CN112180931B

CN112180931B - Cleaning path planning method and device of sweeper and readable storage medium

Info

Publication number: CN112180931B
Application number: CN202011062891.5A
Authority: CN
Inventors: 檀冲; 王颖; 张书新
Original assignee: Beijing Puppy Vacuum Cleaner Group Co Ltd
Current assignee: Beijing Puppy Vacuum Cleaner Group Co Ltd
Priority date: 2020-09-30
Filing date: 2020-09-30
Publication date: 2024-04-12
Anticipated expiration: 2040-09-30
Also published as: CN112180931A

Abstract

The invention discloses a cleaning path planning method and device of a sweeper, a readable storage medium and electronic equipment, wherein the method comprises the following steps: acquiring a position point formed by sweeping the floor sweeping machine along the wall and a room map to be processed corresponding to a room swept by the floor sweeping machine along the wall; determining a map of an area to be cleaned according to the map of the room to be processed and the position points, wherein the position points are positioned in the boundary of the map of the area to be cleaned; determining a wall-following path of an obstacle in the map of the area to be cleaned; and planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle. According to the technical scheme, the map of the area to be cleaned with a larger cleaning range is redetermined based on the map of the room to be processed and the position points formed by the cleaning of the sweeper along the wall, so that the cleaning coverage rate and the cleaning effect are ensured, and then, the cleaning path is quickly planned in the map of the area to be cleaned based on the path along the wall of the obstacle, so that the planning efficiency of the cleaning path is improved.

Description

Cleaning path planning method and device of sweeper and readable storage medium

Technical Field

The present invention relates to the field of sweeper technologies, and in particular, to a method and apparatus for planning a sweeping path of a sweeper, and a readable storage medium.

Background

With the continuous improvement of the living standard of people, the sweeper starts to walk into the sight of people and is accepted and used by more and more people. The existing sweeping robot generally takes an area formed by a closed line formed by sweeping along a wall by a sweeping machine as an area to be swept, plans a sweeping path in the area to be swept, and then sweeps along the sweeping path by the sweeping machine.

However, the robot cannot clean the area beyond the closed line formed by the cleaning of the sweeper along the wall, reducing the cleaning effect.

Disclosure of Invention

The invention provides a cleaning path planning method and device of a sweeper, a readable storage medium and electronic equipment.

In a first aspect, the present invention provides a cleaning path planning method for a sweeper, including:

acquiring a position point formed by sweeping the floor sweeping machine along the wall and a room map to be processed corresponding to a room swept by the floor sweeping machine along the wall;

Determining a map of an area to be cleaned according to the map of the room to be processed and the position points, wherein the position points are positioned in the boundary of the map of the area to be cleaned;

determining a wall-following path of an obstacle in the map of the area to be cleaned;

and planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle.

In a second aspect, the present invention provides a cleaning path planning apparatus for a sweeper, including:

the acquisition module is used for acquiring position points formed by the sweeping of the sweeper along the wall and a to-be-processed room map corresponding to a room swept by the sweeper along the wall;

the map determining module is used for determining a map of an area to be cleaned according to the map of the room to be processed and the position points, and the position points are positioned in the boundary of the map of the area to be cleaned;

the path determining module is used for determining a wall-following path of the obstacle in the map of the area to be cleaned;

and the cleaning module is used for planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle.

In a third aspect, the present invention provides a readable medium comprising execution instructions which, when executed by a processor of an electronic device, perform the method according to any of the first aspects.

In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory storing execution instructions, the processor performing the method according to any one of the first aspects when executing the execution instructions stored in the memory.

The invention provides a cleaning path planning method and device of a sweeper, a readable storage medium and electronic equipment, wherein a to-be-cleaned area map with a larger cleaning range is redetermined based on a to-be-processed room map and a position point formed by cleaning the sweeper along a wall, so that cleaning coverage rate and cleaning effect are ensured, then, the wall-along path of an obstacle in the to-be-cleaned area map is determined, the cleaning path is planned quickly in the to-be-cleaned area map based on the wall-along path of the obstacle, the planning efficiency of the cleaning path is improved, and then, the sweeper can be controlled to clean according to the cleaning path.

Further effects of the above-described non-conventional preferred embodiments will be described below in connection with the detailed description.

Drawings

In order to more clearly illustrate the embodiments of the invention or the prior art solutions, the drawings which are used in the description of the embodiments or the prior art will be briefly described below, it being obvious that the drawings in the description below are only some of the embodiments described in the present invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.

Fig. 1 is a schematic flow chart of a cleaning path planning method of a cleaning machine according to an embodiment of the present invention;

fig. 2 is a flow chart of another cleaning path planning method of a cleaning machine according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of a cleaning path planning device of a sweeper according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to specific embodiments and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The cleaning path planning method of the sweeper provided by the embodiment of the invention can be applied to electronic equipment, in particular to a cleaning robot, a server or a general computer, wherein the cleaning robot can be a sweeper, and the cleaning robot is not limited herein. As shown in fig. 1, a cleaning path planning method of a cleaning machine according to an embodiment of the present invention is provided. In this embodiment, the method specifically includes the following steps:

And 101, acquiring a position point formed by sweeping the sweeper along the wall and a room map to be processed corresponding to a room swept by the sweeper along the wall.

Specifically, the plurality of position points comprise path points generated in the process that the sweeper cleans along the wall edge of a room until the travel route forms a closed line, so that the position points can accurately reflect the real travel route of the sweeper, and the reference value of the map of the area to be cleaned, which is obtained later, is convenient. It should be noted that, considering that the positioning error may occur when the sweeper walks along the wall, some of the path points may be outside the room or too far from the wall, and these path points may reduce the accuracy of the subsequent processing, it is necessary to remove the path points outside the room and too far from the wall, so as to ensure the reference value of the obtained position points.

