An indoor hanging rail type aquaculture feeding robot

Abstract

An embodiment of this specification discloses an indoor hanging rail type aquaculture feeding robot, which includes a feeding rail, a feeding vehicle moving along the feeding rail, a signal transceiver module arranged on the feeding vehicle and periodically emitting positioning signals, at least three positioning base stations receiving the positioning signals periodically emitted by the signal transceiver module, and an integrated positioning system; based on the first positioning information corresponding to the feeding vehicle at a target moment and the second positioning information corresponding to the feeding vehicle at the target moment, obtaining the calibrated positioning information corresponding to the feeding vehicle at the target moment; a control module, based on the calibrated positioning information corresponding to the feeding vehicle at the target moment and the point coordinates corresponding to multiple preset feeding points, generating control signals for controlling the movement and feeding of the feeding vehicle. The calibrated positioning information of the feeding vehicle is obtained through a positioning and navigation method of multi-information fusion, the positioning accuracy is improved, and then the movement and feeding of the feeding vehicle are controlled according to the calibrated positioning information, so as to achieve the effect of accurate feeding of the feeding vehicle.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the technical field of intelligent aquaculture equipment, and specifically relate to an indoor suspended rail type aquaculture feeding robot. Background Art

[0002] The mechanization and intelligentization of aquaculture are the inevitable directions for the development of modern green and efficient aquaculture. Feeding is an essential link in aquaculture production, and it is also the link that takes the longest time and has the greatest labor intensity. Feed, as the nutritional source of aquatic animals, is also the source of solid particles, nitrogen, and phosphorus in the aquaculture water body. Unreasonable feeding not only affects the normal growth of aquatic animals, but also causes problems such as feed waste and water quality pollution deterioration.

[0003] Currently, indoor aquaculture usually uses an intelligent feeding device installed above a breeding area including multiple breeding ponds for feeding operations. The intelligent feeding device includes a track and a feeding vehicle that carries feed and runs on the track. Position identification and positioning technologies such as magnetic steel, RFID tags, two-dimensional codes, and limit switches are used to determine the real-time position of the feeding vehicle, control the feeding vehicle to stop at a designated position and perform feeding, making full use of the space above the breeding ponds to achieve timed, fixed-point, and quantitative feeding operations.

[0004] However, due to the limitation of the refresh frequency of the existing position identification and positioning technology, combined with the easy occurrence of occlusion interference in the indoor environment, the positioning accuracy is not high, resulting in inaccurate positions for the feeding vehicle to stop and perform feeding operations, thus affecting the normal growth of aquatic animals and causing feed waste. Summary of the Invention

[0005] Embodiments of this specification provide an indoor suspended rail type aquaculture feeding robot, which obtains calibrated positioning information of the feeding vehicle through a positioning and navigation method of multi-information fusion, improves the positioning accuracy, and then controls the movement and feeding of the feeding vehicle according to the calibrated positioning information, thereby achieving the effect of accurate feeding of the feeding vehicle.

[0006] The technical solution is as follows:

[0007] Embodiments of this specification provide an indoor suspended rail type aquaculture feeding robot, including:

[0008] An indoor suspended rail type aquaculture feeding robot includes a feeding track, a feeding vehicle moving along the feeding track, a signal transceiver module that periodically emits positioning signals on the feeding vehicle, at least three positioning base stations that receive the positioning signals periodically emitted by the signal transceiver module, and a fusion positioning system;

[0009] The fusion positioning system includes a first data processing module, a data acquisition module, a second data processing module, and a control module;

[0010] The first data processing module obtains the first positioning information of the feeding vehicle at the target moment according to the reception moment of the positioning signal sent by the signal transceiver module received by each positioning base station corresponding thereto at the target moment.

[0011] The data acquisition module acquires the direction information and displacement information corresponding to different moments before the target moment of the feeding vehicle, and obtains the second positioning information of the feeding vehicle at the target moment based on the direction information and displacement information corresponding to different moments before the target moment of the feeding vehicle.

[0012] The second data processing module obtains the calibrated positioning information of the feeding vehicle at the target moment based on the first positioning information of the feeding vehicle at the target moment and the second positioning information of the feeding vehicle at the target moment.

[0013] The control module generates a control signal for controlling the movement and feeding of the feeding vehicle based on the calibrated positioning information of the feeding vehicle at the target moment and the point coordinates corresponding to multiple preset feeding points.

[0014] As a preferred solution, each positioning base station and the signal transceiver module are arranged at the same horizontal height, and the connection lines of at least three positioning base stations form an acute triangle.

[0015] As a preferred solution, the first data processing module includes a time processing unit and a distance processing unit.

[0016] The time processing unit obtains the signal reception duration corresponding to each positioning base station based on the target moment and the reception moment of the positioning signal sent by the signal transceiver module received by each positioning base station corresponding thereto.

[0017] The distance processing unit obtains the interval distance from the signal transceiver module to each positioning base station corresponding thereto at the target moment and uses it as the first positioning information of the feeding vehicle at the target moment based on the signal reception duration corresponding to each positioning base station and the signal reception duration-interval distance fitting function obtained in advance for each positioning base station.

[0018] The obtaining method of the signal reception duration-interval distance fitting function corresponding to any positioning base station in the distance processing unit is as follows:

[0019] Based on the positioning signals sent by the signal transceiver module at multiple known coordinates, obtain the signal reception duration corresponding to the positioning base station when receiving the positioning signal sent by the signal transceiver module at each known coordinate.

[0020] Based on each known coordinate and the fixed coordinate of the positioning base station, determine the interval distance between each known coordinate and the positioning base station corresponding thereto.

[0021] Taking the signal reception duration corresponding to the positioning signals sent by the signal transceiver module at each known coordinate and the interval distance between each known coordinate and the positioning base station as sample data, a signal reception duration-interval distance fitting function is obtained through the least squares method.

