BPNN-based bionic takifugu fin adaptive movement mechanism

Abstract

The invention provides a BPNN-based bionic takifugu fin self-adaptive movement mechanism, and belongs to the technical field of bionic fish driving, the BPNN-based bionic takifugu fin self-adaptive movement mechanism comprises a base, the base comprises a main control cabin section and an energy cabin section, the main control cabin section comprises a control system, the control system comprises a main controller, a BPNN-based pressure monitoring system and a signal conditioning circuit, and the energy cabin section comprises an energy source module; the controller is connected with the bionic flexible fluctuation propulsion system, the signal conditioning circuit is connected with the sensor cluster system, and the sensor cluster system is installed on a fin ray of a bionic fin of the bionic flexible fluctuation propulsion system. According to the invention, fish form and joint linkage is accurately simulated, and flexible and smooth bionic fish fins are molded; the pressure monitor integrated at the fin can sensitively capture real-time pressure data generated by water flow to the fin, accurately sense subtle changes of the water flow, remarkably enhance the adaptive capacity to the environment, promote the energy efficiency, and explore a more accurate and energy-saving technical path for application scenes such as underwater robots and marine scientific investigation equipment.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of bionic fish driving, and particularly relates to a bionic puffer fish fin adaptive motion mechanism based on BPNN. BACKGROUND

[0002] In actual marine exploration application scenarios, the advantages of the bionic fish adaptive motion mechanism based on BPNN are fully demonstrated. When performing a deep-sea hydrothermal zone exploration task, the pressure monitor can quickly perceive the great change in deep-sea water pressure and the complex water flow disturbance near the hydrothermal vent. The BPNN algorithm immediately processes these data, accurately regulates the bionic flexible undulating propulsion system, and enables the bionic fish to stably approach the hydrothermal zone. With the aid of the chemical sensor and high-definition camera equipment carried, the bionic fish can carefully collect valuable information such as hydrothermal composition and microbial ecology, and provide key evidence for studying the origin of deep-sea life and geological activity, greatly expanding the cognitive boundaries of humans on deep-sea extreme environments. In marine biological observation tasks, the mechanism can be quietly integrated into marine biological communities. The realistic appearance and dynamic motion mode of the bionic fish benefit from the real-time optimization of the fin swing parameters by the BPNN, and the bionic fish can adjust its swimming posture according to different flow field data. It can track the activity trajectory of marine organisms at close range, record important data such as their behavior habits and migration routes, contribute unique value to marine ecological protection and biodiversity research, and help to develop scientific and effective marine protection strategies.

[0003] In terms of underwater engineering maintenance, such as inspection of submarine oil pipelines, the bionic fish can rely on its adaptive motion ability to swim in the complex environment around the pipeline. After the pressure monitor detects abnormal pressure changes in the water flow due to pipeline leakage or structural damage, the BPNN drives the bionic fish to conduct multi-angle investigation around the suspicious area, and timely feedback detailed location and condition information to the surface control center, effectively improving the maintenance efficiency and safety of underwater engineering facilities, and reducing maintenance costs and potential risks.

[0004] Further, in the process of continuous optimization and upgrading of the mechanism, the research team is committed to improving the learning efficiency and generalization ability of the BPNN algorithm. By introducing more advanced deep learning architectures and large data training sets, it can adapt to new underwater environmental challenges in a shorter time, and enhance the ability to cope with rare water flow phenomena and complex marine working conditions. At the same time, cooperation in the field of materials science is carried out to develop new bionic flexible materials, further improve the durability, flexibility and hydrodynamic performance of the bionic fish fin, and ensure that the mechanism always maintains excellent performance in long-term and high-intensity underwater operations, laying a more solid technical foundation for future marine exploration and development.

[0005] However, the existing bionic fish adaptive motion mechanism still has some deficiencies, for example, the generalization ability of the algorithm in the complex nonlinear flow field is very limited, the reliability of the pressure sensor in the extreme environment of the deep sea is insufficient, the durability of the flexible fin needs to be improved, and the system autonomy and energy endurance also need to be improved. SUMMARY

[0006] The purpose of the present application is to provide a bionic puffer fish fin adaptive motion mechanism based on BPNN, which adjusts the swing parameters of the fin by transmitting the flow field data in real time to the BPNN neural network through the pressure monitor in different environments, so as to make the fin swing reach the optimal state and improve the efficiency of the underwater vehicle and save the energy of the underwater vehicle.

