achieve the optimal energy allocation of the underwater snake robot. This cost function , is defined as a function reflecting the overall dynamic loss of the AUV, where Δv represents the second norm of the speed change, representing the dynamic loss, and ΔSOC represents the change in battery power, reflecting the loss of battery SOC. The constant k is used to adjust the weight between the dynamic loss and the battery power loss. The cost function evaluates the effects of different energy allocation strategies by considering the power allocation of the thruster and the camera and sonar modules, and Pd, as well as the remaining battery power SOC. In the embodiment, when the Bayesian network infers that the thruster needs an underpower response, the cost function will dynamically adjust the value of k according to the current energy status and task requirements to optimize the trade-off between dynamics and energy loss. Through the model predictive control method, the system continuously performs rolling optimization and updates the energy allocation strategy in real time to adapt to the changing task requirements and energy status.

[0096] The multi-power-consuming module energy management method of the underwater robot in this embodiment allows the system to adjust the energy allocation strategy according to real-time feedback during the dynamic optimization process, improving the flexibility and efficiency of energy use. This method not only improves the operation ability of the robot in complex underwater environments, but also extends its endurance time and enhances its adaptability to task changes and environmental uncertainties.

[0097] Optionally, the optimizing the energy allocation strategy according to the energy management cost function and performing dynamic optimization to update the energy allocation strategy includes:

[0098] Optimize the energy allocation strategy according to the energy management cost function;

[0099] Obtain the underpower response K value in the energy allocation strategy, perform dynamic optimization on the underpower response K value to obtain the target K value, and update the energy allocation strategy according to the target K value.

[0100] Specifically, optimizing and dynamically searching for the energy distribution strategy of an underwater snake-shaped robot are important steps to ensure the effective utilization of energy. First, evaluate and adjust the current energy distribution strategy according to the energy management cost function. The cost function includes a term Δv (two-norm) reflecting the overall dynamic loss of the AUV and a term ΔSOC representing the battery power loss, where k is a constant used to balance the dynamic loss and the battery power loss. Then, obtain the K value of the underpower response from the current energy distribution strategy. This value represents the ratio of the actual power obtained by the thruster to its required power. By dynamically searching for the optimal K value, the minimum value of the cost function can be found, thereby minimizing the battery power loss while ensuring the dynamic performance of the AUV. For example, in the embodiment, if the current K value causes the battery power to be consumed too quickly, the system will adjust the K value through the model predictive control method to extend the battery life and optimize the overall performance. The updated K value will be used to adjust the power distribution of the thruster to achieve better energy management.

[0101] The energy management method for the multi-power-consuming modules of the underwater robot in this embodiment adjusts the power distribution of the thruster in real time according to the current energy status and task requirements, optimizing the energy usage efficiency. This method not only improves the dynamic performance and efficiency of the robot during task execution but also effectively extends the battery usage time and the operation time of the robot by adopting the optimal underpower response under limited energy conditions.

[0102] Optionally, it further includes:

[0103] Construct the longitudinal dynamics model of the underwater robot, where the longitudinal dynamics model includes velocity, position, and direction vectors, as well as the transformation matrix between the body coordinate system and the world coordinate system;

[0104] Construct a lithium-ion battery SOC prediction model, where the lithium-ion battery SOC prediction model is used to predict the remaining power of the battery;

[0105] Simulate the thruster module information through the longitudinal dynamics model and the lithium-ion battery SOC prediction model to obtain simulation results;

[0106] Optimize the energy distribution strategy according to the simulation results.

[0107] Specifically, first, the construction of the longitudinal dynamics model involves velocity, position, and direction vectors, as well as the transformation matrix between the body coordinate system and the world coordinate system. These parameters are crucial for simulating the movement of the robot in water. Through this model, the dynamic responses of the robot under different operating conditions, such as speed changes, accelerations, and steering, can be predicted. Secondly, the lithium-ion battery SOC prediction model focuses on predicting the open-circuit voltage and equivalent internal resistance of the battery, which are crucial for evaluating the health status and remaining power of the battery. In the embodiment, these models are used to simulate the performance of the robot under specific tasks and environmental conditions, as well as the behavior of the battery under different discharge conditions. By combining these two models, the energy requirements and battery performance of the robot in practical applications can be predicted more accurately, providing a scientific basis for energy management.

