Navigation method and system for repeated path in closed-loop space

Abstract

The invention relates to the technical field of mobile robot navigation, and discloses a navigation method and system for a closed-loop space repeated path, and the method comprises the steps: constructing a map of a working area of a mobile robot, carrying out the path planning according to a working task, selecting a reference target on the map for each planned path, extracting the feature information of the reference target, storing feature information of each planned path and the reference target; for each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation precision verification: if the navigation precision meets a preset requirement, the verification is passed; otherwise, updating and setting the reference target and carrying out navigation precision verification again until the verification is passed; and when the mobile robot executes the work task, calling the stored planned path and the feature information of the corresponding reference target according to the work task for navigation. According to the method, the calculation amount can be reduced, the calculation power use efficiency and the repeated path navigation efficiency are improved, and the navigation precision is ensured.

Description

Technical Field

[0001] The present invention relates to the field of mobile robot navigation technology, and in particular to a closed-loop space repeated path navigation method and system. Background Art

[0002] In the field of mobile robots, path planning and feature detection and matching are key technologies for achieving autonomous navigation. Repeated path navigation is a common application of mobile robots, but existing technologies have significant efficiency bottlenecks when implementing repeated path navigation. Specifically, whether the same mobile robot or multiple mobile robots repeatedly perform the same task, they need to re-detect and match environmental features each time. This process is not only computationally intensive, but also requires high computing power for the mobile robot, limiting the application capabilities of mobile robots in resource-constrained environments.

[0003] Traditional path planning methods, such as Rapidly-exploring Random Tree (RRT) and its improved version, RRT star, have been widely used in mobile robot navigation. However, these methods also need to recalculate the path each time a task is executed. Not only can they not effectively utilize historical information, but they also have a large computational burden. Simultaneous Localization and Mapping (SLAM) is a technology that allows mobile robots or self-driving vehicles to navigate autonomously in unknown environments. It can realize the self-positioning of mobile robots and the construction of environmental maps. However, during the navigation process, the SLAM system needs to continuously extract and match features to adapt to environmental changes, which also leads to a high computational burden.

[0004] In order to solve the problem of large amount of calculation in repeated path navigation, the existing technology starts from how to optimize the feature detection and matching process, and studies how to improve the operation efficiency of mobile robots. There are methods to optimize path planning by improving genetic algorithms to reduce the search space and thus reduce the computational cost; there are also methods to improve SLAM algorithms by introducing key frame selection mechanisms and graph optimization technologies to improve the speed and accuracy of feature matching and reduce repeated calculations. Although these existing methods can achieve a certain efficiency improvement effect in the path planning and feature matching of mobile robots, when faced with repeated path navigation in a closed-loop space, the effect of reducing the amount of calculation and improving the efficiency of computing power is limited; and in order to reduce the amount of calculation, the existing technology often sacrifices a certain navigation accuracy, resulting in low accuracy of repeated path navigation. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the deficiencies in the prior art and provide a navigation method and system for repeated paths in a closed-loop space, which can reduce the amount of calculation, improve the efficiency of computing power utilization and the efficiency of repeated path navigation, and ensure navigation accuracy.

[0006] In order to solve the above technical problems, the present invention provides a navigation method for a closed-loop space repetitive path, comprising:

[0007] Construct a map of the mobile robot's working area, perform path planning according to the work task, select a reference target on the map for each planned path, extract feature information of the reference target, and store the feature information of each planned path and the reference target in correspondence; for each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification passes; if the navigation accuracy does not meet the preset requirements, update the reference target, and then re-perform navigation accuracy verification until the navigation accuracy meets the preset requirements;

[0008] When the mobile robot needs to perform a work task, it calls the stored planned path and the feature information of the corresponding reference target for navigation according to the work task to be performed.

[0009] Furthermore, the reference target is selected on the map for each planned path, specifically:

[0010] In the map of the mobile robot's working area, reference targets are selected along both sides of the planned path. The number of reference targets selected is:

[0011]

[0012] Among them, F (S,θ) represents the number of reference targets, f1 represents the coefficient preset based on the moving distance, S represents the moving distance, l represents the preset distance parameter, represents rounding up, f2 represents the coefficient preset based on the direction deflection angle, Δθ ΔS Represents the unit moving distance angle, and ω represents the preset angle parameter.