[0007] The purpose of the present application is achieved by the following technical solutions:

[0008] A bionic puffer fish fin adaptive motion mechanism based on BPNN, comprising: a base, the base comprising a main control cabin section, an energy cabin section, the main control cabin section comprising a control system, the control system comprising a main controller, a pressure monitoring system based on BPNN neural network, a signal conditioning circuit, the controller being connected with a bionic flexible undulating propulsion system, the signal conditioning circuit being connected with a sensor cluster system, the sensor cluster system being installed on the fin bar of the bionic fin of the bionic flexible undulating propulsion system.

[0009] Further, the bionic flexible undulating propulsion system comprises a motor, the motor being installed on a longitudinal profile, the longitudinal profile being installed on the base, the longitudinal profile being connected with a connecting frame, the connecting frame being connected with four toothless tether unit fixed frames, toothless tether units being installed on the toothless tether unit fixed frames, the swing ends of the driving rods on the toothless tether units being connected with the bionic fins respectively, the synchronous pulleys of the four toothless tether units being connected with the motor through synchronous belts respectively, and the two ends of the bionic fins being rotatably connected on the two ends of the connecting frame through hinges respectively.

[0010] Further, the bionic fin comprises four fin bars arranged in parallel, and the fin bars are covered by the flexible fin surface.

[0011] Further, the fin bar tail has a sensor cluster system connected with the controller through a signal line for transmitting control commands and feedback signals.

[0012] Further, the pressure monitoring system of the BPNN neural network comprises a feature extraction module for extracting pressure features; the feature extraction module is connected with the BP neural network for receiving the pressure features to judge the parameters of the fin swing and give the judgment result.

[0013] Further, the sensor cluster system comprises a set of pressure sensors, four pressure sensors are installed on the first fin, three pressure sensors are installed on the second fin, two pressure sensors are installed on the third fin, and one pressure sensor is installed on the fourth fin, and the pressure sensors are connected to the main control cabin through signal lines to transmit the measured pressure data to the main control cabin.

[0014] Further, the toothless system unit comprises a transmission shaft, one end of the transmission shaft is rotatably connected with the toothless shell, a synchronous pulley is sleeved on the other end of the transmission shaft, the transmission shaft passes through the toothless shell and is connected with a crank rocker adapter plate, the crank rocker adapter plate is connected with one end of a crank rocker through a pin shaft, the other end of the crank rocker is connected with a transfer pair through a crank rocker bearing, and the transfer pair is rotatably connected with the toothless shell through bearings, each fin is connected with the crank rocker adapter plate through an output base, and the fins are swingably arranged on the toothless shell.

[0015] Further, the material of the flexible fin surface comprises a hydrophobic material, and the hydrophobic material is a polydimethylsiloxane film.

[0016] Further, the fin is made of shape memory alloy or carbon fiber.

[0017] Further, the sensor cluster system further comprises a temperature sensor and a water flow sensor.

[0018] The beneficial effects of the present application are:

[0019] The present application accurately simulates the morphology and joint linkage of fish, and shapes a flexible and smooth biomimetic fin. The pressure monitor integrated at the fin can sensitively capture real-time pressure data generated by water flow on the fin, accurately perceive subtle changes in water flow, and apply a BPNN algorithm to integrate and analyze the data collected by the pressure monitor with other environmental parameters such as water temperature and water flow speed, follow the back propagation mechanism, quickly iterate and optimize the motion control strategy, and adjust the fin swing parameters in real time.