[0108] Specifically, construct the longitudinal dynamics model of the underwater snake-shaped robot. This model includes parameters such as velocity, position, and direction vectors, the transformation matrix between the body coordinate system and the world coordinate system, the total mass of the underwater snake-shaped robot, Coriolis force, fluid resistance, hydrostatic force, and generalized driving force. Construct the lithium-ion battery SOC prediction model, which predicts the remaining power of the battery for energy allocation. Use the model prediction method to dynamically optimize the energy allocation of the thruster module. This involves optimizing the underpower response of the thruster, that is, the K value of the required power (0 < K ≤ 1). During the optimization process, the following constraint conditions need to be satisfied:

[0109] The relationship between the battery discharge power P and the sum of the power Pm allocated to the thruster and the power Po of the camera and sonar modules;

[0110] The upper and lower limits of SOC.

[0111] In the actual application process, the above constrained optimization problem can be solved by the sequential quadratic programming (SQP) algorithm. Set the initial value to 3.2AH, the prediction step k to 100, and the rolling optimization time domain to 5.

[0112] According to the simulation results, the optimal control variable U(K), that is, the K value, and the change in SOC are solved under the constraint conditions. SOC changes from the initial value of 1 to 0.64, and the average running time per step is 0.6802S, indicating that the proposed multi-module energy management strategy can achieve online real-time operation.

[0113] According to whether the power-consuming module responds or not inferred by the Bayesian network reasoning and whether the thruster responds with full power or underpower, calculate the power actually allocated to the thruster = KPd. After other power-consuming modules and the thruster have done work, the battery state changes. The total power consumption of other power-consuming modules becomes Po', and the optimal control variable K becomes K'.

[0114] Specifically, the key points to be judged in this embodiment include: judging whether to respond to the power consumption requests of the camera and sonar modules; judging whether the thruster performs a full-power response or an under-power response; dynamically adjusting the K value according to the change of SOC and the inference decision result to optimize the energy distribution.

[0115] Through the above steps and key points of judgment, the energy management system of the underwater robot can dynamically adjust the power distribution of each module according to the current state and prediction model to achieve optimal energy management.

[0116] The multi-power-consuming-module energy management method for the underwater robot in this embodiment can optimize the task planning and execution by accurately simulating the dynamic behavior of the robot, reducing the risk of failure caused by insufficient energy. Secondly, the prediction function of the lithium-ion battery SOC prediction model enables the energy distribution strategy to be more flexible and intelligent, adjusting the energy use according to the actual state of the battery, extending the battery life, and improving the success rate of the task.

[0117] As Figure 2 shown, an energy management device for multi-power-consuming modules of an underwater robot provided by an embodiment of the present invention includes:

[0118] An acquisition unit, configured to acquire multi-power-consuming-module information, where the multi-power-consuming-module information includes camera module information, sonar module information, and thruster module information;

[0119] An inference unit, configured to construct a Bayesian network and perform inference and decision-making on the multi-power-consuming-module information according to the Bayesian network to obtain an energy distribution strategy;

[0120] A correction unit, configured to acquire an energy management cost function, optimize the energy distribution strategy according to the energy management cost function, and perform dynamic optimization to update the energy distribution strategy;

[0121] A processing unit, configured to adjust the energy distribution of the underwater robot according to the energy distribution strategy, acquire the actual operating conditions of the underwater robot, and perform real-time feedback and adjustment on the energy distribution strategy according to the actual operating conditions.

[0122] The energy management device for multi-power-consuming modules of the underwater robot in this embodiment is used to implement the multi-power-consuming-module energy management method for the underwater robot as described above, and its advantages compared with the prior art are the same as those of the multi-power-consuming-module energy management method for the underwater robot compared with the prior art, which will not be elaborated here.

[0123] As Figure 3As shown in the figure, an electronic device 300 provided by an embodiment of the present invention includes a memory 310 and a processor 320; the memory 310 is used to store a computer program; the processor 320 is used to implement the energy management method for multiple power-consuming modules of the underwater robot as described above when executing the computer program.

[0124] Alternatively, an electronic device 300 includes a memory 310 and a processor 320 coupled to the memory 310; the memory 310 is configured to store a computer program; the processor 320 is configured to perform the following operations when executing the computer program:

[0125] Obtain information about multiple power-consuming modules, where the information about multiple power-consuming modules includes information about a camera module, information about a sonar module, and information about a thruster module;

[0126] Construct a Bayesian network, perform inference and decision-making on the information about multiple power-consuming modules according to the Bayesian network, and obtain an energy allocation strategy;

[0127] Obtain an energy management cost function, optimize the energy allocation strategy according to the energy management cost function, and perform dynamic optimization to update the energy allocation strategy;

[0128] Adjust the energy allocation of the underwater robot according to the energy allocation strategy, obtain the actual operating conditions of the underwater robot, and perform real-time feedback and adjustment on the energy allocation strategy according to the actual operating conditions.

