AGV navigation method, device, equipment, medium and program product

Abstract

The invention provides an AGV navigation method, device and equipment, a medium and a program product. The method comprises the steps that real-time visual data and target position data are acquired, the real-time visual data represent the real-time environment where an AGV is located, and the target position data represent the destination position of the AGV; positioning the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; performance evaluation is carried out on a preset positioning algorithm according to the first real-time position information, a position evaluation result is obtained, and the position evaluation result represents the closed-loop detection capability of the preset positioning algorithm; correcting the first real-time position information according to the position evaluation result to obtain second real-time position data which represents the real-time position of the AGV; according to the target position data and the second real-time position data, real-time navigation information is generated, and the real-time navigation information is used for controlling the AGV to advance to the destination position from the second real-time position.

Description

Technical Field

[0001] The present application relates to the technical field of indoor navigation, and in particular to an AGV navigation method, device, equipment, medium and program product. Background Art

[0002] In the navigation and positioning process of Automated Guided Vehicle (AGV), the simultaneous localization and mapping algorithm is widely used to determine the real-time position of the vehicle. The algorithm collects environmental information through sensors and combines existing map data to estimate the position and posture of the AGV, thereby achieving autonomous navigation.

[0003] However, in highly uniform environments such as warehouses and production workshops, the lack of significant features in the surrounding environment makes it difficult for simultaneous localization and mapping algorithms based on vision or other sensor inputs to provide accurate location information, and may even produce multiple seemingly reasonable but actually wrong location solutions. The location coordinates output by the simultaneous localization and mapping algorithm may have large errors, affecting the path planning and navigation accuracy of the automated guided vehicle, which may lead to reduced operational efficiency or increased risk of mission failure.

[0004] Therefore, how to perform more accurate path planning for an automated guided vehicle is a technical problem that needs to be urgently solved by those skilled in the art. Summary of the invention

[0005] The present application provides an AGV navigation method, device, equipment, medium and program product to solve the technical problem of inaccurate path planning.

[0006] In a first aspect, the present application provides an AGV navigation method, comprising: Acquire real-time visual data and target position data, wherein the real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV; Positioning the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; Performing a performance evaluation on the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, wherein the position evaluation result represents a closed-loop detection capability of the preset positioning algorithm; Correcting the first real-time position information according to the position assessment result to obtain second real-time position data, wherein the second real-time position data represents the real-time position of the AGV; Real-time navigation information is generated according to the target position data and the second real-time position data, and the real-time navigation information is used to control the AGV to travel from the second real-time position to the destination position.

[0007] Optionally, the correcting the first real-time location information according to the location assessment result to obtain second real-time location data includes: When the position evaluation result is unqualified, determining a target visual feature corresponding to the first real-time position information from pre-stored visual features, wherein the pre-stored visual features can be obtained based on map data; For each of the first real-time position information, a feature comparison is performed according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, wherein the feature comparison result indicates whether the real-time visual data contains the target visual feature; The first real-time location information is screened according to the plurality of feature comparison results to obtain second real-time location data.

[0008] Optionally, the real-time visual data includes a plurality of key frames updated in real time; The performing feature comparison according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain the feature comparison result corresponding to the first real-time position information includes: When a new key frame is detected in the real-time visual data, a neighbor key frame corresponding to the new key frame is determined according to pre-stored visual features and map data, wherein the neighbor key frame is a key frame directly connected to the new key frame among the multiple key frames; According to the new key frame and the plurality of neighbor key frames, two target key frames are obtained, wherein the target key frames are two neighbor key frames that are most similar to the new key frame; A feature comparison result is obtained according to the target visual feature corresponding to the first real-time position information and the new key frame and its corresponding two target key frames.

[0009] Optionally, after determining the neighbor key frame corresponding to the new key frame according to the pre-stored visual features and map data, the method further comprises: Determine whether the neighbor key frame meets the key frame quantity threshold; If not, then obtaining a previous key frame corresponding to the new key frame, and querying the second position data corresponding to the previous key frame, wherein the second position data is the second real-time position data corresponding to the previous key frame; Acquire a steering angle corresponding to the new key frame, and determine a substitute key frame in the map data according to the second position data and the steering angle, wherein the substitute key frame represents the visual data lost between the previous key frame and the new key frame; The neighbor key frame is updated based on the substitute key frame to obtain a new neighbor key frame.