[0013] Furthermore, the characteristic information of the reference target includes center coordinates based on the boundary point coordinates of the reference target.

[0014] Furthermore, for each planned path, the process in which the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification includes:

[0015] The mobile robot sequentially calls the feature information of each planned path and its corresponding reference target for navigation;

[0016] The real-time position and posture of the mobile robot during the navigation process is obtained, and the movement deviation between the actual running path of the mobile robot and the planned path is calculated: if the movement deviation is less than the preset error, the navigation accuracy meets the requirements; otherwise, the navigation accuracy does not meet the requirements, and the number of reference targets is adjusted based on the obtained movement deviation, and then the navigation accuracy verification is performed again until the navigation accuracy meets the requirements.

[0017] Furthermore, the process of navigation of the mobile robot includes:

[0018] The reference target is detected by laser radar, and the relative position and posture of the reference target relative to the mobile robot are obtained. The current position and posture of the mobile robot is obtained by combining the characteristic information of the reference target, which is recorded as in, represents the X-direction coordinate of the mobile robot in the map coordinate system at time t, represents the Y coordinate of the mobile robot in the map coordinate system at time t, yaw t represents the heading angle of the mobile robot at time t;

[0019] Based on the odometer and inertial measurement unit of the mobile robot, the angular velocity, speed, and acceleration of the mobile robot are obtained, and the current state of the mobile robot is obtained and recorded as (ω t ,v t ,a t ), where ω t represents the angular velocity of the mobile robot at time t, v t represents the speed of the mobile robot at time t, a t represents the acceleration of the mobile robot at time t;

[0020] According to the current position and state of the mobile robot, the position point at the next moment is calculated, and the path point closest to the position point on the planned path is used as the preview point, which is recorded as (X t+1 ,Y t+1 ,Yaw t+1 ), where X t+1 Indicates the X coordinate of the preview point in the map coordinate system, Y t+1 Indicates the Y coordinate of the preview point in the map coordinate system, Yaw t+1 Indicates the heading angle of the preview point;

[0021] The difference between the current position of the mobile robot and the moving target of the preview point is calculated, and the mobile robot is controlled to move and navigate to the preview point on the planned path.

[0022] Furthermore, the process of obtaining the real-time position and posture of the mobile robot during navigation and calculating the movement deviation between the actual running path of the mobile robot and the planned path includes:

[0023] The real-time position and posture of the mobile robot is obtained, and the movement deviation between the current position and posture of the mobile robot and the preview point determined by the mobile robot at the previous moment is calculated; the movement deviation includes a distance deviation and an angle deviation, and the preset error includes a distance error and an angle error; when the distance deviation is less than the distance error, and the angle deviation is less than the angle error, it is considered that the navigation accuracy meets the requirements, otherwise it is considered that the navigation accuracy does not meet the requirements.

[0024] Furthermore, the process of updating and setting the reference target includes:

[0025] Determine the path segment type of the planned path corresponding to the current posture of the mobile machine according to the real-time posture of the mobile robot;

[0026] According to the path segment type, and in combination with the current distance deviation and angle deviation of the mobile robot, a preset coefficient based on the moving distance or a preset coefficient based on the direction deflection angle corresponding to the path segment is adjusted;

[0027] The number of reference targets of the mobile robot on the path segment is recalculated according to the adjusted coefficient preset based on the moving distance or the coefficient preset based on the direction deflection angle, and then the reference targets are re-determined and feature information of the reference targets is extracted.

[0028] Furthermore, the process of adjusting the preset coefficient based on the moving distance or the preset coefficient based on the direction deflection angle corresponding to the path segment according to the path segment type and in combination with the current distance deviation and angle deviation of the mobile robot includes:

[0029] If the path segment is a straight line segment, adjusting a coefficient of the path segment that is preset based on the moving distance;

[0030] If the path segment is a curved segment, a coefficient preset based on the direction deflection angle of the path segment is adjusted.