[0022] As a preferred solution, the data acquisition module includes a position estimation unit and an interval estimation unit;

[0023] The position estimation unit obtains the estimated coordinates of the feeding vehicle at the target moment based on the direction information, displacement information corresponding to different moments after the moment when the feeding vehicle sends the positioning signal in the previous cycle before the target moment and before the target moment, and the calibration positioning information of the feeding vehicle corresponding to the moment when the signal transceiver module sends the positioning signal in the previous cycle before the target moment;

[0024] The interval estimation unit obtains the estimated intervals from the signal transceiver module to each positioning base station at the target moment according to the estimated coordinates of the feeding vehicle at the target moment and the fixed coordinates of each positioning base station, and takes the calibration positioning information of the feeding vehicle corresponding to the moment when the signal transceiver module sends the positioning signal in the previous cycle before the target moment, the estimated coordinates of the feeding vehicle at the target moment, and the estimated intervals from the signal transceiver module to each positioning base station as the second positioning information of the feeding vehicle at the target moment.

[0025] As a preferred solution, the second data processing module includes a difference matrix calculation unit, a state matrix calculation unit, a model calculation unit, and a position calibration unit;

[0026] The difference matrix calculation unit obtains the difference matrix of the signal transceiver module to each positioning base station at that moment based on the interval distance and the estimated interval from the signal transceiver module to each positioning base station at any moment , where and are respectively the estimated interval and the interval distance from the signal transceiver module to the i-th positioning base station at the k-th moment;

[0027] The state matrix calculation unit obtains the state matrix of the feeding vehicle at that moment based on the estimated coordinates, direction information, and displacement information of the feeding vehicle at any moment , where represents the coordinate value of the feeding vehicle in the coordinate system, represents the displacement amounts of the feeding vehicle on the X-axis and Y-axis at adjacent moments, and θ is the direction quantity;

[0028] The model calculation unit uses the extended Kalman filter model with the difference matrix at any moment as the observed quantity and the state matrix at that moment as the predicted quantity Obtain the Kalman gain corresponding to this moment;

[0029] A position calibration unit that obtains the calibrated positioning information of the feeding vehicle at this moment based on the Kalman gain at any moment;

[0030] The state equation of the extended Kalman filter model in the model calculation unit is:

[0031]

[0032] Where, is the state matrix corresponding to the calibrated positioning information of the feeding vehicle at the (k - 1)th moment, is the input control quantity of the feeding vehicle at the (k - 1)th moment, is a non - linear state function, is the process noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of Q;

[0033] The observation equation of the extended Kalman filter model in the model calculation unit is:

[0034]

[0035] Where, is a non - linear observation function, is the observation noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of R.

[0036] As a preferred solution, the second data processing module further includes an error processing unit and a noise processing unit:

[0037] An error processing unit that obtains the non - line - of - sight error corresponding to each positioning base station based on the difference matrix;

[0038] A noise processing unit that obtains the measurement noise factor corresponding to each positioning base station based on the magnitude relationship between the non - line - of - sight error corresponding to each positioning base station and a preset value;

[0039] A model calculation unit that corrects the Kalman gain based on the measurement noise factor corresponding to each positioning base station.

[0040] As a preferred solution, the control module, when the feeding vehicle stops at the point coordinates corresponding to any preset feeding point, generates a feeding control signal for controlling the feeding of the feeding vehicle based on the feeding radius corresponding to the preset feeding point.

[0041] As a preferred solution, the feeding vehicle is provided with a spreading turntable;

[0042] A control module, when the feeding vehicle stops at the point coordinates corresponding to any preset feeding point, generates a feeding control signal for controlling the rotation speed of the feeding turntable based on the feeding radius corresponding to the preset feeding point and a preset minimum feeding radius, so that the feeding radius of the feeding vehicle changes reciprocally between the feeding radius corresponding to the preset feeding point and the preset minimum feeding radius.

[0043] As a preferred solution, the feeding vehicle includes a discharging box, and the feeding turntable is arranged at the discharging port of the discharging box;

[0044] The control module generates a feeding control signal for controlling the discharging flow rate of the discharging box based on the feeding radius of the feeding vehicle.

[0045] As a preferred solution, the feeding vehicle includes an image acquisition module for acquiring feeding images;

[0046] The control module generates a feeding control signal for controlling the discharging of the discharging box based on the feeding images.

[0047] The beneficial effects brought by the technical solutions provided in some embodiments of this specification at least include:

[0048] 1. On the one hand, the first positioning information of the feeding vehicle obtained by the positioning base station and the signal transceiver module is subject to environmental interference and errors, and the first positioning information obtained from the periodically emitted positioning signals is discontinuous position information. There is no corresponding first positioning information between the times of two adjacent positioning signals, and only a linear approximate position can be obtained. On the other hand, there are also errors in the second positioning information of the feeding vehicle calculated from the real-time direction information and displacement information of the feeding vehicle, and the cumulative error will also increase as the moving distance of the feeding vehicle increases. Therefore, the present invention fuses the first positioning information and the second positioning information at the same moment, corrects the second positioning information at the same moment of the first positioning information according to the first positioning information, and at the same time corrects the jump value in the first positioning information according to the second positioning information, so as to obtain more accurate calibrated positioning information than the two positioning methods, and on this basis, realizes the precise feeding control of the feeding vehicle;

[0049] 2. All positioning base stations are set at the horizontal height of the signal transceiver module, which can ignore the height positioning of the feeding vehicle, reduce the calculation complexity and errors, and the positional relationship of the three positioning base stations forms an acute triangle. On the one hand, it can provide more stable signal coverage in the positioning area, reduce signal blind spots, and on the other hand, it helps to perform better error correction in the positioning calculation, improving the accuracy and reliability of positioning;