[0020] The present application adjusts the swing parameters of the fin by transmitting the flow field data to the BP neural network in real time through the pressure monitor in different environments, so that the fin swing reaches the optimal state to improve the efficiency of the underwater vehicle and save the energy of the underwater vehicle, solves the adaptive motion problem of underwater equipment, and realizes efficient and intelligent underwater operation. BRIEF DESCRIPTION OF DRAWINGS

[0021] FIG. 1 is a structural schematic diagram of the present application; Figure 1 is a structural schematic diagram of the present application;

[0022] FIG. 3 is a structural schematic diagram of the sensor cluster system of the present application; Figure 2 is a structural schematic diagram of the sensor cluster system of the present application;

[0023] Figure 1 is a structural schematic diagram of the present application; Figure 3 Figure 2 is a structural schematic diagram of the present application from a second angle;

[0024] Figure 3 is a structural schematic diagram of the present application from a third angle; Figure 4 Figure 4 is a structural schematic diagram of the present application bionic flexible wave propulsion system;

[0025] Figure 5 is a structural schematic diagram of the present application base internal structure; Figure 5 Figure 6 is a structural schematic diagram of the present application from a fourth angle;

[0026] In the drawings: 1, flexible fin surface, 2, base, 3, connecting frame, 4, fin strip, 5, toothless tether unit fixing frame, 6, longitudinal profile, 7, sensor cluster system, 8, synchronous belt, 9, toothless tether unit, 10, motor, 11, energy cabin section, 12, four groups of batteries, 13, main control cabin section, 14, control system, 15, main body frame type structure on the base. DETAILED DESCRIPTION

[0027] The present application will be further described below in conjunction with the drawings.

[0028] The present application provides a kind of based on BPNN bionic puffer fish fin self-adapting movement mechanism, as shown in Figure 1, including: base 2, the base 2 includes main control cabin section 13, energy cabin section 11, the main control cabin section 13 includes control system, the control system 14 includes main controller, pressure monitoring system based on BPNN neural network, signal conditioning circuit, the controller is connected with bionic flexible wave propulsion system, the signal conditioning circuit is connected with sensor cluster system 7, the sensor cluster system 7 is installed on the fin strip 4 of bionic fish fin of bionic flexible wave propulsion system. Figures 1-3 The bionic fish fin includes four parallelly arranged fin strips 4, and the fin strip 4 is covered by the flexible fin surface 1.

[0029] The fin strip 4 tail portion is provided with sensor cluster system 7, and is connected with the controller by signal line for transmitting control command and feedback signal.

[0030] The pressure monitoring system based on BPNN neural network includes feature extraction module, for extracting pressure feature;The feature extraction module is connected with BP neural network, for receiving the pressure feature so as to judge the parameter of fish fin swing and give judgment result.

[0031]

[0032] ​In this embodiment, the sensor cluster system 7 includes a set of pressure sensors, four pressure sensors are installed on the first fin 4, three pressure sensors are installed on the second fin 4, two pressure sensors are installed on the third fin 4, and one pressure sensor is installed on the fourth fin 4. The pressure sensors are respectively connected to the main control cabin section 13 through signal lines to transmit the measured pressure data to the main control cabin section 13. The first fin needs to be monitored with high precision because it is the leading fin, and the subsequent fins from the second fin onwards assist in monitoring, and so on, which can reduce the cost of expenditure.

[0033] The toothless system unit 9 includes a transmission shaft, one end of which is rotatably connected to the toothless shell, and the other end of which is sleeved with a synchronous pulley. The transmission shaft passes through the toothless shell and is connected to the crank rocker adapter plate. The crank rocker adapter plate is connected to one end of the crank rocker through a pin shaft. The other end of the crank rocker is connected to the intermediate transfer through a crank rocker bearing. The intermediate transfer is rotatably connected to the toothless shell through bearings. Each fin 4 is connected to the crank rocker adapter plate through an output base. The fin 4 is swingably arranged on the toothless shell.

[0034] The material of the flexible fin surface 1 includes a hydrophobic material, which is a polydimethylsiloxane film.

[0035] The fin uses shape memory alloy or carbon fiber.

[0036] The sensor cluster system 7 also includes temperature sensors and water flow sensors.

[0037] The base is placed at the bottom of the mechanism, and the base provides a stable support structure for the entire mechanism, ensuring that each component maintains a stable relative position in a complex underwater environment. Prevents displacement, deformation or loosening of the mechanism, ensuring the integrity and reliability of the mechanical structure. The base also provides a stable mounting position for the main control cabin and energy cabin, ensuring smooth connection of the power supply line, avoiding power interruption problems caused by poor connection, and ensuring continuous operation of the entire mechanism.

