A motion control method for searching for a charging base signal, a chip, and a robot

Abstract

The application discloses a motion control method for searching for a charging base signal, a chip and a robot. The motion control method comprises the following steps: planning a navigation path based on the boundary of a grid map previously constructed by the robot; controlling the robot to move to a searching starting point position of the navigation path, and then controlling the robot to move along the navigation path in a preset searching direction from the searching starting point position until a charging base signal is detected; when the robot touches an obstacle during the movement along the navigation path, an edge-keeping obstacle-circumventing path with an optimal deviation degree relative to the navigation path is set, and then the robot is controlled to walk along the currently set edge-keeping obstacle-circumventing path until the charging base signal is detected; and after the robot circumvents the currently touched obstacle along the currently set edge-keeping obstacle-circumventing path, the robot is controlled to continue to move along the navigation path in the preset searching direction until the charging base signal is detected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot charging planning, and particularly relates to a motion control method for searching for a charging base signal, a chip and a robot. BACKGROUND

[0002] After the robot cleaner finishes cleaning, it enters a charging base searching process. Usually, the robot cleaner navigates to the vicinity of the charging base coordinate to complete the charging base searching process. In actual charging scenarios, the robot cleaner often fails to return to the charging base due to the inability to plan a path, being blocked by obstacles during navigation, or being trapped by obstacles. Specifically, if the robot records the coverage position of the charging base signal, the robot can be directly controlled to navigate back to the charging base. However, due to the passable factors of the path, including the inability to plan a navigation path, being blocked by obstacles during navigation, or being trapped by obstacles, the robot usually navigates to the charging base along the edge of the global map. However, the speed of the robot along the edge of the global map to the charging base is relatively slow, and the overall charging time is relatively long. SUMMARY

[0003] To solve the above problems, the present application provides a motion control method for searching for a charging base signal. In a pre-created grid map, a navigation path parallel to the boundary of the grid map is obtained through a gradual erosion operation and a neighborhood expansion method. During the movement of the robot along the currently planned navigation path, if an obstacle is encountered, an edge-obstacle-avoiding path with the best fitting degree relative to the navigation path is selected from the average deviation degree on the local map detected by the sensor in combination with the pre-planned navigation path. The success rate of the robot searching for the charging base signal is improved, and the robot is prevented from deviating from the navigation path, thereby speeding up the process of returning to the charging base. The specific technical solutions are as follows.

[0004] A motion control method for searching for a charging base signal, before the motion control method is executed, the robot does not detect the charging base signal. The motion control method comprises: planning a navigation path based on the boundary of the grid map pre-built by the robot, so that the navigation path is parallel to the boundary of the grid map; controlling the robot to move to a search starting point position of the navigation path, and then controlling the robot to move along the navigation path in a preset search direction from the search starting point position until the charging base signal is detected; when the robot encounters an obstacle during the movement along the navigation path, an edge-obstacle-avoiding path with the best deviation degree relative to the navigation path is set, and then the robot is controlled to move along the currently set edge-obstacle-avoiding path until the charging base signal is detected; after the robot moves along the currently set edge-obstacle-avoiding path to bypass the currently encountered obstacle, the robot is controlled to continue to move along the navigation path in the preset search direction until the charging base signal is detected.

[0005] Compared with the prior art global edge-searching charging base signal scheme, the technical scheme guarantees the continuity, smoothness and smoothness of the navigation path of the robot, and makes the robot move on the navigation path as much as possible during the searching of the charging base signal, which is beneficial to guide the robot to complete the obstacle avoidance behavior along the edge around the obstacle path with a reasonable deviation from the originally set navigation path, improve the navigation ability and obstacle avoidance ability of the robot in searching the charging base signal, greatly shorten the time for searching the charging base signal, speed up the speed of the robot returning to the seat, and improve the charging efficiency.

[0006] Further, the setting method of the search starting point position comprises: searching a grid point on the navigation path, which is not connected with the current position of the robot through unknown grid points and obstacle grid points, and configuring the currently searched grid point as a candidate target position; then selecting one of the candidate target positions closest to the current position of the robot as the search starting point position. It is guaranteed that the robot can reach the navigation path in a straight line to speed up the search speed.

[0007] Further, the setting method of the preset search direction comprises: when there is no grid coordinate point of the charging base on the navigation path, then counterclockwise is set as the preset search direction; when there is a grid coordinate point of the charging base on the navigation path, then the extension direction of the shortest connected path segment between the search starting point position and the grid coordinate point of the charging base is selected as the preset search direction; wherein the connected path segment is a continuous path segment extending from the search starting point position to the grid coordinate point of the charging base along the navigation path. This technical scheme takes the position of the charging base as the guide position, and screens a faster starting movement direction for the robot to speed up the speed of the robot in searching the charging base signal along the navigation path.

[0008] Further, the method for planning a navigation path based on the boundary of the grid map pre-constructed by the robot comprises: step 1, constructing a first circular domain with each map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then performing a gradual erosion operation on each first circular domain to obtain a path obstacle degree evaluation value of the grid points in each first circular domain; wherein the map boundary grid point is present on the boundary of the grid map pre-constructed by the robot; wherein the preset multiple is related to the positioning accuracy of the grid map; step 2, traversing every other preset interval map boundary grid point, and constructing a second circular domain with the currently traversed map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then searching for a grid point satisfying a critical condition in combination with the path obstacle degree evaluation value obtained in step 1; wherein all the grid points in the second circular domain have been assigned the path obstacle degree evaluation value in step 1; step 3, connecting the grid points satisfying the critical condition searched in step 2 in sequence as a navigation path, wherein the navigation path is parallel to the boundary of the grid map.

[0009] The technical solution gradually erodes each first circular domain in step 1 to obtain a path obstacle degree evaluation value of any grid point in the first circular domain, which is different from the path obstacle degree evaluation value obtained by a conventional erosion operation, and then searches for a grid point satisfying a critical condition in a second circular domain constructed in step 2 in combination with the path obstacle degree evaluation value obtained in step 1, to connect a navigation path with a detour effect, so as to obtain a boundary line obtained by subtracting the gradual erosion operation from the grid map, thereby planning a navigation path parallel to the boundary of the grid map, which is not affected by the environment, effectively realizing the denoising effect and smoothing effect of the navigation path, and improving the smoothness of the robot in edge obstacle avoidance and search for a charging seat signal.

[0010] Further, the gradual erosion operation in step 1 comprises: calculating a straight-line distance between a current eroded grid point and a center of a first circular domain to which the current eroded grid point belongs, and denoted as a boundary search distance; wherein the current eroded grid point is a grid point covered by a traversal unit starting from the center of the first circular domain; the traversal unit is equivalent to a structural element of the erosion operation; the preset multiple of the robot body diameter is set as an erosion radius; then, a path obstacle degree evaluation value of the current eroded grid point is set as a negative correlation with the boundary search distance within the first circular domain; it is judged whether the current set path obstacle degree evaluation value of the current eroded grid point is greater than a previously set path obstacle degree evaluation value of the same grid point, if yes, the current set path obstacle degree evaluation value of the current eroded grid point is maintained unchanged, and then the current set path obstacle degree evaluation value of the current eroded grid point is set as a latest path obstacle degree evaluation value of the current eroded grid point; otherwise, the previously set path obstacle degree evaluation value of the current eroded grid point is updated to the current set path obstacle degree evaluation value of the current eroded grid point, and then the current set path obstacle degree evaluation value of the current eroded grid point is set as the latest path obstacle degree evaluation value of the current eroded grid point.

[0011] The technical solution belongs to an improvement of a conventional erosion operation. In the process of traversing each map boundary grid point, each first circular domain performs a gradual erosion operation from the corresponding map boundary grid point to the surrounding, and the path obstacle degree evaluation value of the current eroded grid point is set as decreasing with the increase of the boundary search distance within the first circular domain, so that the path obstacle degree evaluation values of the grid points distributed from outside to inside within the first circular domain are sequentially increased. The technical solution also compares the current set path obstacle degree evaluation value of the current eroded grid point with the previously set path obstacle degree evaluation value of the same grid point based on the feature that the same grid point is covered by multiple first circular domains, and then sets the maximum path obstacle degree evaluation value obtained by the comparison as the latest path obstacle degree evaluation value of the current eroded grid point, as the latest path obstacle degree evaluation value of the current eroded grid point obtained under the gradual erosion operation (which can be understood as filling the grid point with the maximum value), so as to obtain the current eroded grid point closest to the corresponding map boundary grid point, which is beneficial to planning the shortest navigation path parallel to the boundary in the grid map.

[0012] Further, before performing the gradual erosion operation in step 1, the initial value of the path obstacle degree evaluation value of the map boundary grid point is set to a preset maximum path obstacle degree evaluation value, the initial value of the path obstacle degree evaluation value of the unknown grid point in the grid map is set to a preset maximum path obstacle degree evaluation value, and the initial value of the path obstacle degree evaluation value of the remaining types of grid points in the grid map is set to a preset minimum path obstacle degree evaluation value. In this way, by setting the path obstacle degree evaluation value, the map boundary grid point is extracted in advance, and different types of grid points, especially unknown grid points, are marked differently.

[0013] Further, step 2 specifically includes: setting a preset multiple of the robot body diameter as the preset interval, traversing the map boundary grid points along the boundary of the robot-prebuilt grid map according to the preset interval; constructing a second circular domain with the current traversed map boundary grid point as the center and the preset multiple of the robot body diameter as the radius, and then performing neighborhood expansion in the second circular domain with the current traversed map boundary grid point as the search center; when the latest path obstacle degree evaluation value of the currently expanded grid point in the second circular domain is greater than the preset minimum path obstacle degree evaluation value, and there is at least one grid point in the eight-neighborhood of the currently expanded grid point whose latest path obstacle degree evaluation value is the preset minimum path obstacle degree evaluation value, the grid point with the minimum path obstacle degree evaluation value in the currently expanded grid point is set as the searched grid point satisfying the critical condition. In this way, the critical position of the grid point with the latest path obstacle degree evaluation value being the preset minimum path obstacle degree evaluation value and the grid point with the latest path obstacle degree evaluation value not being the preset minimum path obstacle degree evaluation value is searched, which is used as the target position for connecting the navigation path.

[0014] Further, in step 3, one of the grid points satisfying the critical condition searched in the second circular domain corresponding to each map boundary grid point traversed according to the preset interval in step 2 is sequentially connected as the navigation path, so that the navigation path is parallel to the boundary of the grid map; one side of the navigation path has a grid point with the latest path obstacle degree evaluation value being the preset minimum path obstacle degree evaluation value, and the other side of the navigation path has a grid point with the latest path obstacle degree evaluation value not being the preset minimum path obstacle degree evaluation value; each of the grid points satisfying the critical condition corresponds to a map boundary grid point traversed at a specific time, so that the acquisition order of the grid points satisfying the critical condition is the same as the traversal order of the map boundary grid points. In this way, the navigation path has the meaning of robot obstacle-avoiding walking, and the feature of being parallel to the boundary of the grid map improves the coverage of the navigation path and reduces the planning of discontinuous and non-smooth path segments.

