CN112357030B - A water quality monitoring machine fish for ocean or inland river lake - Google Patents

Abstract

The invention discloses a water quality monitoring robot fish for oceans or inland lakes. When the robot fish floats and dives, it mainly rotates a duct propeller, and has the characteristics of fast and stable starting speed. In addition, the rotation of the steering gear drives the pectoral fins to rotate, so as to realize free movement in space. The invention adopts the water jet propeller for propulsion. When encountering danger, the water jet propeller is used for main push, and the fish tail swings for auxiliary propulsion, so as to escape the danger. In the safe area, fish tail propulsion is used, which not only saves energy consumption, but also reasonably completes the detection task. The water quality monitoring system and target identification system described in the present invention can obtain its water quality parameters in real time and can identify precious fish for tracking and protection.

Description

A robotic fish for water quality monitoring in oceans or inland lakes

technical field

The invention relates to a robotic fish used for water quality monitoring, belonging to the technical field of ship engineering.

Background technique

With the continuous development of the economy, land resources are increasingly consumed, and the ocean occupies 71% of the earth's surface area, and people's exploration of the ocean is still in its infancy, so the development of marine resources is particularly important. As the main tool for ocean exploration and underwater operations, robotic fish can work in relatively dangerous underwater environments, and can well adapt to changing underwater environments. At present, robotic fish are mainly used in data monitoring of the marine environment, maintenance of underwater equipment, acquisition and identification of marine targets, etc.

However, at present, the main propulsion method of the robotic fish is caudal fin propulsion, and its control method is to rely on the swing of the steering gear to drive the fish tail movement or by applying shape memory alloy (SMA) drive technology to control the force and torque of the fish tail, so as to advance the robotic fish. . The former is slow and the latter is expensive. After investigation, the current speed of the robotic fish is generally 4-5 knots, and the cost is about 200,000.

Aiming at the disadvantages of slow speed and high cost of the current robotic fish, the present invention designs a new type of robotic fish for water quality monitoring. The robotic fish has the advantages of fast speed, high efficiency, low cost and wide application range. After relevant tests, the speed can generally reach more than 5 knots, and the cost is about 100,000 yuan.

The invention designs a robotic fish for water quality monitoring, which can realize free movement in space, realize water quality monitoring by carrying sensors, and can change its own speed according to different scenarios. When detecting organisms such as algae corals and water quality, low-speed sailing is used, and the driving method is to rely on the swing of the fish tail. When encountering dangerous creatures, high-speed sailing is adopted, and the driving mode is water jet propulsion as the main propulsion and fish tail as auxiliary propulsion. On the one hand, energy is saved, on the other hand, the operation is simple and convenient, and the efficiency is high. At the same time, it is equipped with target recognition technology, which can identify some endangered aquatic species, and can detect their living environment by carrying corresponding sensors, so that different schemes can be formulated to better protect them.

SUMMARY OF THE INVENTION

The object of the present invention is achieved through the following technical solutions:

A water quality monitoring robot fish for oceans or inland lakes, comprising a robot fish main body, a sealed cabin, an intelligent maneuvering and propulsion system, a tail propulsion mechanism, an intelligent monitoring system, a target recognition system, a semi-autonomous navigation system and a communication system, and its characteristics are: The main body of the robotic fish is provided with pectoral fins located on both sides of the robotic fish chest and a caudal fin located at the tail of the robotic fish, and the robotic fish body is provided with a sealed cabin;

The intelligent operation and push system is configured to be able to drive the main body of the robotic fish and drive the pectoral fins to rotate, thereby changing the heading angle of the robotic fish;

The tail propulsion mechanism is installed at the rear of the main body of the robotic fish, and the tail propulsion mechanism is configured to drive the tail fin to swing back and forth;

The intelligent monitoring system is configured to at least collect water turbidity, water depth and pH value, and can transmit the information to the remote controller through wireless communication;

The target recognition system can automatically capture images and can automatically identify the category of the target object, so that the robotic fish can make a reaction action corresponding to the category of the target object;

The semi-autonomous navigation system is used for wireless reception and GPS positioning, so as to realize the semi-autonomous navigation of the robotic fish;

The communication system is used for the wireless communication between the robotic fish and the remote controller, so as to facilitate the remote control of the movement and operation of the robotic fish.