[0038] In this embodiment, the main control cabin is placed in the base, and the controller is installed with a hardware control system for controlling the movement of the entire thruster. The BPNN is closely connected with the pressure monitoring system and other sensors (such as temperature sensors, water flow sensors, etc.) to receive data collected from these devices. It quickly integrates and deeply analyzes the pressure data collected by the pressure monitor on the fish fin, as well as environmental parameters such as water temperature and water flow speed, through built-in complex algorithms and powerful computing capabilities. During operation, the BPNN continuously learns and adjusts itself based on new input data, optimizes its model parameters following the backpropagation mechanism, and then generates precise motion control instructions. These instructions are quickly transmitted to the biomimetic flexible undulating propulsion system to drive the joints and fish fins of the system to adjust the swing frequency, amplitude, and phase difference parameters according to the optimal strategy, achieving adaptive and efficient movement of the biomimetic fish underwater, and enabling the entire mechanism to flexibly respond to complex and variable underwater environments.

[0039] The energy cabin is placed in the base, and the energy cabin is used to provide power for the entire adaptive motion mechanism of the biomimetic fish.

[0040] Preferably, the controller is a waterproof controller.

[0041] As shown in the accompanying drawings, Figure 4 The biomimetic flexible undulating propulsion system includes a motor 10 installed on a longitudinal profile 6, the longitudinal profile 6 is installed on the base 2, the longitudinal profile 6 is connected with a connecting frame 3, the connecting frame 3 is connected with four toothless tie unit fixing frames 5, toothless tie units 9 are installed on the toothless tie unit fixing frames 5, the swing ends of the driving rods on the toothless tie units 9 are respectively connected with biomimetic fish fins, the synchronous pulley of the four toothless tie units 9 is respectively connected with the motor 10 through a synchronous belt 8, and the two ends of the biomimetic fish fins are respectively connected with the two ends of the connecting frame 3 through a hinge.

[0042] In this embodiment, the data collected by the pressure sensor is first transmitted to the signal conditioning circuit located in the master cabin section. In the signal conditioning circuit, the original pressure data signal is amplified, filtered, and analog-to-digital converted, and a series of processing operations are performed to convert it into a digital signal format suitable for subsequent processing. The pressure data after signal conditioning is transmitted to the feature extraction module in the pressure monitoring system of the BPNN neural network. This module extracts key feature information from the pressure data. Subsequently, these extracted feature information is input into the BP neural network. The BP neural network, based on the model and algorithm obtained through training, combines relevant environmental parameter information obtained from other sensors (such as temperature sensors, water flow sensors, etc.) to judge the parameters of fish fin oscillation and generate corresponding motion control instructions. These motion control instructions are finally sent to the biomimetic flexible undulating propulsion system to drive the fish fin and joints to move according to the optimal strategy, realizing the adaptive motion of the biomimetic fish underwater, enabling it to adjust its posture and propulsion mode in real time according to changes in water flow and other environmental changes.

[0043] Embodiment 1:

[0044] In combination with the accompanying Figures 1-5 In this embodiment, a BPNN-based adaptive motion mechanism of a biomimetic puffer fish fin is described. The mechanism includes a base, a master cabin section, an energy cabin section, a sensor cluster system, and a biomimetic flexible undulating propulsion system. The sensor cluster system is installed on the fin strip of the flexible fin surface of the biomimetic flexible undulating propulsion system. The master cabin section and the energy cabin section are installed on the main frame structure of the base. The outer shell is wrapped around the outside of the main frame structure. The biomimetic flexible undulating propulsion system is installed on the side of the main frame structure.

[0045] Embodiment 2:

[0046] In combination with the accompanying Figures 1-5 In this embodiment, a BPNN-based adaptive motion mechanism of a biomimetic puffer fish fin is described. The base includes a master cabin section, an energy cabin section, and a main frame structure. The energy cabin section is arranged at the lower middle position of the base and connected to the master cabin section and the biomimetic flexible undulating propulsion system through power lines. The master cabin section is placed above the base, near the center of the base and relatively high. The main frame structure is used to connect the master cabin section and the energy cabin section.