[0010] Optionally, generating real-time navigation information according to the target position data and the second real-time position data includes: Obtain the type of environment in which the AGV is located; According to the environment type, in the path planning algorithm, determining a target planning algorithm; The target position data and the second real-time position data are input into the target planning algorithm to obtain real-time navigation information.

[0011] Optionally, after generating the real-time navigation information according to the target position data and the second real-time position data, the method further comprises: Obtaining real-time road condition features according to the real-time visual data; Speed ​​planning is performed on the real-time navigation information according to the real-time traffic characteristics to obtain real-time navigation information including speed setting.

[0012] In a second aspect, the present application provides an AGV navigation device, comprising: An acquisition module, used to acquire real-time visual data and target position data, wherein the real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV; A positioning module, used to locate the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; An evaluation module, configured to evaluate the performance of the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, wherein the position evaluation result represents the closed-loop detection capability of the preset positioning algorithm; A correction module, used for correcting the first real-time position information according to the position assessment result to obtain second real-time position data, wherein the second real-time position data represents the real-time position of the AGV; A generating module is used to generate real-time navigation information according to the target position data and the second real-time position data, wherein the real-time navigation information is used to control the AGV to travel from the second real-time position to the destination position.

[0013] Optionally, the correction module, when performing the correction of the first real-time location information according to the location assessment result to obtain the second real-time location data, is used to: When the position evaluation result is unqualified, determining a target visual feature corresponding to the first real-time position information from pre-stored visual features, wherein the pre-stored visual features can be obtained based on map data; For each of the first real-time position information, a feature comparison is performed according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, wherein the feature comparison result indicates whether the real-time visual data contains the target visual feature; The first real-time location information is screened according to the plurality of feature comparison results to obtain second real-time location data.

[0014] The correction module, wherein the real-time visual data includes a plurality of key frames updated in real time, is used to: When a new key frame is detected in the real-time visual data, a neighbor key frame corresponding to the new key frame is determined according to pre-stored visual features and map data, wherein the neighbor key frame is a key frame directly connected to the new key frame among the multiple key frames; According to the new key frame and the plurality of neighbor key frames, two target key frames are obtained, wherein the target key frames are two neighbor key frames that are most similar to the new key frame; A feature comparison result is obtained according to the target visual feature corresponding to the first real-time position information and the new key frame and its corresponding two target key frames.

[0015] Optionally, the above device further includes a substitute module; The substitute module is used to: Determine whether the neighbor key frame meets the key frame quantity threshold; If not, then obtaining a previous key frame corresponding to the new key frame, and querying the second position data corresponding to the previous key frame, wherein the second position data is the second real-time position data corresponding to the previous key frame; Acquire a steering angle corresponding to the new key frame, and determine a substitute key frame in the map data according to the second position data and the steering angle, wherein the substitute key frame represents the visual data lost between the previous key frame and the new key frame; The neighbor key frame is updated based on the substitute key frame to obtain a new neighbor key frame.

[0016] Optionally, the generating module, when executing the step of generating the real-time navigation information according to the target position data and the second real-time position data, is used to: Obtain the type of environment in which the AGV is located; According to the environment type, in the path planning algorithm, determining a target planning algorithm; The target position data and the second real-time position data are input into the target planning algorithm to obtain real-time navigation information.

[0017] Optionally, the above device further includes a speed control module; The speed control module is used for: Obtaining real-time road condition features according to the real-time visual data; Speed ​​planning is performed on the real-time navigation information according to the real-time traffic characteristics to obtain real-time navigation information including speed setting.

[0018] In a third aspect, the present application provides an AGV device, comprising: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the AGV navigation method as described in any one of the first aspects.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the AGV navigation method as described in any one of the first aspects.

[0020] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the AGV navigation method as described in any one of the first aspects.

[0021] The AGV navigation method provided by the present application determines the location and destination of the AGV by acquiring real-time visual data and target position data; locates the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; performs a performance evaluation on the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, which represents the closed-loop detection capability of the preset positioning algorithm; when the first real-time position information is not unique, it proves that there is an error. At this time, the erroneous position is screened out by comparing it with the existing map to ensure the positioning accuracy, thereby improving the accuracy of navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0023] Figure 1 A flowchart of an AGV navigation method provided in an embodiment of the present application; Figure 2A schematic diagram of the structure of an AGV navigation device provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an AGV device provided in this application.

