Storage tank cleaning robot and real-time positioning method thereof

Abstract

The invention provides a real-time positioning method of a storage tank cleaning robot and the storage tank cleaning robot, and the real-time positioning method of the storage tank cleaning robot comprises the steps: obtaining a storage tank internal map corresponding to a storage tank inner wall space to be cleaned by the storage tank cleaning robot; based on a first positioning module of the storage tank cleaning robot, first positioning information corresponding to the storage tank cleaning robot is obtained; robot projection information, corresponding to the preset projection direction, of the storage tank cleaning robot is obtained, and a map projection block corresponding to the robot projection information is determined based on the storage tank internal map; and according to the first positioning information and the map projection block, real-time positioning parameters of the storage tank cleaning robot are determined. Therefore, based on the projection information of the modeling map, the positioning information determined by the positioning module can be effectively verified and calibrated, and autonomous high-precision positioning of the storage tank cleaning robot in the inner wall of the storage tank is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of petrochemical equipment, and in particular to a real-time positioning method and device for a tank cleaning robot, a tank cleaning robot and a non-transient computer-readable storage medium. Background Art

[0002] At present, in the field of petrochemicals, the cleaning process of oil storage tanks is mainly mechanical cleaning, assisted by manual cleaning. There are a series of problems such as long cleaning time, high safety risks and poor cleaning effect.

[0003] In the above context, intelligent cleaning robots have developed rapidly. Taking crude oil storage tanks as an example, heavy components such as paraffin and asphalt in the oil products, as well as impurities such as soil and heavy metal salts, will naturally settle and accumulate at the bottom of the oil tank due to the difference in specific gravity, forming sludge, which reduces the effective capacity of the oil tank and affects storage and transportation production. The high-pressure jet and strong suction function of the intelligent cleaning robot can safely and efficiently remove the sludge in the tank, which can effectively ensure the safety of personnel and reduce labor costs.

[0004] However, when the cleaning robot moves in the tank, it needs to actively avoid the protruding anode blocks, heating coils and other components at the bottom of the tank. Since the tank is a closed metal tank with strong electromagnetic shielding, it is impossible to effectively verify the reliability of the cleaning robot's positioning results, which also makes it difficult to achieve displacement control of the cleaning robot in the tank.

[0005] In response to the above problems, the industry has not yet proposed a better technical solution. Summary of the invention

[0006] The present invention provides a real-time positioning method and device for a tank cleaning robot, a tank cleaning robot and a non-transitory computer-readable storage medium, which are used to at least solve the defect in the prior art that the tank cleaning robot operating in the tank space cannot be accurately positioned.

[0007] The present invention provides a real-time positioning method for a tank cleaning robot, the method comprising: obtaining a tank internal map corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot; obtaining first positioning information corresponding to the tank cleaning robot based on a first positioning module of the tank cleaning robot; obtaining robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determining a map projection block corresponding to the robot projection information based on the tank internal map; and determining real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0008] According to a real-time positioning method for a tank cleaning robot provided by the present invention, obtaining an internal map of the tank corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot includes: based on a depth camera, collecting inner wall depth information corresponding to the inner wall space of the tank; and constructing an internal map of the tank according to the collected inner wall depth information.

[0009] According to a real-time positioning method for a tank cleaning robot provided by the present invention, the real-time positioning parameters of the tank cleaning robot are determined based on the first positioning information and the map projection block, including: determining a first position matching degree between the first positioning information and the map projection block; and determining the real-time positioning parameters of the tank cleaning robot based on the first positioning information when it is determined that the first position matching degree is greater than a preset first matching degree threshold.