[0015] Further, the method for setting the edge-around path with the best deviation degree relative to the navigation path when the robot collides with the obstacle during the movement along the navigation path comprises: when the robot collides with and detects the obstacle during the movement along the navigation path, planning all edge-around paths for guiding the robot to go around the obstacle, wherein each edge-around path corresponds to an edge-around direction; then, according to the minimum line segment distance between each grid point on each edge-around path and all grid points on a reference navigation path segment on the pre-planned navigation path that supports the robot to continue to move after going around the obstacle, selecting an edge-around path with the minimum average deviation degree relative to the reference navigation path segment, and setting the currently selected edge-around path as the edge-around path with the best deviation degree relative to the navigation path. Compared with the prior art, the technical scheme uses the reference navigation path segment that supports the robot to continue to move after going around the obstacle to select the edge-around path corresponding to the best edge-around direction, so as to reduce the deviation of the robot from the originally planned navigation path during and after going around the obstacle, and is beneficial to guiding the robot to complete the going-around behavior of the currently detected obstacle along the edge-around path with navigation value, and improves the intelligent navigation ability and the going-around ability of the robot.

[0016] Further, the method for selecting the edge-around path with the minimum average deviation degree relative to the reference navigation path segment according to the minimum line segment distance between each grid point on each edge-around path and all grid points on a reference navigation path segment on the pre-planned navigation path that supports the robot to continue to move after going around the obstacle comprises: respectively counting the number of grid points included in each edge-around path; for each edge-around path, calculating the minimum line segment distance between each grid point on the edge-around path and all grid points on the reference navigation path segment, adding the minimum line segment distances corresponding to each grid point on the edge-around path to obtain a local distance sum in the corresponding edge-around direction, and then averaging the local distance sum in the corresponding edge-around direction using the number of grid points included in the edge-around path to obtain the average deviation of the edge-around path, which is used to represent the average deviation degree of the edge-around path relative to the reference navigation path segment; by comparing the sizes of the average deviations corresponding to each edge-around path, selecting the edge-around path with the minimum average deviation as the path for the robot to actually walk along the edge of the obstacle subsequently.

[0017] Compared with the prior art, the average distance of grid points between each edge- around barrier path and the reference navigation path segment is compared, which is used to measure the deviation of each edge- around barrier path relative to the same reference navigation path segment, and is more representative and can quickly indicate the change of the edge- around barrier path relative to the same reference navigation path segment compared with directly comparing the distance sum.

[0018] Further, the edge- around barrier path is a left edge- around barrier path for guiding the robot to walk along the left edge of the barrier in a clockwise direction, or the edge- around barrier path is a right edge- around barrier path for guiding the robot to walk along the right edge of the barrier in a counterclockwise direction. Thus, the edge- around path required for the robot to walk along the contour edge of the map projection of the barrier is planned. The robot is adapted to detect and walk along the edge between different barriers.

[0019] Further, in the process of comparing the sizes of the average deviations of each edge- around barrier path, if the average deviations of two or more edge- around barrier paths are equal, a random edge- around barrier path is selected as the path for the robot to actually walk along the edge of the barrier subsequently. The efficiency of the robot in processing the edge- around barrier path is improved.

[0020] Further, the specific planning method of the edge- around barrier path for guiding the robot to walk around the barrier includes: step 1, within the detection range of the sensor of the robot, a pair of adjacent free grid points and obstacle grid points are searched along the current moving direction of the robot by means of neighborhood search; then, among the pair of adjacent free grid points and obstacle grid points, the free grid point is configured as an edge- around search center, and the obstacle grid point is configured as an edge- around search starting point; step 2, when a candidate edge- around behavior point is searched in the neighborhood of the edge- around search center along a preset edge- around direction from the edge- around search starting point, the candidate edge- around behavior point is updated as the edge- around search center, the edge- around search center before the update is updated as the edge- around search starting point, and the edge- around search center before the update is connected to the edge- around barrier path in the corresponding edge- around direction to guide the robot to walk along the edge of the barrier; step 3, step 2 is repeatedly executed until the newly connected edge- around barrier path meets a preset edge- around barrier condition.

[0021] Compared with the prior art, the technical scheme is characterized in that after the robot detects and touches the obstacle, the search is performed from the obstacle to the current position of the robot in the current moving direction of the robot, and a pair of adjacent reference idle grid points and reference obstacle grid points are searched out, then the neighborhood search is performed in the predetermined edge-following direction with the reference idle grid point as the center and the reference obstacle grid point as the starting point, the edge-following obstacle-avoiding path supporting the robot to walk along the edge of the obstacle is searched out along each edge-following direction, the edge-following obstacle is crossed faster, and the success rate of the robot to avoid the obstacle is improved by searching the neighborhood in the form of grid by grid.

[0022] Further, in step 1, the method of searching out a pair of adjacent idle grid points and obstacle grid points by neighborhood search specifically comprises the following steps: step 11, obtaining a set of obstacle grid points covered by the obstacle in the two-dimensional grid map, and selecting the obstacle grid point closest to the current position of the robot from the set, and setting the obstacle grid point closest to the current position of the robot as the search center for neighborhood search; step 12, judging whether the neighborhood grid point corresponding to the search center searched in the current moving direction is an idle grid point, if yes, entering step 13, otherwise entering step 14; step 13, configuring the searched idle grid point as the edge-following search center, and configuring the search center as the edge-following search starting point, wherein the newly configured edge-following search center and the newly configured edge-following search starting point are a pair of adjacent grid points; step 14, if the neighborhood grid points searched in the current moving direction in step 12 are all obstacle grid points, updating the neighborhood grid point closest to the search center searched in step 12 as the search center in step 12, and returning to step 12, wherein the grid points searched in step 12 are not the grid points of repeated search.

[0023] Compared with the prior art, the technical scheme is characterized in that based on the set of obstacle grid points detected in real time, the neighborhood search is performed from the obstacle grid point closest to the current position of the robot as the first search center, so as to realize the maximum proximity to the contour line of the obstacle; then steps 12 to 14 are iteratively executed, the search center is constantly updated along the current moving direction of the robot, until a pair of adjacent idle grid point configured as the edge-following search center and obstacle grid point configured as the edge-following search starting point are searched out by the neighborhood search, so as to provide the starting point for searching the grid point for accessing the edge-following obstacle-avoiding path.

[0024] Further, in the step 2, the method for searching the candidate edge behavior point from the edge search starting point along the preset edge direction in the neighborhood of the edge search center comprises: in the neighborhood of the edge search center, from the edge search starting point, judging whether the first searched idle grid point exists in the grid points searched along the preset edge direction one by one, and if yes, determining the first searched idle grid point as the candidate edge behavior point; wherein the edge search starting point is not included in the range of the grid points searched along the preset edge direction.

[0025] The technical scheme determines a candidate edge behavior point in a nine-grid area centered on the edge search center, performs neighborhood search around the edge search center along the preset edge direction, and determines a candidate edge behavior point until the first idle grid point is searched, so as to wait for joining and planning the edge obstacle-avoiding path, thereby realizing the process of constantly updating the neighborhood of the edge search center along the same preset edge direction, connecting the edge search center with the idle grid point in the neighborhood in sequence to form the edge obstacle-avoiding path, and accelerating the planning speed of the edge obstacle-avoiding path.

[0026] Further, the preset edge obstacle-avoiding condition comprises: the edge search center newly connected into the edge obstacle-avoiding path coincides with other grid points on the edge obstacle-avoiding path to which the edge search center belongs, so that a closed planar geometric figure is formed on the edge obstacle-avoiding path in the corresponding edge direction. It is determined that the robot has searched and planned the edge path supporting the robot to avoid obstacles according to the obstacle information. The technical scheme considers the length feature of the navigation path after the robot avoids obstacles, selects the reference navigation path segment supporting the robot to pass through, and uses the reference navigation path segment as the reference navigation path to judge the navigation rationality of the currently planned edge obstacle-avoiding path relative to the navigation path.

[0027] Further, the motion control method further comprises: selecting a reference navigation path segment supporting the robot to pass through after avoiding obstacles from the navigation path which is not occupied by the obstacle, located behind the obstacle and adjacent to the obstacle grid point, along the navigation extension direction of the navigation path, wherein the reference navigation path segment is connected by the grid points arranged continuously in the navigation path. The technical scheme considers the length feature of the navigation path after the robot avoids obstacles, selects the reference navigation path segment supporting the robot to pass through, and uses the reference navigation path segment as the reference navigation path to judge the navigation rationality of the currently planned edge obstacle-avoiding path relative to the navigation path.

[0028] Further, if the navigation path passes through the obstacle, the navigation path is divided into a path segment reaching the front of the obstacle, a path segment covered by the obstacle and a path segment behind the obstacle along the navigation extension direction of the navigation path, wherein each path segment is connected by grid points, and the reference navigation path segment is selected from the path segment behind the obstacle. The technical solution classifies the navigation path based on the relative position relationship of the obstacle on the navigation path to obtain the navigation path segment supporting the robot passing through after the obstacle is bypassed.

[0029] Further, the preset along-edge obstacle bypassing condition includes that none of the line segments between the newly searched candidate along-edge behavior point and any grid point of the reference navigation path segment passes through an obstacle grid point. When the newly searched candidate along-edge behavior point and the reference navigation path segment are passable, the along-edge obstacle bypassing path is stopped from being continuously planned, so as to avoid the along-edge obstacle bypassing path being farther and farther along the preset along-edge direction, and facilitate the robot to return to the navigation path smoothly after the along-edge obstacle bypassing is completed.

[0030] Further, the preset along-edge obstacle bypassing condition includes that the trajectory length of the along-edge obstacle bypassing path is greater than the profile length of the obstacle bypassed in the corresponding along-edge direction. The profile length of the obstacle bypassed in the two-dimensional grid map is greater than the detectable distance of the sensor of the robot. The technical solution limits the trajectory length of the along-edge obstacle bypassing path to a certain range, so as to plan the along-edge obstacle bypassing path with a reasonable length, reduce the search calculation amount, and improve the efficiency of the robot in planning the along-edge obstacle bypassing path.

[0031] A chip, the chip stores a computer program, characterized in that the computer program is used to control the mobile robot to execute the motion control method.

[0032] A robot, the robot is provided with the chip, and the robot is configured to execute the motion control method. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 A flowchart of a motion control method for searching a charging base signal is provided for an embodiment of the present application.

[0034] Figure 2 A flowchart of a navigation path planning method based on a map boundary is provided for another embodiment of the present application.

[0035] Figure 3 A flowchart of an along-edge obstacle bypassing path setting method based on a navigation path is provided for another embodiment of the present application. DETAILED DESCRIPTION

[0036] The accompanying drawings that are described below are merely some examples or embodiments of the present application, and for those of ordinary skill in the art, the present application can also be applied to other similar scenarios according to the drawings without any creative labor. In addition, it can also be understood that although the efforts made in this development process can be complex and lengthy, some design, manufacturing or production changes made on the basis of the technical content disclosed in the present application are only routine technical means for those of ordinary skill in the art related to the disclosure content of the present application, and should not be understood as insufficient disclosure of the present application.