[0031] Furthermore, the adjustment method of the coefficient preset based on the moving distance is:

[0032]

[0033] Among them, f1′ represents the adjusted coefficient based on the preset moving distance, μ d represents the distance deviation, μ α represents the angular deviation, δ d represents the distance error, δ α represents the angle error;

[0034] The adjustment method of the coefficient preset based on the direction deflection angle is:

[0035]

[0036] Wherein, f2′ represents the adjusted coefficient preset based on the direction deflection angle.

[0037] The present invention also provides a closed-loop space repetitive path navigation system, comprising:

[0038] The navigation data preparation module is used to construct a map of the mobile robot's working area, perform path planning according to the work task, select a reference target on the map for each planned path, extract feature information of the reference target, and store the feature information of each planned path and the reference target in correspondence; for each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification passes; if the navigation accuracy does not meet the preset requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements;

[0039] The repeated path navigation module is used to call the stored planned path and the corresponding reference target feature information for navigation according to the work task to be performed when the mobile robot needs to perform a work task.

[0040] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0041] Before the mobile robot performs repeated path navigation, the present invention performs path planning according to the work task, selects reference targets on the map for each planned path, and extracts characteristic information of the reference targets. When performing repeated path navigation of the mobile robot, the stored planned path and the corresponding characteristic information of the reference targets are directly called according to the work task for navigation, which can avoid recalculating and planning the path, effectively reducing the amount of calculation and improving the efficiency of computing power utilization. The navigation accuracy is verified at least once for each planned path, which can improve the efficiency of repeated path navigation while ensuring the path navigation accuracy of the mobile robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0043] Figure 1 Flow chart of the method in the preferred embodiment of the present invention.

[0044] Figure 2 It is a schematic diagram of selecting a reference target on a map for a planned path in a preferred embodiment of the present invention.

[0045] Figure 3Schematic diagram of the radius tangent method used when selecting a preview point on a planned path in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0047] Embodiment 1

[0048] Reference Figure 1 As shown, the present invention discloses a navigation method for a closed-loop space repetitive path, comprising the following steps:

[0049] S1: Build a map of the mobile robot's work area, plan paths according to the work tasks, and select reference targets on the map for each planned path.

[0050] S1-1: Before performing path planning, first construct a map of the mobile robot's working area. The working area of ​​the mobile robot can generally be understood as a closed-loop space with certain boundaries. The mobile robot will not exceed the scope of this space when performing work tasks. In this embodiment, SLAM technology is used to construct a map of the working area. The sensors carried by the mobile robot itself identify characteristic markers in the location environment. For example, point cloud feature information is obtained through a laser sensor, and image, RGB and other depth feature information are obtained through a visual sensor. Both point cloud feature information and depth feature information are characteristic markers; then, based on the relative position between the mobile robot and the characteristic marker and the value of the odometer, the global coordinates of the mobile robot and the characteristic marker are obtained, and the map is constructed according to the global coordinates. There are various existing mapping methods. The map construction method of the present invention can also be other methods such as raster maps, which will not be repeated.

[0051] S1-2: The server performs path planning according to the work tasks of the mobile robot and plans at least one planned path for each work task.

[0052] Each planned path is a set of several path points, where the path points include the starting point, the end point, and the path points between the starting point and the end point of the work task. The planned path is represented by {(X1, Y1, Yaw1), (X2, Y2, Yaw2), ··· (X i ,Y i ,Yaw i ),······,(X m ,Y m ,Yaw m )}, where X i Indicates the X coordinate of the i-th path point in the planned path in the map coordinate system, Y iIndicates the Y coordinate of the i-th path point in the planned path in the map coordinate system, Yaw i represents the heading angle of the i-th path point in the planned path, and m represents the number of path points in the planned path.

[0053] The benefit of planning at least one planned path for each work task is that when a certain planned path is difficult to complete the work task, or the efficiency of completing the work task is low, it can be adjusted in time. Specifically, when a certain planned path is congested with mobile robots, other planned paths can be adjusted to perform the task; or, when the scenario of a certain planned path is more complex and too many reference targets need to be selected, resulting in reduced computing efficiency, a planned path with fewer reference targets can be selected as the planned path for the actual work. In addition, the multiple planned paths for each of the above work tasks can be repeated in some sections.