[0024] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0025] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0026] In the navigation and positioning process of Automated Guided Vehicle (AGV), the Simultaneous Localization and Mapping (SLAM) algorithm is widely used to determine the real-time position of the vehicle. The algorithm collects environmental information through sensors and combines existing map data to estimate the position and posture of the AGV, thereby achieving autonomous navigation.

[0027] However, in some special application scenarios, such as warehouses, production workshops and other highly consistent environments, SLAM algorithms face challenges in identifying the specific location of AGVs. Due to the lack of significant features in the surrounding environment, SLAM algorithms based on vision or other sensor inputs are difficult to provide accurate location information, and may even produce multiple seemingly reasonable but actually wrong location solutions.

[0028] In the prior art, although the SLAM algorithm itself is quite mature, the problem of its performance limitation under the above-mentioned specific conditions has not been completely solved. When encountering extremely similar environments, the position coordinates output by the SLAM algorithm may have large errors, affecting the path planning and navigation accuracy of the AGV, which may lead to reduced operational efficiency or increased risk of mission failure.

[0029] Therefore, an improved method is needed to overcome this limitation and ensure high-precision positioning and navigation capabilities even when environmental features are not obvious. By introducing an additional map verification step, incorrect position solutions can be effectively screened out when position uncertainty is detected, improving overall system reliability and navigation accuracy. This method not only enhances the adaptability and flexibility of AGV in complex environments, but also provides technical support for a wider range of application scenarios.

[0030] An AGV navigation method provided in an embodiment of the present application is intended to solve the above technical problems in the prior art, and is executed by a control component in the above AGV, hereinafter referred to as the AGV as the execution subject.

[0031] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0032] like Figure 1 As shown, Figure 1 A flowchart of an AGV navigation method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, an AGV navigation method may specifically include steps S201 to S204, wherein: S201. Acquire real-time visual data and target position data.

[0033] The real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV.

[0034] In the method for obtaining real-time visual data and target position data, at least one image acquisition device is first configured on an automatic guided vehicle (AGV), which is used to continuously capture visual information of the environment around the AGV, and the visual information accurately represents the actual conditions of the environment in which the AGV is located. Then, the destination position data of the AGV is determined and obtained through a positioning system (such as GPS, indoor positioning system, etc.) or preset target coordinates, and the position data accurately describes the target location that the AGV needs to reach. Then, the obtained real-time visual data and target position data are transmitted to the central processing unit of the AGV, wherein the central processing unit processes and analyzes the received data for subsequent navigation decisions.

[0035] S202: Position the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV.

[0036] In the process of determining the first real-time position information of the automatic guided vehicle (AGV), the previously acquired real-time visual data is first used to identify and extract the information of feature points or landmarks in the environment through image processing technology. Then, the information of these feature points or landmarks is compared and matched with the environmental map data pre-stored in the database to achieve preliminary positioning. Then, a preset positioning algorithm is applied, such as based on the extended Kalman filter (EKF), particle filter or other advanced SLAM (Simultaneous Localization and Mapping) technology, integrating the real-time data stream from the AGV sensor, including but not limited to the data of the laser rangefinder and inertial measurement unit (IMU), to accurately calculate the exact position of the AGV in the current environment. Finally, the process outputs the first real-time position information of the AGV, providing an accurate basis for subsequent path planning and navigation decisions.

[0037] S203: Evaluate the performance of a preset positioning algorithm according to the first real-time position information to obtain a position evaluation result.

[0038] The position assessment result characterizes the closed-loop detection capability of the preset positioning algorithm.

[0039] In the process of evaluating the performance of the preset positioning algorithm, the previously obtained first real-time position information of the AGV is first used as a reference. Then, the reference position is compared and analyzed with the AGV position information calculated by the positioning algorithm at the same time point, and the degree of deviation between the two is calculated, including but not limited to statistical indicators such as average error, maximum error, and standard deviation of error. Then, based on the quantitative analysis results such as standard deviation, combined with the requirements for positioning accuracy, stability, and response speed in actual application scenarios, the performance of the positioning algorithm is comprehensively evaluated. Finally, according to the above analysis and evaluation process, a detailed position evaluation result is formed.

