Underwater robot online positioning method and device based on multiple sensors

Abstract

The embodiment of the invention provides an underwater robot online positioning method and device based on multiple sensors, and relates to the technical field of underwater target positioning technologies. The method comprises the following steps: acquiring sensing data; based on the acoustic data, determining a first suspicious area, and performing detail focusing recognition processing on the first suspicious area through optical data; according to a detail focusing recognition processing result, matching first electromagnetic data and first environment data of the first suspicious area from the sensing data; and determining the position information of the underwater target according to the electromagnetic data and the environment sensing data. According to the invention, the problem of low positioning precision of the underwater target is solved, and the effect of improving the positioning precision is achieved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of underwater target detection, and in particular, to a multi-sensor based online positioning method and device for an underwater robot. Background Art

[0002] With the continuous development of ocean development and underwater detection technology, the application scope of underwater robots is becoming increasingly wide, including marine resource exploration, underwater archaeology, environmental monitoring, military reconnaissance and other fields. The current underwater robots include cable remote-controlled submersibles, cable-free autonomous submersibles and hybrids. Among them, cable remote-controlled submersibles are connected to the mother ship through cables, and transmit data and electricity in real time. They are divided into self-propelled, towed and seabed crawling types, which are suitable for long-term precision operations (such as pipeline maintenance and sample collection); cable-free autonomous submersibles mainly rely on built-in energy and programs to operate autonomously, and their core features include cable-free operation, autonomous navigation and long-range detection capabilities; hybrids have both to adapt to different environments; when underwater robots perform tasks, accurately locating their own position and target position is the key to achieving efficient operations, especially in environments with complex terrain, high water turbidity and salinity.

[0003] However, the existing technology still has many shortcomings in underwater positioning.

[0004] For example, existing underwater positioning technologies mostly rely on a single sensor, such as sonar or optical cameras. Acoustic positioning systems are susceptible to water temperature stratification, salinity changes, and water turbidity, resulting in increased signal attenuation and reflection errors; optical positioning systems are limited by the underwater light propagation characteristics and are easily affected by turbidity, light intensity, and water color, resulting in blurred imaging and reduced recognition accuracy. Therefore, it is difficult for a single sensor to achieve high-precision positioning in a complex underwater environment.

[0005] There is currently no better solution to the above problems. Summary of the invention

[0006] The embodiments of the present invention provide a multi-sensor based online positioning method and device for an underwater robot, so as to at least solve the problem of low positioning accuracy in the related art.

[0007] According to one embodiment of the present invention, a method for online positioning of an underwater robot based on multiple sensors is provided, comprising:

[0008] Acquiring sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data, and environmental perception data;

[0009] Based on the acoustic data, determining a first suspicious area, and performing detail focus recognition processing on the first suspicious area through the optical data;

[0010] According to the result of the detail focus recognition processing, matching the first electromagnetic data and the first environmental data of the first suspicious area from the sensing data, wherein the environmental sensing data includes the first environmental data;

[0011] The position information of the underwater target is determined according to the electromagnetic data and the environmental perception data.

[0012] In an exemplary embodiment, determining a first suspicious area based on the acoustic data, and performing detail focus recognition processing on the first suspicious area through optical data includes:

[0013] Dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious area;

[0014] Convert the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates;

[0015] Performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image;

[0016] The first image is subjected to detail focus recognition processing based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

[0017] In an exemplary embodiment, before acquiring the sensor data, the method further includes:

[0018] Focusing energy to the first suspicious area by minimum variance distortion-free response beamforming;

[0019] Optimizing the emission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area;

[0020] The array gain distribution of the acoustic data is dynamically adjusted based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

[0021] In an exemplary embodiment, determining the position information of the underwater target according to the electromagnetic data and the environmental perception data includes:

[0022] determining electric field gradient change information of the first suspicious area according to the first electromagnetic data;

[0023] Determining object features in the first visible area according to the electric field gradient change information and the first environmental data;

[0024] When the object feature meets the first condition, the pre-obtained target center point coordinates and target area edge coordinates are fused by an extended Kalman filter algorithm to obtain the position information.