[0054] Furthermore, in this embodiment, each planned path may be smoothed to obtain a smooth planned path, and the smoothing of the planned path may be implemented based on algorithms such as high-order Bezier curves or spline curves.

[0055] S1-3: In this embodiment, reference targets are selected according to the complexity of the scene of the planned path. In practice, the scene of the planned path is relatively complex, for example, including straight sections (straight sections), turning sections (curved sections), etc. Based on the preset rules, in the map of the working area of ​​the mobile robot, reference targets are selected along both sides of the planned path. The number of selected reference targets is:

[0056]

[0057] Among them, F (S,θ) represents the number of reference targets, f1 represents the coefficient preset based on the moving distance, S represents the moving distance, l represents the preset distance parameter, represents rounding up, f2 represents the coefficient preset based on the direction deflection angle, Δθ ΔS Indicates the unit moving distance angle, ω indicates the preset angle parameter. Figure 2 It can be seen that for the straight segment in the planned path, the number of reference targets selected is small because there is no corner; for the curved segment in the planned path, the number of reference targets selected is large. In addition, reference targets are selected on both sides of the planned path to reduce the influence of the cumulative error caused by selecting reference targets on one side, so that the driving trajectory of the mobile robot can be corrected during the actual navigation process. Figure 2In the figure, a simplified wall scene is given. In actual applications, the scenes on both sides of the planned path may include machines, shelves, etc. Therefore, the selection of each reference target can be based on the actual scene objects, or the marked points, lines or surfaces on the objects can be selected as reference targets. In actual applications, reference targets are usually objects with point cloud boundaries to facilitate subsequent feature information extraction and fast matching and positioning.

[0058] S2: extracting characteristic information of the reference target, and storing the characteristic information of each planned path and the reference target in correspondence.

[0059] The characteristic information of the reference target is specifically the center coordinates based on the boundary point coordinates of the reference target. In this embodiment, taking the point cloud map as an example, the point cloud information of the reference target and its surroundings is extracted, and the point cloud is clustered to obtain the coordinates of all points of the reference target, and further obtain the boundary point coordinates of the reference target. in, represents the coordinates of the jth boundary point in the i-th reference target in the map; n represents the number of boundary points of the reference target. Based on the obtained boundary point coordinates, the center coordinates of each reference target are further obtained. The center coordinates As the feature information of the i-th reference target. Finally, the selected reference target and its feature information corresponding to the planned path are stored in the server. In this process, the error of the reference target boundary point should be considered, and the error should be within a reasonable range and not greater than the following navigation accuracy requirements. Positioning and matching based on the feature information containing the center coordinates of the reference target can greatly reduce the amount of calculation, thereby further improving the operating efficiency.

[0060] Furthermore, in this embodiment, after extracting the feature information of the reference target, semantic information fusion can be performed on the feature information, which can be the nature of the reference target, the control strategy that the mobile robot needs to execute, etc. Figure 2 Reference target 1 is marked as going straight along the wall, and reference target 5 is marked as turning at the corner.

[0061] S3: For each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification passes; if the navigation accuracy does not meet the preset requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements.

[0062] S3-1: The mobile robot sequentially calls the feature information of each planned path and its corresponding reference target for navigation.

[0063] In this embodiment, the sensor carried by the mobile robot is a laser radar. When the mobile robot is navigating, the laser radar is used to detect the reference target and obtain the relative position of the reference target relative to the mobile robot. The current position of the mobile robot is obtained by calculation based on the feature information of the reference target corresponding to the path, which is recorded as in, represents the X-direction coordinate of the mobile robot in the map coordinate system at time t, represents the Y coordinate of the mobile robot in the map coordinate system at time t, yaw t Represents the heading angle of the mobile robot at time t. Based on the odometer and inertial measurement unit of the mobile robot, the angular velocity, speed, and acceleration of the mobile robot are obtained, and the current state of the mobile robot is obtained, which is recorded as (ω t ,v t ,a t ). Among them, ω t represents the angular velocity of the mobile robot at time t; v t represents the speed of the mobile robot at time t; a t represents the acceleration of the mobile robot at time t.