[0037] In the present application, the reference to "embodiments" means that the specific features, structures or properties described in conjunction with the embodiments can be included in at least one embodiment of the present application, and the appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor is it an independent or alternative embodiment to other embodiments. The person of ordinary skill in the art explicitly or implicitly understands that the embodiments described in the present application can be combined with other embodiments without conflict.

[0038] After the robot completes the task in the current working area, it needs to return to the charging seat. If the robot detects a charging seat signal during the execution of the task in the current working area, the corresponding position is recorded, so that the robot starts from this position to select the return charging algorithm disclosed in the prior art to navigate back to the charging seat. However, sometimes the navigation cannot reach the target position, and the robot is selected to navigate globally along the edge to the charging seat, but the global edge navigation method reduces the speed of the robot returning to the charging seat. In order to speed up the process of the robot returning to the seat, the present embodiment discloses a motion control method for searching for a charging seat signal, which is suitable for executing this motion control method in the scenario where the robot does not detect a charging seat signal. As shown in the figure, the motion control method comprises: Figure 1

[0039] Step M1: planning a navigation path based on the boundary of the grid map previously constructed by the robot, so that the navigation path is parallel to the boundary of the grid map; specifically, a navigation path parallel to the boundary of the grid map is planned; then go to step M2.

[0040] ​Step M2, control the robot to move to the search starting point position of the navigation path, and then control the robot to move along the navigation path in a preset search direction from the search starting point position; and then enter step M3. It should be noted that the search starting point position of the navigation path is set before the robot moves, and the preset search direction is a path extension direction set on the basis of the search starting point position, so as to facilitate the robot to quickly search the charging seat signal along the navigation path, thereby speeding up the completion of the return-to-seat task. It is worth noting that the purpose of the robot moving along the navigation path in this embodiment is to search for the charging seat signal, which is used to speed up the process of the robot returning to the charging seat. In some implementation scenarios, step M2 cannot directly control the robot to return to the charging seat position.

[0041] Step M3, when the robot touches an obstacle during movement along the navigation path, an edge-around-obstacle path with an optimal deviation degree relative to the navigation path is set, and the robot is then controlled to move along the currently set edge-around-obstacle path; and then enter step M4. It should be noted that the setting of the edge-around-obstacle path is based on the movement of the robot along the navigation path, so as to plan or select an edge-around-obstacle path with an optimal deviation degree relative to the navigation path, so as to control the robot to complete the obstacle-avoiding behavior and avoid the robot being blocked during movement along the navigation path

[0042] Between steps M2 and M3, if the robot detects the charging seat signal during movement along the navigation path, the robot is controlled not to continue moving along the navigation path, and it is determined that the robot has detected the charging seat signal (return-to-charge guiding signal) emitted by the charging seat, thereby improving the search speed of the charging seat signal, speeding up the return-to-seat process of the robot, and achieving the purpose of accelerating the return-to-seat process of the application.

[0043] It is worth noting that how to complete the return-to-seat docking charging according to the currently received charging seat signal is not the technical innovation point of the application, and will not be described here. The application has already accelerated the return-to-seat process of the robot by controlling the robot to quickly search the charging seat signal, and the application concept is to plan the path during the movement of the robot from when the robot does not detect the charging seat signal to when the robot detects the charging seat signal, and to control the movement on the planned path, so as to accelerate the stage of performing return-to-seat charging according to the charging seat signal, thereby accelerating the return-to-seat process.

[0044] Step M4, after the robot passes the current touched obstacle along the edge- around-obstacle path set in step M3, the robot is controlled to continue moving along the navigation path in the preset search direction until the charging base signal is detected. In this embodiment, the edge-around-obstacle path set in step M3 is to support the robot to complete the edge-around-obstacle behavior along the edge of the current touched obstacle, so that the robot can pass the current touched obstacle and return to the navigation path in step M1, and then the robot is controlled to continue moving along the navigation path in the preset search direction until the charging base signal is detected, so as to speed up the process of returning to the charging base.

[0045] Between step M3 and step M4, if the robot detects the charging base signal while moving along the edge-around-obstacle path set in step M3, the robot is controlled not to continue moving along the edge-around-obstacle path set in step M3, and it is determined that the robot has detected the charging base signal (return-to-charge guiding signal) emitted by the charging base, which ensures the success rate of the local edge-around-obstacle behavior of the robot, reduces the probability of being trapped, improves the search speed of the charging base signal, and achieves the purpose of accelerating the return-to-charge process of the present application.

[0046] Compared with the prior art of globally searching the charging base signal along the edge, the embodiment ensures the continuity, smoothness and smoothness of the navigation path of the robot, and makes the robot move on the navigation path as much as possible during the search for the charging base signal, which is conducive to guiding the robot to complete the edge-around-obstacle behavior of the obstacle along the edge-around-obstacle path with navigation value, improves the navigation ability and edge-around-obstacle ability of the robot in searching for the charging base signal, greatly shortens the search time, and improves the seat-finding efficiency.

[0047] As an embodiment, the setting method of the search starting point position comprises: searching for a grid point on the navigation path, which is not connected with unknown grid points and obstacle grid points, and configuring the currently searched grid point as a candidate target position, so that the robot can directly reach any candidate target position from the current position without obstacles; then selecting one of the currently configured candidate target positions closest to the current position of the robot as the search starting point position. This ensures that the robot can directly reach the target position on the navigation path to speed up the search for the signal.

[0048] As an embodiment, the setting method of the preset searching direction comprises: when there is no grid coordinate point of the charging base on the navigation path, then the counterclockwise is set as the preset searching direction, so that the robot moves along the navigation path from the searching starting position in the counterclockwise direction until the charging base signal is detected. When there is a grid coordinate point of the charging base on the navigation path, then the extension direction of the shortest connected path segment among all the connected path segments between the searching starting position and the grid coordinate point of the charging base is selected as the preset searching direction; wherein the connected path segment is a continuous path segment extending from the searching starting position along the navigation path to the grid coordinate point of the charging base. Specifically, along the navigation path, the extension direction of the shortest connected path segment among all the connected path segments pointing from the searching starting position to the grid coordinate point of the charging base is selected as the preset searching direction, wherein the aforementioned extension direction is equivalent to the circumnavigation direction extending from the searching starting position to the grid coordinate point of the charging base. This embodiment takes the position of the charging base as the guide position, and screens a faster starting movement direction for the robot, so as to speed up the speed of the robot searching for the charging base signal along the navigation path.

[0049] As an embodiment, as shown in Figure 2 the planning method of the navigation path based on the map boundary is disclosed, before the planning method of the navigation path is executed, the robot pre-constructs a grid map, the grid map has a map boundary, the map boundary is represented by a grid point, which is referred to as a map boundary grid point. In the aforementioned step M1, the method for planning the navigation path based on the boundary of the grid map pre-constructed by the robot specifically comprises the following steps:

[0050] In step P1, a first circular domain is constructed with each map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then a gradual erosion operation is performed on each first circular domain to obtain the path obstacle degree evaluation value of the grid points in each first circular domain. Specifically, the path obstacle degree evaluation value obtained by performing the gradual erosion operation on the first circular domain of the current map boundary grid point is updated by the path obstacle degree evaluation value obtained by the corresponding gradual erosion operation of the adjacent previous map boundary grid point, so that the path obstacle degree evaluation value of the current map boundary grid point is filled with the best value. Wherein, the map boundary grid point is present on the boundary of the grid map pre-constructed by the robot; the preset multiple is related to the positioning accuracy of the grid map, and the preset multiple is set according to the accuracy of the grid map. When the step M1 is started to be executed, the preset multiple is set to 1 by default; then step P2 is entered.

[0051] Step P2, traverse one map boundary grid point every other preset interval, and construct a second circular domain with the currently traversed map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then search for a grid point that meets the critical condition in combination with the path obstacle degree evaluation value obtained in step 1, so as to obtain a grid point filled with a suitable path obstacle degree evaluation value during traversal along the map boundary; then enter step P3; wherein all the grid points in the second circular domain have been assigned the path obstacle degree evaluation value in step 1. The process of executing step P2 is to traverse a corresponding map boundary grid point along the map boundary according to the preset interval, but not all map boundary grid points, saving computing resources.

[0052] Step P3, sequentially connect the grid points that meet the critical condition searched out in step P2 as a navigation path. Wherein each grid point that meets the critical condition is a corresponding map boundary grid point traversed, at least consistent in the order of time extracted, but irrelevant to the size of the path obstacle degree evaluation value of the specific grid point; preferably, the route sequentially connected according to the grid points that meet the critical condition searched out in step P2 is parallel to the boundary of the grid map.

[0053] Compared with the morphological operation of the related map graph in the prior art, the present embodiment performs a gradual erosion operation on each first circular domain in step P1 to obtain a path obstacle degree evaluation value of any grid point in the first circular domain, which is different from the path obstacle degree evaluation value obtained by a conventional erosion operation, and then searches for a grid point that meets the critical condition in the second circular domain constructed in step P2 in combination with the path obstacle degree evaluation value obtained in step P1, to connect a navigation path, so as to obtain an inner coil obtained by subtracting the gradual erosion operation from the grid map, and thus the navigation path parallel to the boundary of the grid map planned by the present embodiment is not affected by the environment, effectively realizing the denoising effect and smoothing effect of the navigation path, and improving the smoothness of the robot in searching for the charging seat signal along the boundary.

[0054] As an embodiment, the gradual erosion operation in step P1 includes:

[0055] Step P11, calculate the straight-line distance between the current eroded grid point and the center of the first circular domain to which the current eroded grid point belongs, and mark it as the boundary search distance; wherein the current eroded grid point is a grid point covered by the traversal unit starting from the center of the first circular domain; the traversal unit is equivalent to the structural element of the erosion operation; then go to step P12. It should be noted that when the structural element scans the grid points of the grid map, part of the neighborhood of the structural element will be outside the boundary of the grid map. For example, when the structural element scans the first row of pixel values above the boundary of the grid map, it is assumed that the structural element and its neighborhood form a 3x3 sliding window, which is used to obtain the outline of the region to be scanned in the grid map; part of the structural element will be outside the grid map. At this time, in order to effectively process the grid points covered by the boundary of the grid map, the embodiment assigns a value to the structural element and its neighborhood in a manner different from the conventional morphological operation, that is, the aforementioned path obstacle degree evaluation value. As known by those skilled in the art, the window used to scan the grid map can be any shape and size of convolution kernel with the structural element as the reference point. This convolution kernel can be a binary matrix representing the neighborhood of the structural element defined in general dilation and erosion operations. Therefore, a two-dimensional planar structural element is composed of a matrix of numerical values of 0 or 1, so that the origin of the structural element specifies the range of the grid region to be processed in the grid map, and the grid points with a value of 1 in the structural element determine whether the neighborhood grid of the structural element needs to participate in the calculation of the aforementioned path obstacle degree evaluation value when performing dilation or erosion operations.