[0025] In an exemplary embodiment, after matching the first electromagnetic data and the first environmental data of the first suspicious area from the sensing data according to the detail focus recognition processing result, and the environmental perception data includes the first environmental data, the method further includes:

[0026] According to the environmental perception data, the sensor data and the electromagnetic data are dynamically adjusted and allocated in terms of weights.

[0027] According to another embodiment of the present invention, a multi-sensor based underwater robot online positioning method and device is provided, comprising:

[0028] A data acquisition module, used to acquire sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data and environmental perception data;

[0029] A first processing module, configured to determine a first suspicious area based on the acoustic data, and perform detail focus recognition processing on the first suspicious area through optical data;

[0030] a matching module, configured to match first electromagnetic data and first environmental data of the first suspicious area from the sensing data according to the result of the detail focus recognition processing, wherein the environmental sensing data includes the first environmental data;

[0031] A position determination module is used to determine the position information of the underwater target based on the electromagnetic data and the environmental perception data.

[0032] In an exemplary embodiment, determining a first suspicious area based on the acoustic data, and performing detail focus recognition processing on the first suspicious area through optical data includes:

[0033] Dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious area;

[0034] Convert the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates;

[0035] Performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image;

[0036] The first image is subjected to detail focus recognition processing based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

[0037] In an exemplary embodiment, the apparatus further comprises:

[0038] A beamforming module, used for focusing energy to the first suspicious area by minimum variance distortion-free response beamforming before acquiring the sensing data;

[0039] A dynamic matching module, used to optimize the transmission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area;

[0040] A dynamic adjustment module is used to dynamically adjust the array gain distribution of the acoustic data based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

[0041] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any one of the above method embodiments when run.

[0042] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0043] Through the present invention, since underwater targets are positioned and detected from multiple dimensions by fusing multiple sensors, the problem of low positioning accuracy of underwater targets can be solved, thereby achieving the effect of improving positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of an online positioning method of an underwater robot based on multiple sensors according to an embodiment of the present invention;

[0045] Figure 2 It is a structural block diagram of an underwater robot online positioning method device based on multiple sensors according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0047] In the following, the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0048] In addition, in the present application, directional terms such as "up", "down", "left" and "right" may be defined including but not limited to the orientation relative to the schematic placement of the components in the drawings. It should be understood that these directional terms may be relative concepts, which are used for relative description and clarification, and may change accordingly according to the change in the orientation of the components in the drawings.

[0049] In this application, unless otherwise specified or limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. In addition, the term "coupling" can be a way of achieving electrical connection for signal transmission.

[0050] As used herein, "about," "substantially," or "approximately" includes the stated value and an average value that is within an acceptable range of variation from the particular value as determined by one of ordinary skill in the art taking into account the measurements in question and the errors associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0051] In the existing underwater target positioning scheme, a single sensor (sonar or visual processing) is generally used for detection, which cannot be effectively used when the underwater turbidity is high and the electromagnetic interference is large. In this embodiment, an underwater robot online positioning method based on multiple sensors is provided, which can avoid the above problems and thus improve the accuracy of underwater target detection.

[0052] Figure 1 FIG. 1 is a flow chart of an online positioning method of an underwater robot based on multiple sensors according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0053] Step S11, acquiring sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data, and environmental perception data;

[0054] In this embodiment, compared with the prior art, the present application also incorporates electromagnetic data and environmental perception data into the content of underwater target detection, thereby effectively adapting to complex underwater scenes and improving detection accuracy.