[0010] According to a real-time positioning method for a tank cleaning robot provided by the present invention, the determining of a first position matching degree between the first positioning information and the map projection block comprises: determining a set of projection position feature points corresponding to the map projection block; the internal map of the tank comprises position feature points evenly distributed on the internal map of the tank; based on a preset positioning feature point matching model, determining a set of positioning feature points corresponding to the first positioning information; the positioning feature point matching model defines a matching relationship between the positioning information in the inner wall space of the tank and the position feature points on the internal map of the tank; and determining a first position matching degree according to a feature point intersection between the projection position feature point set and the positioning feature point set.

[0011] According to a real-time positioning method for a tank cleaning robot provided by the present invention, after determining a first position matching degree between the first positioning information and the map projection block, the method further includes: acquiring second positioning information corresponding to the tank cleaning robot based on a second positioning module of the tank cleaning robot; determining a second position matching degree between the second positioning information and the map projection block; and determining the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the second positioning information when it is determined that both the first position matching degree and the second position matching degree are greater than the first matching degree threshold.

[0012] According to a real-time positioning method for a tank cleaning robot provided by the present invention, the first positioning module is a visual positioning module, and the second positioning module is an inertial navigation positioning module.

[0013] According to a real-time positioning method for a tank cleaning robot provided by the present invention, after determining the second position matching degree between the second positioning information and the map projection block, the method also includes: when it is determined that the first position matching degree or the second position matching degree is less than the first matching degree threshold, determining a corresponding positioning accuracy rating according to the first position matching degree and the second position matching degree; the positioning accuracy rating defines the level of positioning accuracy of the positioning system, and the positioning system includes the first positioning module and the second positioning module; when the positioning accuracy rating is a preset first positioning accuracy rating, re-calling the first positioning module and the second positioning module to update the first positioning information and the second positioning information respectively; when the positioning accuracy rating is a preset second positioning accuracy rating, controlling the tank cleaning robot to move to a preset manhole opening position; wherein the first positioning accuracy rating is higher than the second positioning accuracy rating.

[0014] The present invention also provides a real-time positioning device for a tank cleaning robot, the device comprising: a map acquisition unit, used to acquire a tank internal map corresponding to the tank inner wall space to be cleaned by the tank cleaning robot; a positioning acquisition unit, used to determine first positioning information corresponding to the tank cleaning robot based on a first positioning module; a map projection unit, used to acquire robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determine a map projection block corresponding to the robot projection information based on the tank internal map; a positioning determination unit, used to determine the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0015] The present invention also provides a tank cleaning robot, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a real-time positioning method for the tank cleaning robot as described above is implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the real-time positioning method of the tank cleaning robot as described in any one of the above is implemented.

[0017] The present invention provides a real-time positioning method for a tank cleaning robot and a tank cleaning robot, which obtains the internal map of the tank of the inner wall space of the tank, collects the positioning information of the tank cleaning robot, and determines the map projection block of the tank cleaning robot for the internal map of the tank, and determines the real-time positioning parameters of the tank cleaning robot by integrating the positioning information and the map projection block. Thus, the map projection block of the tank cleaning robot relative to the modeling map of the inner wall of the tank in the preset projection direction is integrated with the positioning information determined based on the positioning module, which can effectively verify and calibrate the positioning information, realize the autonomous high-precision positioning of the tank cleaning robot in the inner wall of the tank, and facilitate the precise displacement control of the mobile operation of the tank cleaning robot in the tank space. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flowchart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown;

[0020] Figure 2 A flowchart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown;

[0021] Figure 3 A flowchart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown;

[0022] Figure 4 A structural block diagram of an example of a real-time positioning system for a tank cleaning robot based on visual inertial navigation SLAM according to an embodiment of the present invention is shown;

[0023] Figure 5 Shown based on Figure 4 An algorithm flow chart of an example of determining the real-time position and posture of a tank cleaning robot by a real-time positioning system in;

[0024] Figure 6 A structural block diagram showing an example of a real-time positioning device for a tank cleaning robot according to an embodiment of the present invention;

[0025] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] It should be noted that when the cleaning robot moves in the tank, it needs to actively avoid the protruding anode blocks, heating coils and other components at the bottom of the tank. The tank is a completely closed scene with strong electromagnetic shielding. Wireless positioning and navigation technologies such as GNSS (Global Navigation Satellite System), wireless local area network (WLAN) positioning, WiFi, Bluetooth, etc. cannot be applied. Furthermore, if the tank cleaning robot cannot complete accurate self-positioning, it is impossible to achieve accurate control of the movement of the tank cleaning robot in the tank space.