[0047] Embodiment 3:

[0048] In combination with the accompanying Figures 1-5 In this embodiment, a BPNN-based adaptive motion mechanism of a biomimetic puffer fish fin is described. The biomimetic flexible undulating propulsion system includes a motor, a flexible fin, a connecting frame, four toothless tie units, four toothless tie unit fixing frames, and four fin strips. The sensor cluster system is installed on the fin strip of the flexible fin surface of the biomimetic flexible undulating propulsion system.

[0049] The motor is fixedly installed on the longitudinal profile, one toothless series unit fixing frame 5 is installed on each toothless series unit, four toothless series units 9 are fixedly installed on the connecting frame in a straight line through four toothless series unit fixing frames, the swing end of the driving rod on each toothless series unit is fixedly connected with the flexible fin 1, any two adjacent toothless series units on the four toothless series units are synchronously controlled, the rotating shaft of the motor 10 is drivingly connected with any one toothless series unit, and the two ends of the flexible fin are respectively rotatably connected and installed on the two ends of the connecting frame 3 through hinges.

[0050] The synchronous pulleys on the two adjacent toothless series units on the four toothless series units are drivingly connected through synchronous belts, the rotating shaft of the motor is fixedly connected with the rotating shaft of any one synchronous pulley, and the two ends of the flexible fin are respectively rotatably connected and installed on the two ends of the connecting frame. The driving connection of the motor rotating shaft and any one toothless series unit is the core driving link of the whole power system. The motor, as a power source, efficiently converts electric energy into mechanical energy and accurately transmits torque to the toothless series unit connected thereto through the rotating shaft. Under the intelligent control of the BPNN algorithm, the motor can dynamically adjust the rotating speed and torque output according to real-time underwater environmental parameters and motion requirements, thereby driving the driving rod of the toothless series unit to produce a corresponding swing action, and finally realizing precise action control of the flexible fin and providing strong and reliable power guarantee for adaptive motion of the bionic fish.

[0051] In the embodiment, the fin strips are covered by the fin surface, the material of the fin strips is carbon fiber, and the material of the fin surface is selected from at least a hydrophobic material of a polydimethylsiloxane film.

[0052] Example 4:

[0053] In combination with the accompanying Figures 1-5 In the embodiment, the bionic puffer fish fin adaptive motion mechanism based on the BPNN includes a transmission shaft, a crank rocker, a crank rocker adapter plate, a transfer pair, a toothless series shell and two synchronous pulleys. One end of the transmission shaft penetrates the toothless series shell and is connected with the crank rocker adapter plate, and the transmission shaft is rotatably connected with the toothless series shell. The other end of the transmission shaft is sleeved with the two synchronous pulleys. The crank rocker adapter plate is connected with one end of the crank rocker through a pin shaft. The other end of the crank rocker is connected with the transfer pair through a crank rocker bearing. The two ends of the transfer pair are rotatably connected with the toothless series shell through bearings. Each swing rod is connected with the crank rocker adapter plate through an output base, and the swing rod is swingably arranged on the toothless series shell.

[0054] In the embodiment, the toothless system shell is provided with a slide for sliding the fin strip 4 upward and downward. One of the transmission shafts is driven to rotate by a motor, and the adjacent transmission shafts are sequentially driven to rotate by the synchronous pulleys on the transmission shafts, so that the transmission shafts of the four toothless system units are driven to rotate in different phases and at the same angular velocity. The swing speed of the undulating fin toothless system transmission mechanism can be controlled in real time by adjusting the rotation speed of the synchronous pulleys. The rotation phase difference of the undulating fin toothless system transmission mechanism can be controlled by changing the phase difference between the inclined crank rocker base shaft and the synchronous pulley. The swing amplitude of the output end can be modified by changing the relative angle of the crank rocker and the crank rocker adapter plate. The flexible fin structure of the bionic vehicle is driven by the swing of the swing rod to realize various sinusoidal undulating propulsion forms. In the embodiment, the four toothless system units are synchronously rotated by the synchronous pulleys and synchronous belts, so that the four toothless system units can simultaneously drive the swing rod to synchronously drive the flexible fin to realize sinusoidal driving work.