[0055] Among them, acoustic data include bathymetric data (measurement of the time difference from the emission to the reception of sound waves through the sonar system), echo signals (using active sonar to emit sound waves and receive echoes reflected by the target), Doppler shift (detection of target movement speed through changes in sound wave frequency), scattering signals (scattered sound waves of suspended particles or bubbles in turbid water bodies, used to evaluate water quality environment (such as the scattering intensity fed back by turbidity sensors)), etc.; optical data include underwater images of target surface details obtained by blue-green light (532nm) cameras or laser scanning, spectral information in different bands, laser stripe projection data for analyzing the three-dimensional contour of the target through laser stripe deformation, and photon counting data corresponding to extremely weak light signals captured by single-photon detectors (such as SNSPD), etc.; electromagnetic data include changes in magnetic field gradient, changes in electric field intensity, electromagnetic spectrum characteristics, low-frequency electric field data, etc.; environmental perception data include water body parameters (turbidity, salinity, temperature, etc.), water flow velocity and direction, Beijing noise, seabed bottom data, etc.

[0056] Step S12, determining a first suspicious area based on the acoustic data, and performing detail focus recognition processing on the first suspicious area through the optical data;

[0057] In this embodiment, the areas where target objects (underwater robots, underwater targets, etc.) may exist are first determined by acoustic data, and then these areas are further identified in detail by optical data, thereby preliminarily determining the areas where target objects may exist.

[0058] Specifically, an 8-channel annular transducer group (center frequency 300kHz, bandwidth ±100kHz) is first arranged to support beamforming and multi-target tracking, and a turbidity sensor (NTU measurement range 0-200) is integrated to provide real-time feedback to the transmit power controller; then a 532nm wavelength laser (output power 1W) is used in conjunction with a galvanometer to achieve 0.1° precision pointing control to scan the data in the possible area; then, when performing acoustic data acquisition, in order to achieve anti-turbidity signal processing, a linear frequency modulation signal (50-500kHz) is used in combination with Costas coding to suppress multipath interference, and minimum variance distortion-free response beamforming is used to focus energy to the suspicious area, and the array gain distribution is dynamically adjusted based on turbidity feedback (NTU value), and the background noise is adjusted according to the background noise. The detection threshold is adjusted dynamically according to the sound level, and the area with a signal-to-noise ratio > 12dB is marked (the first suspicious area). Then, based on the pre-constructed association relationship, the coordinates of the suspicious area are converted into an optical combination table, and combined with the fiber optic gyroscope (INS) and Doppler odometer (DVL) data, the coordinate offset caused by the movement of the equipment is corrected in real time. After that, the laser center line of the laser sensor is extracted, the three-dimensional contour of the target surface is calculated, and the reflective features of the suspicious area are identified through phase consistency analysis, thereby obtaining the detailed focus recognition processing results. By seamlessly connecting the acoustic anti-turbidity coarse positioning and the optical high-precision recognition, the target detection efficiency in complex environments can be effectively improved. At the same time, the real-time parameter adjustment based on environmental parameters (NTU value, motion posture) can adapt to the full scene requirements from nearshore turbid waters to deep-sea environments.

[0059] It should be noted that in this process, the acoustic data and the optical data need to be repeatedly cross-verified. For example, if the optical recognition confidence is <70%, the acoustic secondary scan is triggered (the transmission power is increased by 10dB); or if the acoustic rescan still cannot be confirmed, the magnetoelectric sensor auxiliary detection is started; and when multiple suspicious areas appear at the same time, the targets with high optical recognition confidence and high acoustic echo intensity are given priority; when the optical sensor fails to detect the target for three consecutive frames, it switches to the wide-spectrum imaging mode and expands the search range (±30°). If it still cannot be locked at this time, it falls back to the acoustic global scanning mode, and so on.

[0060] Step S13, matching first electromagnetic data and first environmental data of the first suspicious area from the sensing data according to the detail focus recognition processing result, wherein the environmental sensing data includes the first environmental data;

[0061] In this embodiment, after obtaining the acoustic and optical details, the corresponding electromagnetic data is further determined, and the electromagnetic data is further adjusted according to the environmental data, so as to more accurately determine the position of the target object.

[0062] Specifically, firstly transform the optical coordinate system (x i,y i ,z i ) is converted into an electromagnetic coordinate system, and then the target space-time window (±0.1m spatial range, ±10ms time window) is intercepted in the electromagnetic data stream to extract the magnetic field gradient The target area is then divided into two parts: the target area is divided into two parts, and the target area is divided into two parts. The ...

