Method, system and equipment for positioning humanoid robot and medium

Abstract

The invention discloses a humanoid robot positioning method, system and device and a medium, and the method comprises the steps: obtaining a building structure map and first precise positioning information in a current moving process of a humanoid robot, and obtaining initial positioning information and an environment feature set of the humanoid robot; according to the building structure map, carrying out building environment matching on the environment feature set to obtain reference positioning information; according to a positioning correction condition, performing first condition verification on the first precise positioning information and the initial positioning information to obtain a first condition verification result; and if the first condition verification result is that the positioning correction condition is met, performing positioning correction on the initial positioning information according to the reference positioning information to obtain second accurate positioning information of the humanoid robot in the current moving process. According to the method, the positioning precision of the humanoid robot in a complex building environment can be effectively improved. The invention relates to the technical field of humanoid robots.

Description

Technical Field

[0001] This invention relates to the field of humanoid robot technology, and in particular to a positioning method, system, device and medium for humanoid robots. Background Technology

[0002] With the increasing popularity of humanoid robots in various fields, the demand for high-precision positioning in ultra-large areas and multi-story buildings is becoming more and more urgent.

[0003] Currently, the relevant technologies are usually based on a combination of base stations and inertial navigation to achieve indoor positioning of humanoid robots. Specifically, indoor positioning is achieved by combining Wi-Fi, Bluetooth, or UWB base stations with inertial navigation. However, due to the complex internal environment of complex buildings (such as very large multi-story buildings), there are often many interference sources and network coverage blind spots. In addition, inertial navigation suffers from error accumulation. Therefore, the positioning accuracy of this method in complex building environments is not satisfactory.

[0004] Therefore, the problems with the relevant technologies still need to be solved and optimized. Summary of the Invention

[0005] The purpose of this invention is to at least partially solve one of the technical problems existing in the related art.

[0006] Therefore, one objective of this invention is to provide a positioning method, system, device, and medium for a humanoid robot, wherein the method can effectively improve the positioning accuracy of the humanoid robot in complex building environments.

[0007] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of this application include: In a first aspect, embodiments of this application provide a method for locating a humanoid robot, comprising: During the current movement of the humanoid robot, a building structure map and first precise positioning information are acquired, as well as the initial positioning information and environmental feature set of the humanoid robot are acquired; the first precise positioning information is the second precise positioning information of the humanoid robot in the previous movement process; Based on the building structure map, the environmental feature set is matched with the building environment to obtain reference positioning information; Based on the positioning correction conditions, the first precise positioning information and the initial positioning information are subjected to a first condition verification to obtain the first condition verification result. If the first condition verification result satisfies the positioning correction condition, then the initial positioning information is corrected according to the reference positioning information to obtain the second precise positioning information of the humanoid robot during the current movement process.

[0008] In addition, the method according to the above embodiments of this application may also have the following additional technical features: Furthermore, in one embodiment of this application, the step of performing building environment matching on the environmental feature set based on the building structure map to obtain reference positioning information includes: Based on the environmental feature set, obtain floor height data, environmental structure data, and environmental image data; Based on the floor height data, the building structure map is filtered to obtain a layered structure map; the layered structure map is used to represent the floor structure sub-maps in all floor structure sub-maps of the building structure map, where the floor height range includes the floor height data; Based on the environmental structure data and the environmental image data, the layered structure map is subjected to layered environmental matching to obtain the reference positioning information.

[0009] Furthermore, in one embodiment of this application, the step of performing layered environment matching on the layered structure map based on the environmental structure data and the environmental image data to obtain the reference positioning information includes: Obtain structural thresholds and visual overlap thresholds; Based on the structural threshold and the environmental structural data, the hierarchical structural map is filtered for structural regions to obtain several candidate regions. The candidate regions are used to represent structural regions in the hierarchical structural map whose structural similarity is greater than the structural threshold. The structural similarity is the similarity between the structural region and the environmental structural data. Based on the visual overlap threshold and the environmental image data, visual feature matching is performed on all the candidate regions to obtain the reference positioning information.