[0055] Example 5:

[0056] The accompanying drawings are incorporated in and constitute a part of this specification. Figures 1-5 In the embodiment, the energy cabin section includes four battery cabins, which are fixedly installed on the base and symmetrically arranged.

[0057] Experiments show that the bionic fish equipped with the mechanism has significantly enhanced environmental adaptability compared with traditional similar devices. In complex water flow scenarios, the self-adaptive adjustment time is shortened, and the propulsion energy efficiency is improved, which opens up a more precise and energy-saving technical path for underwater robots, marine scientific research equipment and other application scenarios.

[0058] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A biomimetic pufferfish fin adaptive motion mechanism based on BPNN, characterized in that, include: The base (2) includes a main control section (13) and an energy section (11). The main control section (13) includes a control system. The control system (14) includes a main controller, a pressure monitoring system based on a BPNN neural network, and a signal conditioning circuit. The controller is connected to the biomimetic flexible wave propulsion system. The signal conditioning circuit is connected to the sensor cluster system (7). The sensor cluster system (7) is installed on the fin rays (4) of the biomimetic fish fin of the biomimetic flexible wave propulsion system.

2. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 1, characterized in that, The biomimetic flexible wave propulsion system includes a motor (10), which is mounted on a longitudinal profile (6). The longitudinal profile (6) is mounted on a base (2). The longitudinal profile (6) is connected to a connecting frame (3). The connecting frame (3) is connected to four toothless system unit fixing frames (5). Toothless system units (9) are mounted on the toothless system unit fixing frames (5). The swing ends of the drive rods on the toothless system units (9) are respectively connected to the biomimetic fish fins. The synchronous pulleys of the four toothless system units (9) are respectively connected to the motor (10) through synchronous belts (8). The two ends of the biomimetic fish fins are respectively rotatably connected to the two ends of the connecting frame (3) through hinges.

3. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 1 or 2, characterized in that, The biomimetic fin includes four parallel fin rays (4), which are covered by the flexible fin surface (1).

4. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 3, characterized in that, The tail of the fin (4) has a sensor cluster system (7) which is connected to the controller via signal lines for transmitting control commands and feedback signals.

5. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 4, characterized in that, The pressure monitoring system based on the BPNN neural network includes a feature extraction module for extracting pressure features; The feature extraction module is connected to the BP neural network and is used to receive the pressure features to determine the parameters of the fin oscillation and provide the determination result.

6. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 4 or 5, characterized in that, The sensor cluster system (7) includes a set of pressure sensors. Four pressure sensors are installed on the first fin (4), three pressure sensors are installed on the second fin (4), two pressure sensors are installed on the third fin (4), and one pressure sensor is installed on the fourth fin (4). The pressure sensors are connected to the main control section (13) through signal lines to transmit the measured pressure data to the main control section (13).

7. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 6, characterized in that, The gearless system unit (9) includes a drive shaft. One end of the drive shaft is rotatably connected to the gearless system housing. The other end of the drive shaft is fitted with a synchronous pulley. One end of the drive shaft passes through the gearless system housing and is connected to the crank-rocker adapter plate. The crank-rocker adapter plate is connected to one end of the crank-rocker via a pin. The other end of the crank-rocker is connected to the intermediate rotation pair via a crank-rocker bearing. Both ends of the intermediate rotation pair are rotatably connected to the gearless system housing via bearings. Each fin (4) is connected to the crank-rocker adapter plate via an output base. The fin (4) is oscillating on the gearless system housing.

8. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 7, characterized in that, The flexible fin (1) is made of a hydrophobic material, which is a polydimethylsiloxane film.

9. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 8, characterized in that, The fins are made of shape memory alloy or carbon fiber.

10. The biomimetic pufferfish fin adaptive motion mechanism based on BPNN according to claim 9, characterized in that, The sensor cluster system (7) also includes a temperature sensor and a water flow sensor.

Images