Further, preferably, the robotic fish body (1) includes a fish head, a fish body, and a fish tail, and the robotic fish body is spindle-shaped as a whole, with a total length of 0.61.2m, an aspect ratio of 3:1, and a length-to-height ratio of 3:1. The ratio is 4:1, the length to total length ratio of the caudal fin is about 4:1, and the aspect ratio is about 6:1.

Further, as a preference, the shape of the sealed cabin is cylindrical, and is located in the fish body of the robotic fish, the length of the sealed cabin is one-half of the total length of the robotic fish, and the radius of the sealed cabin is the maximum width of the robotic fish. In one half, the sealed cabin is fixedly connected to the main body of the robotic fish through the upper plate, the front plate, the first lower plate, the second lower plate and the rear plate.

Further, as preferably, the intelligent maneuvering and pushing system comprises a water jet, a duct propeller and a steering gear, wherein the duct propeller is located at the fish head position of the robotic fish, and the outer part of the duct propeller is sleeved with a circular tube, And the circular tube and the duct propeller are fixedly connected by two screws; the water jet propeller is located at the bottom of the robotic fish, and is fixedly connected to the main body of the robotic fish through the first lower plate and the second lower plate, and the two steering gears are It is located on both sides of the fish body, and the output ends are connected to the pectoral fins, and the swing of the steering gear drives the pectoral fins to rotate, thereby changing the heading angle.

Further, preferably, the tail propulsion mechanism includes a brushless deceleration motor, a bevel gear 1, a bevel gear 2, a synchronous wheel 1, a synchronous wheel 2, a timing belt and a tail fin, wherein the brushless deceleration motor is fixed by a motor fixing member On the base plate, the base plate is fixedly connected to the main body of the robotic fish through the support plate 3 and the support plate 4, and the shaft 1 of the output end of the brushless reduction motor is connected to the shaft 2 through the bevel gear 1 and the bevel gear 2 that mesh with each other. Two are rotatably supported through the bottom plate, and the second shaft is fixedly connected with the first synchronizing wheel, the second synchronizing wheel is fixed on the third axis, and the third axis is rotatably supported through the bottom plate, and is located in the At the rear of the second shaft, the synchronous belt is driven and wound between the first and second synchronizing wheels, and the tail fin and the second synchronizing wheel are connected by a connecting plate. By controlling the rotation of the brushless deceleration motor, the Gear 1 and bevel gear 2 drive the synchronizing wheel 1 to rotate, and then drive the synchronizing wheel 2 to rotate through the synchronous belt, thereby driving the tail fin to swing back and forth.

Further, preferably, the intelligent monitoring system includes a turbidity sensor, a water depth sensor, a pH sensor, a single-chip microcomputer and a remote control, wherein the turbidity sensor, the water depth sensor and the pH sensor are arranged at the rear of the sealed cabin, and the The single-chip microcomputer is arranged inside the sealed cabin. The turbidity sensor, water depth sensor and pH value sensor are connected with the single-chip microcomputer through wires. on the remote control.

Further, preferably, the target recognition system includes a camera for capturing images, a wireless serial port module, and a remote control screen, wherein the camera is arranged at the fish head, the wireless serial port module is located inside the sealed cabin, and the remote control screen is arranged on the shore At the same time, the underwater image is transmitted to the screen of the remote controller through the wireless serial port module, and the target recognition system is configured to be able to use the target recognition yolov model for recognition.

Further, preferably, the semi-autonomous navigation system includes a single-chip microcomputer, a GPS module, and a receiver, and the single-chip microcomputer, the GPS module, and the receiver are arranged inside the sealed cabin, and the semi-autonomous navigation system adopts a fuzzy algorithm. The measurement value obtained by the module calculates the error and the error rate of change, quantizes it according to the quantization factor, obtains the quantized error and the error rate of change, and finally obtains the control amount according to the fuzzy control table, so as to realize semi-autonomous navigation.