[0028] Figure 1 A flow chart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown.

[0029] Regarding the execution subject of the method of the embodiment of the present invention, it can be any controller or processor with computing or processing capabilities to achieve the goal of controlling the tank cleaning robot to complete real-time positioning. In some examples, it can be integrated and configured in the tank cleaning robot by software, hardware, or a combination of software and hardware, which should not be limited here.

[0030] The following will take the tank cleaning robot as an exemplary implementation subject to introduce the details of the technical solution involved in the present invention. However, it should be understood that one or more steps involved in the following process can be implemented by one or more controllers or software installed and deployed in the tank cleaning robot.

[0031] like Figure 1 As shown, in step S110, a tank internal map corresponding to the tank inner wall space to be cleaned by the tank cleaning robot is obtained.

[0032] In one example of the embodiment of the present invention, the environment of the inner wall of the tank to be inspected can be simulated in advance by modeling software, and a corresponding tank internal map can be generated, which can then be preset in the tank cleaning robot. In another example of the embodiment of the present invention, the tank cleaning robot can model the environment of the inner wall of the tank in real time while moving on the inner wall of the tank to determine the corresponding tank internal map.

[0033] In step S120, first positioning information corresponding to the tank cleaning robot is acquired based on the first positioning module of the tank cleaning robot.

[0034] Here, the positioning information may include various positioning parameter types, such as absolute coordinate positioning information, relative coordinate positioning information, posture information, etc., which are not limited here. Accordingly, the positioning module may also adopt various types of positioning modules, but considering that the storage tank has strong electromagnetic shielding, a positioning function module that does not require signal interaction with external devices should be preferably adopted.

[0035] In step S130, robot projection information of the tank cleaning robot corresponding to a preset projection direction is obtained, and a map projection block corresponding to the robot projection information is determined based on the internal map of the tank. Here, the map projection block may represent a projection area covered by the projection light onto the inner wall of the tank, and may be determined by matching with the internal map of the tank used to simulate the inner wall space of the tank.

[0036] It should be noted that, on the one hand, the preset projection direction can be a projection corresponding to an absolute direction, such as a projection corresponding to the X-axis or Y-axis of the world coordinate system. On the other hand, the preset projection direction can also be a projection corresponding to a relative direction, such as a projection corresponding to the vertical axis direction of the tank cleaning robot, and so on.

[0037] In step S140, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information and the map projection block. Specifically, the first positioning information of the tank cleaning robot can be matched with the map projection block to determine the real-time positioning parameters of the tank cleaning robot.

[0038] In one example of an embodiment of the present invention, the first positioning information can be calibrated using a map projection block, for example, according to the coordinate axis dimension (for example, the X axis or the Y axis), the average coordinate value between the center position of the map projection block and the first positioning information is calculated to determine the corresponding real-time positioning parameters. In another example of an embodiment of the present invention, the first positioning information can also be verified using a map projection block, for example, whether the first positioning information is in the map projection block.

[0039] Figure 2 A flow chart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown.

[0040] like Figure 2 As shown, in step S210, the inner wall depth information corresponding to the inner wall space of the storage tank is collected based on the depth camera. Preferably, the depth camera can adopt an active ranging structured light RGB-D camera, which has high measurement accuracy (can reach mm level) and long measurement distance (can reach 10m, covering the inner wall space depth of a general storage tank).

[0041] In step S220, the internal map of the tank is constructed based on the collected inner wall depth information. Thus, the tank cleaning robot uses a depth camera and does not rely on external light sources, which solves the problem of being sensitive to ambient light during visual map construction and realizes the automated map construction operation of the inner wall space of the tank with dark ambient light.

