Water leakage detection and protection method for underwater robot multi-sensor information fusion

By adopting a multi-sensor information fusion method in underwater robots, combining water leakage, humidity and pressure sensors, the weighted average algorithm is used to improve detection accuracy and anti-interference ability, the problem of insufficient accuracy of a single sensor is solved, and timely leakage detection and protection is achieved Effect.

CN111307380BActive Publication Date: 2024-11-22FUJIAN SPECIAL EQUIP TESTING RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010239865.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-03-30
Publication Date
2024-11-22
Estimated Expiration
2040-03-30

AI Technical Summary

Technical Problem

The existing underwater robot leak detection device uses a single sensor, which has insufficient accuracy and poor anti-interference ability, resulting in false alarms and waste of resources.

Method used

Multi-sensor information fusion method is adopted, including water leakage sensors, humidity sensors and pressure sensors. Through signal acquisition, signal processing and information fusion, the weighted average algorithm is used to improve detection accuracy and anti-interference ability, and the power-off protection device is cut off when water leakage is detected.

Benefits of technology

It effectively improves the anti-interference ability and accuracy of the detection system, reduces false alarms, prompt power outage protection, and reduces the risk of component damage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111307380B_ABST
    Figure CN111307380B_ABST
Patent Text Reader

Abstract

The present invention provides a water leakage detection and protection method for multi-sensor information fusion of an underwater robot, comprising the following steps: step S1, installing a plurality of water leakage sensors at the edge of a door of a cabin of the underwater robot which is prone to water leakage; step S2, installing humidity sensors for detecting humidity at the head, middle and tail of the cabin of the underwater robot; step S3, installing pressure sensors inside and outside the cabin of the underwater robot; step S4, starting the underwater robot and starting signal collection of multiple sensors; step S5, removing interference signals through signal processing and extracting signal features from the collected signals; step S6, a controller performs decision output, and if water leakage occurs inside the cabin of the underwater robot, an upper computer performs an alarm prompt; the present invention can effectively improve the anti-interference ability and detection accuracy of the detection system through information fusion of multiple sensors.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of underwater robots, in particular to a water leakage detection and protection method for multi-sensor information fusion of underwater robots. Background Art

[0002] Underwater robots are extreme operation robots that work underwater. The underwater environment is harsh and dangerous. Due to the limited diving depth of humans, underwater robots can replace humans to work underwater for a long time in highly dangerous environments, polluted environments, and waters with zero visibility. Therefore, underwater robots have become an important tool for marine development.

[0003] Underwater robots are generally equipped with sonar systems, cameras, lighting, controllers and other devices. In deep water, the sealed cabin may leak due to excessive water pressure, causing the underwater robot to fail to work normally, and even causing short-circuit damage to internal devices and components, which is extremely harmful.

[0004] Currently, water leakage detection devices on the market generally use a single sensor (such as a water leakage sensor or a pressure sensor) to identify whether there is a water leak in the cabin by collecting sensor signals. Due to insufficient sensor accuracy and anti-interference capabilities, false alarms sometimes occur, resulting in a waste of human maintenance resources. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a water leakage detection and protection method for underwater robot multi-sensor information fusion, which can effectively improve the anti-interference ability and detection accuracy of the detection system through multi-sensor information fusion. The present invention adopts the following method to achieve: a water leakage detection and protection method for underwater robot multi-sensor information fusion, characterized in that:

[0006] Step S1, installing a plurality of water leakage sensors at the edge of the door of the underwater robot cabin that is prone to water leakage, and the water leakage sensors are electrically connected to the controller of the underwater robot;

[0007] Step S2, humidity sensors for detecting humidity are installed at the front, middle and rear parts of the cabin of the underwater robot, and the humidity sensors are electrically connected to the controller of the underwater robot;

[0008] Step S3, installing pressure sensors inside and outside the cabin of the underwater robot, and the pressure sensors are electrically connected to the controller of the underwater robot;

[0009] Step S4, starting the underwater robot and starting signal collection of multiple sensors;

[0010] Step S5: After the signal is collected, the interference signal is filtered and removed through signal processing. After the interference signal is removed, the controller fuses the information obtained by the water leakage sensor, the humidity sensor and the pressure sensor;

[0011] Step S6: After fusion, if there is no water leakage inside the underwater robot cabin, return to step S4; if there is a water leakage inside the underwater robot cabin, issue an alarm to the host computer;

[0012] Step S7: the alarm prompt starts, and the water leakage protection device of the underwater robot starts to operate to perform water leakage protection.

