CN107544517B - Control method of intelligent cleaning robot - Google Patents

Abstract

The invention relates to a control method of an intelligent cleaning robot, which can record the current position and the cleaning state when the robot receives an external navigation instruction, so that the robot interrupts cleaning, can accurately find the position points of continuous cleaning according to the recorded current position and the external navigation path in the subsequent action process after walking according to the external navigation instruction, and determine the subsequent cleaning mode according to the recorded cleaning state, thereby avoiding the problem of missed cleaning or repeated cleaning, and improving the cleaning efficiency of the robot and the comprehensiveness and the integrity of the cleaning.

Description

Control method of intelligent cleaning robot

Technical Field

The invention relates to the field of robots, in particular to a control method of an intelligent cleaning robot.

Background

In the actual use process of the sweeping robot, a user needs to interrupt the current sweeping process many times, for example, the robot is navigated to a dirty place by using a remote controller to mainly sweep the area, because all robots do not have garbage detection sensors, the sweeping machine does not know which place is dirty particularly in the sweeping process, and therefore the sweeping machine does not need to mainly sweep the area. In addition, the user can navigate the robot to a certain area for cleaning, such as a certain room, through remote control, and therefore the user can operate flexibly. However, the sweeping robot in the market at present cannot be interrupted at will in the sweeping process, and because the map can be cleared after interruption, the original sweeping process cannot be recovered, so that the sweeping efficiency and the user experience of the robot can be reduced.

Disclosure of Invention

In order to solve the problems, the invention provides a control method of an intelligent cleaning robot, which can improve the cleaning efficiency of the robot and the comprehensiveness and integrity of cleaning. The specific technical scheme of the invention is as follows:

a control method of an intelligent cleaning robot comprises the following steps:

receiving an external navigation instruction;

recording the current position and the cleaning state;

walking according to the external navigation instruction, and recording the external navigation path of walking;

and receiving a functional instruction, and acting according to the functional instruction based on the recorded current position and cleaning state and the external navigation path.

Further, the recording the current position includes the following steps:

recording XY coordinate information of the current position point;

and recording the angle information of the current position point.

Further, the cleaning state is recorded, and the method comprises the following steps:

judging whether the current sweeping stage belongs to a local sweeping stage;

if so, recording a local cleaning stage, recording a cleaning range of a current local area and a cleaning range of a current global area, recording whether the current state is in a bow-shaped cleaning state or a barrier-bypassing cleaning state, and recording a cleaning direction in the bow-shaped cleaning state or a barrier-bypassing direction in the barrier-bypassing cleaning state;

if not, judging whether the current sweeping stage belongs to a global sweeping stage or not;

if so, recording as a global cleaning stage, and recording the cleaning range of the current global area;

if not, recording as a global edge stage, and recording the starting position point of the edge, the angle at the starting position point, the path along the edge and the current edge direction.

Further, the recording of the sweeping range of the current local area includes the following steps:

determining a first leftmost position point positioned at the leftmost side, a first rightmost position point positioned at the rightmost side, a first uppermost position point positioned at the uppermost end and a first lowermost position point positioned at the lowermost end in the walking path based on the walking path of the robot in the current local area;

establishing a virtual first leftmost vertical line based on the first leftmost position point, establishing a virtual first rightmost vertical line based on the first rightmost position point, establishing a virtual first uppermost horizontal line based on the first uppermost position point, and establishing a virtual first lowermost horizontal line based on the first lowermost position point;

recording an area surrounded by the first leftmost vertical line, the first rightmost vertical line, the first uppermost horizontal line and the first lowermost horizontal line as a sweeping range of the current local area;

and/or the presence of a gas in the gas,

the method for recording the sweeping range of the current global area comprises the following steps:

determining a second leftmost position point located at the leftmost side, a second rightmost position point located at the rightmost side, a second uppermost position point located at the uppermost end, and a second lowermost position point located at the lowermost end in the walking path based on the walking path of the robot in the current global area;

establishing a virtual second leftmost vertical line based on the second leftmost position point, establishing a virtual second rightmost vertical line based on the second rightmost position point, establishing a virtual second uppermost horizontal line based on the second uppermost position point, and establishing a virtual second lowermost horizontal line based on the second lowermost position point;

recording an area surrounded by the second leftmost vertical line, the second rightmost vertical line, the second uppermost horizontal line and the second lowermost horizontal line as a sweeping range of the current global area.