[0010] Further, in one embodiment of this application, the positioning correction condition is a displacement distance threshold, and the step of performing a first condition verification on the first precise positioning information and the initial positioning information according to the positioning correction condition to obtain a first condition verification result includes: The displacement time interval is determined based on the positioning timestamp of the first precise positioning information and the positioning timestamp of the initial positioning information; Based on the displacement time interval, obtain the velocity sequence data of the humanoid robot during the current movement process; The velocity sequence data is integrated based on the displacement time interval to obtain the moving distance of the humanoid robot; Based on the displacement distance threshold, the movement distance is compared with the threshold to obtain the first condition verification result.

[0011] Furthermore, in one embodiment of this application, the step of correcting the initial positioning information based on the reference positioning information to obtain the second precise positioning information of the humanoid robot during its current movement includes: Based on the initial positioning information, Kalman prediction processing is performed to obtain the prior state and prior covariance; Based on the prior covariance, Kalman gain analysis is performed on the reference positioning information to obtain gain data; Based on the gain data and the reference positioning information, the prior state is updated to obtain the posterior state, and the posterior state is determined as the second precise positioning information.

[0012] Furthermore, in one embodiment of this application, the method further includes: Obtain network stability conditions, as well as the current signal strength of the humanoid robot and the duration of the current signal strength; Based on the network stability conditions, a second condition verification is performed on the current signal strength and the duration to obtain the second condition verification result; Based on the verification result of the second condition, the second precise positioning information is fused to obtain the fused second precise positioning information. Based on the fused second precise positioning information, the building structure map is updated to obtain an updated building structure map.

[0013] Furthermore, in one embodiment of this application, updating the building structure map based on the fused second precise positioning information to obtain an updated building structure map includes: Obtain the distance difference threshold; Based on the environmental feature set corresponding to the fused second precise positioning information, feature positions are extracted from the building structure map to obtain target location information; the target location information is the location information of the target feature in the building structure map that corresponds to the environmental feature set. Based on the fused second precise positioning information, distance analysis is performed on the target location information to obtain the interval distance; Based on the distance difference threshold, a threshold analysis is performed on the interval distance to obtain the threshold analysis result; If the threshold analysis result indicates that the interval distance is greater than the distance difference threshold, then the target features in the building structure map are updated with location information based on the fused second precise positioning information to obtain the updated building structure map.

[0014] Secondly, embodiments of this application provide a positioning system for a humanoid robot, comprising: The first processing unit is used to acquire a building structure map and first precise positioning information during the current movement of the humanoid robot, as well as to acquire the initial positioning information and environmental feature set of the humanoid robot; the first precise positioning information is the second precise positioning information of the humanoid robot in the previous movement process; The second processing unit is used to perform building environment matching on the environmental feature set based on the building structure map to obtain reference positioning information; The third processing unit is used to perform a first condition verification on the first precise positioning information and the initial positioning information according to the positioning correction conditions, and obtain the first condition verification result. The fourth processing unit is used to correct the initial positioning information based on the reference positioning information when the first condition verification result satisfies the positioning correction condition, so as to obtain the second accurate positioning information of the humanoid robot in the current movement process.

[0015] Thirdly, embodiments of this application also provide an electronic device, including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0016] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a processor-executable program, which, when executed by the processor, is used to implement the above-described method.