[0056] Step P12, set the preset multiple of the robot body diameter as the erosion radius; then go to step P13; wherein the preset multiple is set to 1 by default, and in actual path planning operations, it will be adaptively adjusted according to the positioning error or reading accuracy of the grid map to overcome the interference of sensor collection errors and environmental factors, which is an adjustment method that can be mastered by those skilled in the art.

[0057] Step P13, set the path obstacle degree evaluation value of the current eroded grid point to be in a negative correlation with the boundary search distance within the first circular domain, so that the path obstacle degree evaluation value corresponding to the current eroded grid point farther from the center is smaller, and the path obstacle degree evaluation value corresponding to the current eroded grid point closer to the center is larger. Then go to step P14.

[0058] It should be noted that before performing the gradual erosion operation in step P1, the initial value of the path obstacle degree evaluation value of the map boundary grid point is set to a preset maximum path obstacle degree evaluation value, the initial value of the path obstacle degree evaluation value of the unknown grid point in the grid map is set to a preset maximum path obstacle degree evaluation value, and the initial value of the path obstacle degree evaluation value of the remaining type of grid point in the grid map is set to a preset minimum path obstacle degree evaluation value, so that the map boundary grid point is extracted in advance by setting the path obstacle degree evaluation value, and different types of grid points, especially unknown grid points, are distinguished and marked, and preferably, these grid points with pre-set path obstacle degree evaluation values are stored in the same array for subsequent sequential traversal.

[0059] In the embodiment, the grid map constructed by the robot in advance includes free grid points, obstacle grid points and unknown grid points, which correspond to three states of free, occupied and unknown of the grid map respectively; the free state grid refers to a grid not occupied by an obstacle, which is a reachable grid position point of the robot, and is the free grid point which can constitute an unoccupied area; the occupied state grid refers to a grid occupied by an obstacle, which is the obstacle grid point and can constitute an occupied area; the unknown grid refers to a grid area whose specific situation is not clear during the construction of the map by the robot, and its position point is often blocked by an obstacle, which can constitute an unknown area.

[0060] It should be noted that taking a sweeping robot as an example, a local probability grid can be established based on a single-line laser radar, or a local grid map can be established based on a visual sensor. The global map is a map drawn by the sweeping robot for the first time, which searches each room movement area by using various sensors (such as a collision sensor, an acceleration sensor, a gyroscope, an ultrasonic range finder, a camera, a single-line laser radar, etc.) carried by the sweeping robot, senses the position, shape and size of each room, and the position, shape and size of the encountered obstacles, and draws an environment boundary map according to the above information. The map is drawn by recording the entire indoor map while walking along the edge of the room, and the map includes an occupied area, an unoccupied area (free area) and an unknown area. The occupied area and the unoccupied area can belong to a cleaned area.

[0061] Step P14, judging whether the path obstacle degree evaluation value of the current set current eroded grid point is greater than the path obstacle degree evaluation value of the same grid point set in advance, if yes, entering step P15, otherwise entering step P16.

[0062] Step P15, maintaining the current set of the current eroded grid point path obstacle degree evaluation value unchanged, and setting the current set of the current eroded grid point path obstacle degree evaluation value as the latest path obstacle degree evaluation value of the current eroded grid point; achieving the maximum value in the path obstacle degree evaluation values obtained by the gradual erosion operation of the current eroded grid point through the first circle domain corresponding to the plurality of map boundary grid points as the latest path obstacle degree evaluation value of the current eroded grid point.

[0063] Step P16, updating the previously set of the current eroded grid point path obstacle degree evaluation value to the current set of the current eroded grid point path obstacle degree evaluation value, and setting the current set of the current eroded grid point path obstacle degree evaluation value as the latest path obstacle degree evaluation value of the current eroded grid point.

[0064] Preferably, in the process of performing the aforementioned gradual erosion operation on each first circle domain, the path obstacle degree evaluation value of the current eroded grid point is updated at least once after comparison according to step P14 when the current eroded grid point is not set to the preset maximum path obstacle degree evaluation value (the current eroded grid point is not a map boundary grid point, nor an unknown grid point).

[0065] As can be seen from the aforementioned gradual erosion operation of steps P11 to P16, the embodiment belongs to an improvement of the conventional erosion operation. In the process of traversing each map boundary grid point, each first circle domain performs gradual erosion operation from the corresponding map boundary grid point to the surrounding, and the obtained path obstacle degree evaluation value of the current eroded grid point is set to decrease with the increase of the boundary search distance in the first circle domain, so that the path obstacle degree evaluation values of the grid points distributed from outside to inside in the first circle domain are sequentially increased. The embodiment is also based on the feature that the same grid point is covered by multiple first circle domains. The path obstacle degree evaluation value of the current set of the current eroded grid point is compared with the path obstacle degree evaluation value of the same grid point set previously, and the maximum path obstacle degree evaluation value obtained by the comparison is set as the latest path obstacle degree evaluation value of the current eroded grid point, which is the latest path obstacle degree evaluation value of the current eroded grid point obtained under the gradual erosion operation (it can be understood as filling the grid point with the maximum value). Thus, the current eroded grid point closest to the corresponding map boundary grid point is obtained, which is beneficial to planning the shortest navigation path parallel to the boundary in the grid map.

[0066] On the basis of the foregoing embodiment, when the boundary search distance is greater than the corrosion radius, the path obstacle degree evaluation value of the current corroded grid point is set to a preset minimum path obstacle degree evaluation value, which is equivalent to setting the latest path obstacle degree evaluation value of the current corroded grid point with a straight-line distance greater than the corrosion radius from the center of the corresponding first circular domain to the preset minimum path obstacle degree evaluation value; wherein the path obstacle degree evaluation value of the map boundary grid point is set to a preset maximum path obstacle degree evaluation value. Thus, all grid points outside the first circular domain of the current corrosion (performing the progressive corrosion operation) are set to the preset minimum path obstacle degree evaluation value, and the path obstacle degree evaluation value of any grid point in the first circular domain is constrained to be greater than the preset minimum path obstacle degree evaluation value and less than or equal to the preset maximum path obstacle degree evaluation value, which is beneficial to the composition of the corrosion region. The present embodiment creates a critical grid position at the edge of the first circular domain by filling the preset minimum path obstacle degree evaluation value to the grid points outside the first circular domain.

[0067] As an embodiment, the method of setting the path obstacle degree evaluation value of the current corroded grid point in a negative correlation with the boundary search distance in the first circular domain includes that the path obstacle degree evaluation value of the current corroded grid point is the product of the ratio of the difference between the corrosion radius and the boundary contraction distance and the corrosion radius and the preset maximum path obstacle degree evaluation value, so that the negative correlation is a linear negative correlation. Wherein the difference between the corrosion radius and the boundary contraction distance is the difference between the corrosion radius and the boundary contraction distance, so that the path obstacle degree evaluation value of the current corroded grid point decreases with the increase of the boundary search distance, that is, the greater the boundary search distance, the smaller the difference between the corrosion radius and the boundary contraction distance, the farther the current corroded grid point deviates from the center of the first circular domain (covered in the progressive corrosion operation), resulting in a smaller calculation result of the path obstacle degree evaluation value of the current corroded grid point, and vice versa. In order to ensure the calculation accuracy to match the positioning accuracy of the map, the present embodiment preferably sets the preset maximum path obstacle degree evaluation value to the decimal number 100, and sets the preset minimum path obstacle degree evaluation value to the decimal number 0.

[0068] As an embodiment, the step P2 specifically includes:

[0069] The robot body diameter of the preset multiple is set as the preset interval, and the preset interval is traversed along the boundary of the pre-constructed grid map of the robot and according to the preset interval. That is, every other preset interval is traversed on the boundary of the pre-constructed grid map. The preset multiple is set, which also represents the setting of the preset interval, so that the similarity between the navigation path and the boundary of the grid map and the search repetitiveness of the grid point meeting the critical condition are balanced.

[0070] Every time a grid point on the boundary of the map is traversed, a second circular domain is constructed with the currently traversed grid point on the boundary of the map as the center and the preset multiple of the robot body diameter as the radius. Then, the neighborhood expansion is performed in the second circular domain with the currently traversed grid point on the boundary of the map as the search center, including four-neighborhood expansion or eight-neighborhood expansion.

[0071] In the second circular domain, when the latest path obstacle degree evaluation value of the currently expanded grid point is greater than the preset minimum path obstacle degree evaluation value and there is at least one grid point in the eight-neighborhood of the currently expanded grid point whose latest path obstacle degree evaluation value is the preset minimum path obstacle degree evaluation value, the grid point with the minimum path obstacle degree evaluation value in the currently expanded grid point is set as the searched grid point meeting the critical condition. This also means that the currently expanded grid point is adjacent to the grid point filled with the preset minimum path obstacle degree evaluation value, which is the boundary grid point of the grid point filled with the preset minimum path obstacle degree evaluation value and the grid point not filled with the preset minimum path obstacle degree evaluation value. In this embodiment, it can be used as the target position for connecting the navigation path.

[0072] In this embodiment, in the expansion process, if every other robot body diameter (i.e., the preset multiple is set to 1) is searched for a grid point with the minimum path obstacle degree evaluation value adjacent to the grid point with the path obstacle degree evaluation value set to the preset minimum path obstacle degree evaluation value, although the searched grid point meeting the critical condition is the farthest one relative to the currently searched grid point on the boundary of the map each time, it may be repeated from the previous search. Therefore, preferably, the preset interval is set such that the searched grid point meeting the critical condition in the second circular domain corresponding to different grid points on the boundary of the map is different. The preset interval and the adaptive setting of the preset multiple balance the similarity between the navigation path and the boundary of the grid map and the search repetitiveness of the grid point meeting the critical condition.

[0073] On the basis of the foregoing embodiment, in step P3, one grid point satisfying the critical condition searched out in the second circular domain corresponding to each map boundary grid point traversed according to the preset interval in step P2 is sequentially connected to form the navigation path, so that the navigation path is parallel to the boundary of the grid map; wherein one side of the navigation path has a grid point with the latest path obstacle degree evaluation value being the preset minimum path obstacle degree evaluation value, and the other side of the navigation path has a grid point with the latest path obstacle degree evaluation value not being the preset minimum path obstacle degree evaluation value, so that a pair of grid points filled with the preset minimum path obstacle degree evaluation value and not filled with the preset minimum path obstacle degree evaluation value can be found on both sides of the navigation path, and the navigation path has the significance of the boundary line, so that the navigation path has the significance of the robot walking around the obstacle. Thus, the navigation path has the significance of the robot walking around the obstacle, and the feature of being parallel to the boundary of the grid map improves the coverage rate of the navigation path and reduces the planning of discontinuous and non-smooth path segments.