[0064] Then, the position point at the next moment is calculated based on the current posture and current state of the mobile robot. Figure 3 , the path point closest to the position point on the planned path is taken as the preview point, recorded as (X t+1 ,Y t+1 ,Yaw t+1 ), where X t+1 Indicates the X coordinate of the preview point in the map coordinate system, Y t+1 Indicates the Y coordinate of the preview point in the map coordinate system, Yaw t+1 Indicates the heading angle of the preview point.

[0065] Finally, the difference between the current position of the mobile robot and the moving target of the preview point is calculated, and the mobile robot is controlled to move and navigate to the preview point on the planned path.

[0066] S3-2: During the navigation process of the mobile robot, the position and posture of the mobile robot are obtained in real time and the movement deviation between the actual running path of the mobile robot and the planned path is calculated. If the movement deviation is less than the preset error, the navigation accuracy meets the requirements; otherwise, the navigation accuracy does not meet the requirements, and the number of reference targets is adjusted based on the obtained movement deviation, and then the navigation accuracy is re-verified until the mobile machine completes the navigation process according to the planned path, and the navigation accuracy meets the preset requirements during the entire navigation process.

[0067] S3-2-1: Obtain the real-time position and posture of the mobile robot, and calculate the movement deviation between the current position and posture of the mobile robot and the preview point determined by the mobile robot at the previous moment. Specifically, the position and posture of the mobile robot at time t is According to the navigation process of S3-1, the real-time position and posture of the mobile robot at time t+1 can be obtained as The mobile robot's movement deviation is Among them, (X t+1 ,Y t+1 ,Yaw t+1 ) is the moving deviation μ of the mobile robot at the preview point determined at time t, including the distance deviation μ d and angle deviation μ α , the preset error includes the distance error δ d and the angle error δ α . Preset errors are pre-set and stored.

[0068] S3-2-2: When the distance deviation μ d Less than the distance error δ d , and the angle deviation μ α Less than the angle error δ α If the navigation accuracy is 0, the navigation precision meets the requirement; otherwise, the navigation precision does not meet the requirement.

[0069] When the navigation accuracy meets the requirements, the verification passes.

[0070] When the navigation accuracy does not meet the requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements.

[0071] The process for updating the reference target includes:

[0072] The path segment type of the planned path corresponding to the current posture of the mobile robot is determined according to the real-time posture of the mobile robot.

[0073] According to the path segment type, and in combination with the current distance deviation and angle deviation of the mobile robot, the preset coefficient f1 based on the moving distance or the preset coefficient f2 based on the direction deflection angle corresponding to the path segment is adjusted. Specifically, it includes:

[0074] If the path segment is a straight line segment, the coefficient preset based on the moving distance of the path segment is adjusted:

[0075]

[0076] Among them, f1′ represents the adjusted coefficient based on the preset moving distance, μ d represents the distance deviation, μ α represents the angular deviation, δ d represents the distance error, δα represents the angle error;

[0077] If the path segment is a curved segment, the coefficient preset based on the direction deflection angle of the path segment is adjusted:

[0078]

[0079] Wherein, f2′ represents the adjusted coefficient preset based on the direction deflection angle.

[0080] The number of reference targets of the mobile robot on the path segment is recalculated according to the adjusted coefficient preset based on the moving distance or the coefficient preset based on the direction deflection angle, and then the reference targets are re-determined and feature information of the reference targets is extracted.

[0081] S3-3: storing the updated reference target's feature information and the corresponding planned path in correspondence In this embodiment, the reference target and its feature information and the planned path are stored in the server, which can improve storage and retrieval efficiency and reduce storage medium requirements.

[0082] In practical applications, considering the accidental errors in the navigation accuracy verification process, a single pass of the mobile robot through the navigation accuracy verification does not fully indicate that the mobile robot can meet the navigation accuracy requirements every time. Therefore, in this embodiment, when the mobile robot performs navigation accuracy verification, after the first verification that the entire planned path segment meets the navigation accuracy requirements, the mobile robot is subjected to multiple iterations of navigation accuracy verification, and the subsequent navigation information (feature information of the planned path and the reference target) is based on the updated result of the previous navigation accuracy verification. In this way, the impact of accidental errors on the verification results can be effectively reduced. In addition, if the mobile robot issues an error alarm (such as position loss) during the navigation accuracy verification process, the navigation accuracy verification process is reinitialized.