[0129] A computer-readable storage medium provided by an embodiment of the present invention has a computer program stored thereon. When the computer program is executed by a processor, the energy management method for multiple power-consuming modules of the underwater robot as described above is implemented.

[0130] Alternatively, a non-volatile computer-readable storage medium has a computer program stored thereon. When the computer program is executed by a processor, the processor is caused to perform the following operations:

[0131] Obtain information about multiple power-consuming modules, where the information about multiple power-consuming modules includes information about a camera module, information about a sonar module, and information about a thruster module;

[0132] Construct a Bayesian network, perform inference and decision-making on the information about multiple power-consuming modules according to the Bayesian network, and obtain an energy allocation strategy;

[0133] Obtain an energy management cost function, optimize the energy allocation strategy according to the energy management cost function, and perform dynamic optimization to update the energy allocation strategy;

[0134] Adjust the energy distribution of the underwater robot according to the energy distribution strategy, obtain the actual operating condition of the underwater robot, and perform real-time feedback and adjustment on the energy distribution strategy according to the actual operating condition.

[0135] Now, an electronic device 300 that can be a server or a client of the present invention will be described. It is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 300 is intended to represent various forms of digital electronic computer devices, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 300 can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0136] The electronic device 300 includes a computing unit that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The computing unit, the ROM, and the RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0137] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc. In the present application, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of the present invention. In addition, the functional units in various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0138] Although the present invention is disclosed as above, the scope of protection of the present invention is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will all fall within the scope of protection of the present invention.

Claims

1. An energy management method for multiple power-consuming modules of an underwater robot, characterized in that, Including: Obtain multi-power-consuming module information, where the multi-power-consuming module information includes camera module information, sonar module information, and thruster module information; Construct a Bayesian network, and perform inference and decision-making on the multi-power-consuming module information according to the Bayesian network to obtain an energy allocation strategy; including: performing inference on the multi-power-consuming module information through the Bayesian network; Determine whether the camera module information and sonar module information are responsive to obtain a response result; determine the power response mode of the thruster module information to obtain a target response mode; obtain the energy allocation strategy according to the response result and the target response mode; specifically including: performing inference on the multi-power-consuming module information through the Bayesian network; Obtain sample data, train the Bayesian network according to the sample data, and adjust the conditional probability parameters between network nodes; Obtain the battery node state corresponding to the multi-power-consuming module information, input the battery node state into the Bayesian network, and correct it through the conditional probability parameters; When the conditional probability parameter corresponding to the battery node is greater than or equal to a preset probability threshold, the response result is responsive, and then obtain the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode; including: obtaining the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode, where P is the maximum available discharge power of the battery at present; Po is the sum of the power allocated to the camera module information and the sonar module information; Pd is the thruster demand power; When P - Po ≥ Pd and the target response mode is full-power response, the energy allocation strategy is full-power response, the remaining allocable power is sufficient, and the thruster responds at full power; When P - Po ≥ Pd and the target response mode is under-power response, the energy allocation strategy is under-power response, the remaining allocable power is sufficient, and the thruster responds with under power; When P - Po < Pd and the target response mode is full-power response, the energy allocation strategy is full-power response, the remaining allocable power is insufficient, and the thruster responds with under power; When P - Po < Pd and the target response mode is under-power response, the energy allocation strategy is under-power response, the remaining allocable power is insufficient, and the thruster responds with under power; When the conditional probability parameter corresponding to the battery node is less than the preset probability threshold, the response result is non-responsive, and then no energy is allocated to the camera module information and the sonar module information; Obtain an energy management cost function, optimize the energy allocation strategy according to the energy management cost function, and perform dynamic optimization to update the energy allocation strategy; Adjust the energy allocation of the underwater robot according to the energy allocation strategy, obtain the actual operating condition of the underwater robot, and perform real-time feedback and adjustment on the energy allocation strategy according to the actual operating condition.