[0017] The advantages and beneficial effects of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application: This application discloses a humanoid robot localization method, system, device, and medium. The method acquires a building structure map and first precise positioning information, as well as initial positioning information and an environmental feature set, during the current movement of the humanoid robot. The first precise positioning information is the second precise positioning information of the humanoid robot in the previous movement. Based on the building structure map, the environmental feature set is matched against the building environment to obtain reference positioning information. According to positioning correction conditions, the first precise positioning information and the initial positioning information are subjected to a first condition verification to obtain a first condition verification result. If the first condition verification result satisfies the positioning correction condition, the initial positioning information is corrected based on the reference positioning information to obtain the second precise positioning information of the humanoid robot in the current movement. This method, by correcting the initial positioning information based on the reference positioning information obtained from the building structure map when the first condition verification result satisfies the positioning correction condition, can effectively improve the positioning accuracy of the humanoid robot in complex building environments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of this application or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions in this application. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 A flowchart illustrating a positioning method for a humanoid robot provided in an embodiment of this application; Figure 2 A schematic diagram of the framework of a positioning system for a humanoid robot provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] Currently, related technologies typically rely on the following three technological paths, specifically: 1. Base Station / Inertial Navigation Combined Positioning: Positioning is achieved by combining Wi-Fi, Bluetooth, or UWB base stations with inertial navigation. However, in complex buildings, base station signals are easily blocked by walls and metal equipment (such as metal components in a car factory welding workshop, which can cause UWB signal attenuation rates of over 50%), leading to signal interruptions. Inertial navigation also suffers from "drift" problems; after a network interruption of 5 minutes, the accumulated error can reach more than 1 meter, and after 10 minutes, the error can be ≥2 meters, which cannot meet the requirements for high precision.

[0023] 2. Visual / LiDAR SLAM Localization: This technology constructs a real-time map and locates the target using visual sensors or LiDAR. This method demands high computational resources (consuming over 80% of the robot's embedded computing power) and is prone to localization failure in environments with limited features (such as open warehouses or underground parking lots) due to insufficient feature points. Furthermore, SLAM maps cannot be associated with pre-defined structural information about buildings (such as floor layouts or equipment numbers), making it difficult to accurately locate specific targets (such as specific machine tools in a factory or specific wards in a hospital).

[0024] 3. Emerging UWB+IMU fusion positioning and visual landmark positioning: Although UWB+IMU fusion positioning has high accuracy in the short term, it has poor signal stability and rapid error accumulation in dense metal environments. Visual landmark positioning relies on manually deployed QR codes and reflectors, which is not only costly (thousands of landmarks need to be deployed in very large buildings), but the landmarks are also easily blocked by people and goods (such as promotional booths in shopping malls blocking QR codes), leading to positioning interruption.

[0025] In summary, in practical applications, due to the common blind spots in network coverage of complex buildings (such as multi-story buildings with very large areas) (which usually occur in underground warehouses and high-rise corridors) and frequent signal fluctuations, related technologies struggle to balance positioning accuracy and scene adaptability when the network is unstable.

[0026] It should be noted that the aforementioned related technologies are only used to assist in understanding the technical solutions of this application and do not mean that they belong to the publicly disclosed prior art.

[0027] In view of this, embodiments of this application provide a localization method, system, device, and medium for a humanoid robot. The method corrects the initial localization information based on reference localization information obtained from a building structure map when the first condition verification result satisfies the localization correction condition. Specifically, during each movement of the humanoid robot, environmental features are matched against the building environment based on the building structure map of the current movement (which may be an updated building structure map obtained in the previous movement). The reference localization information obtained from the matching is then used to correct the initial localization information using Kalman filtering when the localization correction condition is met, resulting in second accurate localization information. This effectively improves the localization accuracy of the humanoid robot in complex building environments.

[0028] Reference Figure 1 In this application embodiment, a positioning method, system, device, and medium for a humanoid robot include: Step 110: During the current movement of the humanoid robot, acquire the building structure map and the first precise positioning information, as well as the initial positioning information and environmental feature set of the humanoid robot; the first precise positioning information is the second precise positioning information of the humanoid robot in the previous movement process; In this embodiment of the application, during the current movement of the humanoid robot, a building structure map and first precise positioning information can be acquired. The building structure map can be the updated building structure map of the humanoid robot in the previous movement. The building structure map can be constructed by architectural design CAD drawings and on-site laser mapping data. It records the structural and visual features of several floors of the building where the humanoid robot is located, as well as the height range of each floor. The structural features can specifically be the coordinates of the wall position, the distribution parameters of beams and columns (spacing, diameter), equipment position information (such as machine tool number and corresponding coordinates), etc.; while the visual features can be high-definition image templates of signs in the building and feature point annotation data (such as the corner pixel coordinates of door signs).