[0040] S204: Correct the first real-time location information according to the location assessment result to obtain second real-time location data.

[0041] The second real-time position data represents the real-time position of the AGV.

[0042] After obtaining the position assessment results, the first real-time position information is first analyzed to identify the deviations and error patterns therein. Based on this analysis, the corresponding correction algorithm or model is applied to adjust the first real-time position information. The correction process may involve but is not limited to technical means such as coordinate transformation and error compensation to reduce the errors generated by the previous positioning algorithm. Then, by verifying the matching degree between the corrected data and the actual position, the effectiveness and accuracy of the correction are ensured. Finally, the second real-time position data is generated, which more accurately reflects the actual position status of the AGV.

[0043] S205: Generate real-time navigation information according to the target position data and the second real-time position data.

[0044] The real-time navigation information is used to control the AGV to travel from the second real-time location to the destination location.

[0045] The AGV device first obtains the target position data and its own second real-time position data, and calculates the relative position and direction difference between the two through a built-in algorithm. Based on this information, the AGV performs path planning to determine an optimal route from the current real-time position to the target position, taking into account factors including but not limited to the shortest distance, obstacle avoidance, and traffic rules compliance. Subsequently, based on the selected path, the AGV generates a series of control instructions that accurately guide its speed adjustment, steering operations, and other necessary actions to ensure accurate progress along the predetermined trajectory. During the journey, the AGV continuously uses sensors to update its own position information and adjusts navigation instructions in real time to ensure that it can reach its destination efficiently and safely even in the face of dynamic environmental changes.

[0046] The AGV navigation method provided in the embodiment of the present application determines the location and destination of the AGV by acquiring real-time visual data and target position data; locates the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; performs a performance evaluation on the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, which represents the closed-loop detection capability of the preset positioning algorithm; when the first real-time position information is not unique, it proves that there is an error. At this time, the erroneous position is screened out by comparing it with the existing map to ensure the positioning accuracy, thereby improving the accuracy of navigation.

[0047] In one achievable manner, correcting the first real-time location information according to the location assessment result to obtain the second real-time location data includes: When the position evaluation result is unqualified, the target visual feature corresponding to the first real-time position information is determined in the pre-stored visual features, and the pre-stored visual features can be obtained based on the map data.

[0048] Specifically, when the AGV device (i.e., AGV) evaluates that the accuracy of the current position does not meet the preset standard (there are multiple first real-time position information), the system will automatically identify the position as unqualified. At this time, the AGV searches for the target visual features that match the first real-time position information in the pre-stored visual feature library (i.e., pre-stored visual features) based on the built-in map data. This process involves comparing the currently captured visual information with the feature descriptions in the pre-stored data to find the closest corresponding items. These pre-stored visual features are pre-established based on map data and contain detailed visual identification of multiple possible location points. The map data is a real-time map. Through this matching method, the AGV can accurately determine its relative position in the environment and use this information to make necessary corrections to restore the expected navigation accuracy.

[0049] For each first real-time position information, feature comparison is performed based on the target visual features corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, wherein the feature comparison result indicates whether the real-time visual data contains the target visual features.

[0050] First, the corresponding target visual features are identified based on the first real-time position information. Then, the currently acquired real-time visual data is carefully compared with the previously determined target visual features. This process involves analyzing and matching elements in the real-time visual data with pre-stored target visual features to evaluate the consistency or similarity between the two. The final feature comparison result clearly indicates whether the real-time visual data contains the expected target visual features, thereby verifying the positioning accuracy of the AGV device at a specific location.

[0051] The first real-time location information is screened according to the multiple feature comparison results to obtain the second real-time location data.

[0052] The first real-time position information including the target visual feature is used as the second real-time position data.

[0053] The AGV navigation method provided in the embodiment of the present application determines the target visual feature corresponding to the first real-time position information in the pre-stored visual features when the position assessment result is unqualified, so as to determine the pre-stored visual feature at the first real-time position in the map data; for each of the first real-time position information, the target visual feature corresponding to the first real-time position information is compared with the real-time visual data to determine whether the real-time visual data contains the target visual feature; based on multiple feature comparison results, the first real-time position information is screened to obtain the second real-time position data, thereby filtering out the inconsistent first real-time position data by determining whether the real-time visual data matches the pre-stored visual feature to obtain the second real-time position data.