[0083] S4: When the mobile robot needs to perform a work task, it calls the stored planned path and the feature information of the corresponding reference target for navigation according to the work task to be performed. When the mobile robot performs repeated path navigation, it only needs to read the corresponding planned path and the feature information of the corresponding reference target from the server according to the specific work task of the mobile robot, and there is no need to perform repeated feature extraction and path planning. Its navigation process is the same as the navigation process of the mobile robot for navigation accuracy verification, except that the accuracy verification process is missing.

[0084] The process of mobile robot navigation includes:

[0085] S4-1: Detect the reference target through the laser radar, obtain the relative position of the reference target with respect to the mobile robot, and combine the feature information of the reference target to obtain the current position of the mobile robot, which is recorded as in, represents the X-direction coordinate of the mobile robot in the map coordinate system at time t, represents the Y coordinate of the mobile robot in the map coordinate system at time t, yaw t Represents the heading angle of the mobile robot at time t. Specifically, the mobile robot acquires at least two reference targets through laser radar detection. The relative positions of the two reference targets relative to the mobile robot are acquired according to the detected distance information and the detection angle, that is, the relative position and posture of the two reference targets relative to the mobile robot are determined. The current position and posture of the mobile robot is further acquired by calculating according to the characteristic information (center coordinates) of the two reference targets.

[0086] S4-2: Based on the odometer (Odom) and inertial measurement unit (IMU) of the mobile robot, the angular velocity, speed, and acceleration of the mobile robot are obtained, and the current state of the mobile robot is obtained, which is recorded as (ω t ,v t ,a t ), where ω t represents the angular velocity of the mobile robot at time t, v t represents the speed of the mobile robot at time t, a t represents the acceleration of the mobile robot at time t.

[0087] S4-3: Calculate the position point at the next moment (i.e., time t+1) according to the current position and current state of the mobile robot (specifically, estimate the position point of the robot at the next moment according to the current position, speed, and acceleration of the mobile robot), and take the path point closest to the position point on the planned path as the preview point, recorded as (X t+1 ,Y t+1 ,Yaw t+1 ), where X t+1 Indicates the X coordinate of the preview point in the map coordinate system, Y t+1 Indicates the Y coordinate of the preview point in the map coordinate system, Yaw t+1 Indicates the heading angle of the preview point.

[0088] In this embodiment, the method used when selecting the preview point on the planned path is radius tangent, such as Figure 3 shown.

[0089] S4-4: Calculate the difference between the current position of the mobile robot and the moving target of the preview point (i.e., the deviation between the position at time t and the target position at time t+1), and control the mobile robot to move and navigate to the preview point on the planned path.

[0090] Compared with the prior art, the navigation method provided in this embodiment is based on the stored historical paths (repeated paths) for navigation, avoiding recalculation and planning of paths, and greatly improving efficiency. At the same time, positioning and matching based on the center coordinates of the reference target can not only reduce the storage of point cloud data, but also greatly reduce the amount of calculation, further improving operating efficiency. During the navigation process, the mobile robot only needs to store the planned path and reference target points, which improves storage and retrieval efficiency and reduces storage medium requirements. Therefore, the navigation method provided in this embodiment can achieve cost savings by reducing the amount of calculation and improving the efficiency of computing power while ensuring navigation accuracy.

[0091] Furthermore, the mobile robot in this embodiment can also determine the detection range according to the position of the reference target, and obtain local features containing the characteristic information of the reference target through the directional detection of the sensor thereon, and then extract the characteristic information of the reference target from the local features for matching and positioning. This method is generally applied to radar sensors, which obtain local features containing the characteristic information of the reference target by scanning the environmental information in a smaller range, thereby further improving the accuracy and effectiveness of feature extraction of the reference target. Among them, the local features containing the characteristic information of the reference target obtained by the directional detection of the sensor can be environmental information, and then the local features containing the characteristic information of the reference target are intercepted according to the position of the reference target, and then the characteristic information of the reference target is extracted from the local features for matching and positioning. Extracting and matching the reference target features based on the local environmental characteristic information during the navigation process can further greatly reduce the amount of calculation, improve the efficiency of computing power utilization, and improve the operating efficiency.