[0074] It should be emphasized that each grid point satisfying the critical condition corresponds to a map boundary grid point traversed at a specific time, and the corresponding grid point satisfying the critical condition can be sequentially connected according to the traversal order of the map boundary grid point. The specific direction includes but is not limited to the clockwise direction and the counterclockwise direction. Preferably, the navigation path planned ensures parallel to the boundary of the grid map.

[0075] The progressive corrosion operation performed in the embodiment forms a corrosion region between the navigation path and the boundary of the grid map, wherein the boundary of the grid map is a closed boundary line, so that the navigation path corresponding to the boundary of the grid map connected in the foregoing step P3 is also a closed route, and the navigation path becomes a boundary line of the remaining closed grid map region (including the grid point filled with the preset minimum path obstacle degree evaluation value and the grid point filled with the preset maximum path obstacle degree evaluation value) after the grid map is subtracted by the corrosion region.

[0076] Another embodiment of the present application discloses a method for setting a side-around obstacle path based on a previous navigation path, wherein in step M3, when the robot collides with an obstacle during movement along the navigation path, the method for setting the side-around obstacle path with the best deviation degree relative to the navigation path comprises: when the robot collides with and detects an obstacle during movement along the navigation path, planning all side-around obstacle paths for guiding the robot to go around the obstacle, wherein each side-around obstacle path corresponds to a side-around direction; then, according to the minimum line segment distance between each grid point on each side-around obstacle path and all grid points on a reference navigation path segment supporting the robot to continue to move after going around the obstacle on the previously planned navigation path, selecting a side-around obstacle path with the minimum average deviation degree relative to the reference navigation path segment, and setting the currently selected side-around obstacle path as the side-around obstacle path with the best deviation degree relative to the navigation path. Compared with the prior art, the embodiment selects the side-around obstacle path corresponding to the best side-around direction by using the reference navigation path segment supporting the robot to continue to move after going around the obstacle, so as to reduce the deviation of the robot from the originally planned navigation path during and after going around the obstacle, and is beneficial to guiding the robot to complete the going-around behavior of the currently detected obstacle along the side-around obstacle path with navigation value, thereby improving the intelligent navigation ability and the going-around ability of the robot.

[0077] As an embodiment, as shown in Figure 3 The method for setting the side-around obstacle path specifically comprises:

[0078] Step Sa: when the robot collides with and detects an obstacle during movement along the navigation path, planning all side-around obstacle paths for guiding the robot to go around the obstacle, wherein each side-around obstacle path corresponds to a side-around direction; and then entering step Sb. Step Sa is used to simultaneously plan multiple side-around obstacle paths for guiding the robot to go around the obstacle, and each side-around obstacle path corresponds to a different side-around direction. In the embodiment of the present application, the robot simultaneously searches for a left side-around obstacle path in a left side-around direction and a right side-around obstacle path in a right side-around direction to the left and right edges of the obstacle at a position. Because the mode of the robot walking along the edge of the obstacle is divided into a left side-around mode and a right side-around mode, the robot in the left side-around mode moves clockwise along the left edge of the obstacle, and the robot in the right side-around mode moves counterclockwise along the right edge of the obstacle.

[0079] Step Sb: respectively counting the number of grid points included in each side-around obstacle path, and then entering step Sc. Each side-around obstacle path is connected by free grid points distributed along the edge of the obstacle, and some grid points may coincide but do not affect the counting result of the grid points.

[0080] Step Sc, for each edge around barrier path, respectively calculate the minimum line segment distance between each grid point on the corresponding edge around barrier path and all grid points of the reference navigation path segment, wherein the minimum line segment distance is the length of the line closest to the perpendicular line segment from a grid point on the edge around barrier path to the reference navigation path segment, thus reducing the search amount of the line between grid points, and obtaining the minimum distance information of the corresponding edge around barrier path and the reference navigation path segment, which is conducive to selecting the edge around barrier path with smaller deviation from the original navigation path; the reference navigation path segment is the reference navigation path segment on the navigation path that supports the robot to continue passing after passing around the barrier, providing the path selection direction. Then, add the minimum line segment distance corresponding to each grid point on the corresponding edge around barrier path to obtain the local distance sum in the corresponding edge direction; then enter step Sd.

[0081] The local distance sum relative to the deviation degree processing in the prior art is the total sum of the minimum deviation degree of all grid points on an edge around barrier path and the reference navigation path segment, rather than adding the distance sum of each edge behavior point and all grid points of the reference navigation path segment to obtain a distance total sum, which facilitates the embodiment of the present application to simplify the statistical method for describing the edge change trend of the edge around barrier path, so that the local distance sum described in the embodiment disclosed herein is more representative in describing the deviation degree relative to the reference navigation path segment, reduces the calculation of redundant line segment length, and in particular overcomes the redundant distance information generated by the sum of the line segment distance between a single edge behavior point and all grid points of the reference navigation path segment in the prior art, avoiding affecting the judgment of the overall deviation degree of the edge around barrier path relative to the navigation advancing direction of the reference navigation path segment by those skilled in the art.

[0082] Step Sd, average the local distance sum in the corresponding edge direction by using the number of grid points included in the edge around barrier path in step Sc to obtain the average deviation of the edge around barrier path, which is used to represent the average deviation degree of the edge around barrier path relative to the reference navigation path segment; then enter step Se. Since the local distance sum in the corresponding edge direction obtained by step Sc by summing the minimum line segment distance of each grid point on the edge around barrier path relative to the reference navigation path segment is the sum of all minimum line segment distances between all grid points on the edge around barrier path and the reference navigation path segment, step Sd divides the local distance sum by the number of grid points included in the edge around barrier path in step Sc to obtain the average result, which is the average deviation of the edge around barrier path. Compared with the prior art, the embodiment compares the average distance of the grid points between each edge around barrier path and the reference navigation path segment, which is used to measure the deviation degree of each edge around barrier path relative to the same reference navigation path segment. Compared with the prior art, the embodiment is more representative and can quickly represent the change of the edge around barrier path relative to the same reference navigation path segment by directly comparing the distance sum.

[0083] It should be noted that the embodiment calculates the minimum line segment distance between each grid point on the edge around barrier path and all grid points of the reference navigation path segment, and then adds and averages the corresponding calculated minimum line segment distances of each grid point on the edge around barrier path. Compared with the least square method, which obtains the difference between the actual value and the fitting line function by squaring the sum of the fitting line to represent the fitting effect, the embodiment simplifies the calculation difficulty between grid points and does not need to perform fitting processing on the navigation path.

[0084] Step Se, comparing the size of the average deviation amount corresponding to each edge around the barrier path, selecting the average deviation amount of the edge around the barrier path as the path of the robot to walk along the edge of the barrier in the future, as the best edge around the barrier path. Wherein, the smaller the average deviation amount, the higher the matching degree of the edge around the barrier path in the corresponding edge direction and the reference navigation path segment, including the matching degree of the real-time navigation direction and the overall extension trend, and the edge direction required for the robot to overcome the obstacle from the current position; the larger the average deviation amount, the lower the matching degree of the edge around the barrier path in the corresponding edge direction and the reference navigation path segment, including the matching degree of the real-time navigation direction and the overall extension trend. Thus, according to the minimum line segment distance between each grid point on each edge around the barrier path and all grid points of the reference navigation path segment supporting the robot to continue to pass after the barrier, the average deviation degree of the edge around the barrier path relative to the reference navigation path segment is selected, so that the robot walks along the selected edge around the barrier path after touching the barrier. The foregoing step selects the edge around the barrier path corresponding to the best edge direction by using the reference navigation path segment supporting the robot to continue to pass after the barrier, so as to reduce the deviation of the robot from the originally planned navigation path during and after the barrier, which is beneficial to guide the robot to complete the barrier behavior of the currently detected barrier along the edge around the barrier path with navigation value, and improve the intelligent navigation ability and barrier passing ability of the robot.

[0085] It should be noted that in the present embodiment, the path of the robot to walk along the edge of the barrier in the future is to judge and select the edge prediction path with the smallest difference from the pre-planned navigation path, and then control the robot to walk along the selected edge prediction path after colliding with the barrier, but the actual edge path of the robot may not be executed according to the prediction planning, because the prediction planning path has certain error, and can only represent the approximate path and the clear edge direction, such as counterclockwise or clockwise movement along the edge of the barrier, but it can be determined that the similar selection method of the prior art can select a more effective edge path more quickly, so as to realize the robot to quickly cross the real-time detected barrier, although the barrier has significantly blocked the navigation extension direction of the navigation path.

[0086] Specifically, the edge around the barrier path is a left edge around the barrier path for guiding the robot to walk along the edge of the barrier in a clockwise direction, or the edge around the barrier path is a right edge around the barrier path for guiding the robot to walk along the edge of the barrier in a counterclockwise direction. Thus, the edge path required for the robot to walk along the contour edge of the barrier in the map projection is planned. Adapt to the robot to detect and walk along the edge between different barriers.

[0087] Preferably, when the average deviation of the left edge around barrier path is greater than the average deviation of the right edge around barrier path, the right edge around barrier path is configured as the path along which the robot actually walks along the right edge of the barrier in the future; when the average deviation of the left edge around barrier path is less than the average deviation of the right edge around barrier path, the left edge around barrier path is configured as the path along which the robot actually walks along the left edge of the barrier in the future; when the average deviation of the left edge around barrier path is equal to the average deviation of the right edge around barrier path, one of the left edge around barrier path and the right edge around barrier path is randomly selected as the path along which the robot actually walks along the edge of the barrier in the future. The efficiency of the robot in processing the edge around barrier path is improved.

[0088] It is worth noting that the obstacle currently detected by the sensor of the robot is allowed to change, including changes in coverage area, size and shape, so that the obstacle blocks the movement of the robot along the navigation path; wherein the pre-planned navigation path is only suitable for bypassing fixed obstacles. In the face of constantly changing obstacles, the embodiment plans an effective edge around barrier path by implementing the edge around barrier path planning method or multiple times implementing the edge around barrier path planning method, thereby improving the success rate of local obstacle bypass navigation of the robot; and enhancing the algorithm robustness of the edge around barrier path planning method.

[0089] As can be understood by those skilled in the art, the grid map constructed in real time is marked with environmental information around the current position of the robot, but due to the existence of map drift error, the grid information of the obstacle in the map may not be consistent with the actual environmental obstacle position information. In order to plan a more reasonable edge around barrier path, the invention specially uses the neighborhood grid point search method to plan an edge around barrier path that can maintain proximity to the edge of the obstacle.