Specifically, when the electronic equipment is a sweeper, the environment data can be acquired through the data acquisition sensor arranged on the electronic equipment, when the electronic equipment is not the sweeper, the environment data acquired by the sensor on the sweeper can be acquired, then an environment map is constructed according to the environment data, and operations such as graying, denoising, boundary line extraction and the like are performed on the environment map so as to obtain a to-be-processed room map which is convenient to process and can reflect an actual scene. The data acquisition sensor may be a non-visual sensor such as a structured light sensor or a laser radar sensor, or may be a visual sensor such as a camera, and is preferably a non-visual sensor in view of the fact that the processing of the image acquired by the visual sensor is more complicated and the calculation amount is larger.

Specifically, the room map to be processed is preferably a grid map, and for convenience of image processing, the room map to be processed is preferably a gradation map. Obviously, the room map to be processed should be a room swept along the walls, ensuring that location points can be mapped into the room map to be processed. It should be noted that, considering that the sweeper cleans one room only after cleaning the next room, the map of the room to be processed should include one room.

In one embodiment, the room map to be processed may be determined specifically by:

according to the environmental data collected by the non-visual sensor on the sweeper, constructing an original room gray map corresponding to a room cleaned by the sweeper along the wall; sequentially removing noise points of obstacle boundaries, extracting the obstacle boundaries and removing outlier areas in idle areas from the original room gray map to determine a reference room gray map; when the obstacle boundary in the reference room gray map is closed, carrying out integrity detection on the reference room gray map, and determining the reference room gray map passing the integrity detection as a room map to be processed; when the obstacle boundary in the reference room gray map is not closed, performing expansion processing on the non-closed obstacle boundary in the reference room gray map to enable the boundary line to be closed, performing integrity detection on the expanded reference room gray map, and determining the expanded reference room gray map passing the integrity detection as a room map to be processed.

In the embodiment, the original room map constructed by the environmental data acquired by the non-visual sensor is subjected to noise point removal of the obstacle boundary, extraction of the obstacle boundary, closure detection of the obstacle boundary and closure processing of the non-closure obstacle boundary, so that a room map to be processed with the obstacle boundary closed and the complete room is obtained.

It should be noted that, the construction of the gray map of the room by the environmental data collected by the non-visual sensor belongs to the prior art, and will not be repeated here.

Specifically, the original room gray map includes an obstacle region, an idle region, and an unknown region, the obstacle region refers to a region where an obstacle is located, the idle region refers to a region where no obstacle exists, the unknown region refers to a region where whether an obstacle exists is unknown, and the unknown region can be understood as a region not detected by a non-visual sensor, in other words, the original room gray map includes pixels on the obstacle region, pixels on the idle region, and unknown pixels on the unknown region. The obstacle boundary may be understood as the outline of the obstacle region, and may be understood as a line formed by all pixels on the obstacle region that are directly connected to the free region.

In consideration of the fact that noise points exist in an original room gray map constructed based on environment data acquired by a non-visual sensor, the difference between the original room gray map and a real scene occurs, for the original room gray map, the obstacle is mainly a wall and objects such as a table, a sofa and a chair in a room surrounded by the wall, the obstacle is mainly the wall, the boundary of the wall is an important characteristic for representing the object, the basis for determining an area to be cleaned is also the basis, when the noise points exist on the boundary of the wall, the boundary of the wall is not continuous straight lines any more, the boundary of the wall cannot truly reflect the shape of a wall, and therefore the noise points on the boundary of the obstacle in the original room gray map need to be removed, the boundary of the obstacle after the noise points are removed can be accurately represented. It should be noted that, a noise point on the obstacle boundary may be understood as a pixel point that does not truly reflect the outline of the obstacle, i.e., a pixel point that makes the obstacle boundary not smooth.

For non-visual sensors such as a structured light sensor or a laser radar, the surrounding environment is usually detected by using a laser line, so that the thickness of a detected obstacle, for example, the thickness of a wall cannot be known, and here, after removing noise points on the obstacle boundary in an original room gray level map, the obstacle boundary is extracted, pixel points inside the obstacle region are removed, and an important feature of the obstacle boundary characterizing the obstacle region is reserved, so that the subsequent processing is facilitated.

The original room gray map has a plurality of idle areas, the areas of the idle areas are large and small, and for a room, only one idle area with a larger area normally needs to be available, therefore, for the idle area with a smaller area, noise is needed, namely, an outlier area, which can be located outside the room or inside an obstacle, so that the original room gray map is different from a real scene, and specifically, the gray values of pixels on an unknown area or the gray values of pixels on an obstacle area can be replaced by the gray values of pixels on the outlier area, so that the outlier area in the original gray room gray map can be removed. Since the respective free areas are not in contact with each other, the gray value of the pixel point in the outlier area is not required to be the gray value of the pixel point in the free area, and therefore the outlier area can be removed by the gray value of the pixel point in the unknown area or the gray value of the pixel point in the obstacle. Specifically, the other idle regions except the idle region with the largest area are outliers, for example, N idle regions are assumed, the areas corresponding to 1 to N idle regions are S1, S2, …, SN-1, SN, respectively, and if SN is the largest, the idle regions corresponding to S1, S2, …, SN-1 are determined as outliers.

Specifically, detecting a boundary noise point of an original room gray map to determine a noise point on an obstacle boundary, and when the noise point is judged to be the obstacle boundary, replacing the gray value of the noise point by the gray value of a pixel point on the obstacle boundary to remove the noise point; when the noise point is judged to be a noise point outside the obstacle boundary, and the noise point is positioned inside the obstacle boundary, namely in a room, replacing the gray value of the noise point by the gray value of the pixel point on the idle area so as to remove the noise point; when the noise point is judged to be a noise point outside the obstacle boundary, and the noise point is positioned outside the obstacle boundary, namely outside a room, the gray value of the noise point is replaced by the gray value of the pixel point on the position area so as to remove the noise point; and then, reserving all the pixel points on the barrier area connected with the pixel points of the idle area, and replacing the gray values of other pixel points on the barrier area by the gray values of the pixel points of the unknown area. And then, replacing the gray values of the pixel points on other idle areas except the idle area with the largest area by the gray values of the pixel points of the unknown area, and finally obtaining the gray map of the reference room.