[0042] In step S230, first positioning information corresponding to the tank cleaning robot is acquired based on the first positioning module.

[0043] In step S240, robot projection information of the tank cleaning robot corresponding to a preset projection direction is obtained, and a map projection block corresponding to the robot projection information is determined based on the internal map of the tank.

[0044] In step S250, a first position matching degree between the first positioning information and the map projection block is determined.

[0045] Here, the first positioning information may be one or more positioning coordinate values. In one example of an embodiment of the present invention, the first positioning information may represent a single coordinate value, which may be matched with a map projection block to determine whether there is a position match. In another example of an embodiment of the present invention, the first positioning information may also represent multiple coordinate values, such as a coordinate set corresponding to the robot's appearance, such as coordinates of various contour edges of the robot, so as to facilitate more refined motion control or path planning for the robot.

[0046] In some embodiments, the tank cleaning robot can determine a set of projection position feature points corresponding to the map projection block, and the internal map of the tank includes position feature points evenly distributed on the internal map of the tank. Exemplarily, after completing the map modeling, the internal map of the tank is segmented according to a preset feature point interval, thereby obtaining position feature points evenly distributed on the internal map of the tank, and obtaining a corresponding set of projection position feature points according to the coverage area of ​​the map projection block.

[0047] Furthermore, the tank cleaning robot can determine the set of positioning feature points corresponding to the first positioning information based on a preset positioning feature point matching model. It should be noted that the positioning feature point matching model defines the matching relationship between the positioning information in the inner wall space of the tank and the position feature points on the internal map of the tank. In some embodiments, each position feature point has corresponding positioning information, which is managed and maintained by the positioning feature point matching model, thereby realizing the association and conversion of feature points and positioning information.

[0048] Then, the tank cleaning robot determines the first position matching degree according to the intersection of the feature points between the projection position feature point set and the positioning feature point set. In some embodiments, the proportion of feature points included in the projected position feature point set in the positioning feature point set is counted, and the first position matching degree is determined according to the proportion. Thus, by analyzing the map segmentation feature points, the high accuracy of the determined first position matching degree can be guaranteed.

[0049] In step S260, when it is determined that the first position matching degree is greater than a preset first matching degree threshold, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information. Thus, the reliability of the first positioning information is verified by the map projection block, which can ensure the high accuracy of the determined real-time positioning parameters.

[0050] Figure 3 A flow chart showing an example of a real-time positioning method for a tank cleaning robot according to an embodiment of the present invention is shown.

[0051] like Figure 3 As shown, in step S310, a tank internal map corresponding to the tank inner wall space to be cleaned by the tank cleaning robot is obtained.

[0052] In step S320, based on the visual positioning module, the first positioning information corresponding to the tank cleaning robot is obtained. Here, the first positioning information is depth positioning information, which does not rely on signal transmission with the outside world and can achieve a better self-positioning effect.

[0053] In step S330, robot projection information of the tank cleaning robot corresponding to a preset projection direction is obtained, and a map projection block corresponding to the robot projection information is determined based on the internal map of the tank.

[0054] In step S340, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information and the map projection block.

[0055] In step S350, based on the inertial navigation positioning module, the second positioning information corresponding to the tank cleaning robot is obtained. Here, the second positioning information is inertial navigation positioning information, which does not rely on signal transmission with the outside world and can achieve a better self-positioning effect.

[0056] In step S360, the second position matching degree between the second positioning information and the map projection block is determined. Specifically, more details for calculating the second position matching degree can also refer to the implementation method for calculating the first position matching degree above, which will not be repeated here.

[0057] In step S370, the first position matching degree and the second position matching degree are respectively compared with the first matching degree threshold.

[0058] In step S381, when it is determined that both the first position matching degree and the second position matching degree are greater than the first matching degree threshold, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information and the second positioning information.