[0063]

[0064] In the formula, k Env is the correction coefficient, NTU is turbidity, S is salinity, T is temperature, α=0.2, β=0.1, δ=0.4. Of course, if the salinity is >40PSU or the turbidity is >100NTU, the weight of electromagnetic data is reduced by 50% and optical data is used first.

[0065] In particular, if the electromagnetic signal is submerged by strong interference, the historical electromagnetic fingerprint library matching is enabled (similarity > 80% is considered valid).

[0066] Step S14: determining the position information of the underwater target according to the electromagnetic data and the environmental perception data.

[0067] In this embodiment, after the electromagnetic data is determined and corrected by the environmental perception data, the data is fused by the extended Kalman algorithm to output the precise position information of the underwater target.

[0068] Among them, for low-frequency electric field detection, it is necessary to deploy high-sensitivity charge sensing electrodes (such as copper / iron materials, input impedance>100GΩ) to capture the axial frequency electric field (1-10Hz) or magnetic field gradient (0.1nT resolution) generated by the underwater target, convert it into a voltage signal through a charge sensing chip and filter and reduce noise, and determine the water depth of the data acquisition module through a flexible cable to achieve the spatial distribution measurement of the underwater target electric field.

[0069] Through the above steps, by fusing multiple sensors, underwater targets are located and detected from multiple dimensions, which solves the problem of low underwater target positioning accuracy and improves positioning accuracy.

[0070] The execution entities of the above steps may be [base stations, terminals], etc., but are not limited thereto.

[0071] In an optional embodiment, determining the first suspicious area based on the acoustic data, and performing detail focus recognition processing on the first suspicious area through the optical data includes:

[0072] Step S121, dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious region;

[0073] Step S122, converting the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates;

[0074] Step S123, performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image;

[0075] Step S124: performing detail focus recognition processing on the first image based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

[0076] In this embodiment, the background noise intensity (such as noise power spectral density) is monitored in real time, and the detection threshold is dynamically adjusted according to the background noise level, and the area with a signal-to-noise ratio > 12dB is marked as a suspicious area. Of course, the threshold can also be dynamically updated by calculating the local signal-to-noise ratio (SNR) and combining Kalman filtering to predict noise changes, which is not limited here.

[0077] Then, a density clustering algorithm such as DBSCAN is used to cluster the acoustic cloud points (such as sonar point clouds), identify high-density areas as potential targets, and determine the center point coordinates (x c ,y c ,z c ) and the region edge coordinates (x ce ,y ce ,z ce ); among them, key parameters such as neighborhood radius (eps, assumed to be 5m) and minimum number of points (minPts, assumed to be 5) need to be optimized according to the sonar resolution and target size. For example, for small targets, eps can be reduced to capture details, and isolated points with points below the threshold in the cluster are eliminated to reduce noise interference.

[0078] Convert the center point coordinates and area edge coordinates in the acoustic sensor coordinate system to the optical camera coordinate system. Assuming that the relative position and posture of the acoustic sensor and the optical camera are known, the conversion can be completed through the coordinate transformation matrix:

[0079]

[0080] In the formula, R is the rotation matrix, T is the translation vector, is the optical coordinate, is the acoustic coordinate, and then the center point coordinate and the area edge coordinate are substituted into the calculation to obtain the target center point coordinate and the target area edge coordinate in the optical coordinate.

[0081] Then based on the underwater light transmission model I(x) = J(x)e -βz +B(1-e -βz ), the transmittance t(x) is estimated by the dark channel prior, and the clear image J(x) (i.e., the first image mentioned above) is restored. During the restoration process, it is necessary to compensate for the image distortion caused by water absorption and scattering.