[0050] 3. After obtaining more accurate calibration and positioning information, the spreading radius of the feeding vehicle can be controlled according to the feeding pool radius corresponding to the feeding point. As long as the spreading radius does not exceed the upper limit of the feeding radius corresponding to the feeding pool, the feed will not be spilled outside the pool, avoiding feed waste and achieving accurate and uniform spreading effects. The uniform distribution of the feed helps all animals obtain sufficient nutrition, thereby promoting their balanced growth, reducing size differences, and also preventing animals from fighting for food, increasing stress and affecting their health. It can also reduce the accumulation of feed residues and feces in the pool, contribute to maintaining water quality, reduce the frequency of water changes and the cost of water quality treatment, and improve the utilization rate of the feed. Brief Description of the Drawings

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0052] Figure 1 is a schematic structural diagram of an indoor hanging rail type aquaculture feeding robot provided by an embodiment of this specification (the positioning base station is not shown);

[0053] Figure 2 is a schematic structural diagram of a fusion positioning system in an indoor hanging rail type aquaculture feeding robot provided by an embodiment of this specification;

[0054] Figure 3 is a schematic structural diagram of a fusion positioning system in an indoor hanging rail type aquaculture feeding robot provided by another embodiment of this specification.

[0055] In the figure: 1. Feeding track; 2. Feeding vehicle; 21. Signal transceiver module; 22. Spreading turntable; 221. Driving motor; 23. Feeding box; 24. Rotating motor; 241. Drum; 300. Fusion positioning system; 301. First data processing module; 3011. Time processing unit; 3012. Distance processing unit; 302. Data acquisition module; 3021. Position estimation unit; 3022. Interval estimation unit; 303. Second data processing module; 3031. Difference matrix calculation unit; 3032. State matrix calculation unit; 3033. Model calculation unit; 3034. Position calibration unit; 3035. Error processing unit; 3036. Noise processing unit; 304. Control module. Detailed Embodiments

[0056] The following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the drawings in the embodiments of this specification.

[0057] The terms "first", "second", "third", etc. in the description, claims and the above-mentioned drawings of this specification are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices.

[0058] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes can be made to the functions and arrangements of the described elements without departing from the scope of the content of this specification. Various processes or components can be appropriately omitted, substituted or added in each example. For example, the described method can be executed in a different order from the described order, and various steps can be added, omitted or combined. In addition, the features described for some examples can be combined into other examples.

[0059] Referring to the attached Figure 1 With the attached Figure 2 , Figure 1 is a schematic structural diagram of an indoor hanging track type aquaculture feeding robot provided by an embodiment of this specification, Figure 2 is a schematic structural diagram of the fusion positioning system 300.

[0060] An indoor hanging track type aquaculture feeding robot includes a feeding track 1, a feeding vehicle 2 moving along the feeding track 1, a signal transceiver module 21 arranged on the feeding vehicle 2 that periodically emits positioning signals, at least three positioning base stations that receive the positioning signals periodically emitted by the signal transceiver module 21, and a fusion positioning system 300. Figure 1 The positioning base stations are not shown in , and the positioning base stations are arranged around the feeding track 1.

[0061] Illustratively, in theory, two positioning base stations can determine a position on a two-dimensional plane, but three base stations can provide more observation data, which helps to correct various errors, such as systematic errors, random errors, etc. By comparing the data of different base stations, these errors can be corrected more accurately.

[0062] In multiple embodiments of this specification, each positioning base station is arranged at the same horizontal height as the signal transceiver module 21, and the connection lines of at least three positioning base stations form an acute triangle.

[0063] Explanatory, since the overall height of the feeding track 1 is consistent, each positioning base station is set at the horizontal height of the signal transceiver module 21, so the height positioning can be ignored, reducing the computational complexity and error. Taking the setting of three positioning base stations as an example, the positional relationship of the three positioning base stations forms an acute triangle, making the angle between any two base stations and the third base station relatively large. On the one hand, it can provide more stable signal coverage within the positioning area and reduce signal blind spots. On the other hand, it helps to perform better error correction in the positioning calculation. If the angle between the base stations is too small, the positioning algorithm may have difficulty distinguishing the error source due to similar or overlapping signal paths, resulting in deviation of the positioning result, improving the accuracy and reliability of the positioning.

[0064] The integrated positioning system 300 includes a first data processing module 301, a data acquisition module 302, a second data processing module 303, and a control module 304.

[0065] The first data processing module 301 obtains the first positioning information of the feeding vehicle 2 at the target moment according to the reception moment of the positioning signal sent by the signal transceiver module 21 received by each positioning base station at the corresponding target moment.

[0066] Illustratively, the signal transceiver module 21 on the feeding vehicle 2 periodically sends out positioning signals. For the same positioning signal, the time when each positioning base station receives the positioning signal is the reception moment corresponding to that positioning base station. By calculating the difference between the reception moment and the transmission moment of the positioning signal and the propagation speed of the positioning signal, the distance from the positioning base station to the signal transceiver module 21 can be obtained, thereby obtaining the first positioning information of the feeding vehicle 2. Each positioning signal sent in each cycle corresponds to a first positioning information.

[0067] The data acquisition module 302 obtains the direction information and displacement information of the feeding vehicle 2 at different moments before the target moment, and obtains the second positioning information of the feeding vehicle 2 at the target moment based on the direction information and displacement information of the feeding vehicle 2 at different moments before the target moment.