[0092] Embodiment 2

[0093] The invention also discloses a closed-loop space repeated path navigation system, comprising a navigation data preparation module and a repeated path navigation module.

[0094] The navigation data preparation module is used to construct a map of the mobile robot's working area, perform path planning according to the work task, select a reference target on the map for each planned path, extract the characteristic information of the reference target, and store the characteristic information of each planned path and the reference target in correspondence; for each planned path, the mobile robot calls the characteristic information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification passes; if the navigation accuracy does not meet the preset requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements.

[0095] The repeated path navigation module is used to call the stored planned path and the corresponding reference target feature information for navigation according to the work task to be performed when the mobile robot needs to perform a work task.

[0096] Embodiment 3

[0097] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the navigation method of a closed-loop space repetitive path in the first embodiment is implemented.

[0098] Embodiment 4

[0099] The present invention also discloses a device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a navigation method for a repeated path in a closed-loop space in the first embodiment is implemented.

[0100] Before the mobile robot performs repeated path navigation, the present invention performs path planning according to the work task, selects reference targets on the map for each planned path, and extracts characteristic information of the reference targets. When performing repeated path navigation of the mobile robot, the stored planned path and the corresponding characteristic information of the reference targets are directly called according to the work task for navigation, which can avoid recalculating and planning the path, effectively reducing the amount of calculation and improving the efficiency of computing power utilization. The navigation accuracy is verified at least once for each planned path, which can improve the efficiency of repeated path navigation while ensuring the path navigation accuracy of the mobile robot.

[0101] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0102] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A navigation method for a closed-loop space repetitive path, characterized in that: include: Constructing a map of the working area of ​​the mobile robot, performing path planning according to the work task, selecting a reference target on the map for each planned path, extracting feature information of the reference target, and storing the feature information of each planned path and the reference target in correspondence; For each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification is passed; if the navigation accuracy does not meet the preset requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements; When the mobile robot needs to perform a work task, it calls the stored planned path and the feature information of the corresponding reference target for navigation according to the work task to be performed.

2. The closed-loop space repetitive path navigation method according to claim 1, characterized in that: The reference target is selected on the map for each planned path, specifically: In the map of the mobile robot's working area, reference targets are selected along both sides of the planned path. The number of reference targets selected is: Among them, F (S,θ) represents the number of reference targets, f1 represents the coefficient preset based on the moving distance, S represents the moving distance, l represents the preset distance parameter, represents rounding up, f2 represents the coefficient preset based on the direction deflection angle, Δθ ΔS Represents the unit moving distance angle, and ω represents the preset angle parameter.

3. The closed-loop space repetitive path navigation method according to claim 1, characterized in that: The characteristic information of the reference target includes the center coordinates based on the boundary point coordinates of the reference target.

4. The closed-loop space repetitive path navigation method according to claim 2, characterized in that: The process of the mobile robot calling the feature information of the corresponding reference target to perform at least one navigation accuracy verification for each planned path includes: The mobile robot sequentially calls the feature information of each planned path and its corresponding reference target for navigation; The real-time position and posture of the mobile robot during the navigation process is obtained, and the movement deviation between the actual running path of the mobile robot and the planned path is calculated: if the movement deviation is less than the preset error, the navigation accuracy meets the requirements; otherwise, the navigation accuracy does not meet the requirements, and the number of reference targets is adjusted based on the obtained movement deviation, and then the navigation accuracy verification is performed again until the navigation accuracy meets the requirements.