[0059] It should be noted that the first position matching degree and the second position matching degree are both greater than the first matching degree threshold, indicating that the positioning accuracy of the visual positioning module and the inertial navigation positioning module are both high. Furthermore, by fusing the first positioning information and the second positioning information, accurate perception of the position and posture of the tank robot is achieved.

[0060] In step S383, when it is determined that the first position matching degree or the second position matching degree is less than the first matching degree threshold, a corresponding positioning accuracy rating is determined according to the first position matching degree and the second position matching degree. Here, the positioning accuracy rating defines the level of positioning accuracy of the positioning system, and the positioning system includes a visual positioning module and an inertial navigation positioning module.

[0061] In step S390, when the positioning accuracy rating is a preset first positioning accuracy rating, the first positioning module and the second positioning module are recalled to update the first positioning information and the second positioning information respectively.

[0062] In step S3110, when the positioning accuracy rating is a preset second positioning accuracy rating, the tank cleaning robot is controlled to move to a preset manhole opening position, and the first positioning accuracy rating is higher than the second positioning accuracy rating.

[0063] In some embodiments, by analyzing the first position matching degree and the second position matching degree, when it is found that the positioning accuracy of the visual positioning module or the inertial navigation positioning module cannot meet the requirements, the first position matching degree and the second position matching degree are used to calculate the corresponding positioning accuracy rating, for example, by matching using a preset threshold interval. If the positioning accuracy rating is good, the historical positioning data can be cleared, and the positioning system can be re-called to update the positioning information. In addition, if the positioning accuracy rating is poor, the tank cleaning robot is controlled to move to the manhole opening position of the tank, so that the operator can calibrate or repair the positioning system of the tank cleaning robot.

[0064] Figure 4 A structural block diagram of an example of a real-time positioning system of a tank cleaning robot based on visual inertial navigation SLAM (Simultaneous Localization and Mapping) according to an embodiment of the present invention is shown.

[0065] like Figure 4As shown, the real-time positioning system of the tank cleaning robot includes an RGB-D camera module 410, an IMU (Inertial measurement unit) sensor module 420, a control module 430, a data interaction module 440 and an operation processing module 450. The real-time positioning system of the tank cleaning robot can be installed inside the tank cleaning robot, and by receiving visual image signals, the robot is positioned in real time and a global map of the tank is constructed, and visual navigation SLAM positioning data is output.

[0066] The RGB-D camera module 410 adopts an active ranging structured light RGB-D camera with a measurement distance of up to 10m and a positioning accuracy of mm level. At the same time, the camera does not rely on external light sources to work, solving the problem of sensitivity to ambient light.

[0067] The IMU sensor module 420 calculates the motion trajectory information of the robot device through the path, including the robot's speed / acceleration, rotation / tilt angle and displacement data, and outputs inertial navigation SLAM positioning data. Specifically, through the IMU sensor, acceleration, tilt, impact, vibration, rotation and multi-degree-of-freedom (DoF) motion can be detected and measured. It contains 3 acceleration sensors and 3 angular velocity sensors (gyroscopes) inside, which can realize the position state measurement of 6 degrees of freedom in space.

[0068] The control module 430 establishes two-way communication between the RGB-D camera module 410, the IMU sensor module 420 and the operation processing module 450, and can receive the control signal sent by the operation processing module 450 and the feedback signal sent by the sensor module.

[0069] The data interaction module 440 performs data preprocessing and data relay, segments the feature point map within the local range of the current position, and simultaneously analyzes the feature point data corresponding to the first pose information and the second pose information respectively collected by the RGB-D camera module 410 and the IMU sensor module 420.

[0070] It should be noted that the number of feature points in the feature point map and the number of reference points corresponding to the first pose information and the second pose information are determined by the computing power of the data interaction module 440 and the operation processing module 450. For example, the density distribution of feature points can be 1500 to 2000 per square meter.