2. The energy management method for multiple power-consuming modules of an underwater robot according to claim 1, characterized in that, Obtaining the energy management cost function, optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy, including: Obtaining the energy management cost function, where the energy management cost function is as follows: ； where, L is the energy management cost function, , reflecting the loss of the overall dynamic performance of the AUV, is the two-norm, , representing the lost battery power, is a constant, Pd is the required power of the thruster, is the power actually allocated to the thruster, and SOC is the remaining battery power; Optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy.

3. The energy management method for multiple power-consuming modules of an underwater robot according to claim 2, characterized in that The optimizing the energy allocation strategy according to the energy management cost function, and performing dynamic optimization to update the energy allocation strategy includes: Optimizing the energy allocation strategy according to the energy management cost function; Obtaining the under-power response K value in the energy allocation strategy, performing dynamic optimization on the under-power response K value to obtain a target K value, and updating the energy allocation strategy according to the target K value.

4. The energy management method for multiple power-consuming modules of an underwater robot according to claim 1, characterized in that, It further includes: Constructing a longitudinal dynamics model of the underwater robot, where the longitudinal dynamics model includes speed, position, and direction vectors, and a transformation matrix between the body coordinate system and the world coordinate system; Constructing a lithium-ion battery SOC prediction model, where the lithium-ion battery SOC prediction model is used to predict the remaining power of the battery; Performing simulation on the thruster module information through the longitudinal dynamics model and the lithium-ion battery SOC prediction model to obtain a simulation result; Optimizing the energy allocation strategy according to the simulation result.

5. An energy management device for multiple power-consuming modules of an underwater robot, characterized in that, It includes: An acquisition unit for acquiring multi-power-consuming module information, where the multi-power-consuming module information includes camera module information, sonar module information, and thruster module information; An inference unit for constructing a Bayesian network, and performing inference and decision-making on the multi-power-consuming module information according to the Bayesian network to obtain an energy allocation strategy; including: performing inference on the multi-power-consuming module information through the Bayesian network; Determining whether the camera module information and the sonar module information respond to obtain a response result; determining the dynamic response mode of the thruster module information to obtain a target response mode; and obtaining the energy allocation strategy according to the response result and the target response mode; Specifically including: performing inference on the multi-power-consuming module information through the Bayesian network; Obtaining sample data, training the Bayesian network according to the sample data, and adjusting the conditional probability parameters between network nodes; Obtaining the battery node state corresponding to the multi-power-consuming module information, inputting the battery node state into the Bayesian network, and correcting it through the conditional probability parameters; When the conditional probability parameter corresponding to the battery node is greater than or equal to a preset probability threshold, the response result is a response, and then the energy allocation strategy is obtained according to the camera module information, the sonar module information, and the target response mode; including: obtaining the energy allocation strategy according to the camera module information, the sonar module information, and the target response mode, where P is the maximum available discharge power of the battery currently; Po is the sum of the powers allocated to the camera module information and the sonar module information; Pd is the thruster demand power; When P - Po ≥ Pd and the target response mode is full - power response, the energy allocation strategy is full - power response. With sufficient remaining allocable power, the thruster responds with full power; When P - Po ≥ Pd and the target response mode is under - power response, the energy allocation strategy is under - power response. With sufficient remaining allocable power, the thruster responds with under - power; When P - Po < Pd and the target response mode is full - power response, the energy allocation strategy is full - power response. With insufficient remaining allocable power, the thruster responds with under - power; When P - Po < Pd and the target response mode is under - power response, the energy allocation strategy is under - power response. With insufficient remaining allocable power, the thruster responds with under - power; When the conditional probability parameter corresponding to the battery node is less than the preset probability threshold, the response result is non - response, and no energy is allocated to the camera module information and the sonar module information; A correction unit, configured to obtain an energy management cost function, optimize the energy allocation strategy according to the energy management cost function, and perform dynamic optimization to update the energy allocation strategy; A processing unit, configured to adjust the energy allocation of the underwater robot according to the energy allocation strategy, obtain the actual operating condition of the underwater robot, and perform real - time feedback and adjustment on the energy allocation strategy according to the actual operating condition.

6. An electronic device, characterized in that, It includes a memory and a processor; The memory is used to store computer programs; The processor, when executing the computer program, implements the multi - power - consuming module energy management method for the underwater robot according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer program is stored on the storage medium. When the computer program is executed by the processor, the multi - power - consuming module energy management method for the underwater robot according to any one of claims 1 to 4 is implemented.

Images