Further, preferably, the communication system includes a receiver, an antenna, and a remote controller, the receiver is arranged inside the sealed cabin, the antenna is arranged on the water surface, and the remote controller is arranged on the shore. The receiver of the fish is connected to the antenna, so that the remote control communicates with the robotic fish, controls the movement and operation of the robotic fish, and can receive underwater images and sensor information in real time.

Further, as a preference, the establishment of the target recognition yolov3 model model is specifically: select Baidu flying paddle and marine fish data set as the training data set, first through the labelImg label tool, the objects on each picture are marked, and made into xml format label file, import the prepared dataset into the model, YOLO divides the input image into SxS grids, if the coordinates of the center position of an object fall into a grid, then this grid is responsible for detecting the object , using the entire image as the input of the network, directly regressing the position of the bounding box and the category to which it belongs in the output layer, and then using batch normalization as the scale training, using regularization, accelerating convergence and avoiding overfitting, the BN layer After the leaky relu layer is connected to each convolutional layer, the final training result is obtained through multi-scale training, and finally the trained weight is used for target detection, so as to identify the more precious fish, so as to better protect the wildlife. .

Compared with the prior art, the beneficial effects of the present invention are:

(1) The manipulation system of the present invention is different from the conventional robotic fish manipulation system. It is first reflected in the heave motion. The current robotic fish mainly relies on the swing of the pectoral fins when it floats up and down, but it requires a very fast initial speed to can be realised. This results in a slow and unstable response. When the robotic fish of the present invention floats and dives, it is mainly through the rotation of the duct propeller, and has the characteristics of fast and stable starting speed. In addition, the rotation of the steering gear drives the pectoral fins to rotate, so as to realize free movement in space.

(2) Secondly, it is reflected in the horizontal movement. The current robotic fish relies on the swing of the tail fin to move forward, and the gap between the maximum speed and the minimum speed is not large. When encountering dangerous creatures, it is difficult to escape the danger only by swinging the tail fin. In view of this point, the present invention adopts water jet propulsion for propulsion. When encountering danger, the water jet propulsion is used for main push, and the fish tail swings for auxiliary propulsion, so as to escape the danger. In the safe area, fish tail propulsion is used, which not only saves energy consumption, but also reasonably completes the detection task.

(3) The water quality monitoring system and target identification system described in the present invention can obtain its water quality parameters in real time and can identify precious fish for tracking protection.

Description of drawings

Fig. 1 is the overall oblique side view of the present invention;

Fig. 2 is the antenna schematic diagram of the present invention;

FIG. 3 is a schematic diagram of the sealed cabin support system of the present invention;

Fig. 4 is the schematic diagram of the operating push system of the present invention;

5 is an oblique second view of the caudal fin propulsion system of the present invention;

6 is a schematic side view of the caudal fin propulsion system of the present invention;

FIG. 7 is a schematic plan view of the caudal fin propulsion system of the present invention;

Figure 8 is a schematic side view of the robotic fish of the present invention;

Fig. 9 is the schematic plan view of the robotic fish of the present invention;

10 is a schematic diagram of the front view of the robotic fish of the present invention;

11 is a schematic diagram of two oblique sides of the remote controller of the present invention;

Figure 12 is a layout view of the capsule of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Please refer to Figures 1-12, the present invention provides a technical solution: a water quality monitoring robot fish for oceans or inland lakes of the present invention, which includes a robot fish body 1, a sealed cabin 2, an intelligent operation and push system 3, a tail The propulsion mechanism 4, the intelligent monitoring system 5, the target recognition system 6, the semi-autonomous navigation system 7 and the communication system 8 are characterized in that, the robot fish main body 1 is provided with pectoral fins 3-3 located on both sides of the robot fish chest and The caudal fins 4-17 at the tail of the robotic fish, the robotic fish body 1 is provided with a sealed cabin 2;