When the obstacle boundary is not a closed area, the idle area is very probably connected with the unknown area, and whether the room is complete or not is judged to be still whether the idle area is connected with the unknown area or not, so that the reference room gray level map is subjected to closed detection of the obstacle boundary.

As a possible scenario, if the obstacle boundary in the reference room gray map is closed, an integrity check may be performed on the reference room gray map, and after the integrity check passes, the reference room gray map is determined as the room map to be processed.

As another possible case, when the obstacle boundary in the reference room gradation map is not closed, the expansion processing is performed on the non-closed obstacle boundary in the reference room gradation map to close the obstacle boundary, the integrity detection is performed on the expanded reference room gradation map, and after the integrity detection is passed, the expanded reference room gradation map is determined as the room map to be processed.

Specifically, when the idle area in the map of the room to be processed is only connected with the obstacle boundary and not connected with the unknown area, the room exploration is completed, and the map construction can be stopped at the moment.

And 102, determining a map of the area to be cleaned according to the map of the room to be processed and the position points, wherein the position points are positioned in the boundary of the map of the area to be cleaned.

In the embodiment, the to-be-cleaned area map is determined through the to-be-processed room map and the position points, and the position points are located in the boundary of the to-be-cleaned area map, so that the cleaning area is enlarged, the cleaning coverage rate is improved, the area between the position points and the wall can be cleaned, and the cleaning effect is improved.

Here, the boundary of the area map to be cleaned refers to an outline in the area map to be cleaned, and can also be understood as a closed line forming the area to be cleaned.

In one embodiment, the map of the area to be cleaned may be determined specifically by:

determining that the position points correspond to reference pixel points on a map of the room to be processed, and determining an area outside a closed area formed by the reference pixel points in the room to be processed as a legal area; searching according to the sequence from small radius to large radius by taking the reference pixel point as the circle center, and determining a candidate pixel point closest to the reference pixel point from the legal area, wherein the candidate pixel point is a pixel point on the boundary of the obstacle; determining a target pixel point from pixel points on a connecting line of the reference pixel point and the candidate pixel point according to the radius of the sweeper; when the distance between two adjacent target pixel points is larger than the distance threshold value, pixel point interpolation is carried out between the two adjacent target pixel points so as to obtain interpolation pixel points; and determining a closed area formed by the target pixel points and the interpolation pixel points in the room map to be processed as a map of the area to be cleaned.

Specifically, each position point corresponds to a reference pixel point, the reference pixel points in the map of the room to be processed are connected or fitted to obtain a closed area, the area outside the closed line is determined to be a legal area, so that the maximum probability that the boundary of an obstacle in the legal area is a wall body, namely the maximum boundary of the room is ensured, and then the map of the area to be cleaned can be determined in the legal area, so that the cleaning coverage area is enlarged. Optionally, a closed line obtained by connecting or fitting a plurality of reference pixel points is determined, an obstacle boundary outside the closed line is determined, and a region formed by the closed line and the obstacle boundary is determined as a legal region.

Then, searching is performed by taking the reference pixel point as the center of a circle according to the sequence from small radius to large radius, and the candidate pixel point closest to the reference pixel point is determined from the legal area, namely, the pixel point on the obstacle boundary closest to the position point is found, so that the maximum boundary of the room is known, for example, the obstacle boundary is a wall surface, the candidate pixel point can be understood as the pixel point closest to the position point on the wall surface, and in consideration of the processing of the room map to be processed, all pixel points on the room map to be processed need to be traversed, so that the calculation force is high.

Optionally, determining a preset distance according to the width of the pixel point in the room map to be processed and the radius of the sweeper, for example, if the width of the pixel point is 5cm, the preset distance may be 10cm, that is, the widths of two pixel points, searching with the reference pixel point as the center of a circle, taking the preset distance as the radius, without searching the pixel points within the preset distance, reducing the calculated amount, determining, for each searched pixel point, whether the gray value of the pixel point is the gray value of the pixel point on the obstacle boundary, if yes, determining whether the pixel point is in the legal area, if no, continuing searching, if yes, determining the pixel point as a candidate pixel point, stopping searching, if no candidate pixel point is found, increasing the width of one pixel point on the basis of the preset distance, and then continuing searching in a similar manner.

After knowing the pixel point on the boundary of the obstacle closest to the position point, namely, the candidate pixel point, determining the target pixel point from the pixel points on the connecting line of the reference pixel point and the candidate pixel point according to the radius of the sweeper, wherein the sweeper cannot contact with the wall when the target pixel point corresponds to the actual position of the room, so that collision friction cannot occur between the sweeper and the wall, wherein when the target pixel point is positioned on the left side or the right side of the candidate pixel point, the horizontal distance between the target pixel point and the candidate pixel point is not smaller than the radius of the sweeper, and when the target pixel point is positioned on the upper side or the lower side of the candidate pixel point, the vertical distance between the target pixel point and the candidate pixel point is not smaller than the radius of the sweeper.