Further, the receiving a function instruction, acting according to the function instruction based on the recorded current position and cleaning state and the external navigation path, includes the following steps:

receiving a functional instruction, and judging whether the functional instruction is a key cleaning instruction or not;

if yes, performing key cleaning;

if not, judging whether the functional instruction is a current point cleaning instruction or not;

if so, cleaning is started from the current position point;

if not, judging whether the functional instruction is an automatic recharging instruction or not;

if yes, automatically returning to the seat for charging;

if not, judging whether the functional instruction is a recording point continuous scanning instruction or not;

if so, starting subsequent cleaning from the current position or returning to the recorded current position to start subsequent cleaning according to the recorded cleaning state;

if not, the next functional instruction is waited for.

Further, the performing of the focused cleaning includes the following steps:

taking a current position point as a first reference point, walking forward from the first reference point along a first direction for a set distance, then backing to the first reference point, then backing from the first reference point along a second direction opposite to the first direction for a set distance, and then walking forward to the first reference point;

walking from the first reference point along a third direction perpendicular to the first direction for a set width, then taking the current position point as a second reference point, walking from the second reference point forward along the first direction for a set distance, then backing to the second reference point, then backing from the second reference point along the second direction for a set distance, and then walking forward to the second reference point;

walking twice the set width in a fourth direction opposite to the third direction from the second reference point, then walking forward for a set distance in the first direction from the third reference point by taking the current position point as a third reference point, then backing to the third reference point, then backing for a set distance in the second direction from the third reference point, and then walking forward to the third reference point;

walking the set width three times along the third direction from the third reference point, then walking the set distance forward along the first direction from the fourth reference point by taking the current position point as the fourth reference point, then backing to the fourth reference point, backing the set distance along the second direction from the fourth reference point, and then walking the set distance forward to the fourth reference point;

walking the set width four times along the fourth direction from the fourth reference point, then walking the set distance forward along the first direction from the fifth reference point by taking the current position point as the fifth reference point, then backing to the second direction from the fifth reference point by the set distance, and then walking the set distance forward to the fifth reference point;

in this way, until the distance from the first reference point to the third direction reaches N times of the set width, the key cleaning is finished;

wherein N is a natural number greater than 1.

Further, after the key point cleaning is finished, the method further comprises the following steps: and according to the recorded cleaning state, starting to perform subsequent cleaning from the current position, or returning to the recorded current position to perform subsequent cleaning.

Further, if the robot detects an obstacle or cliff during the forward walking in the first direction, directly backing to the corresponding reference point;

if the robot detects an obstacle or cliff during the backward movement in said second direction, it walks straight forward to the corresponding reference point.

Further, the cleaning from the current position point includes the following steps:

determining an initial cleaning angle from an initial position point when the robot initially starts cleaning, based on the initial position point;

adjusting the current cleaning angle of the robot at the current position point to enable the current cleaning angle to be the same as the initial cleaning angle;

starting the arch-shaped cleaning, and stopping cleaning the current route and switching to the next route of the arch-shaped to continue cleaning if M continuous grid units are detected as cleaned units in the arch-shaped cleaning process;

wherein M is a natural number greater than or equal to 3.