[0013] Furthermore, the signal acquisition of the multi-channel sensors in step S4 is that the water leakage sensor outputs high and low level signals, the high level is 3.3V, and the low level is 0V; the pressure sensor and the humidity sensor are converted into analog voltage signals of 0-3.3V by the signal conditioning circuit, the output of the water leakage sensor is connected to the IO port of the DSP, and the outputs of the pressure sensor and the humidity sensor are connected to the ADC module of the DSP, and the ADC sampling channels are ADCINA0-ADCINA3 respectively.

[0014] Furthermore, the signal processing in step S5 is used for data preprocessing after signal acquisition, and a digital filter is used to remove the collected interference signal, leaving the signal that actually matches it.

[0015] Furthermore, the information fusion in step S5 adopts a weighted average algorithm, and the three types of sensors, namely, the pressure sensor, the humidity sensor and the water leakage sensor, are first normalized and then weighted averaged. Each sensor has a weight value; the data of the sensor with high precision has a small error and a large weight; while the data of the sensor with low precision has a large error and a small weight; the digital quantities collected by the pressure sensor are x1, x2, x3....x i , get the average value n is a natural number, K1 is the inverse of the hardware conditioning circuit gain, and the collected average value is converted into the real pressure value. P is the pressure value of the pressure sensor; the digital values ​​collected by the humidity sensor are y1, y2, y3....y i , get the average value n is a natural number, K2 is the inverse of the gain of the hardware conditioning circuit, and the collected average value is converted into the true value of humidity. H is the humidity value of the humidity sensor; and the result of the water leakage sensor is Where R i is the result of each water leakage sensor, L is the set of all water leakage sensor results, and each water leakage sensor outputs a high level, then R i =1, output low level, R i=0; convert the sampled values ​​of the three types of sensors, namely, pressure sensor, humidity sensor and water leakage sensor, into recognizable results and perform weighting;

[0016] 1) Pressure sensor judgment criteria: P 外 -P 内 ≥P0 (P0 is the threshold), it is judged as water leakage; P 外 ≤P 内 , judged to be watertight;

[0017] 2) Humidity sensor judgment criteria: H≥H1, H1 is generally taken as 85%, which is considered to be leaking; otherwise, H≤H2, which is considered to be not leaking, and H2 is taken as 75%;

[0018] 3) Leakage sensor judgment criteria: L = 1, judged as water leakage; L = 0, judged as no water leakage; the judgment results of the above sensors are output as 0 or 1, 1 represents water leakage, and 0 represents no water leakage;

[0019] Perform weighted summation on the three types of sensors and output the final result; among them, the pressure sensor, humidity sensor, and water leakage sensor are judged by the criteria as T1, T2, and T3 respectively; the weight coefficients of the pressure, humidity, and water leakage sensors are w1, w2, and w3 respectively, among which w1+w2+w3=1, and the output result is T=w1T1+w2T2+w3T3, 0≤T≤1, and the comparison is made based on T.

[0020] Furthermore, the water leakage protection in step S7 is to determine whether the system is leaking after the decision output, and the controller outputs an audible and visual alarm, the host computer lights up a red light, and the buzzer sounds a long sound; about 5 seconds after the alarm, the controller outputs a control signal to control the underwater robot to cut off the power, and the alarm of the host computer continues until the power supply of the underwater robot is restored and the water leakage fault is eliminated.

[0021] The beneficial effects of the present invention are as follows: the present invention adds a water leakage sensor, a humidity sensor and a pressure sensor, and adopts information fusion of multiple sensors, which can effectively improve the anti-interference ability and detection accuracy of the detection system; overcome the problem of insufficient accuracy of a single sensor; after the underwater robot detects a water leakage, the present invention effectively cuts off the power supply to reduce the degree of damage to components. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the present invention.