[0082] Then, according to the coordinates of the target center point and the edge coordinates of the target area, the target area is cropped from the first image, and the cropped image is preprocessed, including grayscale, denoising (such as Gaussian filtering), contrast enhancement and other operations to improve the image quality; then, the cropped image is subjected to detail focus recognition using an image processing algorithm (such as edge detection and feature extraction). For example, the Canny edge detection algorithm is used to extract the contour information of the target, and the extracted edge information is analyzed to identify the shape, texture and other features of the target. At the same time, the type and confidence of the target are judged according to a preset feature library (such as the known shape and texture pattern of the target), and the type, texture features and confidence of the target are output as the result of detail focus recognition to a subsequent processing module for further target confirmation and positioning.

[0083] In an optional embodiment, before acquiring the sensor data, the method further includes:

[0084] Step S101, focusing energy to the first suspicious area by minimum variance distortion-free response beamforming;

[0085] Step S102, optimizing the transmission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area;

[0086] Step S103: dynamically adjusting the array gain distribution of the acoustic data based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

[0087] In this embodiment, assuming that the acoustic sensor array is a uniform linear array or annular array, its output signal can be expressed as:

[0088]

[0089] In the formula, is the beam weight vector, is the input signal vector of the sensor array, express The conjugate transpose of ; then the beam weight vector It can be calculated by the following formula:

[0090]

[0091] In the formula, is the covariance matrix of the input signal, is the guidance vector in the target direction, and θ is the target direction angle.

[0092] Then, according to the preset target direction or the estimated position of the suspicious area, the corresponding guidance vector is calculated. , and use the optimized beam weight w opt The acoustic sensor array is weighted so that the beam is directed toward the first suspicious area, thereby increasing the signal strength in the area.

[0093] The Bayesian probability model is used to describe the relationship between the emission waveform parameters and the turbidity of the water body. Assume that the emission waveform parameters are , the water turbidity is T, then the Bayesian model can be expressed as:

[0094]

[0095] In the formula, is the conditional probability, indicating that the emission waveform parameters are probability; is the likelihood function, which means that given the transmission waveform parameters The probability of water turbidity T under the condition of ; is the prior probability, which represents the prior distribution of the transmission waveform parameters; P(T) is the normalization constant.

[0096] According to the Bayesian probability model, the optimal emission waveform parameter popt is calculated under the current water turbidity T. For example, the optimal parameter can be selected by maximizing the posterior probability P(p|T):

[0097]

[0098] The transmitting signal of the sonar system is then adjusted according to the optimized transmitting waveform parameters, such as adjusting the frequency, bandwidth or modulation method of the signal to adapt to the current water environment.

[0099] It should be noted that dynamically adjusting the gain distribution of the acoustic sensor array includes:

[0100]

[0101] Where G(T) is the adjusted array gain distribution, To adjust the function.

[0102] In an optional embodiment, determining the position information of the underwater target according to the electromagnetic data and the environmental perception data includes:

[0103] Step S141, determining electric field gradient change information of the first suspicious area according to the first electromagnetic data;

[0104] Step S142, determining the object features in the first visible area according to the electric field gradient change information and the first environmental data;

[0105] Step S143, when the object feature meets the first condition, the pre-obtained target center point coordinates and target area edge coordinates are fused by an extended Kalman filter algorithm to obtain the position information.

[0106] In this embodiment, the electric field and the magnetic field are recorded by time flow to facilitate the establishment of an association relationship with the acoustic data and the optical data.

[0107] Specifically, the measured values ​​of the electric field strength (E) and the magnetic field strength (B) are expressed as:

[0108] E(t)=[E x (t),E y (t),E z (t)] (Formula 8)

[0109] B(t)=[B x (t),B y (t),B z (t)] (Formula 9)

[0110] In the formula, E x (t),E y (t),E z (t) and B x (t),B y (t),B z (t) represents the components of the electric field and magnetic field on the three coordinate axes respectively.

[0111] At the same time, the spatial gradient of the electric field intensity data is calculated to obtain the electric field gradient change information; the electric field gradient can be calculated by the finite difference method:

[0112]

[0113] The specific calculation formula is:

[0114]

[0115] Among them, Δx, Δy, Δz are the spatial resolution of the sensor.