[0054] In an implementable manner, the real-time visual data includes a plurality of key frames updated in real time, and feature comparison is performed according to the target visual features corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, including: When a new keyframe is detected in real-time visual data, a neighbor keyframe corresponding to the new keyframe is determined based on pre-stored visual features and map data. The neighbor keyframe is a keyframe directly connected to the new keyframe among multiple keyframes.

[0055] Among them, keyframes are of great significance in animation, video editing and computer graphics. They refer to specific frames in a sequence or timeline that are used to define significant changes in the properties of an object or scene. Keyframe technology allows creators to specify the start and end states, as well as any important state change points in the middle, and the software automatically calculates the transitions between these keyframes (a process usually called interpolation) to generate smooth animation effects.

[0056] It is understandable that in the related art, neighbor keyframes are directly connected, where the direct connection in the related art means that there are overlapping points (the same reference object) between the two. However, if the camera is disconnected when the AGV turns, and returns to normal after turning, the field of view before and after the turn is a blind spot for each other. At this time, the keyframe will be lost, and the neighbor keyframe cannot be found in the real-time visual data.

[0057] According to the new keyframe and multiple neighbor keyframes, two target keyframes are obtained, and the target keyframes are the two neighbor keyframes that are most similar to the new keyframe.

[0058] There is a temporal sequence relationship between the new key frame and the two target key frames, and the time corresponding to the new key frame is between the times corresponding to the two target key frames.

[0059] A feature comparison result is obtained according to the target visual feature corresponding to the first real-time position information and the new key frame and its corresponding two target key frames.

[0060] Specifically, the similarity scores of the target visual features corresponding to the first real-time position information and the new key frame and its corresponding two target key frames are calculated, and then the average of these three similarity scores is taken. If the above average is greater than the preset average, it indicates that the feature comparison result is qualified, otherwise it is unqualified.

[0061] In one possible implementation, after determining the neighbor key frame corresponding to the new key frame according to the pre-stored visual features and map data, the method further includes: Determine whether the neighbor keyframes meet the keyframe quantity threshold.

[0062] If not, then the previous key frame corresponding to the new key frame is obtained, and the second position data corresponding to the previous key frame is queried, wherein the second position data is the second real-time position data corresponding to the previous key frame.

[0063] The previous key frame and the new key frame are adjacent to each other in terms of their acquisition time.

[0064] The steering angle corresponding to the new key frame is obtained, and a substitute key frame is determined in the map data according to the second position data and the steering angle, where the substitute key frame represents the visual data lost between the previous key frame and the new key frame.

[0065] The steering angle can be obtained based on the IMU module.

[0066] Specifically, among the preset visual features of the map data, multiple preset visual features included in the position corresponding to the second position data are determined, and the above multiple preset visual features correspond to different acquisition angles. In the preset visual feature preparation stage, the AGV uses the camera equipment and the IMU module to collect visual data including historical key frames. It should be noted that the historical key frames carry three labels: steering angle, position data and visual features. Among the historical key frames, the substitute key frames corresponding to the steering angle corresponding to the new key frames are determined.

[0067] The neighbor keyframe is updated based on the substitute keyframe to obtain a new neighbor keyframe.

[0068] Specifically, between the acquisition moments corresponding to the previous key frame and the new key frame, a filling time is determined for the substitute key frame; the filling time is entered into the substitute key frame as a time tag; based on the filling time, the substitute key frame is added into the original neighbor key frame as a new neighbor key frame, thereby completing the update of the neighbor key frame.

[0069] The AGV navigation method provided in the embodiment of the present application maintains the stability of the indoor positioning process by supplementing the lost visual data between the previous key frame and the new key frame.

[0070] In an implementable manner, the above S205 generates real-time navigation information according to the target position data and the second real-time position data, which may specifically include: The environment type of the AGV is obtained, where the environment type is an indoor environment or an outdoor environment, which represents the static environment in which the AGV device is located.

[0071] According to the environment type, in the path planning algorithm, the target planning algorithm is determined, wherein the path planning algorithm includes A* algorithm, Dijkstra, and RRT.

[0072] The target position data and the second real-time position data are input into the target planning algorithm to obtain real-time navigation information.