[0023] Figure 2 This is the water leakage detection control block diagram.

[0024] Figure 3 A block diagram of a fault is identified for the decision output.

[0025] Figure 4 This is a fault identification table.

[0026] Figure 5 is a circuit schematic diagram of the LED driving circuit.

[0027] Figure 6 The schematic diagram of the buzzer alarm driving circuit is shown in FIG.

[0028] Figure 7 4 is a circuit schematic diagram of the low-pass filtering circuit.

[0029] Figure 8 is a circuit schematic diagram of the main control module. DETAILED DESCRIPTION

[0030] The present invention will be further described below in conjunction with the accompanying drawings.

[0031] See also Figures 1 to 4 As shown, the present invention provides an embodiment: a water leakage detection and protection method for multi-sensor information fusion of an underwater robot, characterized in that:

[0032] Step S1, installing multiple water leakage sensors at the edge of the door of the underwater robot cabin where water leakage is easy to occur, and the water leakage sensors are electrically connected to the controller of the underwater robot; so that N water leakage sensors are arranged at the location where water leakage is easy to occur at the bottom of the cabin (the value of N is determined according to the actual control system accuracy requirements and hardware), when water leakage occurs, the water leakage sensor outputs a high level signal, otherwise it is a low level signal;

[0033] Step S2, humidity sensors for detecting humidity are installed at the head, middle and tail of the cabin of the underwater robot, and the humidity sensors are electrically connected to the controller of the underwater robot; the humidity sensors are installed in the cabin to detect the humidity of the cabin, and if the humidity exceeds 85%, there is a possibility of water entering the cabin, and if the humidity is within 50%, it is within the normal range, and the value is determined based on experience;

[0034] Step S3, installing pressure sensors inside and outside the cabin of the underwater robot, and the pressure sensors are electrically connected to the controller of the underwater robot; the pressure sensors are composed of an in-cabin pressure sensor and an out-cabin pressure sensor, and are mainly used to measure the pressure inside and outside the cabin, and then calculate the pressure difference between the inside and outside of the cabin. When the pressure outside the cabin is greater than the pressure inside the cabin by a certain threshold value P0 (P0 is an engineering experience value), it indicates that water leakage may occur;

[0035] Step S4, starting the underwater robot and starting signal collection of multiple sensors;

[0036] Step S5: After the signal is collected, the interference signal is filtered and removed through signal processing. After the interference signal is removed, the controller fuses the information obtained by the water leakage sensor, the humidity sensor and the pressure sensor;

[0037] Step S6: After fusion, if there is no water leakage inside the underwater robot cabin, return to step S4; if there is a water leakage inside the underwater robot cabin, issue an alarm to the host computer;

[0038] Step S7: the alarm prompt starts, and the water leakage protection device of the underwater robot starts to operate to perform water leakage protection.

[0039] Please continue reading Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the signal acquisition of the multi-channel sensor in step S4 is the high and low level signals output by the water leakage sensor, the high level is 3.3V, and the low level is 0V; the pressure sensor and the humidity sensor are converted into 0-3.3V analog voltage signals through the signal conditioning circuit, the water leakage sensor output is connected to the IO port of the DSP, the pressure sensor and the humidity sensor output are connected to the ADC port of the DSP, and the ADC channels are ADCINA0-ADCINA3 respectively. The pressure sensor and the humidity sensor are converted into 0-3.3V analog voltage signals through the signal conditioning circuit, and the controller adopts TI's digital signal processor (DSP for short) TMS320F28033, which is a 32-bit processor with a main frequency of 60MHz, 16-way dual-channel 16-bit ADC sampler, but not limited to this; this system contains 10 water leakage sensors, 2 pressure sensors and 2 humidity sensors, wherein: 10 water leakage sensor outputs are connected to 10 IO ports of the DSP, 2 pressure sensors and 2 humidity sensor outputs are connected to 4 ADC ports of the DSP, and the ADC channels are ADCINA0-ADCINA3 respectively.