[0068] Explanatory, the direction information of the feeding vehicle 2 can be obtained in real time by setting an inertial sensor on the feeding vehicle 2, and the displacement information of the feeding vehicle 2 can be obtained in real time by setting an encoder on the driving wheel of the feeding vehicle 2. Taking the position of the feeding vehicle 2 determined at the previous moment of the target moment as the starting point, the position of the feeding vehicle 2 at the target moment can be calculated according to the direction information and displacement information obtained in real time, thereby obtaining the second positioning information of the feeding vehicle 2. The transmission moment of each cycle of positioning signals can correspond to the second positioning information at the same moment.

[0069] The second data processing module 303 obtains the calibrated positioning information of the feeding vehicle 2 at the target moment based on the first positioning information of the feeding vehicle 2 corresponding to the target moment and the second positioning information of the feeding vehicle 2 corresponding to the target moment.

[0070] Illustratively, two sets of positioning information obtained by two different methods at the target moment are fused and calibrated to obtain the calibrated positioning information as the basis for determining the position of the feeding vehicle 2.

[0071] The control module 304 generates a control signal for controlling the movement and feeding of the feeding vehicle 2 based on the calibrated positioning information of the feeding vehicle 2 corresponding to the target moment and the point coordinates corresponding to each of the multiple preset feeding points.

[0072] Illustratively, the breeding area includes multiple breeding ponds, and the feeding track 1 is arranged above the breeding area and passes through each breeding pond. In order to feed evenly, when the feeding track 1 is arranged, it will pass through the position directly above the center of each breeding pond. Ideally, the feeding vehicle 2 stops directly above the center position of the breeding pond for the best feeding effect. At this time, the distance from the feeding vehicle 2 to the edge of the breeding pond is equal in all directions, which is conducive to uniform feeding and also helps to avoid feed spilling out of the breeding pond.

[0073] Explanatorily, the preset feeding point is the position corresponding to the center of the breeding pond at the height of the feeding track 1. A movement command is issued to the feeding vehicle 2 according to the calibrated positioning information, so that the feeding vehicle 2 accurately stops at the preset feeding point, and feeding operations are carried out at the preset feeding point, thereby achieving the effect of accurate feeding of the feeding vehicle 2 and avoiding feed waste caused by deviation of the feeding position of the feeding vehicle 2. After feeding, continue to control the feeding vehicle 2 to move on the feeding track 1, and judge whether it reaches the preset feeding point according to the new calibrated positioning information.

[0074] On the one hand, the first positioning information of the feeding vehicle 2 obtained by the positioning base station and the signal transceiver module 21 is subject to environmental interference and errors, and the first positioning information obtained by the periodically emitted positioning signals is discontinuous position information. There is no corresponding first positioning information between the times of two adjacent positioning signals, and only linearly approximate positions can be obtained; on the other hand, the second positioning information of the feeding vehicle 2 calculated from the real-time direction information and displacement information of the feeding vehicle 2 also has errors, and the cumulative error will become larger as the moving distance of the feeding vehicle 2 increases. Therefore, the first positioning information and the second positioning information at the same moment are fused, the second positioning information at the same moment of the first positioning information is corrected according to the first positioning information, and at the same time, the jump value in the first positioning information is corrected according to the second positioning information, so as to obtain more accurate calibrated positioning information than the two positioning methods, and on this basis, accurate feeding control of the feeding vehicle 2 is realized.

[0075] In an embodiment of this specification, refer to the attached Figure 3, the first data processing module 301 includes a time processing unit 3011 and a distance processing unit 3012;

[0076] The time processing unit 3011 obtains the signal reception duration corresponding to each positioning base station based on the target time and the reception time of the positioning signal sent by the target time signal transceiver module 21 corresponding to each positioning base station.

[0077] The distance processing unit 3012 obtains the interval distance from the target time signal transceiver module 21 to each positioning base station based on the signal reception duration corresponding to each positioning base station and the pre-obtained signal reception duration-interval distance fitting function corresponding to each positioning base station, and uses it as the first positioning information of the feeding vehicle 2 at the target time.

[0078] The obtaining method of the signal reception duration-interval distance fitting function corresponding to any positioning base station in the distance processing unit 3012 is as follows:

[0079] Based on the positioning signals sent by the signal transceiver module 21 at multiple known coordinates, obtain the signal reception duration corresponding to the positioning base station when receiving the positioning signals sent by the signal transceiver module 21 at each known coordinate.

[0080] Based on each known coordinate and the fixed coordinate of the positioning base station, determine the interval distance between each known coordinate and the positioning base station.

[0081] Using the signal reception duration corresponding to the positioning base station when receiving the positioning signals sent by the signal transceiver module 21 at each known coordinate and the interval distance between each known coordinate and the positioning base station as sample data, obtain the signal reception duration-interval distance fitting function through the least squares method.

[0082] Explanatorily, a linear regression function between the signal reception duration and the interval distance of each positioning base station in the use environment is pre-fitted by the least squares method, so as to correct the relative error at different interval distances, improve the calculation accuracy and anti-interference ability. The interval distance from the signal transceiver module 21 to each positioning base station is directly used as the first positioning information. In theory, the interval distance between two positioning base stations can determine the position coordinates of a signal transceiver module 21. However, due to the existence of noise errors, three or more positioning base stations will determine many position coordinates instead of a single definite point. The multiple position coordinates enclose an area. In the prior art, generally, an optimal estimated position is calculated directly within the enclosed area based on multiple interval distances as the position coordinates of the signal transceiver module 21. In the present invention, the interval distance is not processed, and the interval distance from the signal transceiver module 21 to each positioning base station is directly used as the first positioning information.