[0071] In addition, each module can also achieve electrical connection and data communication with the output host through other auxiliary accessories such as data cables.

[0072] Specifically, the robot cleaning program near the manhole opening of the tank is first started, and the RGB-D camera module 410 is triggered to receive the visual image signal. The camera performs real-time positioning of the robot to obtain the first pose information, and at the same time establishes the initial feature map of the internal environment of the tank, and synchronously transmits the first pose information and the initial feature map to the data interaction module 440. At the same time, the IMU sensor module 420 is synchronously triggered to record the robot's motion trajectory information in real time, and the robot's speed / acceleration, rotation / tilt angle and displacement data are solved by the internal processor to obtain its real-time positioning information, that is, the second pose information, and the second pose information is synchronously transmitted to the data interaction module.

[0073] The data interaction module 440 divides the initial feature map through an internal processor to obtain a current feature map that changes accordingly with the robot's step length, and simultaneously marks a number of evenly distributed feature points in the feature point map. In the feature point map, the feature points corresponding to the first pose information and the second pose information of the tank cleaning robot are determined.

[0074] Then, the internal processor of the data interaction module 440 performs preliminary preprocessing on the data, matches the feature points and reference points, records the number of first feature points corresponding to the first pose information as S1, records the number of second feature points corresponding to the second reference point number as S2, and records the number of projection feature points of the projection block corresponding to the vertical axis direction of the tank robot for the constructed feature point map as S.

[0075] Furthermore, the matching degree between the first number of feature points and the number of projected feature points is recorded as K1 (K1=S1 / S), and the matching degree between the second number of feature points and the number of projected feature points is recorded as K2 (K2=S2 / S). Finally, the matching degree K value is transmitted to the operation processing module 450, and the matching result K value is comprehensively analyzed and optimized according to the algorithm inside its processor.

[0076] In some embodiments, if K1≥0.98, it indicates that the visual positioning performance is excellent; if 0.96≤K1<0.98, it indicates that the visual positioning performance is good; if 0.95≤K1<0.96, it indicates that the visual positioning performance is average; if K1<0.95, it indicates that the visual positioning performance is poor. In addition, if K2≥0.98, it indicates that the inertial navigation positioning performance is excellent; if 0.97≤K2<0.98, it indicates that the inertial navigation positioning performance is good; if 0.96≤K2<0.97, it indicates that the inertial navigation positioning performance is average; if K2<0.96, it indicates that the inertial navigation positioning performance is poor.

[0077] Therefore, the matching results are comprehensively analyzed and the feedback is optimized, integrating data collection, calculation processing, closed-loop control, and map construction. It has more robust performance, can better adapt to the environment, and can achieve repeatable and precise positioning of the robot.

[0078] In some implementations, K1 and K2 need to be considered comprehensively to evaluate the current positioning performance of the positioning system of the tank cleaning robot.

[0079] Table 1 Evaluation table of current positioning performance of the positioning system of the tank cleaning robot

[0080]

[0081] Figure 5 Shown based on Figure 4 An algorithm flow chart of an example of determining the real-time position and posture of a tank cleaning robot by a real-time positioning system in FIG.

[0082] Specifically, the calculation processing module 450 can use the K1 value and the K2 value corresponding to the current position to query the above Table 1 to obtain the evaluation result of the current positioning performance of the positioning system.

[0083] On the one hand, when the control module 430 receives the "excellent" signal provided by the operation processing module 450, it means that the construction of the projection feature points of the robot's current position for the map is accurate, the robot's visual positioning and motion trajectory positioning are correct, and the operation processing module 450 continuously updates and outputs the positioning information.

[0084] On the other hand, when the control module 430 receives a "good" signal from the operation processing module 450, it indicates that there is a deviation in the construction of the projection feature points of the map for the current position of the robot. This signal is fed back to the RGB-D camera module 450, the robot is repositioned and the current feature map is constructed. If the operation processing module 450 judges that it is "excellent", it means that the deviation has been corrected, and the corrected current feature map, robot visual positioning information and motion trajectory information are transmitted through the data cable. Otherwise, this process is repeated.