[0029] It is understandable that the initial positioning information can be the position information determined by the base station positioning system such as UWB base station or Wi-Fi base station during the current movement of the humanoid robot in a stable network state; or the position information output by the inertial navigation system when the network is interrupted or the network signal is weak (such as signal strength < -80dBm).

[0030] The environmental feature set can be data collected by three types of sensors on a humanoid robot and integrated into an environmental feature set corresponding to the current movement process. Specifically, structural data based on the humanoid robot's own coordinate system can be collected by LiDAR, including the horizontal distance to the wall, the cross-sectional dimensions and spacing of beams and columns, and the position coordinates of doors and windows; visual sensors can be used to capture images of signs inside the building (door signs, equipment nameplates, safety slogans, etc.), and feature points of the RGB images can be extracted using the SIFT algorithm; and an absolute height value based on the ground floor of the building can be collected by a barometric altimeter to calculate the height difference between adjacent floors for subsequent floor matching.

[0031] Step 120: Based on the building structure map, perform building environment matching on the environmental feature set to obtain reference positioning information; In this embodiment of the application, building environment matching can be based on the feature data recorded in the building structure map and matched with the feature data of the environmental feature set to determine the reference positioning information of the humanoid robot.

[0032] In some embodiments, the step of performing building environment matching on the environmental feature set based on the building structure map to obtain reference positioning information includes: Based on the environmental feature set, obtain floor height data, environmental structure data, and environmental image data; Based on the floor height data, the building structure map is filtered to obtain a layered structure map; the layered structure map is used to represent the floor structure sub-maps in all floor structure sub-maps of the building structure map, where the floor height range includes the floor height data; In this embodiment, the floor height data can be the absolute height value collected by the aforementioned barometric altimeter; the environmental structure data can be the structural data collected by the aforementioned lidar; and the environmental image data can be the image feature points extracted by the aforementioned visual sensor.

[0033] Understandably, floor map filtering can compare the floor height data with the pre-stored height ranges of each floor in the building structure map (floor 1: 0-3m, floor 2: 3-6m) to determine the current floor of the humanoid robot. Then, based on the current floor of the humanoid robot, the floor structure sub-graph of the humanoid robot's current floor is determined from all floor structure sub-graphs and recorded as a layered structure sub-graph. This can avoid the increased computation time caused by the humanoid robot searching the entire map.

[0034] Based on the environmental structure data and the environmental image data, the layered structure map is subjected to layered environmental matching to obtain the reference positioning information.

[0035] Further, the step of performing layered environment matching on the layered structure map based on the environmental structure data and the environmental image data to obtain the reference positioning information includes: Obtain structural thresholds and visual overlap thresholds; Based on the structural threshold and the environmental structural data, the hierarchical structural map is filtered for structural regions to obtain several candidate regions. The candidate regions are used to represent structural regions in the hierarchical structural map whose structural similarity is greater than the structural threshold. The structural similarity is the similarity between the structural region and the environmental structural data. Based on the visual overlap threshold and the environmental image data, visual feature matching is performed on all the candidate regions to obtain the reference positioning information.

[0036] In this embodiment, the structural region is a region representation of the aforementioned structural features. Structural region selection can be based on the Euclidean distance algorithm, calculating the similarity between environmental structural data and structural regions in a hierarchical structural map, and identifying structural regions with a similarity greater than a structural threshold as candidate regions.

[0037] For example, taking a structural threshold of 0.9 and the wall distance d and beam-column spacing s of a certain structural region as an example, structural regions with a structural similarity greater than 0.9 can be identified as candidate regions, and the structural similarity of a certain structural region can be expressed as:

[0038] in, For structural similarity; This refers to the wall distance of a specific structure within the environmental structure data. This refers to the distance between walls in a layered map. The beam-column spacing in the environmental structural data; This refers to the beam-column spacing in the layered structure map.

[0039] Understandably, for any candidate region, the image feature points of that candidate region in the environmental image data can be compared with the landmark templates in the hierarchical structure map to obtain the feature point overlap rate of the candidate region. The feature point overlap rates of other candidate regions are obtained in the same way. Then, the positions corresponding to the candidate regions with feature point overlap rates greater than the visual overlap threshold are used as reference positioning information. The specific value of the visual overlap threshold can be flexibly set, such as any one of 0.8, 0.85, 0.93, etc.