[0090] Based on the foregoing embodiment, a specific planning method for the edge around barrier path for guiding the robot to bypass the obstacle is also disclosed, and specifically includes:

[0091] Step S1, when the robot touches and detects an obstacle during movement along the navigation path, the set of obstacle grid points covered by the obstacle in the two-dimensional grid map is obtained within the detection range of the sensors of the robot, and then step S2 is entered. As for the obstacle used in this embodiment, it is detected and obtained by the sensors of the robot (including but not limited to collision sensors, 3dtof, laser radar), and the point cloud data of the obstacle will be obtained. The projection of the real-time obtained point cloud data on the two-dimensional grid map covers the grid markers, which become the obstacle grid points covered by the obstacle in the two-dimensional grid map and form the set of obstacle grid points. Then, the coordinate values of the obstacle grid points corresponding to the markers in the grid map previously constructed by the robot are obtained by combining the coordinates of the current position of the robot, the point cloud data of the obstacle and the configuration parameters of the sensor, and the specific conversion method is well known to those skilled in the art and will not be described here. Among them, when the robot touches and detects an obstacle during movement along the navigation path, the current moving direction of the robot is saved.

[0092] Step S2, from the set of obstacle grid points based on the obstacle obtained in step S1, based on the coordinate values of the obstacle grid points, the obstacle grid point closest to the current position of the robot is selected, and the obstacle grid point closest to the current position of the robot is set as the search center for neighborhood search to start neighborhood search in the neighborhood with the obstacle grid point closest to the current position of the robot as the search center, and then step S3 is entered. Among them, the neighborhood search here includes searching in the adjacent grid positions of the obstacle grid point closest to the current position of the robot, left, left up, down, left down, right, right up, down, right down, the two adjacent grid points are connected and are adjacent to each other, in general, the adjacent grid points above, below, left and right of a search center are the closest adjacent grid points. It should be noted that the current position of the robot belongs to the idle grid point.

[0093] Step S3, determine whether the search center in the current moving direction corresponding to the searched neighborhood grid point is an idle grid point, if yes, enter step S5, otherwise enter step S4. Step S3 determines that the obstacle grid point is the search center and then searches the neighborhood grid point along the current moving direction to obtain the grid marked as idle state, so as to search the edge behavior point of the robot approaching the obstacle contour line as much as possible based on the edge of the obstacle.

[0094] Step S4, if all the neighbor grid points searched in the current moving direction of the robot are determined as obstacle grid points, the nearest neighbor grid point to the search center in the current moving direction of the robot is updated as the search center for the next iteration of the neighbor search, and then the process returns to step S3. When step S3 fails to search for a free grid point in the current moving direction of the robot, the nearest obstacle grid point to the search center in the current moving direction of the robot is updated as the search center for the next iteration of the neighbor search, and the iteration of steps S3 to S4 is implemented to search for a free grid point near the edge of the obstacle to be circumvented along the edge, but not limited to a specific edge-following direction. It is worth noting that in the process of repeating step S3 to search for a grid point, the grid point searched in the current iteration of step S3 is different from the grid point searched in the previous iteration of step S3, so as to avoid the robot repeatedly walking along the edge-following path that has been planned, and to improve the efficiency of the robot circumventing the obstacle along the edge.

[0095] Step S5, the searched free grid point is configured as an edge-following search center, and the search center is configured as an edge-following search starting point, i.e., a pair of adjacent free grid points configured as edge-following search centers and a pair of adjacent obstacle grid points configured as edge-following search starting points are found, wherein the edge-following search center configured in step S5 and the edge-following search starting point configured in step S5 form a pair of adjacent grid points that are fixed and unchanged in the planning method of the edge-following circumvention path, and are used as a starting point for searching for an edge-following behavior point and laying the foundation for edge-following guidance. Then step S6 is entered.

[0096] Compared with the prior art, the embodiment is based on a set of obstacle grid points detected in real time, and a neighbor search is performed from the obstacle grid point closest to the current position of the robot as a first search center to maximize the proximity to the contour line of the obstacle. Then, by iteratively performing steps S3 to S4, the search center is constantly updated along the current moving direction of the robot until a pair of adjacent free grid points configured as edge-following search centers and a pair of adjacent obstacle grid points configured as edge-following search starting points are searched out by this neighbor search method, which provides a guidance condition for searching for grid points for accessing the edge-following circumvention path. Preferably, the obstacle grid points mentioned in steps S1 to S5 belong to the obstacle detected in step S1.

[0097] Step S6, from the newly configured edge search starting point, search the neighborhood of the newly configured edge search center along the preset edge direction to search for the grid points one by one in order to search for the grid points suitable for edge following according to the contour distribution characteristics of the obstacle; and then enter step S7. When the preset edge direction is the clockwise direction, step S6 is used to search for the candidate edge following behavior points close to the left side edge of the obstacle, which are subsequently connected to become the edge following obstacle-avoiding path in the left edge direction; when the preset edge direction is the counterclockwise direction, step S6 is used to search for the candidate edge following behavior points close to the right side edge of the obstacle, which are subsequently connected to become the edge following obstacle-avoiding path in the right edge direction.

[0098] Step S7, in the neighborhood of the edge search center, from the edge search starting point, determine whether there is a first searched idle grid point among the grid points searched one by one along the preset edge direction, if yes, enter step S9, otherwise enter step S8. It should be noted that if there is no first searched idle grid point among the grid points searched one by one along the preset edge direction, then in the neighborhood of the edge search center, there is no idle grid point except the edge search starting point; that is, in the nine-grid neighborhood centered on the edge search center, there is no idle grid point except the edge search starting point. It should be noted that the edge search starting point is not included in the range of the grid points searched along the preset edge direction.

[0099] Step S8, in the neighborhood of the edge search center, if no idle grid point is searched along the preset edge direction from the edge search starting point, stop the search operation of the previous steps, determine that an edge following obstacle-avoiding path corresponding to the preset edge direction has been searched, and then enter step S12. Thus, under the premise of not having the robot passing conditions, the invalid edge search is reduced as much as possible.

[0100] Preferably, in step S8, if no idle grid point is searched along the preset edge direction from the edge search starting point, the edge search center in step S8 can also be connected to become one grid point of the edge following obstacle-avoiding path in the corresponding edge direction. Then enter step S12. As many idle grid points as possible are provided for the edge following obstacle-avoiding path, and the environmental adaptability of the edge following obstacle-avoiding path is improved.

[0101] Step S9, the first idle grid point searched in step S7 is determined as the candidate edge-following behavior point, and then step S10 is entered. There can be multiple idle grid points in the neighborhood of the edge-following search center, but the first idle grid point searched in step S7 from the given edge-following search starting point along the preset edge-following direction is the idle grid point closest to the edge of the obstacle, and is also the idle grid point closest to the edge-following search starting point configured in step S6, so as to control the edge-following obstacle-avoiding path connected to be close to the edge of the obstacle.

[0102] Step S10, the candidate edge-following behavior point is updated as the edge-following search center, i.e., the candidate edge-following behavior point searched currently is updated as the edge-following search center configured next time; meanwhile, the edge-following search center before the update is updated as the edge-following search starting point, i.e., the edge-following search center configured currently is updated as the edge-following search starting point configured next time; and the edge-following search center before the update is connected to a grid point in the edge-following obstacle-avoiding path in the corresponding edge-following direction, so as to guide the robot to walk along the edge of the obstacle, i.e., the edge-following search center configured in step S6 (including the idle grid point searched in step S3) is connected to a grid point in the edge-following obstacle-avoiding path in the corresponding edge-following direction; and then step S11 is entered.

[0103] Step S11, it is judged whether the edge-following obstacle-avoiding path connected out newly satisfies the preset edge-following obstacle-avoiding condition, if yes, step S12 is entered, otherwise, step S6 is returned. Step S11 is equivalent to judging whether the candidate edge-following behavior point searched newly satisfies the preset edge-following obstacle-avoiding condition, and is also equivalent to judging whether the edge-following search center existing in the neighborhood for the idle grid point searched along the preset edge-following method satisfies the preset edge-following obstacle-avoiding condition. If not, step S6 to step S11 are repeatedly executed until the edge-following obstacle-avoiding path connected out newly satisfies the preset edge-following obstacle-avoiding condition.

[0104] Step S12, the planning of the edge-following path along the obstacle is completed. Step S12 is specifically the planning of the edge-following path corresponding to the preset edge-following direction. Preferably, for each preset edge-following direction, steps S6 to S12 are simultaneously performed, specifically including: starting from the newly configured same edge-following search starting point, searching each adjacent grid point in the neighborhood of the newly configured same edge-following search center in the clockwise direction to start performing steps S6 to S12, thereby connecting the edge-following search center before the update in sequence into an edge-following path in the left edge-following direction (the edge-following path corresponding to the left edge of the obstacle) by repeatedly performing steps S6 to S11; simultaneously starting from the newly configured same edge-following search starting point, searching each adjacent grid point in the neighborhood of the newly configured same edge-following search center in the counterclockwise direction to perform steps S6 to S12 synchronously, thereby connecting the edge-following search center before the update in sequence into an edge-following path in the right edge-following direction (the edge-following path corresponding to the right edge of the obstacle) by repeatedly performing steps S6 to S11; and finally completing the planning of the edge-following path corresponding to the clockwise direction and the edge-following path corresponding to the counterclockwise direction.

[0105] The foregoing steps perform neighborhood search around the edge-following search center along the preset edge-following direction in a nine-square grid region centered on the edge-following search center, and determine a candidate edge-following behavior point as a candidate edge-following grid point to wait to join and plan an edge-following path around the obstacle when the first free grid point is searched, thereby achieving the process of connecting the edge-following search center with a neighborhood existing a free grid point as an effective edge-following behavior point in sequence into the edge-following path along the same preset edge-following direction to search the constantly updated neighborhood of the edge-following search center, thereby accelerating the planning speed of the edge-following path around the obstacle.

[0106] As an embodiment, the preset edge-around-obstacle condition includes: the edge-searching center newly connected into the edge-around-obstacle path coincides with other grid points on the edge-around-obstacle path to which the edge-searching center belongs, so that a closed planar geometric figure is formed on the edge-around-obstacle path in the corresponding edge direction. Thus, it is determined that the robot has searched and planned an edge-around-obstacle path supported by the obstacle information, the search for candidate edge behavior points is stopped, and the robot is prevented from repeatedly searching the same edge-around-obstacle path, thereby reducing the search amount. Preferably, when the edge-around-obstacle path newly connected satisfies the preset edge-around-obstacle condition, the start point of the edge-around-obstacle path is the same as the edge-searching center newly connected into the edge-around-obstacle path, so that the start point and the end point of the edge-around-obstacle path in the corresponding edge direction coincide, forming a closed planar geometric figure. On this basis, the edge-around-obstacle path in the left edge direction and the edge-around-obstacle path in the right edge direction are both the same closed planar geometric figure, and the closed edge-around-obstacle path can be divided into a left half edge-around-obstacle path and a right half edge-around-obstacle path along the left edge and the right edge of the obstacle based on the number of grid points and the coordinate position characteristics of the grid points, wherein the number of grid points occupied by the left half edge-around-obstacle path and the number of grid points occupied by the right half edge-around-obstacle path are equal.