[0040] Please continue reading Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the signal processing in step S5 is used for data preprocessing after signal acquisition, and a digital filter is used to remove the collected interference signal to leave the signal that actually matches it.

[0041] Please continue reading Figure 1 and Figure 2As shown, in one embodiment of the present invention, the step S5 also includes signal feature extraction, and the signal feature extraction is to extract the time domain features of the signal. The time domain feature extraction methods include average value, mean square value, effective value, and peak value. The time domain feature extraction methods adopted for different signal sampling amounts are also different; water leakage sensor: pulse signal extraction, using continuous filtering method; pressure sensor: extraction in the form of average value, calculating the average pressure inside and outside the cabin; humidity sensor: extraction in the form of average value, calculating the humidity value inside the cabin. After the interference signal is removed by signal preprocessing, the signal feature is extracted; the time domain features of the signal are extracted. The time domain feature extraction methods include average value, mean square value, effective value, peak value, etc., but are not limited to this; the time domain feature extraction methods adopted for different signal sampling amounts are also different.

[0042] Please continue reading Figure 1 and Figure 2 As shown, in one embodiment of the present invention, the information fusion in step S5 adopts a weighted average algorithm, and the three types of sensors, namely, the pressure sensor, the humidity sensor and the water leakage sensor, are first normalized and then weighted averaged. Each sensor has a weight value; the data of the sensor with high precision has a small error and a large weight; while the data of the sensor with low precision has a large error and a small weight; the digital quantities collected by the pressure sensor are x1, x2, x3....x i , get the average value n is a natural number, K1 is the inverse of the hardware conditioning circuit gain, and the collected average value is converted into the real pressure value. P is the pressure value of the pressure sensor; the digital values ​​collected by the humidity sensor are y1, y2, y3....y i , get the average value n is a natural number, K2 is the inverse of the gain of the hardware conditioning circuit, and the collected average value is converted into the true value of humidity. H is the humidity value of the humidity sensor; and the result of the water leakage sensor is Where R i is the result of each water leakage sensor, L is the set of all water leakage sensor results, and each water leakage sensor outputs a high level, then R i =1, output low level, R i =0; convert the sampled values ​​of the three types of sensors, namely, pressure sensor, humidity sensor and water leakage sensor, into recognizable results and perform weighting;

[0043] 1) Pressure sensor judgment criteria: P 外 -P 内 ≥P0 (P0 is the threshold), it is judged as water leakage; P 外 ≤P 内 , judged to be watertight;

[0044] 2) Humidity sensor judgment criteria: H≥H1, H1 is generally taken as 85%, which is considered to be leaking; otherwise, H≤H2, which is considered to be not leaking, and H2 is taken as 75%;

[0045] 3) Leakage sensor judgment criteria: L = 1, judged as water leakage; L = 0, judged as no water leakage; the judgment results of the above sensors are output as 0 or 1, 1 represents water leakage, and 0 represents no water leakage;

[0046] Perform weighted summation on the three types of sensors and output the final result; among them, the pressure sensor, humidity sensor, and water leakage sensor are judged by the criteria as T1, T2, and T3 respectively; the weight coefficients of the pressure, humidity, and water leakage sensors are w1, w2, and w3 respectively, among which w1+w2+w3=1, and the output result is T=w1T1+w2T2+w3T3, 0≤T≤1, and the comparison is made based on T.

[0047] The decision output in the present invention is to identify the fault in accordance with the formula T = w1T1 + w2T2 + w3T3, and make the following final judgment, please refer to Figure 3 and Figure 4 As shown, the present invention provides a block diagram for identifying faults and a fault identification table. In the first step, when T is calculated to be less than or equal to 0.8, the detection calculation is continued downward, and T is calculated again to be greater than or equal to 0.5. In the third step, when at least two of T1, T2 and T3 are 1, the output result is that water leakage is considered; when T is calculated to be greater than 0.8 in the first step, water leakage can be directly determined. In the second step, when T is less than 0.5, it can be directly determined to be water leakage.