[0085] On the other hand, when the control module 430 receives the "normal" signal provided by the operation processing module 450, it indicates that the robot's current visual SLAM positioning and inertial SLAM positioning have both large deviations, and this signal is fed back to the RGB-D camera module and the IMU sensor module. At this time, the robot actively deletes the historical positioning data, sets the robot's current position as the starting point, and re-performs visual positioning and inertial positioning of the robot. If the operation processing module 450 judges it as "excellent", it means that the deviation has been corrected, and the corrected current feature map, robot visual positioning information and trajectory information are output.

[0086] On the other hand, when the control module 430 receives a "poor" signal from the operation processing module 450, it indicates that the robot's current visual SLAM positioning and inertial SLAM positioning are seriously deviated, and the visual positioning information and motion trajectory information are unreliable. At this time, the robot actively returns to the original manhole opening position according to the initial preset positioning information and is manually recovered.

[0087] Through the embodiments of the present invention, the tank cleaning robot can integrate multiple positioning methods while overcoming the inherent defects and positioning errors of the corresponding positioning methods, maintaining the ability to respond to environmental changes in unfamiliar environments, and relying on its own visual inertial SLAM positioning system to achieve real-time positioning and autonomous navigation of the robot.

[0088] Figure 6 A structural block diagram of an example of a real-time positioning device for a tank cleaning robot according to an embodiment of the present invention is shown.

[0089] like Figure 6 As shown, the real-time positioning device 600 of the tank cleaning robot includes a map acquisition unit 610 , a positioning acquisition unit 620 , a map projection unit 630 and a positioning determination unit 640 .

[0090] The map acquisition unit 610 is used to acquire a tank internal map corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot.

[0091] The positioning acquisition unit 620 is used to acquire first positioning information corresponding to the tank cleaning robot based on the first positioning module of the tank cleaning robot.

[0092] The map projection unit 630 is used to obtain the robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determine the map projection block corresponding to the robot projection information based on the internal map of the tank.

[0093] The positioning determination unit 640 is used to determine the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0094] For more details about the real-time positioning device of the tank cleaning robot in the embodiment of the present invention, reference can be made to the description of the corresponding operations in the embodiment of the real-time positioning method of the tank cleaning robot described above, and the same or corresponding technical effects can be achieved.

[0095] Figure 7 The following is a schematic diagram of the physical structure of a tank cleaning robot. Figure 7As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the real-time positioning method of the tank cleaning robot, the method comprising: obtaining a tank internal map corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot; obtaining first positioning information corresponding to the tank cleaning robot based on the first positioning module of the tank cleaning robot; obtaining robot projection information corresponding to a preset projection direction of the tank cleaning robot, and determining the map projection block corresponding to the robot projection information based on the tank internal map; determining the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0096] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0097] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the real-time positioning method of the tank cleaning robot provided by the above methods, the method including: obtaining an internal map of the tank corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot; based on a first positioning module of the tank cleaning robot, obtaining first positioning information corresponding to the tank cleaning robot; obtaining robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determining a map projection block corresponding to the robot projection information based on the internal map of the tank; determining the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0098] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the real-time positioning method of a tank cleaning robot provided by the above-mentioned methods, the method comprising: obtaining an internal map of the tank corresponding to the inner wall space of the tank to be cleaned by the tank cleaning robot; based on a first positioning module of the tank cleaning robot, obtaining first positioning information corresponding to the tank cleaning robot; obtaining robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determining a map projection block corresponding to the robot projection information based on the internal map of the tank; and determining the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

[0099] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time positioning method for a tank cleaning robot, characterized in that: The method comprises: Acquire a tank internal map corresponding to the tank inner wall space to be cleaned by the tank cleaning robot; Based on the first positioning module of the tank cleaning robot, obtaining first positioning information corresponding to the tank cleaning robot; Acquire robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determine a map projection block corresponding to the robot projection information based on the internal map of the tank; According to the first positioning information and the map projection block, the real-time positioning parameters of the tank cleaning robot are determined.