[0116] Then calculate the modulus of the electric field gradient:

[0117]

[0118] The changing characteristics of the electric field gradient are analyzed, such as the peak position of the gradient, the direction of the gradient change, etc.; then, the characteristics of the object in the suspicious area are judged based on the corrected electric field gradient change information. For example: if the difference between the magnetic field intensity and the background magnetic field exceeds 50nT, it is judged as a metal target; if a 1-10Hz bioelectric signal is detected, it is judged as an underwater creature; if the peak position of the electric field gradient is consistent with the target position in the optical data, further confirmation is performed, such as matching the extracted features with the preset target feature library to determine the type and confidence of the target. For example, the degree of matching is evaluated by calculating feature similarity (such as Euclidean distance or cosine similarity).

[0119] The correction formula is:

[0120]

[0121] The fusion processing of the pre-obtained target center point coordinates and target area edge coordinates by using the extended Kalman filter algorithm includes:

[0122] Define the state vector , including the target position coordinates and velocity vector:

[0123]

[0124] And initialize the state estimate and the covariance matrix For example, assume that the initial position is the target center point coordinate in the optical data and the initial velocity is zero.

[0125]

[0126] Among them, σ x ,σ y ,σ z is the initial uncertainty of the position, is the initial uncertainty of the velocity.

[0127] Assuming that the target moves in a uniform linear motion, its discrete time motion model is:

[0128]

[0129] in, is the state transfer matrix:

[0130]

[0131] w k-1 is the process noise, which is assumed to be zero-mean Gaussian white noise.

[0132] Combining the target position information from electromagnetic data and optical data, define the observation model:

[0133]

[0134] in, is the observation vector, including the coordinates of the target center point and the area edge coordinates; is the observation matrix, which is used to map the state vector to the observation space v k is the observation noise, which is assumed to be zero-mean Gaussian white noise.

[0135] Then, the position prediction is performed based on the extended Kalman filter:

[0136]

[0137] Among them, Q is the process noise covariance matrix, k-1 is the data at the previous moment, and k is the data at the current moment.

[0138] In an optional embodiment, after matching the first electromagnetic data and the first environmental data of the first suspicious area from the sensing data according to the detail focus recognition processing result, and the environmental perception data includes the first environmental data, the method further includes:

[0139] According to the environmental perception data, the sensor data and the electromagnetic data are dynamically adjusted and allocated in terms of weights.

[0140] In this embodiment, the dynamic adjustment of weights can adapt to different usage environments, thereby further improving the detection accuracy.

[0141] Specifically, when the turbidity NTU>100, the weight of optical data is reduced and the weight of acoustic data is increased; when the salinity PSU>40, the weight of electromagnetic data is reduced and the weight of optical data is increased; when the background noise exceeds 60dB, the weight of acoustic data is reduced and the weight of electromagnetic data is increased.

[0142] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0143] In this embodiment, a multi-sensor based underwater robot online positioning device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0144] Figure 2 is a structural block diagram of an underwater robot online positioning device based on multiple sensors according to an embodiment of the present invention. Figure 2 As shown, the device comprises:

[0145] A data acquisition module 21, used to acquire sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data and environmental perception data;

[0146] A first processing module 22, configured to determine a first suspicious area based on the acoustic data, and perform detail focus recognition processing on the first suspicious area through optical data;

[0147] A matching module 23, configured to match first electromagnetic data and first environmental data of the first suspicious area from the sensing data according to the result of the detail focus recognition processing, wherein the environmental sensing data includes the first environmental data;

[0148] The position determination module 24 is used to determine the position information of the underwater target according to the electromagnetic data and the environmental perception data.