[0040] Step 130: Based on the positioning correction conditions, perform a first condition verification on the first precise positioning information and the initial positioning information to obtain the first condition verification result; In this embodiment of the application, the first condition verification may be to verify whether the first precise positioning information and the initial positioning information meet the positioning correction conditions, thereby obtaining the first condition verification result.

[0041] In some embodiments, the positioning correction condition is a displacement distance threshold, and the step of performing a first condition verification on the first precise positioning information and the initial positioning information according to the positioning correction condition to obtain a first condition verification result includes: The displacement time interval is determined based on the positioning timestamp of the first precise positioning information and the positioning timestamp of the initial positioning information; Based on the displacement time interval, obtain the velocity sequence data of the humanoid robot during the current movement process; The velocity sequence data is integrated based on the displacement time interval to obtain the moving distance of the humanoid robot; Based on the displacement distance threshold, the movement distance is compared with the threshold to obtain the first condition verification result.

[0042] In the first implementation, if the positioning correction condition is a displacement distance threshold, this displacement distance threshold can be set according to the scenario (e.g., in an industrial scenario with dense equipment, the displacement distance threshold can be 0.5m; while in a commercial scenario with open space, the displacement distance threshold can be 2m). The displacement time interval can be the time period between the positioning timestamp of the first precise positioning information and the initial positioning information and the positioning timestamp; while the velocity sequence data can be the set of velocity data output by the inertial navigation system carried by the humanoid robot within this displacement time interval, with each velocity data corresponding to a different timestamp.

[0043] Understandably, the velocity sequence data can be integrated based on the position-time interval to obtain the distance the humanoid robot moves within that displacement-time interval. This distance is used to characterize the distance the humanoid robot has moved since the last positioning correction, and can be expressed as:

[0044] in, The distance traveled; This is the time of the last location correction, which is also the timestamp of the first accurate location information; This is the current timestamp, which is also the timestamp of the initial location information; This refers to the velocity data at the current time t in the velocity sequence data.

[0045] The threshold comparison can be a comparison between the displacement distance threshold and the movement distance. If the movement distance is greater than the displacement distance threshold, a first condition verification result indicating that the positioning correction condition is met can be generated, and step 140 can be executed; or, if the movement distance is less than or equal to the displacement distance threshold, a first condition verification result indicating that the positioning correction condition is not met can be generated, and step 110 can be returned to be executed.

[0046] In the second embodiment, the positioning correction condition can be a displacement time threshold, which can also be determined according to actual conditions. In this case, the first condition verification can be to verify the relationship between the displacement time threshold and the displacement time. Specifically, the displacement time can be the absolute time difference between the positioning timestamp of the first precise positioning information and the initial positioning information and the positioning timestamp. The subsequent content is similar to that of the first embodiment and can be easily deduced by analogy, so it will not be repeated here.

[0047] Step 140: If the first condition verification result satisfies the positioning correction condition, then the initial positioning information is corrected according to the reference positioning information to obtain the second accurate positioning information of the humanoid robot in the current movement process.

[0048] In this embodiment, the positioning correction can be based on the Kalman equation, using reference positioning information to correct the initial positioning information, thereby obtaining the second accurate positioning information.

[0049] In some embodiments, the step of correcting the initial positioning information based on the reference positioning information to obtain the second precise positioning information of the humanoid robot during its current movement includes: Based on the initial positioning information, Kalman prediction processing is performed to obtain the prior state and prior covariance; Based on the prior covariance, Kalman gain analysis is performed on the reference positioning information to obtain gain data; Based on the gain data and the reference positioning information, the prior state is updated to obtain the posterior state, and the posterior state is determined as the second precise positioning information.