Considering that there is a great probability that there is an error in the process of converting the data on the sensor to the map for representation, for example, the width of each pixel point in the grid map is 5cm for the grid map, then the error is unavoidable in the process of obtaining the grid map, and as a possible implementation manner, the confidence level of the candidate pixel point, that is, the probability that the pixel point is an obstacle boundary, and the confidence level of the reference pixel point, that is, the probability that the pixel point is a position point, are obtained, and the target pixel point is determined from the reference pixel point and the pixel point on the connecting line of the candidate pixel point based on the confidence level of the candidate pixel point, the confidence level of the reference pixel point and the radius of the sweeper, so that the reference value of the determined target pixel point is ensured. Here, the distance errors corresponding to different confidence levels may be preset, the higher the confidence level is, the smaller the distance error is, the distance error corresponding to the confidence level of the obstacle pixel point may be different from the distance error corresponding to the confidence level corresponding to the reference pixel point, which may be the same, preferably different, or a function between the confidence level, the width of the pixel point and the distance error corresponding to the confidence level may be constructed to represent the relationship between the confidence level and the distance error corresponding to the confidence level, so that the confidence level is substituted into the function, and the distance error corresponding to the confidence level may be obtained. Specifically, for each candidate pixel point, determining the sum of a distance error corresponding to the confidence level of the candidate pixel point, a distance error corresponding to the confidence level of the reference pixel point and the radius of the sweeper as a confidence distance, so as to ensure cleaning safety, determining a target pixel point with a distance value which is horizontally away from the candidate pixel point as the confidence distance from a pixel point on a connecting line of the reference pixel point and the candidate pixel point when the candidate pixel point is on the left side or the right side of the target pixel point, and determining a target pixel point with a distance value which is vertically away from the candidate pixel point as the confidence distance from a pixel point on a connecting line of the reference pixel point and the candidate pixel point when the candidate pixel point is on the upper side or the lower side of the target pixel point. Note that, when the confidence distance and the width of the pixel do not have an integer multiple relationship, the confidence distance needs to be rounded so that the confidence distance and the width of the pixel have an integer multiple relationship, for example, assuming that the width of the pixel is 5cm and the confidence distance is 8cm, the confidence distance may be adjusted to be 10cm, that is, 2 times the width of the pixel.

When two adjacent target pixel points are not connected, the situation that the two target pixel points are blank is described, at the moment, pixel point interpolation processing is needed to be carried out between the two target pixel points to obtain interpolation pixel points, when the two adjacent target pixel points are not located at corners, the two target pixel points are directly connected, the pixel points on a connecting line are taken as the interpolation pixel points, when the two adjacent target pixel points are located at the corners, the pixel points formed by the maximum horizontal coordinates and the maximum vertical coordinates of the two target pixel points are determined, then the pixel points are respectively connected with the two target pixel points, and the pixel points on the two connecting lines are taken as the interpolation pixel points. Thus, the distance threshold refers to the distance occupied by one pixel in the room map to be processed, i.e., the width of one pixel.

At this time, two adjacent target pixel points are in contact connection, two adjacent interpolation pixel points are in contact, and two adjacent target pixel points are in contact with the interpolation pixel points, so that a closed area can be obtained by connecting the target pixel points and the interpolation pixel points, the area of the closed area corresponding to the area in the to-be-cleaned room map is determined as the to-be-cleaned room map, and the obtained to-be-cleaned room map is subjected to expansion processing, so that cleaning coverage rate is ensured.

And step 103, determining the wall-following path of the obstacle in the map of the area to be cleaned.

The along-wall path represents the path trace of the sweeper after entering a mode of along-wall motion (simply referred to as an along-wall mode). Specifically, the wall-following mode refers to: the cleaning robot simulates the boundary of the obstacle after the boundary is enlarged into an actual wall, the cleaning robot simulates to walk along the wall for a certain distance in a mode of not contacting the obstacle, and the path track formed finally approximates to one section or the whole section of the boundary of the obstacle, so that the path track is the path along the wall.

It should be noted that, an obstacle in the map of the area to be cleaned may be understood as a closed area, and the gray values of the pixels on the closed area may represent the obstacle.

In one embodiment, the path along the wall may be determined specifically by:

determining the minimum boundary of the obstacle according to the acquired safety distance of the obstacle in the map of the area to be cleaned and the radius of the sweeper; determining the maximum boundary of the obstacle on the basis of the minimum boundary according to the acquired cleaning path planning speed interval of the sweeper; estimating a dirty area corresponding to the obstacle according to the historical cleaning record of the sweeper, wherein the dirty area is positioned in an area between the minimum boundary and the maximum boundary; and determining a wall-following path of the obstacle according to the dirty area corresponding to the obstacle, and passing through the dirty area along the wall path.

Specifically, the safety distance of the obstacle represents the minimum distance between the sweeper and the obstacle, and the distance between the minimum boundary and the obstacle is the sum of the safety distance of the obstacle and the radius of the sweeper, so that the risk or negative influence of the obstacle and/or the cleaning robot is reduced, and the cleaning safety is improved.

Considering that the sweeper cleans around obstacles, it is unlikely to maintain consistency of cleaning speed, and has a cleaning speed interval. Here, the lower the speed is, the slower the sweeper moves, the easier the wall-following path is realized, the faster the speed is, the faster the sweeper moves, the greater the difficulty of the wall-following path is, the longer distance is required to be reserved for buffering, so that on the basis of the minimum boundary, a maximum boundary is set, and the area between the maximum boundary and the minimum boundary is used for buffering of the sweeper, thereby ensuring the realization of the wall-following path.

Then, a dirty region corresponding to the obstacle is estimated from the historical cleaning record of the sweeper, wherein the dirty region is located in a region between the minimum boundary and the maximum boundary. The history cleaning record includes information such as cleaning speed, cleaning time, cleaning route, cleaning area, etc., and it should be noted that the history cleaning record should be recorded in a preset period before the current time to ensure the reference value of the history cleaning record. Specifically, the area between the minimum boundary and the maximum boundary may be divided to obtain a plurality of candidate areas, then, for each candidate area, the number of cleaning times of the candidate area is determined from the history cleaning record, and then, the candidate areas with the number of cleaning times meeting the preset value are respectively determined as dirty areas.