[0048] Please continue reading Figures 1 to 4 As shown, in one embodiment of the present invention, the water leakage protection in step S7 is to determine whether the system is leaking after the decision output, and the controller outputs an audible and visual alarm, the host computer lights up a red light, and the buzzer sounds for a long time; about 5 seconds after the alarm, the controller outputs a control signal to control the underwater robot to cut off the power, and the alarm of the host computer continues until the power supply of the underwater robot is restored and the water leakage fault is eliminated.

[0049] It is worth mentioning that the controller of the underwater robot in the present invention is provided with a circuit board, and the circuit board is provided with a main control module. The main control module is connected to a buzzer alarm drive circuit, an LED drive circuit, a water leakage sensor and a low-pass filter circuit, and the humidity sensor and the pressure sensor are both connected to the main control module via the low-pass filter circuit.

[0050] See also Figure 5As shown, in one embodiment of the present invention, the LED driving circuit includes a VCC port, a PWM IN port, a diode D1, a diode D2, a resistor R1, a resistor R2, a field effect transistor NPN, a capacitor C1 and a capacitor C2, one end of the resistor R1 is connected to the PWM IN port, the other end of the resistor R1 is connected in series with one end of the resistor R2 and then connected to the gate of the field effect transistor NPN, the other end of the resistor R2 is grounded, the drain of the field effect transistor NPN is connected to the cathode of the diode D2 and the cathode of the diode D1, the anode of the diode D1 and the anode of the diode D2 are connected in series and then connected to the VCC port, and the field effect transistor NPN is arranged in parallel with the capacitor C1 and the capacitor C2. The controller can turn on or off the front LED lamp through the action of the LED driving circuit.

[0051] See also Figure 6 As shown, in one embodiment of the present invention, the buzzer alarm driving circuit includes a diode D3, a field effect transistor Q1, a resistor R3, a resistor R4, a capacitor C3, a Bell-Ctrl port and a buzzer Bell, one end of the resistor R3 is connected to the Bell-Ctrl port, the other end of the resistor R3 is connected to the gate of the field effect transistor Q1, the source of the field effect transistor Q1 is connected to one end of the resistor R4, the other end of the resistor R4 is grounded, the drain of the field effect transistor Q1 is connected to the positive electrode of the diode D3 and then connected to the buzzer Bell, the cathode of the diode D3 is connected to the buzzer Bell, the cathode of the diode D3 is connected in series with one end of the capacitor C3 and then connected to the VCC port, and the other end of the capacitor C3 is grounded. The buzzer alarm driving circuit can play a role in driving or closing the buzzer alarm.

[0052] See also Figure 7 As shown, in one embodiment of the present invention, the low-pass filter circuit includes a Signal-OUT port, a capacitor C4, a resistor R5, a resistor R6, a capacitor C5, an operational amplifier U1 and a capacitor C6, one end of the resistor R5 is connected to the Signal-OUT port, the other end of the resistor R5 is connected to one end of the capacitor C4 and one end of the resistor R6, the other end of the capacitor C4 is connected to pins 1 and 2 of the operational amplifier U1, the other end of the resistor R6 is connected to one end of the capacitor C5 and pin 3 of the operational amplifier U1, and the other end of the capacitor C5 is connected to the operational amplifier U1 and then grounded; the operational amplifier U1 is connected to the VCC port, the VCC port is connected to one end of the capacitor C6, and the other end of the capacitor C6 is grounded. The output signal of the pressure sensor 25 and the humidity sensor 24 after passing through the low-pass filter circuit. The operational amplifier U1 used in the present invention is an integrated chip with a four-way operational amplifier, and the model can be LM324J, but is not limited to this.