[0050] In this embodiment of the application, the Kalman prediction processing can be based on the state equation in the Kalman equation to process the initial positioning information to obtain the prior state and the prior covariance, which can be expressed as:

[0051] in, The prior state includes the state vector of the humanoid robot at time k. , The x-axis coordinates of a humanoid robot. For a humanoid robot, the y-axis coordinate is... For the height coordinates of a humanoid robot, The speed at which a humanoid robot moves; It is a 4×4 state transition matrix, in the form of , The sampling time interval (ranging from 0.05s to 0.2s, determined based on the sensor's sampling frequency); The state vector at time k-1 (i.e., the positioning timestamp of the first precise positioning information) is output by the inertial navigation system; It is a 4×2 control input matrix, in the form of This is used to map the effect of control inputs on the state; , which is the control input at time k (i.e., the positioning timestamp of the initial positioning information), For motor angular velocity, This refers to the steering angle; For prior covariance; Let be the posterior covariance at time k-1; This is the transpose of A; The process noise covariance matrix is ​​determined based on the static error calibration results of the inertial navigation system.

[0052] Kalman gain analysis can be based on prior covariance to calculate the Kalman gain of the humanoid robot during its current movement, denoted as gain data. This gain data can be expressed as:

[0053] in, This refers to the Kalman gain, i.e., the gain data. The observation matrix H is the transpose of the observation matrix H. , is used to map the state vector to the observation space; R is the observation noise covariance matrix, which is determined based on the accuracy of the lidar and the factory parameters of the vision sensor.

[0054] Understandably, the state update can be based on gain data, reference positioning information, and prior state to calculate the posterior state of the humanoid robot during the current movement process, and based on gain data and prior covariance to calculate the posterior covariance during the current movement process. This posterior state and posterior covariance can be expressed as:

[0055] in, This is the posterior state; The observation vector at time k is the reference positioning information during the current movement, which includes the x-axis coordinate, y-axis coordinate, and altitude coordinate. The posterior covariance during the current movement process; It is an identity matrix.

[0056] In some embodiments, the method further includes: Obtain network stability conditions, as well as the current signal strength of the humanoid robot and the duration of the current signal strength; Based on the network stability conditions, a second condition verification is performed on the current signal strength and the duration to obtain the second condition verification result; Based on the verification result of the second condition, the second precise positioning information is fused to obtain the fused second precise positioning information. In this embodiment, the network stability condition is used to determine whether the communication network currently connected to the humanoid robot has returned to stability. Specifically, it can be that the signal strength is greater than a strength threshold (e.g., -70dBm) and the duration is greater than a certain time (e.g., 5s). The current signal strength of the humanoid robot can be the signal strength when the humanoid robot determines the initial positioning information and / or reference positioning information, while the duration can be the cumulative time of the humanoid robot at the current signal strength.

[0057] Understandably, the second condition verification can determine whether the current signal strength and duration meet the network stability conditions, thereby obtaining the second condition verification result. Specifically, if the second condition verification result indicates that the signal strength and duration meet the network stability conditions, then the base station's positioning information for the humanoid robot can be obtained, denoted as base station positioning information; weights can be assigned to the base station positioning information and the second precise positioning information, and the base station positioning information and the second precise positioning information can be fused using a weighted average method to obtain the fused second precise positioning information, which can be represented as:

[0058] in, This is the second precise positioning information after fusion; To accurately determine the weights; This is the second precise positioning information before fusion; Assign positioning weights to base stations; Location information for the base station.

[0059] For example, in the first implementation, when the network is stable (e.g., signal strength ≥ -60dBm). In the second implementation, when the network is relatively stable (e.g., -70dBm ≤ signal strength < -60dBm),... .

[0060] Based on the fused second precise positioning information, the building structure map is updated to obtain an updated building structure map.

[0061] Further, the step of updating the building structure map based on the fused second precise positioning information to obtain an updated building structure map includes: Obtain the distance difference threshold; Based on the environmental feature set corresponding to the fused second precise positioning information, feature positions are extracted from the building structure map to obtain target location information; the target location information is the location information of the target feature in the building structure map that corresponds to the environmental feature set. Based on the fused second precise positioning information, distance analysis is performed on the target location information to obtain the interval distance; Based on the distance difference threshold, a threshold analysis is performed on the interval distance to obtain the threshold analysis result; If the threshold analysis result indicates that the interval distance is greater than the distance difference threshold, then the target features in the building structure map are updated with location information based on the fused second precise positioning information to obtain the updated building structure map.