The intelligent operation and push system 3 is configured to be able to drive the main body 1 of the robotic fish and drive the pectoral fins 3-3 to rotate, thereby changing the heading angle of the robotic fish;

The tail propulsion mechanism 4 is installed at the rear of the main body of the robotic fish, and the tail propulsion mechanism 4 is configured to drive the tail fins 4-17 to swing back and forth;

The intelligent monitoring system 5 is configured to at least collect water turbidity, water depth and pH value, and can transmit the information to the remote controller 5-5 by means of wireless communication;

The target recognition system 6 can automatically capture images and automatically identify the category of the target object, so that the robotic fish can make a reaction action corresponding to the category of the target object;

The semi-autonomous navigation system 7 is used for wireless reception and GPS positioning, so as to realize the semi-autonomous navigation of the robotic fish;

The communication system 8 is used for the wireless communication between the robotic fish and the remote controllers 5-5, so as to facilitate the remote control of the movement and operation of the robotic fish.

As shown in FIG. 1 , the main body 1 of the robotic fish includes a fish head, a fish body, and a fish tail. It is spindle-shaped as a whole, with a total length of 0.6-1.2m, an aspect ratio of 3:1, and a length-to-height ratio of 4:1.

As shown in FIG. 3 , the sealed cabin 2 is cylindrical in shape, located in the body of the robotic fish, and its length is about half of the length of the ship. The radius is about one-half of the ship's width, which is fixed to the main body of the robotic fish through the upper plate 2-1, the front plate 2-2, the lower plate 1 2-3, the lower plate 2-4, and the rear plate 2-5. connect.

As shown in Figure 4, the intelligent operation and push system includes water jet thrusters 3-5, ducted propellers 3-1, and steering gears 3-4. Use two screws 3-6 to fix the round tube 3-2 and the duct propeller 3-1. The water jet 3-5 is located at the bottom of the robotic fish, which passes through the lower plate 1 2-3 and the lower plate 2 2- 4 Make a fixed connection with the main body of the robotic fish. The two steering gears 3-4 are located on both sides of the fish body, and the output ends are connected to the pectoral fins 3-3. The working process mainly relies on the swing of the steering gear 3-4 to drive the pectoral fins 3-3 to rotate. , so as to change the heading angle.

As shown in Figure 5-7, the tail propulsion mechanism includes a brushless reduction motor 4-1, a bevel gear 4-4, a bevel gear 2 4-5, a synchronous wheel 1 4-12, a synchronous wheel 2 4-15, The timing belt 4-8 and the tail fin 4-17, wherein the brushless deceleration motor 4-1 is fixed on the base plate 4-16 through the motor fixing member 4-2, and the base plate 4-16 is supported by the support plate 3 4-10. The plate four 4-11 is fixedly connected with the main body 1 of the robotic fish, and the shaft one 4-3 at the output end of the brushless deceleration motor 4-1 is connected to the shaft through the bevel gear one 4-4 and the two bevel gears 4-5 that mesh with each other. Two 4-6, the second shaft 4-6 is rotatably supported and disposed through the bottom plate 4-8, and the second shaft is fixedly connected with the first synchronizing wheel 4-12, and the second synchronizing wheel 4-15 is fixed on the shaft On the third axis 4-9, the third axis 4-9 is rotatably supported through the bottom plate 4-8, and is located at the rear of the second axis. The first synchronizing wheel 4-12 and the second synchronizing wheel 4-15 The synchronous belt 4-16 is arranged around the transmission between the two, and the connecting plate 4-14 is used to connect the tail fin 4-17 and the second synchronous wheel 4-15. By controlling the brushless deceleration motor 4-1 to rotate, the cone Gear 1 4-4 and bevel gear 2 4-5 drive the synchronizing wheel 1 4-12 to rotate, and then drive the synchronizing wheel 2 4-15 to rotate through the synchronous belt 4-8, thereby driving the tail fin 4-17 to swing back and forth.