[0053] See also Figure 8As shown, in one embodiment of the present invention, the main control module includes a TMS320F28033 chip, a Lou-Signal1-IN port, a Lou-Signal2-IN port, a Lou-Signal3-IN port, a Lou-Signal4-IN port, a Lou-Signal5-IN port, a Lou-Signal6-IN port, a Lou-Signal7-IN port, a Lou-Signal8-IN port, a Lou-Signal9-IN port, a Lou-Signal10-IN port, a Signal1-OUT port, a Signal2-OUT port, a Signal3-O The Signal1-OUT port, Signal2-OUT port, Signal3-OUT port and Signal4-OUT port are respectively the Signal-OUT port outputs of the pressure sensor and the humidity sensor signal output after passing through the low-pass filter circuit, the Signal1-OUT port is connected to the No. 18 pin of the TMS320F28033 chip, the Signal2-OUT port is connected to the No. 17 pin of the TMS320F28033 chip, the Signal3-OUT port is connected to the No. 16 pin of the TMS320F28033 chip, and the Signal4-OUT port is connected to the No. 18 pin of the TMS320F28033 chip. The Signal4-OUT port is connected to the No. 15 pin of the TMS320F28033 chip, the No. 61 pin of the TMS320F28033 chip is connected to the Lou-Signal1-IN port, the No. 62 pin of the TMS320F28033 chip is connected to the Lou-Signal2-IN port, the No. 63 pin of the TMS320F28033 chip is connected to the Lou-Signal3-IN port, the No. 64 pin of the TMS320F28033 chip is connected to the Lou-Signal4-IN port, the No. 65 pin of the TMS320F28033 chip is connected to the Lou-Signal1-IN port, the No. 62 pin of the TMS320F28033 chip is connected to the Lou-Signal2-IN port, the No. 63 pin of the TMS320F28033 chip is connected to the Lou-Signal3-IN port, the No. 64 pin of the TMS320F28033 chip is connected to the Lou-Signal4-IN port, and the No. 65 pin of the TMS320F28033 chip is connected to the Lou-Signal1-IN port. The pin 66 of the TMS320F28033 chip is connected to the Lou-Signal6-IN port, the pin 67 of the TMS320F28033 chip is connected to the Lou-Signal7-IN port, the pin 68 of the TMS320F28033 chip is connected to the Lou-Signal8-IN port, the pin 69 of the TMS320F28033 chip is connected to the Lou-Signal9-IN port, and the pin 74 of the TMS320F28033 chip is connected to the Lou-Signal10-IN port.The Lou-Signal1-IN port, Lou-Signal2-IN port, Lou-Signal3-IN port, Lou-Signal4-IN port, Lou-Signal5-IN port, Lou-Signal6-IN port, Lou-Signal7-IN port, Lou-Signal8-IN port, Lou-Signal9-IN port and Lou-Signal10-IN port are connected to the water leakage sensor; pin 75 of the TMS320F28033 chip is connected to the Bell-Ctrl port, and pin 76 of the TMS320F28033 chip is connected to the PWM IN port. The number of the water leakage sensor 25 and the humidity sensor 24 of the present invention is two, the Signal1-OUT port and the Signal2-OUT port are the output ports of the water leakage sensor 25 after low-pass filtering; the Signal3-OUT port and the Signal4-OUT port are the output ports of the humidity sensor 24 after low-pass filtering, and the Signal1-OUT port, the Signal2-OUT port, the Signal3-OUT port and the Signal4-OUT port are output through four low-pass filtering circuits respectively.

[0054] The controller, pressure sensor, humidity sensor and water leakage sensor in the present invention are all prior arts, and those skilled in the art are already able to clearly understand them, so they will not be described in detail here. The controller in the present invention can adopt TI's digital signal processor (DSP for short) TMS320F28033, which is a 32-bit processor with a main frequency of 60MHz, a 16-way dual-channel 16-bit ADC sampler, the pressure sensor can be PT124G-125, the water leakage sensor can be SPLDS-2000, and the humidity sensor can be HS12P, but it is not limited to this. The present invention is based on the hardware structure of the underwater robot, and with the corresponding algorithm, it can achieve the collection and transmission of underwater robot detection related information.