[0062] In this embodiment, the specific value of the distance difference threshold can be flexibly set according to the actual situation, such as any one of 0.05m, 0.1m, 0.2m, etc. Feature location extraction can be based on the target features indicated by the fused second precise positioning information in the environmental feature set to determine the location information of the corresponding target features in the building structure map, which is denoted as target location information. The target feature can be a specific feature in structural features or visual features, such as the location of a wall in structural features or a landmark in visual features.

[0063] It is understandable that distance analysis can be used to calculate the Euclidean distance between the fused second precise positioning information and the target location information, thereby obtaining the interval distance; threshold analysis can be used to compare the magnitude relationship between the interval distance and the distance difference threshold, thereby obtaining the threshold analysis result.

[0064] Specifically, if the threshold analysis result is that the interval distance is greater than the distance difference threshold, the location information of the target features in the building structure map can be updated to the fused second precise positioning information, thereby obtaining the updated building structure map; or, if the threshold analysis result is that the interval distance is less than or equal to the distance difference threshold, the location information of the target features in the building structure map can be retained.

[0065] Reference Figure 2The positioning system for a humanoid robot proposed in this application includes: The first processing unit 101 is used to acquire a building structure map and first precise positioning information during the current movement of the humanoid robot, as well as to acquire the initial positioning information and environmental feature set of the humanoid robot; the first precise positioning information is the second precise positioning information of the humanoid robot in the previous movement process; The second processing unit 102 is used to perform building environment matching on the environmental feature set according to the building structure map to obtain reference positioning information; The third processing unit 103 is used to perform a first condition verification on the first precise positioning information and the initial positioning information according to the positioning correction conditions, and obtain the first condition verification result. The fourth processing unit 104 is used to perform positioning correction on the initial positioning information according to the reference positioning information when the first condition verification result satisfies the positioning correction condition, so as to obtain the second accurate positioning information of the humanoid robot in the current movement process.

[0066] Reference Figure 3 This application also provides an electronic device, including: At least one processor 201; At least one memory 202 is used to store at least one program; When the at least one program is executed by the at least one processor 201, the at least one processor 201 implements the method embodiment described above.

[0067] Similarly, it can be understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0068] This application also provides a computer-readable storage medium storing a program executable by a processor 201, which, when executed by the processor 201, is used to implement the above-described method embodiments.

[0069] Similarly, the content of the above method embodiments is applicable to the present computer-readable storage medium embodiments. The specific functions implemented by the present computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0070] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0071] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods.

[0072] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0073] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional technology for an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0074] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0076] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0077] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0078] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0079] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

[0080] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A positioning method of a humanoid robot, characterized by, Comprise: Obtaining building structure map and first accurate positioning information in current movement of humanoid robot, and obtaining initial positioning information and environment feature set of the humanoid robot; the first accurate positioning information is the second accurate positioning information of the humanoid robot in previous movement; According to the building structure map, the environment feature set is matched with the building environment, and the reference positioning information is obtained; According to the positioning correction condition, the first condition verification of the first accurate positioning information and the initial positioning information is carried out, and the first condition verification result is obtained; If the first condition verification result is to meet the positioning correction condition, the initial positioning information is corrected according to the reference positioning information, and the second accurate positioning information of the humanoid robot in the current movement is obtained.

2. The method of claim 1, wherein, According to the building structure map, the environment feature set is matched with the building environment, and the reference positioning information is obtained, comprising: According to the environment feature set, floor height data, environment structure data and environment image data are obtained; According to the floor height data, the floor map screening is carried out on the building structure map, and the hierarchical structure map is obtained; the hierarchical structure map is used to represent the floor structure subgraph in all floor structure subgraphs of the building structure map, and the floor height range contains the floor height data; According to the environment structure data and the environment image data, the hierarchical environment matching is carried out on the hierarchical structure map, and the reference positioning information is obtained.

