CN-121383985-B - Robot map exploration method and chip based on reachable boundary points

CN121383985BCN 121383985 BCN121383985 BCN 121383985BCN-121383985-B

Abstract

The application discloses a robot map exploration method and a robot map exploration chip based on reachable boundary points, wherein the method comprises the steps of A, fitting a preset boundary line by a robot by utilizing boundary points in a current detection area of the robot, selecting contour lines used for delineating all communicated detectable areas in a map to be optimized as target contour lines, B, respectively controlling two ends of the preset boundary line to expand to obtain preset boundary expansion line segments, and executing a step C if the two ends of the preset boundary expansion line segments are intersected with the target contour lines under the condition that the expansion lengths of the two ends of the preset boundary line segments are smaller than or equal to a preset perception radius, wherein the step C is based on the fact that the robot of the preset boundary expansion line segments cuts out the contour lines to be tracked from the target contour lines along the direction communicated with an unknown area, and then searching front edge boundary points from the area surrounded by the contour lines to be tracked and the preset boundary expansion line segments, and D, screening the reachable boundary points according to the trafficability of the front edge boundary points searched out in the step C.

Inventors

CHEN JINJIE
HUANG HUIBAO
ZHOU HEWEN
ZHANG ZIQIAN
Ma Danyun

Assignees

珠海一微科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20240715

Claims (13)