If other factors do not need to be considered, optionally, acquiring key points on a plurality of cleaning paths corresponding to each dirty area, for example, inflection points or points which are necessary for forming the cleaning paths, and fitting the plurality of key points corresponding to each dirty area, so as to determine the wall-following paths corresponding to the obstacles.

In one embodiment, when knowing the planned speed of the cleaning path of the sweeper, i.e. knowing in advance the speed at which the sweeper is to clean, the wall-following path corresponding to the obstacle may be determined by:

according to the dirty area corresponding to the obstacle, planning a candidate path corresponding to the obstacle in the area between the minimum boundary and the maximum boundary, wherein different candidate paths correspond to different cleaning speeds, and the candidate paths pass through the dirty area; and determining a candidate path matched with the cleaning path planning speed of the sweeper from the candidate paths, and determining the matched candidate path as a wall-following path of the obstacle.

Specifically, for each preset cleaning speed, key points of each dirty region corresponding to a plurality of cleaning paths at the preset cleaning speed are obtained, and the plurality of key points corresponding to each dirty region are fitted, so that candidate paths of the obstacle corresponding to the preset cleaning speed are determined, wherein the preset cleaning speed can be determined according to the actual condition of the sweeper. Then, determining a candidate path matched with the cleaning path planning speed of the sweeper, determining the candidate path as a wall-following path of the obstacle, for example, determining a preset cleaning speed which is the same as the cleaning path planning speed of the sweeper from preset cleaning speeds, and determining the corresponding candidate path as a wall-following path, wherein if the preset cleaning speed which is the same as the cleaning path planning speed of the sweeper does not exist, a plurality of preset cleaning speeds which are similar to the cleaning path planning speed of the sweeper can be determined from the preset cleaning speeds, and the candidate paths which are respectively corresponding to the preset cleaning speeds are fused to obtain the wall-following path. In some possible scenes, when the cleaning speed of the sweeper reaching the periphery of the obstacle changes, the wall-following path of the obstacle can be timely adjusted according to a plurality of preset candidate paths, so that the cleaning path adjustment of the sweeper is quickly realized, and the cleaning efficiency is ensured. In other possible scenarios, when the sweeper needs to avoid an obstacle, the obstacle's avoidance may be performed along a wall path of the obstacle.

In one embodiment, the safe distance may be determined specifically by:

acquiring a confidence level of an obstacle boundary in a map of an area to be cleaned and a shape of the obstacle boundary; the safe distance of the obstacle is determined according to the confidence level of the obstacle boundary and the shape of the obstacle boundary.

A low confidence level for the obstacle boundary indicates an increased likelihood of an error in the obstacle boundary, i.e., greater difficulty in achieving a path along the wall, so that a longer distance needs to be reserved for buffering, and therefore, a greater safety distance for the obstacle.

The more inflection points of the obstacle boundary shape, the more complex the obstacle boundary shape, the more difficult it is to realize the path along the wall, so that a longer distance needs to be reserved as a buffer, and the greater the safety distance of the obstacle.

Here, the obstacle recognition is not required, and the shape of the obstacle boundary and the confidence level of the obstacle boundary need only be known, and the smaller the shape of the obstacle boundary and the smaller the confidence level of the obstacle boundary, the larger the safety distance of the obstacle is, whereas the smaller the safety distance is, and here, the safety distance can be preset in combination with the actual situation. The process does not need to carry out semantic recognition on the obstacle, reduces the processing difficulty and the calculated amount, but can realize that different safety distances are set for different obstacle boundaries due to different obstacle boundaries, so that the cleaning coverage rate is improved to a certain extent, and meanwhile, the risk or negative influence of the obstacle and/or the robot is greatly reduced, and the cleaning safety is improved.

And 104, planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle.

Specifically, path planning can be performed on the map of the area to be cleaned according to the arcuate path, or path planning can be performed on the map of the area to be cleaned according to the return-to-the-figure path, so that a cleaning path is determined, and then the sweeper is controlled to clean according to the cleaning path.

According to the technical scheme, the beneficial effects of the embodiment are as follows: and (3) based on the complete room map to be processed and the position points formed by the sweeping of the sweeper along the wall, determining a map of the area to be swept with a larger sweeping range again, and ensuring the sweeping coverage rate and the sweeping effect. And planning a wall-following path in an area between the minimum boundary and the maximum boundary of the obstacle in the map of the area to be cleaned, so that the road can adapt to the changes of different scenes and the cleaning effect is ensured.

Fig. 1 shows only a basic embodiment of the method according to the invention, on the basis of which certain optimizations and developments are made, but other preferred embodiments of the method can also be obtained.

In order to more clearly describe the technical solution of the present invention, please refer to fig. 2, another cleaning path planning method of a cleaning machine is provided in the embodiment of the present invention, and the embodiment is further described with reference to a specific application scenario based on the foregoing embodiment. In this embodiment, the method specifically includes the following steps:

Step 201, constructing an original room gray map corresponding to a room cleaned by the sweeper along a wall according to environmental data acquired by a non-visual sensor on the sweeper; and sequentially removing noise points of obstacle boundaries, extracting the obstacle boundaries and removing outliers in idle areas from the original room gray map to determine a reference room gray map.

Specifically, the non-visual sensor is a structured light sensor, and the original room gray map includes an obstacle area, an idle area and an unknown area in one room, where the obstacle area is black, formed by a plurality of black dots, the idle area is white, formed by a plurality of white dots, and the unknown area is gray, formed by a plurality of gray dots, and where the black dots, the white dots and the gray dots can be understood as pixel dots. After removing noise points on the boundary of the black area, retaining the black points connected with the white points, changing all the black points which are not connected with the white points into gray points, and changing all the other white areas except the white area with the largest area into black or gray to obtain a reference room gray map.

And 202, when a boundary line in the reference room gray level map is closed, performing integrity detection on the reference room gray level map, and determining the reference room gray level map passing the integrity detection as a room map to be processed.