As shown in Figure 9, Figure 11, Figure 12, the intelligent monitoring system consists of a turbidity sensor 5-1, a water depth sensor 5-2, a pH sensor 5-3, a microcontroller 5-4, and a remote control 5-5. Among them, the turbidity sensor 5-1, the water depth sensor 5-2, and the ph value sensor 5-3 are placed in the rear of the airtight chamber 2, and the single-chip microcomputer 5-4 is placed in the airtight chamber 2. Turbidity sensor 5-1, water depth sensor 5-2, ph value sensor 5-3 are connected with microcontroller 5-4 through wires. way to transmit information to the remote control 5-5.

As shown in FIG. 8 , the target recognition system is composed of a camera 6-1, a wireless serial port module 6-2, and a remote control screen 6-3. Among them, the camera 6-1 is placed at the head of the fish, the wireless serial port module 6-2 is placed inside the sealed cabin, and the remote control screen 6-3 is placed on the shore. The image is captured by the camera 6-1, and the underwater image can be transmitted to the remote control screen 6-3 through the wireless serial port module 6-2, and the modern target recognition yolov3 model is used for recognition. In order to be able to identify categories such as fish, garbage bags, plastic bottles, etc., the Baidu Flying Paddle and Ocean Fish datasets are selected as training datasets. First, use the labelImg label tool to label the objects on each image and make a label file in xml format. Import the prepared dataset into the model, and YOLO divides the input image into SxS grids. If the coordinates of the center position of the Ground truth of an object fall into a grid, then this grid is responsible for detecting the object. Using the entire image as the input of the network, the position of the bounding box and the category to which it belongs are directly returned at the output layer. Then batch normalization is used as scale training. The BN layer and the leaky relu layer are connected after each convolutional layer using regularization, accelerating convergence and avoiding overfitting. The final training results are obtained through multi-scale training. Finally, the trained weights are used for target detection, so that more precious fish can be identified, so that these wildlife can be better protected.

As shown in Figure 12, the semi-autonomous navigation system is composed of a single chip 5-4, a GPS module 7-1, and a receiver 7-2. The single chip 5-4 and the GPS module 7-1, the receiver 7-2 is placed inside the sealed cabin 2, the algorithm used is a fuzzy algorithm, and the error and the error change are calculated by obtaining the expected value and the measurement value obtained by the GPS module 7-1 According to the quantization factor, it is quantized to obtain the quantized error and the error rate of change, and finally the control amount is obtained according to the fuzzy control table, so as to realize semi-autonomous navigation.

As shown in FIG. 2, FIG. 11, and FIG. 12, the communication system is composed of a receiver 7-2, an antenna 8-1, and a remote control 5-5. The receiver 7-2 is located inside the sealed cabin 2, the antenna 8-1 is placed on the water surface, and the remote control 5-5 is placed on the shore. Its working principle is to connect the receiver 7-2 of the robotic fish to the antenna 8-1 by means of a signal wire, so that the remote control 5-5 communicates with the robotic fish, controls the movement and operation of the robotic fish, and can control the movement and operation of the robotic fish. Receive underwater images and sensor information in real time.

When the robot fish of the present invention floats and dives, it is mainly through the rotation of the duct propeller, and has the characteristics of fast and stable starting speed. In addition, the rotation of the steering gear drives the pectoral fins to rotate, so as to realize free movement in space. The invention adopts the water jet propulsion for propulsion. When encountering danger, the water jet propulsion is used for the main push, and the fish tail swings for auxiliary propulsion, so as to escape the danger. In the safe area, fish tail propulsion is used, which not only saves energy consumption, but also reasonably completes the detection task. The water quality monitoring system and target identification system described in the present invention can obtain its water quality parameters in real time and can identify precious fish for tracking protection.

Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, and substitutions can be made in these embodiments without departing from the principle and spirit of the invention and modifications, the scope of the present invention is defined by the appended claims and their equivalents.