[0073] In an implementable manner, after the above S205 generates the real-time navigation information according to the target position data and the second real-time position data, the method further includes: According to the real-time visual data, the real-time road condition features are obtained, wherein the real-time road condition features represent the dynamic environment in which the AGV equipment is located, for example, moving objects such as pedestrians and vehicles.

[0074] Speed ​​planning is performed on the real-time navigation information according to the real-time road condition characteristics to obtain real-time navigation information including speed setting.

[0075] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0076] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0077] Figure 2 A schematic diagram of the structure of an AGV navigation device provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the AGV navigation device 40 provided in the embodiment of the present application includes: An acquisition module 401 is used to acquire real-time visual data and target position data, wherein the real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV; A positioning module 402 is used to locate the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; An evaluation module 403 is used to evaluate the performance of the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, wherein the position evaluation result represents the closed-loop detection capability of the preset positioning algorithm; A correction module 404, used to correct the first real-time position information according to the position evaluation result to obtain second real-time position data, where the second real-time position data represents the real-time position of the AGV; The generating module 405 is used to generate real-time navigation information according to the target position data and the second real-time position data, and the real-time navigation information is used to control the AGV to travel from the second real-time position to the destination position.

[0078] Optionally, the correction module 404, when correcting the first real-time location information according to the location assessment result to obtain the second real-time location data, is used to: When the position evaluation result is unqualified, determining the target visual feature corresponding to the first real-time position information from the pre-stored visual features, where the pre-stored visual features can be obtained based on the map data; For each first real-time position information, a feature comparison is performed according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, wherein the feature comparison result indicates whether the real-time visual data contains the target visual feature; The first real-time location information is screened according to the multiple feature comparison results to obtain the second real-time location data.

[0079] The correction module 404, wherein the real-time visual data includes a plurality of key frames updated in real time, is used to: When a new keyframe is detected in the real-time visual data, a neighbor keyframe corresponding to the new keyframe is determined according to the pre-stored visual features and map data, where the neighbor keyframe is a keyframe directly connected to the new keyframe among multiple keyframes; According to the new keyframe and multiple neighbor keyframes, two target keyframes are obtained, and the target keyframes are the two neighbor keyframes that are most similar to the new keyframe; A feature comparison result is obtained according to the target visual feature corresponding to the first real-time position information and the new key frame and its corresponding two target key frames.

[0080] Optionally, the AGV navigation device 40 further includes a substitute module; Substitute modules for: Determine whether the neighbor key frame meets the key frame quantity threshold; If not, then obtain the previous key frame corresponding to the new key frame, and query the second position data corresponding to the previous key frame, wherein the second position data is the second real-time position data corresponding to the previous key frame; Obtaining a steering angle corresponding to the new key frame, and determining a substitute key frame in the map data according to the second position data and the steering angle, the substitute key frame representing the visual data lost between the previous key frame and the new key frame; The neighbor keyframe is updated based on the substitute keyframe to obtain a new neighbor keyframe.

[0081] Optionally, the generating module 405, when generating the real-time navigation information according to the target position data and the second real-time position data, is used to: Get the type of environment the AGV is in; According to the type of environment, in the path planning algorithm, determine the target planning algorithm; The target position data and the second real-time position data are input into the target planning algorithm to obtain real-time navigation information.

[0082] Optionally, the AGV navigation device 40 further includes a speed control module; Constant speed module for: According to the real-time visual data, the real-time road condition characteristics are obtained; Speed ​​planning is performed on the real-time navigation information according to the real-time road condition characteristics to obtain real-time navigation information including speed setting.

[0083] The AGV navigation device provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be described in detail here.

[0084] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0085] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0086] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to a transistor, a memristor, etc.

[0087] Figure 3 This is a schematic diagram of the structure of the AGV equipment provided in this application. Figure 3 As shown, the AGV device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the AGV device 50 also includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected via a bus 504.

[0088] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that at least one processor 501 executes the above method.

[0089] The specific implementation process of the processor 501 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0090] If not otherwise specified, the processor 501 may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. If not otherwise specified, the memory 502 may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory (RRAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), an enhanced dynamic random access memory (EDRAM), a high-bandwidth memory (HBM), a hybrid memory cube (HMC), etc.

[0091] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0092] An embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above redundant field updating method for a distributed system is implemented.