[0083] In one embodiment of this specification, the data acquisition module 302 includes a position estimation unit 3021 and an interval estimation unit 3022;

[0084] The position estimation unit 3021 obtains the estimated coordinates of the feeding vehicle 2 at the target moment based on the direction information, displacement information corresponding to different moments after the moment when the positioning signal was sent by the feeding vehicle 2 in the previous cycle before the target moment and before the target moment, and the calibration positioning information of the feeding vehicle 2 corresponding to the moment when the positioning signal was sent by the signal transceiver module 21 in the previous cycle before the target moment;

[0085] The interval estimation unit 3022 obtains the estimated interval from the signal transceiver module 21 to each positioning base station at the target moment according to the estimated coordinates of the feeding vehicle 2 at the target moment and the fixed coordinates of each positioning base station, and uses the calibration positioning information of the feeding vehicle 2 corresponding to the moment when the positioning signal was sent by the signal transceiver module 21 in the previous cycle before the target moment, the estimated coordinates of the feeding vehicle 2 at the target moment, and the estimated interval from the signal transceiver module 21 to each positioning base station at the target moment as the second positioning information of the feeding vehicle 2 at the target moment.

[0086] Explanatorily, taking the calibration positioning information of the feeding vehicle 2 corresponding to the moment when the positioning signal was sent in the previous cycle before the target moment as the starting coordinates, and cumulatively calculating the direction information and displacement information obtained in real time after the moment when the positioning signal was sent in the previous cycle before the target moment, theoretically, the real-time position of the feeding vehicle can be obtained. However, since the position calculated in this way will gradually deviate from the real position as the data accumulates, it cannot be used as the basis for determining the position of the feeding vehicle for a long time. Therefore, only the direction information and displacement information obtained in real time after the moment when the positioning signal was sent in the previous cycle before the target moment and before the target moment are cumulatively calculated to obtain the estimated coordinates of the feeding vehicle 2 at the target moment. The interval estimation unit 3022 calculates the distance between two points through the estimated coordinates and the fixed coordinates of each positioning base station, and can obtain the estimated interval from the signal transceiver module 21 to each positioning base station at the target moment, so as to match the first positioning information. In addition, the second positioning information also includes the starting coordinates and estimated coordinates corresponding to the target moment.

[0087] In one embodiment of this specification, the second data processing module 303 includes a difference matrix calculation unit 3031, a state matrix calculation unit 3032, a model calculation unit 3033, and a position calibration unit 3034;

[0088] The difference matrix calculation unit 3031 obtains the difference matrix of the signal transceiver module 21 to each positioning base station at this moment based on the interval distance and estimated interval from the signal transceiver module 21 to each positioning base station at any moment , where and are respectively the estimated interval and the interval distance from the signal transceiver module 21 to the i-th positioning base station at the k-th moment;

[0089] The state matrix calculation unit 3032 obtains the state matrix corresponding to the feeding vehicle 2 at that moment based on the estimated coordinates, direction information, and displacement information of the feeding vehicle 2 corresponding to any moment , where represents the coordinate value of the feeding vehicle 2 in the coordinate system, represents the displacement amounts of the feeding vehicle 2 on the X-axis and Y-axis at adjacent moments, and θ is the direction quantity;

[0090] The model calculation unit 3033 uses the difference matrix at any moment as the observed quantity based on the extended Kalman filter model and the state matrix at that moment as the predicted quantity to obtain the Kalman gain corresponding to that moment;

[0091] The position calibration unit 3034 obtains the calibrated positioning information corresponding to the feeding vehicle 2 at that moment based on the Kalman gain at any moment;

[0092] The state equation of the extended Kalman filter model in the model calculation unit 3033 is:

[0093]

[0094] where is the state matrix corresponding to the calibrated positioning information of the feeding vehicle 2 at the (k - 1)-th moment, is the input control quantity to the feeding vehicle 2 at the (k - 1)-th moment, is the non-linear state function, is the process noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of Q;

[0095] The observation equation of the extended Kalman filter model in the model calculation unit 3033 is:

[0096]

[0097] where is the non-linear observation function, is the observation noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of R.

[0098] Explanatory, the Extended Kalman Filter is a state estimation method for nonlinear systems. It adapts to the state estimation problem of nonlinear systems by linearizing nonlinear functions. In the basic model, there are state equations and observation equations. The state equation describes the law of the system state evolving over time, showing how the system state transfers from one time step to the next. In the ideal case (without process noise ), the state transition function completely determines the evolution of the state. The observation equation describes how to obtain observation data from the system state, showing how the observation data is generated from the system state. In practical applications, observations are usually incomplete or noisy, which means that we cannot directly obtain the complete state of the system, but can only infer the state through noisy observations.

[0099] More accurate calibration and positioning information is obtained according to the Extended Kalman Filter algorithm. The specific process of the Extended Kalman Filter algorithm is divided into five steps: state prediction, error covariance prediction, Kalman gain update, state update, and error covariance update.

[0100] Taking the target moment as the k-th moment, and the moment when the feeding vehicle 2 sends a positioning signal in the previous cycle before the target moment as the (k - 1)-th moment. First, the state matrix at the k-th moment is obtained based on the estimated coordinates, displacement information, and direction information corresponding to the feeding vehicle 2 at the k-th moment, and is used as the prediction at the k-th moment , is the state matrix corresponding to the calibration and positioning information of the feeding vehicle 2 at the (k - 1)-th moment. Taking the direction information and displacement information of the feeding vehicle 2 obtained between the (k - 1)-th moment and the k-th moment as the input control quantity at the (k - 1)-th moment , from which the Jacobian matrix of the state equation can be obtained according to the state equation of the Extended Kalman Filter model . Then, taking the difference matrix corresponding to the k-th moment as the observed quantity , from which the Jacobian matrix of the observation equation can be obtained according to the observation equation of the Extended Kalman Filter model .

[0101] Predict the error covariance at the k-th moment according to the formula , where is the error covariance at the (k - 1)-th moment, is the transpose of .

[0102] Calculate the Kalman gain at the k-th moment according to the formula , where is the transpose of .