Claims (5)

1. A water quality monitoring robot fish for oceans or inland lakes comprises a robot fish main body (1), a sealed cabin (2), an intelligent pushing system (3), a tail pushing mechanism (4), an intelligent monitoring system (5), a target identification system (6), a semi-autonomous navigation system (7) and a communication system (8), and is characterized in that pectoral fins (3-3) positioned on two sides of the chest of the robot fish and tail fins (4-17) positioned at the tail of the robot fish are arranged on the robot fish main body (1), and the sealed cabin (2) is arranged in the robot fish main body (1);

the intelligent control and push system (3) is arranged to be capable of driving the robot fish main body (1) to be pushed and driving the pectoral fins (3-3) to rotate, so that the course angle of the robot fish is changed;

the tail propulsion mechanism (4) is arranged at the rear part of the robot fish main body, and the tail propulsion mechanism (4) is arranged to drive tail fins (4-17) to swing back and forth;

the intelligent monitoring system (5) is at least capable of collecting water turbidity, water depth and PH value and transmitting information to the remote controller (5-5) in a wireless communication mode;

the target recognition system (6) can automatically capture images and automatically recognize the category of a target object so as to facilitate the robot fish to make a reaction action corresponding to the category of the target object;

the semi-autonomous navigation system (7) is used for wireless receiving and GPS positioning so as to realize semi-autonomous navigation of the robot fish;

the communication system (8) is used for the robot fish to wirelessly communicate with the remote controller (5-5) so as to be convenient for remotely controlling the movement and operation of the robot fish;

the robot fish main body (1) comprises a fish head, a fish body and a fish tail, and the whole robot fish main body is in a spindle shape, the total length is 0.6-1.2m, the length-width ratio is 3:1, and the length-height ratio is 4: 1;

the sealed cabin (2) is cylindrical and is positioned in the fish body of the robot fish, the length of the sealed cabin (2) is one half of the total length of the robot fish, the radius of the sealed cabin (2) is one half of the maximum width of the robot fish, and the sealed cabin (2) is fixedly connected with the robot fish main body through an upper plate (2-1), a front plate (2-2), a lower plate I (2-3), a lower plate II (2-4) and a rear plate (2-5);

the intelligent monitoring system comprises a turbidity sensor (5-1), a water depth sensor (5-2), a pH value sensor (5-3), a singlechip (5-4) and a remote controller (5-5), wherein the turbidity sensor (5-1), the water depth sensor (5-2) and the PH value sensor (5-3) are arranged at the rear part of the sealed cabin (2), the single chip microcomputer (5-4) is arranged in the sealed cabin (2), the turbidity sensor (5-1), the water depth sensor (5-2) and the PH value sensor (5-3) are connected with the single chip microcomputer (5-4) through leads, the water turbidity, the water depth and the PH value are collected through a programming singlechip, so that information is transmitted to a remote controller (5-5) in a wireless communication mode;

the target recognition system comprises a camera (6-1) for capturing images, a wireless serial port module (6-2) and a remote controller screen (6-3), wherein the camera (6-1) is arranged at the position of the head of a fish, the wireless serial port module (6-2) is positioned in a sealed cabin, the remote controller screen (6-3) is arranged on the shore, underwater images are transmitted to the remote controller screen (6-3) through the wireless serial port module (6-2), and the target recognition system is constructed to be capable of recognizing by adopting a target recognition yolov3 model;

the specific method for establishing the target identification yolov3 model comprises the following steps: a Baidu flying oar and marine fish data set is selected as a training data set, objects on each picture are labeled through a labelImg labeling tool to manufacture a label file in an xml format, the manufactured data set is led into a model, an input image is divided into SxS grids through YOLO, if the coordinate of the center position of a certain object falls into a certain grid, the grid is responsible for detecting the object, the whole picture is used as the input of a network, the position of a boundary frame and the category of the boundary frame are directly regressed in an output layer, then batch normalization is used as scale training, a regularization, accelerated convergence and over-fitting avoidance method is adopted, a BN layer and a leaky relu layer are layered on each layer, a final training result is obtained through multi-scale training, and finally the trained weight is subjected to target detection.