[0055] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A water leakage detection and protection method for underwater robot multi-sensor information fusion, characterized by: Step S1, installing a plurality of water leakage sensors at the edge of the door of the underwater robot body that is prone to water leakage, and the water leakage sensors are electrically connected to the controller of the underwater robot; Step S2, humidity sensors for detecting humidity are installed at the front, middle and rear parts of the cabin of the underwater robot, and the humidity sensors are electrically connected to the controller of the underwater robot; Step S3, installing pressure sensors inside and outside the cabin of the underwater robot, and the pressure sensors are electrically connected to the controller of the underwater robot; Step S4, starting the underwater robot and starting signal collection of multiple sensors; Step S5: After the signal is collected, the interference signal is filtered and removed through signal processing. After the interference signal is removed, the controller fuses the information obtained by the water leakage sensor, the humidity sensor and the pressure sensor; Step S6: After fusion, if there is no water leakage inside the underwater robot cabin, return to step S4; if there is a water leakage inside the underwater robot cabin, issue an alarm to the host computer; Step S7: the alarm prompt starts, and the water leakage protection device of the underwater robot starts to operate to perform water leakage protection; The information fusion in step S5 adopts a weighted average algorithm, and the three types of sensors, namely, the pressure sensor, the humidity sensor and the water leakage sensor, are first normalized and then weighted averaged. Each sensor has a weight value; the data with high sensor accuracy has a small error and a large weight; while the data with low accuracy has a large error and a small weight; the digital quantities collected by the pressure sensor are x1, x2, x3, .... x i , get the average value n is a natural number, K1 is the inverse of the hardware conditioning circuit gain, and the collected average value is converted into the real pressure value. P is the pressure value of the pressure sensor; the digital values ​​collected by the humidity sensor are y1, y2, y3....y i , get the average value n is a natural number, K2 is the inverse of the gain of the hardware conditioning circuit, and the collected average value is converted into the true value of humidity. H is the humidity value of the humidity sensor; and the result of the water leakage sensor is Where R i is the result of each water leakage sensor, L is the set of all water leakage sensor results, and each water leakage sensor outputs a high level, then R i =1, output low level, R i =0; convert the sampled values ​​of the three types of sensors, namely, pressure sensor, humidity sensor and water leakage sensor, into recognizable results and perform weighting; 1) Pressure sensor judgment criteria: P 外 -P 内 ≥P0, P0 is the threshold, which means it is judged as water leakage; P 外 ≤P 内 , judged to be watertight; 2) Humidity sensor judgment criteria: H≥H1, H1 value is 85%, it is judged as water leakage; otherwise, H≤H2, it is judged as no water leakage, H2 value is 75%; 3) Leakage sensor judgment criteria: L = 1, judged as water leakage; L = 0, judged as no water leakage; the judgment results of the above sensors are output as 0 or 1, 1 represents water leakage, and 0 represents no water leakage; Perform weighted summation on the three types of sensors and output the final result; among them, the pressure sensor, humidity sensor, and water leakage sensor are judged by the criteria as T1, T2, and T3 respectively; the weight coefficients of the pressure, humidity, and water leakage sensors are w1, w2, and w3 respectively, among which w1+w2+w3=1, and the output result is T=w1T1+w2T2+w3T3, 0≤T≤1, and the comparison is made based on T.

2. The water leakage detection and protection method for underwater robot multi-sensor information fusion according to claim 1 is characterized in that: The signal acquisition of the multi-channel sensors in step S4 is that the water leakage sensor outputs high and low level signals, the high level is 3.3V, and the low level is 0V; the pressure sensor and the humidity sensor are converted into 0-3.3V analog voltage signals through the signal conditioning circuit, the water leakage sensor output is connected to the IO port of the DSP, the pressure sensor and the humidity sensor output are connected to the ADC module of the DSP, and the ADC sampling channels are ADCINA0-ADCINA3 respectively.

3. The water leakage detection and protection method for underwater robot multi-sensor information fusion according to claim 1 is characterized in that: The signal processing in step S5 is used for data preprocessing after signal acquisition, and a digital filter is used to remove the collected interference signal, leaving the signal that actually matches it.

4. The water leakage detection and protection method for underwater robot multi-sensor information fusion according to claim 1 is characterized in that: The water leakage protection in step S7 is to determine whether the system is leaking after the decision output, and the controller outputs an audible and visual alarm, the host computer lights up a red light, and the buzzer sounds for a long time; 5 seconds after the alarm, the controller outputs a control signal to control the underwater robot to cut off the power, and the alarm of the host computer continues until the power supply of the underwater robot is restored and the water leakage fault is eliminated.