If the black border line in the reference room gray map is closed, the target gray map is determined as the room map to be processed.

Step 203, determining that the position point corresponds to a reference pixel point on the map of the room to be processed, and determining an area outside a closed area formed by the reference pixel point in the room to be processed as a legal area.

And determining that the position points correspond to reference pixel points on the map of the room to be processed, connecting all the reference pixel points to form a closed line, and determining the area outside the closed line as a legal area, wherein the legal area comprises black boundary lines.

Step 204, searching according to the order of radius from small to large by taking the reference pixel point as the center of a circle, so as to determine a candidate pixel point closest to the reference pixel point from the legal area, wherein the candidate pixel point is a pixel point on the boundary of an obstacle; and determining a target pixel point from the pixel points on the connecting line of the reference pixel point and the candidate pixel point according to the radius of the sweeper.

Determining a nearest black point from the reference pixel point as a candidate pixel point, assuming that a distance error corresponding to a confidence level of the candidate pixel point is D1, a distance error corresponding to a confidence level of the reference pixel point is D2, a radius of the sweeper is R, a width of the pixel point is D, determining a confidence distance D of the candidate pixel point as R+d1 x d+d2 x D, and determining a target pixel point with a distance value which is a confidence distance D in a horizontal direction from the pixel point on a connecting line of the reference pixel point and the candidate pixel point when the reference pixel point is positioned on the left side or the right side of the candidate pixel point; when the reference pixel point is located on the upper side or the lower side of the candidate pixel point, determining a target pixel point with a distance value which is a confidence distance D from the reference pixel point and the candidate pixel point in the vertical direction from the pixel points on the connecting line of the reference pixel point and the candidate pixel point.

Step 205, when the distance between two adjacent target pixel points is greater than a distance threshold, performing pixel point interpolation between the two adjacent target pixel points to obtain interpolation pixel points; and determining a closed area formed by the target pixel point and the interpolation pixel point in the room map to be processed as an area map to be cleaned.

The distance threshold is the width d of the pixel point, the distance between the adjacent target pixel points p1 and p2 is larger than the width d of the pixel point, the pixel coordinates of the p1 and p2 are respectively (x 1, y 3) and (x 3, y 1), x1 is smaller than x3, y1 is smaller than y3, the pixel coordinates of the pixel point p 'formed by the maximum abscissa x3 and the maximum ordinate y3 in the p1 and p2 are determined to be (x 3, y 3), then, the p' and p1, p 'and p2 are connected, and if the pixel point on the connecting line of the p' and p1 is p '1, the pixel point on the connecting line of the p' and p2 is p '2, the interpolation pixel point is p' and p '1 and p' 2.

After the processing, the target pixel point and the interpolation pixel point are connected to form a closed area, and the area of the closed area corresponding to the room map to be processed is determined as an area map to be cleaned.

Step 206, determining the safety distance of the obstacle according to the confidence level of the obstacle boundary and the shape of the obstacle boundary in the acquired map of the area to be cleaned; determining the minimum boundary of the obstacle according to the safety distance of the obstacle in the map of the area to be cleaned and the radius of the sweeper; and amplifying the minimum boundary according to the acquired sweeping speed interval of the sweeper around the obstacle so as to determine the maximum boundary of the obstacle.

The shapes of the barrier boundaries are preset, such as a circle, an ellipse, a rectangle, a round corner rectangle, an irregular wave shape and the like, and for each barrier boundary shape, the confidence level of the barrier boundary is taken as an independent variable, the safety distance of the barrier is taken as a dependent variable, and a function of the confidence level of the barrier boundary and the safety distance of the barrier is established; and then, determining the shape of the obstacle boundary, determining a function corresponding to the shape, and substituting the confidence level of the obstacle boundary into the function, thereby obtaining the safety distance of the obstacle boundary.

Here, the sweeping speed interval is understood as a sweeping speed variation range when the sweeper sweeps around an obstacle, wherein the boundary expansion multiples corresponding to different sweeping speed intervals can be preset, the boundary expansion multiples are not needed to consider the size of the obstacle, and the application condition of different scenes is ensured, and the boundary expansion multiples should not be excessively large, for example, the ratio range of the line length of the second boundary to the line length of the first boundary is 1.01-1.1. The larger the difference between the maximum value of the cleaning speed interval and the speed is, the larger the distance that the robot needs to buffer is, and at this time, the larger the expansion multiple is. Specifically, a plurality of maximum cleaning speeds may be preset, for each maximum cleaning speed, a speed variation value of the maximum cleaning speed is taken as an independent variable, a boundary expansion multiple is taken as a dependent variable, a relationship between the speed variation value of the maximum cleaning speed and the boundary expansion multiple is established, a maximum speed corresponding to a speed variation interval is subsequently determined, a function corresponding to the maximum speed is determined, an absolute value of a difference value between the maximum speed and the minimum speed of the speed variation interval is substituted into the function, so as to obtain a boundary expansion multiple, a first boundary is processed according to the boundary expansion multiple, so as to obtain a second boundary, and the first boundary is amplified by taking a center point of an obstacle boundary as a reference point.

Step 207, estimating a dirty area corresponding to the obstacle according to the historical cleaning record of the sweeper, wherein the dirty area is located in an area between the minimum boundary and the maximum boundary.

The relevant content may refer to the description of step 103, and will not be repeated here.

And step 208, planning candidate paths corresponding to the obstacle in the area between the minimum boundary and the maximum boundary according to the dirty area corresponding to the obstacle, wherein different candidate paths correspond to different cleaning speeds, and the candidate paths pass through the dirty area.

Step 209, determining a candidate path matched with the cleaning path planning speed of the sweeper from the candidate paths, and determining the matched candidate path as a wall-following path of the obstacle; and planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle.

The relevant content may refer to the descriptions of step 103 and step 104, and will not be repeated here. Subsequently, the robot can be controlled to clean according to the cleaning path.