Further, the cleaning process starting from the current position or returning to the current position for subsequent cleaning according to the recorded cleaning state includes the following steps:

judging whether the recorded cleaning state is a local cleaning stage, if so, returning the robot to the recorded current position according to the recorded XY coordinate information and angle information, judging whether the recorded current state is in a bow-shaped cleaning state, if so, continuing cleaning according to the recorded cleaning direction, otherwise, determining that the recorded current state is in an obstacle-detouring cleaning state, and continuing cleaning according to the recorded obstacle-detouring direction;

if not, judging whether the recorded cleaning state is a global cleaning stage or not, if so, searching a global map by the robot, and then automatically navigating to the searched missed cleaning area from the current position to continue cleaning;

if not, determining that the recorded cleaning state is a global edgewise stage, returning the robot to the recorded current position according to the recorded XY coordinate information and angle information, and then continuing cleaning according to the recorded edgewise direction.

Further, the robot searches a global map, automatically navigates to the searched missed-scanning area from the current position and continues to perform cleaning, and the method comprises the following steps:

determining entry boundaries of the cleaned area and the uncleaned area based on the searched global map;

respectively taking end points at two ends of each entry boundary as two entry reference points;

establishing an XY axis coordinate system by taking the current position as a coordinate origin; analyzing the coordinate positions of two inlet reference points of the same inlet boundary; when the X-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the Y-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the X-axis value and the Y-axis value of the two inlet reference points are different, judging the position relation between the non-cleaned area and the cleaned area; selecting the entry reference point having a large X-axis value as an optimal entry point if the unswept area is to the left of the swept area; selecting said entry reference point having a small X-axis value as an optimal entry point if the unswept area is to the right of the swept area;

selecting one optimal entry point closest to the current position as a priority cleaning reference point in an unswept area, and selecting two priority cleaning reference points closest to the current position; judging whether the difference value between the current position and the two priority cleaning reference points is smaller than a preset distance value; if so, selecting the priority cleaning reference point corresponding to the inlet boundary with longer length as a cleaning starting point; if not, selecting the priority cleaning reference point closest to the current position as a cleaning starting point;

walking from the current position to the cleaning starting point, and judging the position relation of two end points of the inlet boundary where the cleaning starting point is located; when the X-axis values of the two end points are the same, directly walking towards the other end point of the inlet boundary from the cleaning starting point, and cleaning an uncleaned area according to a zigzag track form; when the Y-axis values of the two endpoints are the same; if the non-cleaned area is positioned above the inlet boundary, walking along the positive direction of the Y axis from the cleaning starting point, and cleaning the non-cleaned area according to the track form of the Chinese character 'gong'; if the non-cleaning area is positioned below the inlet boundary, walking along the Y-axis negative direction from the cleaning starting point, and cleaning the non-cleaning area in a zigzag track form; when the X-axis value and the Y-axis value of the two end points are different, walking along the direction vertical to the X-axis of the other end point from the cleaning starting point, and cleaning an uncleaned area according to a track form of a Chinese character 'gong';

and after the sweeping is finished, determining the next optimal entry point as the sweeping starting point of the next preferential sweeping area, and repeating the steps until all the non-sweeping areas meeting the sweeping conditions of the non-sweeping areas are completely swept, and finishing the global sweeping stage.

Further, after the step of continuing the cleaning in the recorded edgewise direction, the method further includes the steps of:

judging whether the sum of the length of the path traveled when the edgewise cleaning is continued and the length of the path already along is greater than or equal to the total perimeter; judging whether the difference between the current angle of the robot and the recorded angle at the initial position point is larger than 360 degrees or not in the process of continuing the edgewise cleaning; judging whether the current position point of the robot is within the preset range of the recorded initial position point of the edge in the process of continuing the edge cleaning;

if the judgment results are all yes, the edge is ended;

if the judgment result is negative, continuing the edgewise cleaning, and if the difference between the current angle of the robot and the recorded angle at the initial position point is more than 540 degrees in the process of continuing the edgewise cleaning, or the sum of the path length traveled when continuing the edgewise cleaning and the path length already along the edge is more than 2.5 times of the total circumference, finishing the edgewise cleaning.