1. The robot map exploration method based on the reachable boundary points comprises the steps that a robot obtains laser point clouds through laser sensor scanning, and then builds a map to be optimized by utilizing the laser point clouds, and is characterized by further comprising the following steps: step A, the robot fits a preset boundary line by utilizing boundary points in the current detection area of the robot, and selects contour lines used for delineating all communicated detectable areas in a map to be optimized as target contour lines, and then executes the step B; B, respectively controlling two ends of a preset boundary line to expand to obtain a preset boundary expansion line segment, and executing the step C if the two ends of the preset boundary expansion line segment are intersected with the target contour line under the condition that the expansion lengths of the two ends of the preset boundary line are smaller than or equal to the preset perception radius; Step C, based on a preset demarcation extension line segment, the robot segments a contour line to be tracked from the target contour line along the direction communicated with the unknown area, and then the robot searches a front edge boundary point from the area surrounded by the contour line to be tracked and the preset demarcation extension line segment; Step D, screening out reachable boundary points according to the trafficability of the front boundary points searched out in the step C, and orderly storing the reachable boundary points so that the direction of extending the connecting lines among all the reachable boundary points visited by the robot sequentially simulates the walking direction of the robot along the target contour line; the step B specifically comprises the following steps: Step B1, using two end points of a preset boundary as two extension starting points, respectively extending for one time according to the corresponding extension directions to obtain two extension line segments and two corresponding extension position points, and forming the preset boundary and the two extension line segments into the preset extension boundary; Step B2, judging whether the length of at least one expansion line segment is larger than a preset perception radius or not in the lengths of the two expansion line segments, if yes, determining that the preset boundary line expansion fails, otherwise, executing step B3; step B3, judging whether the length of at least one expansion line segment is equal to a preset perception radius or not in the lengths of the two expansion line segments, if yes, executing a step B4, otherwise, executing a step B5; Step B4, judging whether the two extension line segments are intersected with the target contour line or not, if yes, obtaining intersection points of the two extension line segments and the target contour line, and executing step C, otherwise, determining that the preset dividing line fails to extend; And B5, judging whether the two expansion line segments are intersected with the target contour line, if yes, acquiring intersection points of the two expansion line segments and the target contour line, and executing the step C, otherwise, updating two corresponding expansion position points into the two expansion starting points in the step B1, and then executing the step B1.
2. The method according to claim 1, wherein in the step C, when the preset demarcation expansion line segment is determined to intersect with the target contour line, the target contour line is divided into a traversed contour line and a contour line to be tracked by the preset demarcation expansion line segment; the area surrounded by the contour line to be tracked and the preset demarcation extension line segment is the area of one side of the contour line to be tracked.
3. The robot map exploration method according to claim 2, wherein before each execution of step a, if the robot does not travel from the current position to the reachable boundary point, before executing the current map patching operation, the robot marks its current detection area as an unknown area, and after executing the current map patching operation, the robot marks its current detection area in the area on the side of the traversed contour line as a known area; After the robot walks from the current position to the reachable boundary point and performs a new map repair operation, a known area is newly added to the area on one side where the contour line to be tracked which is segmented last time is located, and the contour line of the contour line to be tracked which is segmented last time in the newly added known area is configured as a traversed contour line so as to reduce the contour line to be tracked which is segmented last time, wherein the newly added known area comprises a current detection area generated after the robot walks to the reachable boundary point to perform the new map repair operation.
4. A method of exploring a map of a robot according to claim 3, wherein after the robot walks from an initial position, each time a map patching operation is performed when the robot walks to an accessible boundary point in the corridor area, a next accessible boundary point is screened out by performing steps a to D after the map patching operation is performed once, so as to guide the robot to walk to the next accessible boundary point along the contour line to be tracked; The current accessible boundary point of the robot is in a preset demarcation expansion line segment expanded in the step B at the accessible boundary point of the last walking of the robot.
5. The robot map exploration method according to claim 4, wherein when the robot walks from the current position to the reachable boundary point screened out in the last executed step D, the robot currently re-divides the target contour line into a traversed contour line and a contour line to be tracked by executing the steps a to C, so that the region covered by the current detection region is marked as a known region by executing a map patching operation in the region on the side of the contour line to be tracked which is divided in the last executed step C; In the process that the robot walks in the corridor area, the distance from the accessible boundary point screened in the last execution step D of the robot to the preset demarcation expansion line segment expanded in the current execution step B is preset.
6. The robot map exploration method according to claim 2, wherein in said step a, the method of fitting a preset boundary line by the robot using boundary points in its current detection area comprises: the robot scans out boundary points in the current detection area through the laser sensor, wherein the distance between the boundary points and the current position of the robot is smaller than or equal to the scanning distance of the laser sensor; and fitting the preset dividing line by utilizing boundary points in the current detection area according to a straight line fitting algorithm.
7. The robotic map exploration method of claim 3, wherein said method of map repair operation comprises: Step 1, after a robot builds a map to be optimized by utilizing laser point clouds, acquiring pose data of the robot and the laser point clouds in a current detection area from the map to be optimized; step 2, processing the pose data obtained in the step 1 according to a digital integration method, converting the pose data into a mask, and then executing the step 3; And step 3, converting the region covered by the mask in the map to be optimized in the step 1 into a passable region, and configuring the map to be optimized converted into the passable region by the mask into the map to be optimized in the step A.
8. The robot map exploration method according to claim 7, wherein said step 2 specifically comprises: Processing the pose data obtained in the step 1 according to a digital integration method to obtain a template; executing a filling algorithm in the corresponding empty map area according to the template, converting the corresponding empty map area into a mask, and enabling pixel values of all position points in the area covered by the mask to be assigned to preset values; The corresponding empty map area consists of position points corresponding to pose data in the current detection area, and the area of the mask is larger than or equal to that of the corresponding empty map area.
9. The robot map exploration method according to claim 8, wherein said step 3 specifically comprises: extracting an overlapped map region in a map to be optimized by using a mask, marking each position point in the overlapped map region as a passable position point to form a passable region, marking an unknown region in the overlapped map region as a known region, and determining that a region covered by a current detection region in the map to be optimized is marked as the known region; and (C) replacing the map region which is correspondingly covered in the map to be optimized by the composed passable region, so that the map to be optimized which is replaced by the map region which is correspondingly covered is updated into the map to be optimized in the step (A).
10. The robot map exploration method according to claim 8, wherein the preset demarcation extension line segment and the boundaries of the current exploration area on two sides of the robot walking direction are connected into a triangular area, and the mask is triangular in shape; The detection angle of the laser sensor is set in front of the robot by taking the walking direction of the robot as a central axis.
11. The method of claim 1, wherein in the step D, the method of screening the reachable boundary points according to the trafficability of the leading edge boundary points searched in the step C comprises: The robot corrodes the obstacle points in the contour line to be tracked according to a preset template image so as to enlarge the area occupied by the obstacle where the obstacle points are located, wherein the obstacle points are pixel points marked as the positions occupied by the obstacle, and the expansion radius of the preset template image is equal to the radius of the robot body; and the robot executes a filling algorithm in the map to be optimized along the extending direction of the contour line to be tracked, judges whether the leading edge boundary point exists in the neighborhood of the current filled passable position point, and sets the leading edge boundary point as an accessible boundary point if the leading edge boundary point exists in the neighborhood of the current filled passable position point, wherein the leading edge boundary point filled by the robot through the filling algorithm is communicated with the current position of the robot.
12. The method of claim 11, wherein in the step D, the method of sequentially storing the reachable boundary points so that a direction in which a line between the reachable boundary points visited by the robot extends simulates a direction in which the robot walks along the target contour line includes: The robot searches the contour line to be tracked according to a preset time direction from a searching starting point in the contour line to be tracked, marks the searched point as an accessed point every time the point in the contour line to be tracked is searched, detects whether the accessible point exists in the neighborhood of the accessed point, stores the accessible point into a linear storage space according to the sequence in the preset time direction if the accessible point exists in the neighborhood of the accessed point, and configures the accessible point stored in the linear storage space preferentially to be accessed preferentially, so that the sequence of reading the accessible point from the linear storage space simulates the sequence of the robot walking along each point in the contour line to be tracked in the process of walking of the robot according to the preset time direction; and B, determining a search starting point in the contour line to be tracked, wherein the search starting point in the contour line to be tracked comprises an intersection point of the preset demarcation extension line segment and the target contour line, which is determined in the step B.
13. A chip for storing program code for executing the robot map exploration method according to any one of claims 1 to 12.