5. The closed-loop space repetitive path navigation method according to claim 4, characterized in that: The process of navigation performed by the mobile robot includes: The reference target is detected by laser radar, and the relative position and posture of the reference target relative to the mobile robot are obtained. The current position and posture of the mobile robot is obtained by combining the characteristic information of the reference target, which is recorded as in, represents the X-direction coordinate of the mobile robot in the map coordinate system at time t, represents the Y coordinate of the mobile robot in the map coordinate system at time t, yaw t represents the heading angle of the mobile robot at time t; Based on the odometer and inertial measurement unit of the mobile robot, the angular velocity, speed, and acceleration of the mobile robot are obtained, and the current state of the mobile robot is obtained and recorded as (ω t ,v t ,a t ), where ω t represents the angular velocity of the mobile robot at time t, v t represents the speed of the mobile robot at time t, a t represents the acceleration of the mobile robot at time t; According to the current position and state of the mobile robot, the position point at the next moment is calculated, and the path point closest to the position point on the planned path is used as the preview point, which is recorded as (X t+1 ,Y t+1 ,Yaw t+1 ), where X t+1 Indicates the X coordinate of the preview point in the map coordinate system, Y t+1 Indicates the Y coordinate of the preview point in the map coordinate system, Yaw t+1 Indicates the heading angle of the preview point; The difference between the current position of the mobile robot and the moving target of the preview point is calculated, and the mobile robot is controlled to move and navigate to the preview point on the planned path.

6. The closed-loop space repetitive path navigation method according to claim 5, characterized in that: The process of obtaining the real-time position and posture of the mobile robot during navigation and calculating the movement deviation between the actual running path of the mobile robot and the planned path includes: The real-time position and posture of the mobile robot is obtained, and the movement deviation between the current position and posture of the mobile robot and the preview point determined by the mobile robot at the previous moment is calculated; the movement deviation includes a distance deviation and an angle deviation, and the preset error includes a distance error and an angle error; when the distance deviation is less than the distance error, and the angle deviation is less than the angle error, it is considered that the navigation accuracy meets the requirements, otherwise it is considered that the navigation accuracy does not meet the requirements.

7. The closed-loop space repetitive path navigation method according to claim 6, characterized in that: The process of updating and setting the reference target includes: Determine the path segment type of the planned path corresponding to the current posture of the mobile machine according to the real-time posture of the mobile robot; According to the path segment type, and in combination with the current distance deviation and angle deviation of the mobile robot, a preset coefficient based on the moving distance or a preset coefficient based on the direction deflection angle corresponding to the path segment is adjusted; The number of reference targets of the mobile robot on the path segment is recalculated according to the adjusted coefficient preset based on the moving distance or the coefficient preset based on the direction deflection angle, and then the reference targets are re-determined and feature information of the reference targets is extracted.

8. The closed-loop spatial repetitive path navigation method according to claim 7, characterized in that: The process of adjusting the preset coefficient based on the moving distance or the preset coefficient based on the direction deflection angle corresponding to the path segment according to the path segment type and in combination with the current distance deviation and angle deviation of the mobile robot includes: If the path segment is a straight line segment, adjusting a coefficient of the path segment that is preset based on the moving distance; If the path segment is a curved segment, a coefficient preset based on the direction deflection angle of the path segment is adjusted.

9. The closed-loop spatial repetitive path navigation method according to claim 8, characterized in that: The adjustment method of the coefficient preset based on the moving distance is: Among them, f1′ represents the adjusted coefficient based on the preset moving distance, μ d represents the distance deviation, μ α represents the angular deviation, δ d represents the distance error, δ α represents the angle error; The adjustment method of the coefficient preset based on the direction deflection angle is: Wherein, f2′ represents the adjusted coefficient preset based on the direction deflection angle.

10. A closed-loop spatial repetitive path navigation system, characterized in that: include: A navigation data preparation module is used to construct a map of the working area of ​​the mobile robot, perform path planning according to the work task, select a reference target on the map for each planned path, extract feature information of the reference target, and store the feature information of each planned path and the reference target in correspondence; For each planned path, the mobile robot calls the feature information of the corresponding reference target to perform at least one navigation accuracy verification: if the navigation accuracy meets the preset requirements, the verification is passed; if the navigation accuracy does not meet the preset requirements, the reference target is updated and set, and then the navigation accuracy verification is performed again until the navigation accuracy meets the preset requirements; The repeated path navigation module is used to call the stored planned path and the corresponding reference target feature information for navigation according to the work task to be performed when the mobile robot needs to perform a work task.

Images