[0103] Finally, according to the formula Update the state matrix at time k , to obtain the optimal estimated value of the state matrix at time k, and at the same time, according to the formula Update the error covariance at time k for the operation of the next cycle, where is the identity matrix. The state matrix is the calibrated positioning information corresponding to time k, which includes the position coordinates of the feeding vehicle 2.

[0104] Explanatorily, the initial states of the covariance matrix P, the process noise matrix Q, and the observation noise matrix R are all identity matrices. The estimated coordinates are corrected according to the differences between the estimated intervals and the interval distances corresponding to different positioning base stations by the signal transceiver module 21 at time k, so as to obtain the optimal estimated value at time k .

[0105] Illustratively, existing suspended-rail aquaculture feeding devices all use common spatial positioning technologies to determine the position of the feeding vehicle 2 and perform feeding after reaching the preset feeding point. Since the update frequency of spatial positioning technologies is not high, the positioning data is actually discrete point data, and the position data of the feeding vehicle 2 between two adjacent positioning data is unknown. After obtaining the calibrated positioning information corresponding to each periodically emitted positioning signal in this embodiment, the real-time position coordinates of the feeding vehicle 2 within the interval time between two positioning signals are obtained according to the calibrated positioning information of each positioning signal and the displacement information and direction information of the feeding vehicle 2 obtained in real time, so as to achieve the effect of real-time and accurate positioning of the position of the feeding vehicle 2.

[0106] In an embodiment of this specification, the second data processing module 303 further includes an error processing unit 3035 and a noise processing unit 3036:

[0107] The error processing unit 3035 obtains the non-line-of-sight error corresponding to each positioning base station based on the difference matrix;

[0108] The noise processing unit 3036 obtains the measurement noise factor corresponding to each positioning base station based on the magnitude relationship between the non-line-of-sight error corresponding to each positioning base station and a preset value;

[0109] The model calculation unit 3033 corrects the Kalman gain based on the measurement noise factor corresponding to each positioning base station.

[0110] Illustrative, the non-line-of-sight error means that before the positioning signal reaches the positioning base station, it does not propagate along a straight-line path, but due to encountering obstacles such as buildings and terrain, it undergoes reflection, refraction, or scattering, resulting in the signal path deviating from the direct line-of-sight path. This will cause the receiving time of the positioning signal sent by the signal transceiver module 21 at the target moment to be delayed when received by the positioning base station. Therefore, steps for discriminating and mitigating non-line-of-sight errors are added, that is, the set non-line-of-sight error discrimination threshold m. When the feeding vehicle 2 is running normally on the feeding track 1, the difference between the estimated interval and the interval distance of the same positioning base station should change within a certain range with the positioning signals corresponding to each cycle, and there should be no jumps.

[0111] According to the formula , the non-line-of-sight error of the i-th positioning base station is obtained .

[0112] According to the magnitude relationship with the preset value m, the measurement noise factor corresponding to the i-th positioning base station is obtained , when it is greater than m, it is determined that non-line-of-sight error has occurred at this positioning base station, and the larger it is, the greater the non-line-of-sight interference on this positioning base station. When , when , , where , using the measurement noise factors of all positioning base stations to form a residual matrix , and adding the residual matrix to the formula for calculating the Kalman gain to mitigate the impact of non-line-of-sight error on positioning, obtaining , thereby correcting the Kalman gain at time k.

[0113] In an embodiment of this specification, the control module 304, when the feeding vehicle 2 stops at the point coordinates corresponding to any preset feeding point, generates a feeding control signal for controlling the feeding of the feeding vehicle 2 based on the feeding radius corresponding to the preset feeding point.

[0114] Explanatory, after obtaining more accurate calibration positioning information, the upper limit of the feeding radius can be set according to the feeding pool radius corresponding to the feeding point. Most feeding pools are circular pools. If it is a square pool, the radius of its largest inscribed circle is used as the upper limit of the feeding radius. When the feeding vehicle 2 is performing the feeding operation, it controls the spreading radius of the feeding operation according to the upper limit of the feeding radius corresponding to the current feeding pool. As long as the spreading radius does not exceed the upper limit of the feeding radius corresponding to the feeding pool, the feed will not be scattered outside the pool, avoiding feed waste and achieving a precise and uniform spreading effect. It should be noted that uniform spreading is the goal pursued by the feeding operation and an important means to improve the feed utilization rate. The uniform distribution of the feed helps all animals obtain sufficient nutrition, thereby promoting their balanced growth, reducing size differences, and can also prevent animals from fighting for food, increasing stress and affecting their health. It can also reduce the accumulation of feed residues and feces in the pool, help maintain water quality, reduce the frequency of water changes and the cost of water quality treatment, etc.

[0115] In an embodiment of the present specification, the feeding vehicle 2 is provided with a spreading turntable 22;

[0116] A control module 304, when the feeding vehicle 2 stops at the point coordinates corresponding to any preset feeding point, generates a feeding control signal for controlling the rotation speed of the spreading turntable 22 based on the feeding radius corresponding to the preset feeding point and the preset minimum feeding radius, so that the feeding radius of the feeding vehicle 2 reciprocally changes between the feeding radius corresponding to the preset feeding point and the preset minimum feeding radius.

[0117] Explanatory, when the feeding vehicle 2 is directly above the center of the feeding pool, since the height of the feeding track 1 is fixed, the rotation speed of the spreading turntable 22 determines the spreading distance. The feeding control signal controls the rotation speed of the driving motor 221 of the spreading turntable 22, driving the spreading turntable 22 to rotate to generate different centrifugal forces, so that the spreading distance of the spreading turntable 22 reciprocally changes between the minimum feeding radius and the upper limit of the feeding radius corresponding to the feeding pool. When the set minimum feeding radius is zero, that is, the rotation speed of the spreading turntable 22 is zero, the size of the minimum feeding radius can be specifically set according to actual needs, thereby achieving uniform spreading of almost the entire feeding pool and further improving the feed utilization rate.