According to the technical scheme, the beneficial effects of the embodiment are as follows: noise point removal of an obstacle boundary, extraction of the obstacle boundary, outlier region removal in an idle region, obstacle boundary closing detection, obstacle boundary expansion, integrity detection and the like are carried out on an original room gray map, so that the obstacle boundary in the room map to be processed is closed, image processing is facilitated, and further the area map to be cleaned is ensured to be capable of expanding the cleaning coverage area, meanwhile, the area map to be processed is not required to be traversed, only the nearby region of a position point is required to be detected, the calculation force is reduced, the calculation efficiency is improved, the area map to be cleaned can be rapidly determined, the radius of a sweeper and the distance between walls are considered by the area map to be cleaned, and cleaning safety is ensured; the determined wall-following path comprehensively considers the factors such as the safety distance of the obstacle, the radius of the sweeper, the history cleaning record and the like, so that the sweeping safety and the sweeping effect can be ensured when the sweeper realizes the wall-following path.

Based on the same concept as the method embodiment of the present invention, please refer to fig. 3, the embodiment of the present invention further provides a cleaning path planning device of a sweeper, including:

an obtaining module 301, configured to obtain a map of a room to be processed corresponding to a room cleaned by a sweeper along a wall, where the map of the room is formed by the sweeper along the wall;

the map determining module 302 is configured to determine a map of an area to be cleaned according to the map of the room to be processed and the location point, where the location point is located within a boundary of the map of the area to be cleaned;

a path determining module 303, configured to determine a wall-following path of an obstacle in the map of the area to be cleaned;

and the cleaning module 304 is configured to plan a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle.

In one embodiment, the map determination module 302 includes: the system comprises a mapping unit, a searching unit, a selecting unit, an interpolation unit and a map determining unit; wherein,

the mapping unit is used for determining that the position points correspond to reference pixel points on the map of the room to be processed, and determining an area outside a closed area formed by the reference pixel points in the room to be processed as a legal area;

The searching unit is used for searching according to the sequence from the small radius to the large radius by taking the reference pixel point as the circle center so as to determine a candidate pixel point closest to the reference pixel point from the legal area, wherein the candidate pixel point is a pixel point on the boundary of an obstacle;

the selecting unit is used for determining a target pixel point from the pixel points on the connecting line of the reference pixel point and the candidate pixel point according to the radius of the sweeper;

the interpolation unit is used for interpolating the pixel points between the two adjacent target pixel points when the distance between the two adjacent target pixel points is greater than a distance threshold value, so as to obtain interpolation pixel points;

the map determining unit is used for determining a closed area formed by the target pixel points and the interpolation pixel points in the room map to be processed as a map of the area to be cleaned.

In one embodiment, the path determination module 303 includes: a first boundary determination unit, a second boundary determination unit, a dirty region determination unit, and a path determination unit; wherein,

the first boundary determining unit is used for determining the minimum boundary of the obstacle according to the acquired safety distance of the obstacle in the area map to be cleaned and the radius of the sweeper;

The second boundary determining unit is used for amplifying the minimum boundary according to the acquired sweeping speed interval of the sweeper around the obstacle so as to determine the maximum boundary of the obstacle;

the dirty area determining unit is used for estimating a dirty area corresponding to the obstacle according to the historical cleaning record of the sweeper, and the dirty area is positioned in an area between the minimum boundary and the maximum boundary;

the path determining unit is used for determining a wall-following path of the obstacle according to the dirty area corresponding to the obstacle, and the wall-following path passes through the dirty area.

In one embodiment, the path determining unit includes: a path planning subunit and a selection subunit; wherein,

the path planning subunit is configured to plan, according to a dirty area corresponding to the obstacle, a candidate path corresponding to the obstacle in an area between the minimum boundary and the maximum boundary, different candidate paths corresponding to different cleaning speeds, and the candidate paths pass through the dirty area;

and the selecting subunit is used for determining a candidate path matched with the cleaning path planning speed of the sweeper from the candidate paths, and determining the matched candidate path as the wall-following path of the obstacle.

In one embodiment, the first boundary determining unit includes: an acquisition subunit and a distance determination subunit; wherein,

the acquisition subunit is used for acquiring the confidence level of the obstacle boundary and the shape of the obstacle boundary in the map of the area to be cleaned;

the distance determining subunit is configured to determine a safe distance of the obstacle according to the confidence level of the obstacle boundary and the shape of the obstacle boundary.

In one embodiment, the acquisition module includes: the system comprises a map construction unit, a map processing unit, a first detection unit and a second detection unit; wherein,

the map construction unit is used for constructing an original room gray map corresponding to a room cleaned by the sweeper along a wall according to the environmental data acquired by the non-visual sensor on the sweeper;

the map processing unit is used for sequentially removing noise points of obstacle boundaries, extracting the obstacle boundaries and removing outlier areas in idle areas from the original room gray map so as to determine a reference room gray map;

the first detection unit is used for carrying out integrity detection on the reference room gray level map when a boundary line in the reference room gray level map is closed, and determining the reference room gray level map passing the integrity detection as a room map to be processed;

And the second detection unit is used for performing expansion processing on the non-closed boundary line in the reference room gray level map to close the boundary line when the boundary line in the reference room gray level map is not closed, performing integrity detection on the expanded reference room gray level map, and determining the expanded reference room gray level map passing the integrity detection as a room map to be processed.

In one embodiment, the free area in the room map to be processed is connected only with obstacle boundaries and is not connected with unknown areas.

Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. At the hardware level, the electronic device comprises a processor 401 and a memory 402 storing executable instructions, optionally together with an internal bus 403 and a network interface 404. The Memory 402 may include a Memory 4021, such as a Random-Access Memory (RAM), and may also include a nonvolatile Memory 4022 (non-volatile Memory), such as at least 1 disk Memory; the processor 401, the network interface 404, and the memory 402 may be interconnected by an internal bus 403, which internal bus 403 may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or an EISA (Extended Industry Standard Architecture ) bus, etc.; the internal bus 403 may be divided into an address bus, a data bus, a control bus, etc., and is represented by only one double-headed arrow in fig. 4 for convenience of illustration, but does not represent only one bus or one type of bus. Of course, the electronic device may also include hardware required for other services. When the processor 401 executes the execution instructions stored in the memory 402, the processor 401 performs the method in any one of the embodiments of the present invention and is at least used to perform the method as shown in fig. 1 or fig. 2.

In one possible implementation manner, the processor reads the corresponding execution instruction from the nonvolatile memory to the memory and then runs the execution instruction, and can also acquire the corresponding execution instruction from other devices to form a cleaning path planning device of the sweeper on a logic level. The processor executes the execution instructions stored in the memory to realize the cleaning path planning method of the sweeper provided by any embodiment of the invention through the execution instructions.

The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The embodiment of the invention also provides a computer readable storage medium, which comprises execution instructions, when the processor of the electronic device executes the execution instructions, the processor executes the method provided in any embodiment of the invention. The electronic device may specifically be an electronic device as shown in fig. 4; the execution instruction is a computer program corresponding to a cleaning path planning device of the sweeper.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method or a computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware aspects.

The embodiments of the present invention are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The foregoing is merely exemplary of the present invention and is not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are to be included in the scope of the claims of the present invention.

Claims

1. The cleaning path planning method of the sweeper is characterized by comprising the following steps of:

planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle;

the determining the map of the area to be cleaned according to the map of the room to be processed and the position point comprises the following steps:

determining a reference pixel point corresponding to the position point on the map of the room to be processed, and determining an area outside a closed area formed by the reference pixel point in the room to be processed as a legal area;

Searching according to the sequence from small radius to large radius by taking the reference pixel point as the circle center to determine a candidate pixel point closest to the reference pixel point from the legal area, wherein the candidate pixel point is a pixel point on the boundary of an obstacle;

determining a target pixel point from the pixel points on the connecting line of the reference pixel point and the candidate pixel point according to the radius of the sweeper;

when the distance between two adjacent target pixel points is larger than a distance threshold value, pixel point interpolation is carried out between the two adjacent target pixel points so as to obtain interpolation pixel points;

determining a closed area formed by the target pixel points and the interpolation pixel points in the room map to be processed as a map of the area to be cleaned;

the determining the wall-following path of the obstacle in the map of the area to be cleaned comprises:

determining the minimum boundary of the obstacle according to the acquired safety distance of the obstacle in the map of the area to be cleaned and the radius of the sweeper;

amplifying the minimum boundary according to the acquired sweeping speed interval of the sweeper around the obstacle to determine the maximum boundary of the obstacle;

Estimating a dirty area corresponding to the obstacle according to the historical cleaning record of the sweeper, wherein the dirty area is positioned in an area between the minimum boundary and the maximum boundary;

and determining a wall-following path of the obstacle according to the dirty area corresponding to the obstacle, wherein the wall-following path passes through the dirty area.

2. The method of claim 1, wherein determining the along-wall path of the obstacle from the dirty region corresponding to the obstacle comprises:

according to a dirty area corresponding to the obstacle, planning a candidate path corresponding to the obstacle in an area between the minimum boundary and the maximum boundary, wherein different candidate paths correspond to different cleaning speeds, and the candidate paths pass through the dirty area;

and determining a candidate path matched with the cleaning path planning speed of the sweeper from the candidate paths, and determining the matched candidate path as a wall-following path of the obstacle.

3. The method of claim 1, wherein the obtaining the safe distance of the obstacle in the map of the area to be cleaned comprises:

Acquiring the confidence level of the obstacle boundary in the map of the area to be cleaned and the shape of the obstacle boundary;

and determining the safety distance of the obstacle according to the confidence level of the obstacle boundary and the shape of the obstacle boundary.

4. The method of claim 1, wherein the acquiring the map of the room to be processed corresponding to the room swept by the sweeper along the wall comprises:

according to the environmental data collected by the non-visual sensor on the sweeper, constructing an original room gray map corresponding to a room cleaned by the sweeper along a wall;

sequentially removing noise points of obstacle boundaries, extracting the obstacle boundaries and removing outlier areas in idle areas from the original room gray map to determine a reference room gray map;

when an obstacle boundary in the reference room gray map is closed, performing integrity detection on the reference room gray map, and determining the reference room gray map passing the integrity detection as a room map to be processed;

and when the obstacle boundary in the reference room gray level map is not closed, performing expansion processing on the non-closed obstacle boundary in the reference room gray level map to enable the boundary line to be closed, performing integrity detection on the expanded reference room gray level map, and determining the expanded reference room gray level map passing the integrity detection as a room map to be processed.

5. The method according to claim 4, characterized in that the free area in the room map to be processed is connected only with obstacle boundaries and not with unknown areas.

6. A cleaning path planning apparatus for a sweeper, comprising:

the cleaning module is used for planning a cleaning path in the map of the area to be cleaned according to the wall-following path of the obstacle;

the map determining module comprises a mapping unit, a searching unit, a selecting unit, an interpolation unit and a map determining unit; wherein,

the map determining unit is used for determining a closed area formed by the target pixel points and the interpolation pixel points in the room map to be processed as a map of the area to be cleaned;

the path determining module comprises a first boundary determining unit, a second boundary determining unit, a dirty area determining unit and a path determining unit; wherein,

7. A computer readable storage medium comprising execution instructions which, when executed by a processor of an electronic device, perform the method of any one of claims 1 to 5.

8. An electronic device comprising a processor and a memory storing execution instructions that, when executed by the processor, perform the method of any of claims 1-5.