2. The water quality monitoring robotic fish for oceans or inland lakes according to claim 1, which is characterized in that: the intelligent control and push system (3) comprises a water jet propeller (3-5), a ducted propeller (3-1) and a steering engine (3-4), wherein the ducted propeller (3-1) is located at the position of the fish head of the robot fish, a round pipe is sleeved outside the ducted propeller (3-1), and the round pipe is fixedly connected with the ducted propeller (3-1) through two screws (3-6); the water jet propeller (3-5) is located at the bottom of the robot fish and fixedly connected with the robot fish main body through a first lower plate (2-3) and a second lower plate (2-4), the two steering engines (3-4) are located on two sides of the fish body, output ends of the two steering engines are connected to the pectoral fins (3-3), and the swing of the steering engines (3-4) drives the pectoral fins (3-3) to rotate so as to change course angles.

3. The water quality monitoring robotic fish for oceans or inland lakes according to claim 1, which is characterized in that: the tail propelling mechanism comprises a brushless speed reducing motor (4-1), a bevel gear I (4-4), a bevel gear II (4-5), a synchronizing wheel I (4-12), a synchronizing wheel II (4-15), a synchronous belt (4-8) and a tail fin (4-17), wherein the brushless speed reducing motor (4-1) is fixed on a bottom plate (4-16) through a motor fixing piece (4-2), the bottom plate (4-16) is fixedly connected with a robot fish main body (1) through a support plate III (4-10) and a support plate IV (4-11), a shaft I (4-3) at the output end of the brushless speed reducing motor (4-1) is in transmission connection with a shaft II (4-6) through the bevel gear I (4-4) and the bevel gear II (4-5) which are meshed with each other, the second shaft (4-6) is rotatably supported and penetrates through the bottom plate (4-16) and is fixedly connected with the first synchronous wheel (4-12), the second synchronous wheel (4-15) is fixed on the third shaft (4-9), the third shaft (4-9) is rotatably supported and penetrates through the bottom plate (4-16) and is positioned at the rear part of the second shaft, the synchronous belt (4-8) is wound between the first synchronous wheel (4-12) and the second synchronous wheel (4-15), the tail fin (4-17) and the second synchronous wheel (4-15) are connected by a connecting plate (4-14), the first synchronous wheel (4-12) is rotated by controlling the rotation of the brushless speed reducing motor (4-1) and by the transmission of the first bevel gear (4-4) and the second bevel gear (4-5), and then the synchronous wheels II (4-15) are driven to rotate through the synchronous belts (4-8), so that the tail fins (4-17) are driven to swing back and forth.

4. The water quality monitoring robotic fish for oceans or inland lakes according to claim 1, which is characterized in that: the semi-autonomous navigation system comprises a single chip microcomputer (5-4), a GPS module (7-1) and a receiver (7-2), wherein the single chip microcomputer (5-4), the GPS module (7-1) and the receiver (7-2) are arranged inside a sealed cabin (2), the semi-autonomous navigation system adopts a fuzzy algorithm, calculates errors and error change rates by obtaining expected values and measured values obtained by the GPS module (7-1), performs quantization processing on the errors and the error change rates according to quantization factors to obtain quantized errors and error change rates, and finally obtains control quantities according to a fuzzy control table, so that semi-autonomous navigation is realized.

5. The water quality monitoring robotic fish for oceans or inland lakes according to claim 1, which is characterized in that: the communication system comprises a receiver (7-2), an antenna (8-1) and a remote controller (5-5), wherein the receiver (7-2) is arranged inside the sealed cabin (2), the antenna (8-1) is arranged on the water surface, the remote controller (5-5) is arranged on the shore, and the receiver (7-2) of the robot fish is connected with the antenna (8-1) through a lead wire in a signal wire mode, so that the remote controller (5-5) is communicated with the robot fish, the movement and operation of the robot fish are controlled, and underwater images and sensor information can be received in real time.

Images