[0118] In an embodiment of the present specification, the spreading turntable 22 is detachable and has various different types to adapt to different spreading requirements.

[0119] In an embodiment of the present specification, the feeding vehicle 2 includes a discharge box 23, and the spreading turntable 22 is arranged at the discharge port of the discharge box 23;

[0120] A control module 304 generates a feeding control signal for controlling the discharge flow rate of the discharge box 23 based on the feeding radius of the feeding vehicle 2.

[0121] Explanatory, a roller 241 can be arranged at the discharge port of the discharge bin 23 and a rotating motor 24 for driving the roller 241 to rotate. A groove parallel to the rotation axis of the rotating motor 24 is provided on the outer surface of the roller 241. When the rotating motor 24 rotates, the feed in the groove is transferred out of the discharge bin 23 and then leaves the groove due to gravity. The discharge flow rate is controlled by controlling the rotation speed of the rotating motor 24.

[0122] Illustrative, at different rotation speeds of the spreading turntable 22, the spreading radius is different, so the circle formed by the spreading radius is also larger. Therefore, according to the change of the feeding radius, the discharge flow rate of the discharge bin 23 is controlled. The larger the feeding radius, the larger the discharge flow rate of the discharge bin 23, so that the spreading density for different feeding radii tends to be consistent, making the feed in the feeding pond more uniform and further improving the feed utilization rate.

[0123] In an embodiment of the present specification, the feeding vehicle 2 includes an image acquisition module for acquiring feeding images;

[0124] A control module 304 for generating a feeding control signal for controlling the discharge of the discharge bin 23 based on the feeding images.

[0125] Explanatory, the image acquisition module is used to acquire video information during fish feeding. The feeding progress can be analyzed and calculated by analyzing the number of water splashes and residual bait on the water surface in the video. When there are fewer water splashes and more residual bait in the video and it exceeds the threshold, the feeding vehicle 2 is controlled to stop feeding. This can avoid over-spreading, reduce the accumulation of feed residues and feces in the pond, help maintain water quality, reduce the frequency of water change and the cost of water quality treatment, and also improve the feed utilization rate.

[0126] Implementation principle: Taking the starting point coordinates of the feeding track 1 as the starting point, the real-time position of the feeding vehicle 2 is calculated according to the real-time displacement information and direction information of the feeding vehicle 2 obtained. At the same time, the signal transceiver module 21 on the feeding vehicle 2 periodically emits positioning signals.

[0127] The first positioning information and the second positioning information corresponding to the first positioning signal are obtained, and the calibrated positioning information corresponding to the first positioning signal is obtained through the extended Kalman filter model. Then, taking the calibrated positioning information corresponding to the first positioning signal as a new starting point, the real-time position of the feeding vehicle 2 is calculated according to the real-time displacement information and direction information of the feeding vehicle 2 obtained.

[0128] Continue to obtain the first positioning information and the second positioning information corresponding to the second positioning signal and perform the same steps, thereby obtaining the real-time position coordinates of the feeding vehicle 2.

[0129] Control the feeding vehicle 2 to stop at a preset feeding point for feeding operations according to the real-time position coordinates of the feeding vehicle 2, and control the rotation speed of the feeding disc 22 to achieve reciprocating changes in the feeding distance between the feeding radius corresponding to the current feeding point and the preset minimum feeding radius. Synchronously control the discharge flow rate of the discharge box 23 according to the change of the feeding radius. Analyze the video information of the fish feeding collected by the image acquisition module. When there are fewer water splashes and more residual bait in the video and it exceeds the threshold, control the feeding vehicle 2 to stop feeding.

[0130] Re-control the feeding vehicle 2 to move along the feeding track 1 and obtain the real-time position coordinates of the feeding vehicle 2.

[0131] The above embodiments are only described in the preferred embodiment mode of this specification, and do not limit the scope of this specification. Without departing from the design spirit of this specification, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of this specification shall fall within the protection scope determined by the claims of this specification.

Claims

1. An indoor hanging rail type aquaculture feeding robot, characterized in that: It includes a feeding track, a feeding vehicle moving along the feeding track, a signal transceiver module arranged on the feeding vehicle and periodically sending positioning signals, at least three positioning base stations receiving the positioning signals periodically sent by the signal transceiver module, and an integrated positioning system; The integrated positioning system includes a first data processing module, a data acquisition module, a second data processing module, and a control module; The first data processing module obtains the first positioning information of the feeding vehicle at the target moment according to the reception moment of the positioning signal sent by the signal transceiver module received by each positioning base station corresponding to it at the target moment; The data acquisition module obtains the direction information and displacement information corresponding to different moments before the target moment of the feeding vehicle, and obtains the second positioning information of the feeding vehicle at the target moment based on the direction information and displacement information corresponding to different moments before the target moment of the feeding vehicle; The second data processing module obtains the calibrated positioning information of the feeding vehicle at the target moment based on the first positioning information of the feeding vehicle at the target moment and the second positioning information of the feeding vehicle at the target moment; The control module generates a control signal for controlling the movement and feeding of the feeding vehicle based on the calibrated positioning information of the feeding vehicle at the target moment and the point coordinates corresponding to multiple preset feeding points; The data acquisition module includes a position estimation unit and an interval estimation unit; The position estimation unit obtains the estimated coordinates of the feeding vehicle at the target moment based on the direction information, displacement information corresponding to different moments after the moment when the signal transceiver module sent the positioning signal in the previous cycle before the target moment and before the target moment, and the calibrated positioning information of the feeding vehicle corresponding to the moment when the signal transceiver module sent the positioning signal in the previous cycle before the target moment; The interval estimation unit obtains the estimated interval from the signal transceiver module to each positioning base station at the target moment according to the estimated coordinates of the feeding vehicle at the target moment and the fixed coordinates corresponding to each positioning base station, and uses the calibrated positioning information of the feeding vehicle corresponding to the moment when the signal transceiver module sent the positioning signal in the previous cycle before the target moment, the estimated coordinates of the feeding vehicle at the target moment, and the estimated interval from the signal transceiver module to each positioning base station at the target moment as the second positioning information of the feeding vehicle at the target moment.