[0107] As an embodiment, when it is determined in step S7 that no idle grid point is searched in the neighborhood of the newly configured edge-searching center along the preset edge direction starting from the newly configured edge-searching start point, the newly configured edge-searching center is not added to the corresponding edge-around-obstacle path in the preset edge direction. Then, when step S7 is performed for the first time, the grid points are searched one by one in the neighborhood of the edge-searching center along the preset edge direction starting from the edge-searching start point, and the first idle grid point is searched, and then the newly configured edge-searching center is added to the corresponding edge-around-obstacle path in the preset edge direction. When step S7 is performed for the second time, the grid points are searched one by one in the neighborhood of the updated edge-searching center along the preset edge direction starting from the updated edge-searching start point, and no idle grid point is searched, so that the newly configured edge-searching center is not added to the corresponding edge-around-obstacle path in the preset edge direction, and the search is ended, and it is determined that the corresponding edge-around-obstacle path in the preset edge direction composed of one grid point is planned.

[0108] As another implementation, step S7 judges that when no free grid point is found in the neighborhood of the newly configured edge-search center along the preset edge-search direction from the newly configured edge-search starting point, the newly configured edge-search center is added to the corresponding edge-obstacle-avoiding path in the preset edge-search direction. Then, when step S7 is performed for the first time, from the edge-search starting point, the grid points are searched one by one in the neighborhood of the edge-search center along the preset edge-search direction, and the first free grid point is found, and then the newly configured edge-search center is added to the corresponding edge-obstacle-avoiding path in the preset edge-search direction. When step S7 is performed for the second time, from the updated edge-search starting point, the grid points are searched one by one in the neighborhood of the updated edge-search center along the preset edge-search direction, and no free grid point is found, and then the newly configured edge-search center is also added to the corresponding edge-obstacle-avoiding path in the preset edge-search direction, and the search is ended, and it is determined that the planning of the corresponding edge-obstacle-avoiding path in the preset edge-search direction composed of two grid points is completed.

[0109] It should be noted that the judgment result involved in step S8 is that when no free grid point is found in the neighborhood of the newly configured edge-search center along the preset edge-search direction from the newly configured edge-search starting point, the newly configured edge-search center is the end position in the edge-obstacle-avoiding path that does not have the condition of continuing to pass through.

[0110] As an embodiment, the method for planning the edge-obstacle-avoiding path further comprises: selecting a reference navigation path segment supporting the robot to continue to pass through after obstacle avoidance along a navigation extension direction of the navigation path from the navigation path not occupied by the obstacle, located behind the obstacle and adjacent to the obstacle grid point, wherein the reference navigation path segment is connected by the grid points arranged continuously in the navigation path; the length of the trajectory of the reference navigation path segment changes with the type of the sensor of the robot, the length of the trajectory of the reference navigation path segment is set to have the function of guiding the robot to continue to pass through after obstacle avoidance, and in order to ensure the comparison effect between paths or the matching degree of the paths, the grid area close to the obstacle is selected as much as possible to generate the guiding value. Therefore, the embodiment considers the length characteristics of the navigation path after the robot avoids the obstacle, selects the reference navigation path segment supporting the robot to pass through as the reference navigation path, judges the navigation rationality of the currently planned edge-obstacle-avoiding path relative to the navigation path, and can filter out the difference degree between paths based on the length characteristics of the navigation path and facilitate the robot to return to the edge-obstacle-avoiding path. It should be noted that the selection method of the reference navigation path segment mentioned in the embodiment can be performed between step S5 and step S6, or after step S9.

[0111] It should be noted that if the navigation path passes through the obstacle, the navigation path is divided into a path segment reaching the front of the obstacle, a path segment covered by the obstacle and a path segment behind the obstacle along the navigation extension direction of the navigation path. Each path segment is connected by grid points. The reference navigation path segment is selected from the path segment behind the obstacle, and the grid points included in the reference navigation path segment serve as reference grid points with navigation significance and play a guiding role. The navigation path is classified based on the relative position relationship of the obstacle on the navigation path to obtain a navigation path segment supporting the robot to pass after obstacle avoidance.

[0112] Therefore, the embodiment also sets the preset edge-avoiding condition as that a line segment between the newly searched candidate edge behavior point and any grid point of the reference navigation path segment does not pass through an obstacle grid point. When the newly searched candidate edge behavior point can directly reach all the grid points on the reference navigation path segment, the edge-avoiding path along the preset edge direction is stopped from being continuously planned, so as to avoid the edge-avoiding path along the preset edge direction from being farther and farther, thereby reserving space for the robot to return to the navigation path successfully after walking along the edge-avoiding path.

[0113] As an embodiment, the preset edge-avoiding condition includes that the trajectory length of the edge-avoiding path is greater than the edge-avoided contour length of the obstacle in the corresponding edge direction. The edge-avoided contour length of the obstacle in the two-dimensional grid map is the edge-avoided contour length of the obstacle in the two-dimensional grid map, is greater than the detectable distance of the sensor of the robot, and changes with the change of the actual edge direction of the robot. Preferably, the edge-avoided contour length of the obstacle in the two-dimensional grid map is equal to the trajectory length of the reference navigation path segment, so as to create sufficient and reasonable matching grid points to facilitate the calculation of the average deviation of the edge-avoiding path in the corresponding edge direction from the reference navigation path segment according to the steps of the foregoing embodiment.

[0114] In an actual experimental scene, the contour length of the obstacle detected by the sensor of the robot is less than the contour length of the obstacle projected in the two-dimensional grid map. The embodiment limits the trajectory length of the edge-avoiding path within a certain range, so as to plan an edge-avoiding path with a reasonable length, reduce the search calculation amount, and improve the efficiency of the robot in planning the edge-avoiding path.

[0115] The chip for processing the pre-created grid map, specifically, the navigation path parallel to the map boundary is obtained through the method of progressive erosion operation and neighborhood expansion, and when the robot moves along the currently planned navigation path, if an obstacle is encountered, then on the local map detected by the sensor, in combination with the pre-planned navigation path, the edge obstacle-avoiding path with the best fitting degree relative to the navigation path is selected from the average deviation degree, the success rate of the robot searching for the charging base signal is improved, and the robot is prevented from deviating from the navigation path; the chip selects the search starting point position and the preset search direction on the navigation path, reduces the calculation amount, and thus speeds up the process of the robot returning to the seat. It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered a list of executable instructions for implementing logic functions, and can be specifically implemented in any computer-readable medium for use by or in connection with an instruction execution system, device or apparatus, such as a computer-based system, a system including a processor, or other system that can fetch the instructions from the instruction execution system, device or apparatus and execute the instructions, or in conjunction with these instruction execution systems, devices or apparatus. For the purpose of this specification, the "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport programs for use by or in connection with an instruction execution system, device or apparatus, or in conjunction with these instruction execution systems, devices or apparatus. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connections having one or more wires (electronic devices), portable computer diskettes (magnetic devices), random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memories), fiber optic devices, and portable compact disc read-only memories (CDROMs). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by electronic editing, interpretation or processing, or if necessary, in other suitable ways, to be electronically obtained and then stored in the computer memory.

[0116] The robot is provided with the chip, and the robot is configured to execute the motion control method. In the constructed grid map, on the basis of acquiring a navigation path parallel to the boundary of the grid map, the robot moves quickly along the navigation path, selects an edge-following path corresponding to the edge-following direction with the highest fitting degree with the reference navigation path segment from the planned left and right edge-following paths, controls the robot to execute quick obstacle crossing along the matched edge-following path, and facilitates the robot to return to the navigation path after successful obstacle crossing, until the robot detects a charging base signal in the movement process, so that the effect of accelerating the return-to-base process is achieved.

[0117] Preferably, the left and right sides of the body of the robot are provided with collision detectors, and the body of the robot is further provided with infrared sensors, visual sensors and / or laser sensors, which are used to detect the coverage area of the obstacle when the collision detector detects that one side of the advancing direction of the robot collides with the obstacle, and then search for edge-following behavior points (grid points that can be passed along the edge of the obstacle) to connect the edge-following paths in each edge-following direction, and select an edge-following path with the smallest average deviation from the reference navigation path segment, so that the robot walks along the selected edge-following path after touching the obstacle.

[0118] Specifically, the left front part of the body of the robot is provided with a left collision detector for detecting the obstacle colliding with the left side of the advancing direction of the robot, and the detection result is used to assist the robot to walk along the right edge of the obstacle in a counterclockwise direction; the right front part of the body of the robot is provided with a right collision detector for detecting the obstacle colliding with the right side of the advancing direction of the robot, and the detection result is used to assist the robot to walk along the left edge of the obstacle in a clockwise direction; and the chip is connected with the left collision detector and the right collision detector, respectively.

[0119] Preferably, the left collision detector and the right collision detector are symmetrically arranged on the left and right sides of the body of the robot.

[0120] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] The above merely describes preferred embodiments of the present application, but is not intended to limit the present application in other forms. Any person skilled in the art can make changes or modifications to the above disclosed technical contents into equivalent embodiments with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution of the present application and according to the technical essence of the present application still belongs to the protection scope of the technical solution of the present application.

Claims

1. A motion control method for searching for a charging station signal, characterized by, Before the motion control method is executed, the robot does not detect the charging base signal; the motion control method comprises: planning a navigation path based on the boundary of the grid map previously constructed by the robot, so that the navigation path is parallel to the boundary of the grid map; controlling the robot to move to a search starting point position of the navigation path, and then controlling the robot to move along the navigation path in a preset search direction from the search starting point position until the charging base signal is detected; when the robot touches an obstacle during movement along the navigation path, setting an edge-avoiding obstacle-circumventing path with an optimal deviation degree relative to the navigation path, and then controlling the robot to walk along the currently set edge-avoiding obstacle-circumventing path until the charging base signal is detected; after the robot circumvents the currently touched obstacle along the currently set edge-avoiding obstacle-circumventing path, controlling the robot to continue moving along the navigation path in the preset search direction until the charging base signal is detected; the method for planning a navigation path based on the boundary of the grid map previously constructed by the robot comprises: Step 1: constructing a first circular domain with each map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then performing a gradual erosion operation on each first circular domain to obtain a path obstacle degree evaluation value of the grid points in each first circular domain; wherein the map boundary grid point is present on the boundary of the grid map previously constructed by the robot; wherein the preset multiple is related to the positioning accuracy of the grid map; Step 2: traversing every other preset interval to a map boundary grid point, and constructing a second circular domain with the currently traversed map boundary grid point as the center and a preset multiple of the robot body diameter as the radius, and then searching for a grid point satisfying a critical condition in combination with the path obstacle degree evaluation value obtained in Step 1; wherein all the grid points in the second circular domain have been assigned the path obstacle degree evaluation value in Step 1; Step 3: sequentially connecting the grid points satisfying the critical condition searched in Step 2 as a navigation path, wherein the navigation path is parallel to the boundary of the grid map; the gradual erosion operation in Step 1 comprises: calculating the straight-line distance between the currently eroded grid point and the center of the first circular domain to which the grid point belongs, denoted as the boundary search distance; wherein the currently eroded grid point is a grid point covered by a traversal unit starting from the center of the first circular domain; the traversal unit is equivalent to the structural element of the erosion operation; setting the robot body diameter of the preset multiple as the erosion radius; then, setting the path obstacle degree evaluation value of the currently eroded grid point in a negative correlation with the boundary search distance in the first circular domain; If the path obstacle degree evaluation value of the current corrosion grid point set currently is greater than the path obstacle degree evaluation value of the same grid point set previously, the path obstacle degree evaluation value of the current corrosion grid point set currently is maintained, and the path obstacle degree evaluation value of the current corrosion grid point set currently is set as the latest path obstacle degree evaluation value of the current corrosion grid point; otherwise, the path obstacle degree evaluation value of the current corrosion grid point set previously is updated to the path obstacle degree evaluation value of the current corrosion grid point set currently, and the path obstacle degree evaluation value of the current corrosion grid point set currently is set as the latest path obstacle degree evaluation value of the current corrosion grid point.