[0149] In an optional embodiment, determining the first suspicious area based on the acoustic data, and performing detail focus recognition processing on the first suspicious area through the optical data includes:

[0150] Dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious area;

[0151] Convert the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates;

[0152] Performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image;

[0153] The first image is subjected to detail focus recognition processing based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

[0154] In an optional embodiment, the device further comprises:

[0155] A beamforming module, used for focusing energy to the first suspicious area by minimum variance distortion-free response beamforming before acquiring the sensing data;

[0156] A dynamic matching module, used to optimize the transmission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area;

[0157] A dynamic adjustment module is used to dynamically adjust the array gain distribution of the acoustic data based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

[0158] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0159] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0160] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0161] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0162] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0163] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0164] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0165] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0166] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0167] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0168] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A multi-sensor based underwater robot online positioning method, characterized in that: include: Acquiring sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data, and environmental perception data; Based on the acoustic data, determining a first suspicious area, and performing detail focus recognition processing on the first suspicious area through the optical data; According to the result of the detail focus recognition processing, matching the first electromagnetic data and the first environmental data of the first suspicious area from the sensing data, wherein the environmental sensing data includes the first environmental data; The position information of the underwater target is determined according to the electromagnetic data and the environmental perception data.

2. The method according to claim 1, characterized in that The determining of a first suspicious area based on the acoustic data and performing detail focus recognition processing on the first suspicious area through optical data includes: Dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious area; Convert the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates; Performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image; The first image is subjected to detail focus recognition processing based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

3. The method according to claim 2, characterized in that Before acquiring the sensor data, the method further includes: Focusing energy to the first suspicious area by minimum variance distortion-free response beamforming; Optimizing the emission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area; The array gain distribution of the acoustic data is dynamically adjusted based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

4. The method according to claim 1, characterized in that Determining the position information of the underwater target according to the electromagnetic data and the environmental perception data includes: determining electric field gradient change information of the first suspicious area according to the first electromagnetic data; Determining object features in the first visible area according to the electric field gradient change information and the first environmental data; When the object feature meets the first condition, the pre-obtained target center point coordinates and target area edge coordinates are fused by an extended Kalman filter algorithm to obtain the position information.

5. The method according to claim 1, characterized in that: After matching the first electromagnetic data and the first environmental data of the first suspicious area from the sensing data according to the result of the detail focus recognition processing, and the environmental perception data includes the first environmental data, the method further includes: According to the environmental perception data, the sensor data and the electromagnetic data are dynamically adjusted and allocated in terms of weights.

6. An online positioning device for underwater robots based on multiple sensors, characterized in that: include: A data acquisition module, used to acquire sensor data, wherein the sensor data includes acoustic data, optical data, electromagnetic data and environmental perception data; A first processing module, configured to determine a first suspicious area based on the acoustic data, and perform detail focus recognition processing on the first suspicious area through optical data; a matching module, configured to match first electromagnetic data and first environmental data of the first suspicious area from the sensing data according to a result of detail focus recognition processing, wherein the environmental sensing data includes the first environmental data; A position determination module is used to determine the position information of the underwater target based on the electromagnetic data and the environmental perception data.

7. The device according to claim 6, characterized in that The determining of a first suspicious area based on the acoustic data and performing detail focus recognition processing on the first suspicious area through optical data includes: Dynamically adjusting the noise based on the acoustic data, and performing density clustering and isolated point removal on the acoustic cloud points in the acoustic data based on the detection threshold, so as to obtain the center point coordinates and the region edge coordinates of the first suspicious area; Convert the center point coordinates and the area edge coordinates into optical camera coordinates to obtain the target center point coordinates and the target area edge coordinates; Performing restoration processing on the image data included in the optical data according to a preset underwater light transmission model to obtain a first image; The first image is subjected to detail focus recognition processing based on the target center point coordinates and the target area edge coordinates to obtain the detail focus recognition processing result.

8. The device according to claim 7, characterized in that The device also includes: A beamforming module, used for focusing energy to the first suspicious area by minimum variance distortion-free response beamforming before acquiring the sensing data; A dynamic matching module, used to optimize the transmission waveform parameters through a Bayesian probability model to dynamically match the water turbidity of the first suspicious area; A dynamic adjustment module is used to dynamically adjust the array gain distribution of the acoustic data based on the water turbidity, wherein the detection threshold is obtained based on the array gain distribution.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 5 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 5.

Images