2. The indoor hanging track type aquaculture feeding robot according to claim 1, wherein: Each of the positioning base stations and the signal transceiver module are arranged at the same horizontal height, and the connection lines of at least three of the positioning base stations form an acute triangle.

3. The indoor hanging rail type aquaculture feeding robot according to claim 1, characterized in that: The first data processing module includes a time processing unit and a distance processing unit; The time processing unit obtains the signal reception duration corresponding to each positioning base station based on the target moment and the reception moment of the positioning signal sent by the signal transceiver module received by each positioning base station corresponding to it; The distance processing unit obtains the interval distance from the signal transceiver module to each positioning base station at the target moment based on the signal reception duration corresponding to each positioning base station and the signal reception duration - interval distance fitting function obtained in advance for each positioning base station, and uses it as the first positioning information of the feeding vehicle at the target moment; The method for obtaining the signal reception duration-interval distance fitting function corresponding to any positioning base station in the distance processing unit is as follows: Based on the positioning signals sent by the signal transceiver module at multiple known coordinates, obtain the signal reception duration corresponding to the positioning base station when it receives the positioning signals sent by the signal transceiver module at each known coordinate; Based on each known coordinate and the fixed coordinate of the positioning base station, determine the interval distance between each known coordinate and the positioning base station; Taking the signal reception duration corresponding to the positioning base station when it receives the positioning signals sent by the signal transceiver module at each known coordinate and the interval distance between each known coordinate and the positioning base station as sample data, obtain the signal reception duration-interval distance fitting function through the least squares method.

4. The indoor hanging rail type aquaculture feeding robot according to claim 3, characterized in that, The second data processing module includes a difference matrix calculation unit, a state matrix calculation unit, a model calculation unit, and a position calibration unit; The difference matrix calculation unit obtains the difference matrix of the signal transceiver module to each positioning base station at a certain moment [(d′ i，k ) 2 -(d″ i，k ) 2 T , where d′ i，k and d″ i，k are respectively the estimated interval and the interval distance of the signal transceiver module to the i-th positioning base station at the k-th moment;​ The state matrix calculation unit obtains the state matrix (x, y, Δx, Δy, θ) corresponding to the feeding vehicle at any moment based on the estimated coordinates, direction information, and displacement information of the feeding vehicle at that moment T , where (x, y) represents the coordinate values of the feeding vehicle in the coordinate system, (Δx, Δy) represents the displacement amounts of the feeding vehicle on the X-axis and Y-axis at adjacent moments, and θ is the direction quantity; The model calculation unit uses the difference matrix at any moment as the observed quantity Z based on the extended Kalman filter model k and the state matrix at this moment as the predicted quantity to obtain the Kalman gain corresponding to this moment; The position calibration unit obtains the calibrated positioning information corresponding to the feeding vehicle at any moment based on the Kalman gain at that moment; The state equation of the extended Kalman filter model in the model calculation unit is: where, x k-1 is the state matrix corresponding to the calibration positioning information of the feeding vehicle at the (k - 1)th moment, u k-1 is the input control quantity of the feeding vehicle at the (k - 1)th moment, f(·) is a nonlinear state function, W k is the process noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of Q; The observation equation of the extended Kalman filter model in the model calculation unit is: where \(h(\cdot)\) is a non - linear observation function, and \(V\) k is the observation noise, which follows a multivariate normal distribution with a mean of 0 and a covariance matrix of \(R\).

5. An indoor hanging rail type aquaculture feeding robot according to claim 4, characterized in that, The second data processing module further includes an error processing unit and a noise processing unit: The error processing unit obtains the non-line-of-sight error corresponding to each positioning base station based on the difference matrix; The noise processing unit obtains the measurement noise factor corresponding to each positioning base station based on the magnitude relationship between the non-line-of-sight error corresponding to each positioning base station and a preset value; The model calculation unit corrects the Kalman gain based on the measurement noise factor corresponding to each positioning base station.

6. The indoor hanging rail type aquaculture feeding robot according to claim 1, wherein: When the feeding vehicle stops at the point coordinates corresponding to any preset feeding point, the control module generates a feeding control signal for controlling the feeding of the feeding vehicle based on the feeding radius corresponding to the preset feeding point.

7. The indoor suspended track type aquaculture feeding robot according to claim 6, characterized in that: The feeding vehicle is provided with a feeding turntable; When the feeding vehicle stops at the point coordinates corresponding to any preset feeding point, the control module generates a feeding control signal for controlling the rotation speed of the feeding turntable based on the feeding radius corresponding to the preset feeding point and a preset minimum feeding radius, so that the feeding radius of the feeding vehicle changes reciprocally between the feeding radius corresponding to the preset feeding point and the preset minimum feeding radius.

8. An indoor suspended track type aquaculture feeding robot according to claim 7, characterized in that: The feeding vehicle includes a discharge box, and the feeding turntable is arranged at the discharge port of the discharge box; The control module generates a feeding control signal for controlling the discharge flow rate of the discharge box based on the feeding radius of the feeding vehicle.

9. The indoor hanging rail type aquaculture feeding robot according to claim 8, wherein: The feeding vehicle includes an image acquisition module for acquiring feeding images; The control module generates a feeding control signal for controlling the discharge of the discharge box based on the feeding images.

Images