[0093] An embodiment of the present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the above redundant field updating method for a distributed system.

[0094] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0095] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. An AGV navigation method, characterized in that: Applied to AGV, the method comprises: Acquire real-time visual data and target position data, wherein the real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV; Positioning the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; Performing a performance evaluation on the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, wherein the position evaluation result represents a closed-loop detection capability of the preset positioning algorithm; Correcting the first real-time position information according to the position assessment result to obtain second real-time position data, wherein the second real-time position data represents the real-time position of the AGV; Real-time navigation information is generated according to the target position data and the second real-time position data, and the real-time navigation information is used to control the AGV to travel from the second real-time position to the destination position.

2. The AGV navigation method according to claim 1, characterized in that: The step of correcting the first real-time location information according to the location assessment result to obtain second real-time location data includes: When the position evaluation result is unqualified, determining a target visual feature corresponding to the first real-time position information from pre-stored visual features, wherein the pre-stored visual features can be obtained based on map data; For each of the first real-time position information, a feature comparison is performed according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain a feature comparison result corresponding to the first real-time position information, wherein the feature comparison result indicates whether the real-time visual data contains the target visual feature; The first real-time location information is screened according to the plurality of feature comparison results to obtain second real-time location data.

3. The AGV navigation method according to claim 2, characterized in that: The real-time visual data includes a plurality of key frames updated in real time; The performing feature comparison according to the target visual feature corresponding to the first real-time position information and the real-time visual data to obtain the feature comparison result corresponding to the first real-time position information includes: When a new key frame is detected in the real-time visual data, a neighbor key frame corresponding to the new key frame is determined according to pre-stored visual features and map data, wherein the neighbor key frame is a key frame directly connected to the new key frame among the multiple key frames; According to the new key frame and the plurality of neighbor key frames, two target key frames are obtained, wherein the target key frames are two neighbor key frames that are most similar to the new key frame; A feature comparison result is obtained according to the target visual feature corresponding to the first real-time position information and the new key frame and its corresponding two target key frames.

4. The AGV navigation method according to claim 3, characterized in that: After determining the neighbor key frame corresponding to the new key frame according to the pre-stored visual features and map data, the method further includes: Determine whether the neighbor key frame meets the key frame quantity threshold; If not, then obtaining a previous key frame corresponding to the new key frame, and querying the second position data corresponding to the previous key frame, wherein the second position data is the second real-time position data corresponding to the previous key frame; Acquire a steering angle corresponding to the new key frame, and determine a substitute key frame in the map data according to the second position data and the steering angle, wherein the substitute key frame represents the visual data lost between the previous key frame and the new key frame; The neighbor key frame is updated based on the substitute key frame to obtain a new neighbor key frame.

5. The AGV navigation method according to claim 1, characterized in that: The generating real-time navigation information according to the target position data and the second real-time position data comprises: Obtain the type of environment in which the AGV is located; According to the environment type, in the path planning algorithm, determining a target planning algorithm; The target position data and the second real-time position data are input into the target planning algorithm to obtain real-time navigation information.

6. The AGV navigation method according to any one of claims 1 to 5, characterized in that: After generating the real-time navigation information according to the target position data and the second real-time position data, the method further includes: Obtaining real-time road condition features according to the real-time visual data; Speed ​​planning is performed on the real-time navigation information according to the real-time traffic characteristics to obtain real-time navigation information including speed setting.

7. An AGV navigation device, characterized in that: include: An acquisition module, used to acquire real-time visual data and target position data, wherein the real-time visual data represents the real-time environment where the AGV is located, and the target position data represents the destination position of the AGV; A positioning module, used to locate the AGV according to the real-time visual data and a preset positioning algorithm to obtain first real-time position information of the AGV; An evaluation module, configured to evaluate the performance of the preset positioning algorithm according to the first real-time position information to obtain a position evaluation result, wherein the position evaluation result represents the closed-loop detection capability of the preset positioning algorithm; A correction module, used for correcting the first real-time position information according to the position assessment result to obtain second real-time position data, wherein the second real-time position data represents the real-time position of the AGV; A generating module is used to generate real-time navigation information according to the target position data and the second real-time position data, wherein the real-time navigation information is used to control the AGV to travel from the second real-time position to the destination position.

8. An AGV device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images