The invention has the beneficial effects that: when the robot receives an external navigation instruction, the current position and the cleaning state can be recorded, so that the robot intermittently cleans, after walking according to the external navigation instruction, the position points which are continuously cleaned can be accurately found according to the recorded current position and the external navigation path in the subsequent action process, and the subsequent cleaning mode is determined according to the recorded cleaning state, so that the problem of missing cleaning or repeated cleaning is avoided, and the cleaning efficiency of the robot and the cleaning comprehensiveness and integrity are improved.

Drawings

Fig. 1 is a flowchart illustrating a control method of an intelligent cleaning robot according to the present invention.

Fig. 2 is a schematic diagram of recording the current sweeping area of the local area according to the present invention.

FIG. 3 is a schematic view of the focused cleaning according to the present invention.

FIG. 4 is a schematic diagram of a global sweeping phase according to the present invention.

Detailed Description

The following further describes embodiments of the present invention with reference to the accompanying drawings:

the robot mainly refers to a floor sweeping robot, also called a cleaning robot, an automatic cleaner, an intelligent dust collector and the like, is one of intelligent household appliances, and can automatically finish floor cleaning work in a room by means of certain artificial intelligence. Generally, the floor cleaning machine adopts a brushing and vacuum mode, and firstly absorbs the impurities on the floor into the garbage storage box, so that the function of cleaning the floor is achieved. Generally, robots that perform sweeping, dust collection, floor wiping, and the like are also collectively called sweeping robots. The body of the sweeping robot is a wireless machine, mainly in a disc shape. The rechargeable battery is used for operation, and the operation mode is remote control or an operation panel on the machine. Generally, the time can be set for cleaning in a reserved mode, and the automatic charging is realized. The machine body is provided with various sensors which can detect the walking distance, the walking angle, the machine body state, obstacles and the like, and can turn automatically when touching a wall or other obstacles, and walk different routes according to different settings, thereby cleaning the area in a planned place. The robot at least comprises the following structures: the robot body with the driving wheels and capable of walking independently is provided with a human-computer interaction interface, the periphery of the robot body is provided with an obstacle detection unit (which can be an infrared sensor or an ultrasonic sensor and the like), and the lower part of the robot body is provided with a main brush (also called a middle brush) and a side brush. The body is internally provided with an inertial sensor comprising an accelerometer, a gyroscope and the like, the driving wheel is provided with a speedometer (generally a coded disc) for detecting the walking distance of the driving wheel, and the body is also provided with a control module which can process the parameters of the related sensor and can output a control signal to an execution component.

As shown in fig. 1, the control method of the intelligent cleaning robot according to the present invention includes the following steps: receiving an external navigation instruction; recording the current position and the cleaning state; walking according to the external navigation instruction, and recording the external navigation path of walking; and receiving a functional instruction, and acting according to the functional instruction based on the recorded current position and cleaning state and the external navigation path. The external navigation instruction may be a navigation instruction generated by a direction key on a robot remote controller, or a navigation instruction generated by a smart phone. Through the mode, the robot can record the current position and the cleaning state when receiving an external navigation instruction, so that the robot can intermittently clean, can accurately find the position points to be continuously cleaned according to the recorded current position and the external navigation path in the subsequent action process after walking according to the external navigation instruction, and determine the subsequent cleaning mode according to the recorded cleaning state, thereby avoiding the problem of missing cleaning or repeated cleaning, and improving the cleaning efficiency of the robot and the cleaning comprehensiveness and integrity.

Preferably, the recording the current position includes the following steps: recording XY coordinate information of the current position point; and recording the angle information of the current position point. Based on the XY coordinate system in the map, the robot records the walking distance detected by the odometer and the angle value detected by the gyroscope while walking, and converts the detected walking distance and angle value into XY coordinate information and angle information. By recording XY coordinate information and angle information of the current position point, the robot can quickly and accurately find the position point which is continuously swept in the subsequent movement, and the problem of repeated sweeping or missing sweeping caused by inaccurate position points which are continuously swept is avoided.