2. The real-time positioning method of the tank cleaning robot according to claim 1, characterized in that: The step of obtaining the internal map of the storage tank corresponding to the inner wall space of the storage tank to be cleaned by the storage tank cleaning robot includes: Based on the depth camera, the inner wall depth information corresponding to the inner wall space of the storage tank is collected; Based on the collected inner wall depth information, a map of the interior of the tank is constructed.

3. The real-time positioning method of the tank cleaning robot according to claim 1, characterized in that: Determining the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block includes: Determining a first position matching degree between the first positioning information and the map projection block; When it is determined that the first position matching degree is greater than a preset first matching degree threshold, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information.

4. The real-time positioning method of the tank cleaning robot according to claim 3 is characterized in that: The determining a first position matching degree between the first positioning information and the map projection block includes: Determine a set of projection position feature points corresponding to the map projection block; the internal map of the storage tank includes position feature points evenly distributed on the internal map of the storage tank; Based on a preset positioning feature point matching model, determining a positioning feature point set corresponding to the first positioning information; the positioning feature point matching model defines a matching relationship between the positioning information in the inner wall space of the storage tank and the position feature points on the internal map of the storage tank; A first position matching degree is determined according to an intersection of feature points between the projection position feature point set and the positioning feature point set.

5. The real-time positioning method of the tank cleaning robot according to claim 3, characterized in that: After determining a first position matching degree between the first positioning information and the map projection block, the method further includes: Based on the second positioning module of the tank cleaning robot, obtaining second positioning information corresponding to the tank cleaning robot; Determining a second position matching degree between the second positioning information and the map projection block; When it is determined that both the first position matching degree and the second position matching degree are greater than the first matching degree threshold, the real-time positioning parameters of the tank cleaning robot are determined according to the first positioning information and the second positioning information.

6. The tank cleaning robot according to claim 5, characterized in that: The first positioning module is a visual positioning module, and the second positioning module is an inertial navigation positioning module.

7. The real-time positioning method of the tank cleaning robot according to claim 5, characterized in that: After determining a second position matching degree between the second positioning information and the map projection block, the method further includes: In a case where it is determined that the first position matching degree or the second position matching degree is less than the first matching degree threshold, determining a corresponding positioning accuracy rating according to the first position matching degree and the second position matching degree; the positioning accuracy rating defines a level of positioning accuracy of a positioning system, the positioning system comprising the first positioning module and the second positioning module; When the positioning accuracy rating is a preset first positioning accuracy rating, re-calling the first positioning module and the second positioning module to update the first positioning information and the second positioning information respectively; When the positioning accuracy rating is a preset second positioning accuracy rating, the tank cleaning robot is controlled to move to a preset manhole opening position; wherein the first positioning accuracy rating is higher than the second positioning accuracy rating.

8. A real-time positioning device for a tank cleaning robot, characterized in that: The device comprises: A map acquisition unit, used to acquire an internal map of the storage tank corresponding to the inner wall space of the storage tank to be cleaned by the storage tank cleaning robot; A positioning acquisition unit, configured to acquire first positioning information corresponding to the tank cleaning robot based on a first positioning module; A map projection unit, used to obtain robot projection information of the tank cleaning robot corresponding to a preset projection direction, and determine a map projection block corresponding to the robot projection information based on the internal map of the tank; A positioning determination unit is used to determine the real-time positioning parameters of the tank cleaning robot according to the first positioning information and the map projection block.

9. A tank cleaning robot, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the real-time positioning method of the tank cleaning robot according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the real-time positioning method of the tank cleaning robot according to any one of claims 1 to 7 is implemented.

Images