Description

Robot map exploration method and chip based on reachable boundary points Technical Field The application relates to the technical field of robot detection maps, in particular to a robot map exploration method and a chip based on reachable boundary points. Background The robot actively explores an unknown environment and builds an environment map, and the method has important significance for realizing autonomous navigation of the mobile robot, and many technologies related to the mobile robot are developed around the map. Before a cleaning robot performs cleaning work, an environment map of a complete house needs to be quickly constructed, typically, the cleaning robot navigates to an unknown area by identifying the unknown area and scans the unknown area by using a laser radar to expand the environment map, however, the map construction of the cleaning robot is gradually constructed in the travelling process, and the map construction mode of the cleaning robot is not suitable for a scene where the cleaning robot needs to quickly acquire the map of the unknown area, so that the navigation work efficiency of the cleaning robot is affected. The Chinese patent application No. CN202110264567.X obtains boundary points meeting preset passing conditions based on a boundary detector of a fast exploration random tree algorithm, obtains optimal boundary points under the current view through complex profit calculation functions, takes the navigation cost of the boundary points and corresponding profit information into consideration, selects boundary points with highest profit from boundary points stored in a boundary point list as target points, and controls a robot to move from a current position to the target points so as to guide the robot to build a map in an unknown area, so that the method for constructing the map by using the boundary points in the unknown area mainly considers factors such as exploration repeatability of the boundary points, information gain of the boundary points, navigation cost and the like, screens the optimal boundary points under the current view through complex profit calculation functions, takes longer overall time consumption, influences the exploration efficiency, and does not consider the influence of contour lines in the local map on the cleaning robot map building. Disclosure of Invention The application discloses a robot map exploration method based on reachable boundary points, which comprises the following specific technical scheme: A robot map exploration method based on reachable boundary points comprises the steps of acquiring laser point clouds through laser sensor scanning by a robot, constructing a map to be optimized by utilizing the laser point clouds, fitting a preset boundary line from boundary points in a current detection area of the robot by the robot, selecting contour lines used for defining all communicated detectable areas in the map to be optimized as target contour lines, executing a step B, respectively controlling two ends of the preset boundary line to expand to obtain preset boundary expansion line segments, executing a step C if two ends of the preset boundary expansion line segments are intersected with the target contour lines under the condition that the expansion lengths of the two ends of the preset boundary line segments are smaller than or equal to a preset perception radius, and executing a step C based on the preset boundary expansion line segments, wherein the robot searches a front boundary point in an area surrounded by the boundary expansion to be tracked from the boundary line to the preset boundary line segments, and then executing a step D, and storing the boundary points of the robot along the boundary expansion line segments in the direction which the boundary points can reach the corresponding directions of the boundary line segments to be reached to the corresponding object boundary lines. The application also discloses a chip for storing program codes for executing the robot map exploration method. Compared with the prior art, the method has the advantages that the detected boundary points are adopted to fit the preset boundary line, the intersection condition of the preset boundary line after the extension and the target contour line is determined by the extension of the preset boundary line, the contour line to be tracked is segmented from the target contour line, the boundary points along the front edge are searched only in the area range related to the contour line to be tracked, and the boundary points are sequentially distributed to the robot for navigation and map construction after the trafficability test, so that the method is suitable for the scene of a map of an unknown area needing to be quickly explored by the robot, the map construction process is optimized, and the effect of quick map construction is achieved. The target contour line is segmented by utilizing the result of expanding the preset boundary line to obtain t