3. The method of claim 2, wherein, According to the building structure map, the environment feature set is matched with the building environment, and the reference positioning information is obtained, comprising: Obtaining structure threshold and visual coincidence threshold; According to the structure threshold and the environment structure data, the structure region screening is carried out on the hierarchical structure map, and a plurality of candidate regions are obtained; candidate region is used to represent the structure region in all structure regions of the hierarchical structure map, and the structure similarity is greater than the structure threshold; the structure similarity is the similarity of the structure region and the environment structure data; According to the visual coincidence threshold and the environment image data, the visual feature matching is carried out on all the candidate regions, and the reference positioning information is obtained.

4. The method of claim 1, wherein, The positioning correction condition is displacement distance threshold, and the first condition verification of the first accurate positioning information and the initial positioning information is carried out according to the positioning correction condition, and the first condition verification result is obtained, comprising: According to the positioning time stamp of the first accurate positioning information and the positioning time stamp of the initial positioning information, the displacement time interval is determined; According to the displacement time interval, the speed sequence data of the humanoid robot in the current movement is obtained; According to the displacement time interval, the speed sequence data is integrated and calculated, and the moving distance of the humanoid robot is obtained; According to the displacement distance threshold, the threshold comparison of the moving distance is carried out, and the first condition verification result is obtained.

5. The method of claim 1, wherein, According to the reference positioning information, the initial positioning information is corrected, and the second accurate positioning information of the humanoid robot in the current movement is obtained. According to the initial positioning information, Kalman prediction processing is performed to obtain a prior state and a prior covariance; According to the prior covariance, Kalman gain analysis is performed on the reference positioning information to obtain gain data; According to the gain data and the reference positioning information, the prior state is updated to obtain a posterior state, and the posterior state is determined as the second accurate positioning information.

6. The method according to any one of claims 1 to 5, characterized in that, The method further comprises: obtaining a network stable condition, and a current signal strength of the humanoid robot and a duration of the current signal strength; According to the network stable condition, the current signal strength and the duration are subjected to second condition verification to obtain a second condition verification result; According to the second condition verification result, the second accurate positioning information is subjected to positioning fusion to obtain fused second accurate positioning information; According to the fused second accurate positioning information, the building structure map is subjected to map updating to obtain an updated building structure map.

7. The method of claim 6, wherein, According to the fused second accurate positioning information, the building structure map is subjected to map updating to obtain an updated building structure map, comprising: obtaining a distance difference threshold; According to the environment feature set corresponding to the fused second accurate positioning information, feature position extraction is performed on the building structure map to obtain target position information; the target position information is the position information of the target feature in the building structure map corresponding to the environment feature set; According to the fused second accurate positioning information, distance analysis is performed on the target position information to obtain an interval distance; According to the distance difference threshold, threshold analysis is performed on the interval distance to obtain a threshold analysis result; If the threshold analysis result is that the interval distance is greater than the distance difference threshold, then according to the fused second accurate positioning information, the position information of the target feature in the building structure map is updated to obtain the updated building structure map.

8. A positioning system for a humanoid robot, characterized by Comprise: A first processing unit is configured to obtain a building structure map and first accurate positioning information during current movement of a humanoid robot, and obtain initial positioning information and an environment feature set of the humanoid robot; the first accurate positioning information is second accurate positioning information of the humanoid robot in a previous movement process; A second processing unit is configured to perform building environment matching on the environment feature set according to the building structure map to obtain reference positioning information; A third processing unit is configured to perform first condition verification on the first accurate positioning information and the initial positioning information according to a positioning correction condition to obtain a first condition verification result; A fourth processing unit is configured to, when the first condition verification result satisfies the positioning correction condition, perform positioning correction on the initial positioning information according to the reference positioning information to obtain second accurate positioning information of the humanoid robot in the current movement process.

9. An electronic device, comprising: Comprise: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the method of any one of claims 1-7.

10. A computer readable storage medium having stored therein a program which is executable by a processor, characterized in that, The program executable by the processor, when executed by the processor, is for implementing the method of any one of claims 1-7.

Images