2. The motion control method of claim 1, wherein, The method for setting the search starting position comprises: searching for a grid point on the navigation path, which is not connected to the current position of the robot through an unknown grid point or an obstacle grid point, and configuring the currently searched grid point as a candidate target position; then selecting one of the candidate target positions closest to the current position of the robot as the search starting position.

3. The motion control method of claim 1, wherein, The method for setting the preset search direction comprises: when there is no grid coordinate point of the charging base on the navigation path, the counterclockwise direction is set as the preset search direction; when there is a grid coordinate point of the charging base on the navigation path, the extension direction of the shortest connected path segment between the search starting position and the grid coordinate point of the charging base is set as the preset search direction; wherein the connected path segment is a continuous path segment extending from the search starting position to the grid coordinate point of the charging base along the navigation path.

4. The motion control method of claim 1, wherein, Before performing the gradual corrosion operation in step 1, the initial value of the path obstacle degree evaluation value of the map boundary grid point is set as a preset maximum path obstacle degree evaluation value, the initial value of the path obstacle degree evaluation value of the unknown grid point in the grid map is set as a preset maximum path obstacle degree evaluation value, and the initial value of the path obstacle degree evaluation value of the remaining types of grid points in the grid map is set as a preset minimum path obstacle degree evaluation value.

5. The motion control method of claim 4, wherein, The step 2 specifically comprises: setting a preset multiple of the robot body diameter as the preset interval, traversing the map boundary grid points along the boundary of the robot pre-constructed grid map according to the preset interval; constructing a second circular domain with the currently traversed map boundary grid point as the center and the preset multiple of the robot body diameter as the radius, and performing neighborhood expansion in the second circular domain with the currently traversed map boundary grid point as the search center; If the latest path obstacle degree evaluation value of the grid point currently expanded to in the second circular domain is greater than the preset minimum path obstacle degree evaluation value, and there is at least one grid point in the eight-neighbor domain of the grid point currently expanded to whose latest path obstacle degree evaluation value is the preset minimum path obstacle degree evaluation value, then the grid point with the minimum path obstacle degree evaluation value among the grid points currently expanded to is set as the grid point satisfying the critical condition searched out.

6. The motion control method of claim 5, wherein, In step 3, one grid point satisfying the critical condition searched out in the second circular domain corresponding to each map boundary grid point traversed according to the preset interval in step 2 is sequentially connected as the navigation path, so that the navigation path is parallel to the boundary of the grid map; Wherein, there is a grid point with the latest path obstacle degree evaluation value being the preset minimum path obstacle degree evaluation value on one side of the navigation path, and there is a grid point with the latest path obstacle degree evaluation value not being the preset minimum path obstacle degree evaluation value on the other side of the navigation path; Wherein, each grid point satisfying the critical condition corresponds to a map boundary grid point traversed at a specific time, so that the acquisition order of the grid point satisfying the critical condition is the same as the traversal order of the map boundary grid point.

7. The motion control method according to any one of claims 1 to 3, characterized by, The method for setting the edge-obstacle-avoiding path with the best deviation degree relative to the navigation path when the robot collides with an obstacle during movement along the navigation path comprises: When the robot collides with and detects an obstacle during movement along the navigation path, all edge-obstacle-avoiding paths for guiding the robot to avoid the obstacle are planned, wherein each edge-obstacle-avoiding path corresponds to an edge direction; Then, according to the minimum line segment distance between each grid point on each edge-obstacle-avoiding path and all grid points of a reference navigation path segment on the pre-planned navigation path supporting the robot to continue to pass after avoiding the obstacle, an edge-obstacle-avoiding path with the minimum average deviation degree relative to the reference navigation path segment is selected, and the currently selected edge-obstacle-avoiding path is set as the edge-obstacle-avoiding path with the best deviation degree relative to the navigation path.

8. The motion control method of claim 7, wherein, The method for selecting an edge-obstacle-avoiding path with the minimum average deviation degree relative to a reference navigation path segment according to the minimum line segment distance between each grid point on each edge-obstacle-avoiding path and all grid points of the reference navigation path segment on the pre-planned navigation path supporting the robot to continue to pass after avoiding the obstacle comprises: The number of grid points included in each edge-obstacle-avoiding path is counted respectively; The number of grid points included in each edge-obstacle-avoiding path is counted respectively; For each edge-keeping path, the minimum line segment distances between each grid point on the edge-keeping path and all grid points of the reference navigation path segment are calculated respectively, and then the local distance sums in the corresponding edge-keeping directions are obtained by adding the minimum line segment distances corresponding to each grid point on the edge-keeping path, and then the average deviation of the edge-keeping path is obtained by averaging the local distance sums in the corresponding edge-keeping directions using the number of grid points included in the edge-keeping path, which is used to represent the average deviation degree of the edge-keeping path relative to the reference navigation path segment; By comparing the average deviations corresponding to each edge-keeping path, the edge-keeping path with the minimum average deviation is selected as the path for the robot to actually walk along the edge of the obstacle in the future.

9. The motion control method of claim 8, wherein, The edge-keeping path is a left edge-keeping path for guiding the robot to walk along the left edge of the obstacle in a clockwise direction, or the edge-keeping path is a right edge-keeping path for guiding the robot to walk along the right edge of the obstacle in a counterclockwise direction.

10. The motion control method of claim 8, wherein, In the process of comparing the average deviations corresponding to each edge-keeping path, if there are two or more edge-keeping paths with equal average deviations, a random edge-keeping path is selected as the path for the robot to actually walk along the edge of the obstacle in the future.

11. The motion control method of claim 7, wherein, The specific planning method of the edge-keeping path for guiding the robot to keep away from the obstacle includes: Step 1: Within the detection range of the sensor of the robot, a pair of adjacent free grid points and obstacle grid points are searched in the current moving direction of the robot by neighborhood search, and then the free grid point is configured as the edge-keeping search center and the obstacle grid point is configured as the edge-keeping search starting point; Step 2: When a candidate edge-keeping behavior point is searched in the neighborhood of the edge-keeping search center along the preset edge-keeping direction from the edge-keeping search starting point, the candidate edge-keeping behavior point is updated as the edge-keeping search center, the edge-keeping search center before the update is updated as the edge-keeping search starting point, and the edge-keeping search center before the update is connected to the edge-keeping path in the corresponding edge-keeping direction to guide the robot to walk along the edge of the obstacle; Step 3: Step 2 is repeatedly executed until the newly connected edge-keeping path meets the preset edge-keeping obstacle condition.

12. The motion control method of claim 11, wherein, In step 1, the method of searching a pair of adjacent free grid points and obstacle grid points by neighborhood search specifically includes: Step 11: The set of obstacle grid points covered by the obstacle in the two-dimensional grid map is obtained, and the obstacle grid point closest to the current position of the robot is selected from the set, and the obstacle grid point closest to the current position of the robot is set as the search center for neighborhood search; Step 12: It is judged whether the neighborhood grid points corresponding to the search center searched in the current moving direction are free grid points, if yes, step 13 is entered, otherwise step 14 is entered; Step 13: The obstacle grid point closest to the current position of the robot is set as the edge-keeping search starting point, and step 2 is entered; Step 14: The obstacle grid point adjacent to the obstacle grid point closest to the current position of the robot is set as the edge-keeping search starting point, and step 2 is entered. Step 13, configure the searched idle grid point as the edge-searching center, and configure the searched center as the edge-searching starting point, wherein the newly configured edge-searching center and the newly configured edge-searching starting point are a pair of adjacent grid points; Step 14, if all the neighborhood grid points searched in step 12 in the current moving direction are obstacle grid points, update the nearest neighborhood grid point to the searched center in step 12 to the searched center in step 12, and then return to step 12, wherein the grid points searched in step 12 are not the grid points of repeated search.

13. The motion control method of claim 11, wherein, In the step 2, the method for searching the candidate edge-behavior grid point in the neighborhood of the edge-searching center along the preset edge direction from the edge-searching starting point comprises: In the neighborhood of the edge-searching center, from the edge-searching starting point, determine whether there is a first searched idle grid point in the grid points searched along the preset edge direction, and if yes, determine the first searched idle grid point as the candidate edge-behavior grid point; wherein the edge-searching starting point is not included in the range of the grid points searched along the preset edge direction.

14. The motion control method of claim 11, wherein, The preset edge-obstacle-avoiding condition comprises: The edge-searching center newly connected to the edge-obstacle-avoiding path coincides with other grid points on the edge-obstacle-avoiding path, so that a closed planar geometric figure is formed on the edge-obstacle-avoiding path in the corresponding edge direction.

15. The motion control method of claim 11, wherein, The motion control method further comprises: selecting a reference navigation path segment supporting the robot to continue passing after avoiding the obstacle from the navigation path which is not occupied by the obstacle, located behind the obstacle and adjacent to the obstacle grid point, along the navigation extension direction of the navigation path, wherein the reference navigation path segment is connected by the grid points arranged continuously in the navigation path.

16. The motion control method of claim 15, wherein, If the navigation path passes through the obstacle, the navigation path is divided into a path segment reaching the front of the obstacle, a path segment covered by the obstacle and a path segment behind the obstacle along the navigation extension direction of the navigation path; wherein each path segment is connected by grid points, and the reference navigation path segment is selected from the path segment behind the obstacle.

17. The motion control method of claim 16, wherein, The preset edge-obstacle-avoiding condition comprises: The line segment between the newly searched candidate edge-behavior grid point and any grid point of the reference navigation path segment does not pass through the obstacle grid point.

18. The motion control method of claim 11, wherein, The preset edge-obstacle-avoiding condition comprises: The trajectory length of the edge-obstacle-avoiding path is greater than the edge-obstacle profile length of the obstacle in the corresponding edge direction.

19. A chip storing a computer program, characterized in that, The computer program is used to control the mobile robot to execute the motion control method of any one of claims 1 to 18.

20. A robot, characterized in that The robot is provided with the chip of claim 19, and the robot is configured to execute the motion control method of any one of claims 1 to 18.

Images