Preferably, the recording of the cleaning state includes the steps of: judging whether the current sweeping stage belongs to a local sweeping stage; if so, recording as a local cleaning stage, recording the cleaning range of the current local area and the cleaning range of the current global area, recording whether the current state is in a bow-shaped cleaning state or in a barrier cleaning state, and recording the cleaning direction in the bow-shaped cleaning state or the barrier cleaning direction in the barrier cleaning state. If not, judging whether the current sweeping stage belongs to a global sweeping stage or not; if yes, recording as a global sweeping stage, and recording a sweeping range of the current global area. If not, recording as a global edge stage, and recording the starting position point of the edge, the angle at the starting position point, the path along the edge and the current edge direction. Because the robot finishes one-time complete cleaning, local cleaning, global cleaning and global edge stages are needed, and the cleaning modes of the robot are different in different stages, on the basis of recording the position points of the robot to be continuously cleaned, the cleaning stage in which the robot is located at the moment is also needed to be recorded, and thus the problem of repeated cleaning or missed cleaning can be further avoided. After the cleaning stage is determined, some relevant data of the current stage also needs to be recorded, otherwise, after the robot reaches the continuous-scanning position point, it is still unknown how to start cleaning, and thus, the phenomenon of repeated cleaning or missed cleaning is caused due to the disordered scanning of the robot.

Preferably, the recording of the current sweeping range of the local area includes the following steps: determining a first leftmost position point positioned at the leftmost side, a first rightmost position point positioned at the rightmost side, a first uppermost position point positioned at the uppermost end and a first lowermost position point positioned at the lowermost end in the walking path based on the walking path of the robot in the current local area; establishing a virtual first leftmost vertical line based on the first leftmost position point, establishing a virtual first rightmost vertical line based on the first rightmost position point, establishing a virtual first uppermost horizontal line based on the first uppermost position point, and establishing a virtual first lowermost horizontal line based on the first lowermost position point; recording an area surrounded by the first leftmost vertical line, the first rightmost vertical line, the first uppermost horizontal line and the first lowermost horizontal line as a sweeping range of the current local area. As shown in fig. 2, the grid cells where the robot hits an obstacle are indicated by small squares marked with X, the largest rectangular frame on the outer periphery indicates the range of a complete local area, and the robot performs cleaning along the zigzag path (the line with the arrow) of ABCDEFGH. A first leftmost position A located at the leftmost side, a first rightmost position H located at the rightmost side, a first uppermost position G located at the uppermost end, and a first lowermost position E located at the lowermost end are determined in the walking path of the bow-shaped figure. And then establishing a virtual first leftmost vertical line K1K2 based on the first leftmost position point A, establishing a virtual first rightmost vertical line K3K4 based on the first rightmost position point H, establishing a virtual first uppermost horizontal line K1K4 based on the first uppermost position point G, and establishing a virtual first lowermost horizontal line K2K3 based on the first lowermost position point E. Recording the square area formed by K1, K2, K3 and K4 as the sweeping range of the current local area. When the robot returns to the position point of continuous sweeping again, the robot continues to sweep along the current sweeping direction (from left to right), and sweeps to the right boundary of the local area, and then the sweeping missing place in the local area (namely the blank area without the zigzag path behind the obstacle in the figure) is swept again, so as to complete the overall sweeping of the local area. The cleaning range which is cleaned is determined by the position point which is positioned at the extreme, and the robot can directly search the position and the path in the determined area, so that the map searching efficiency of the robot is improved, meanwhile, the robot cannot repeatedly clean during continuous cleaning, and the cleaning efficiency of the robot is improved.

Similarly, the step of recording the cleaning range of the current global area comprises the following steps: determining a second leftmost position point located at the leftmost side, a second rightmost position point located at the rightmost side, a second uppermost position point located at the uppermost end, and a second lowermost position point located at the lowermost end in the walking path based on the walking path of the robot in the current global area; establishing a virtual second leftmost vertical line based on the second leftmost position point, establishing a virtual second rightmost vertical line based on the second rightmost position point, establishing a virtual second uppermost horizontal line based on the second uppermost position point, and establishing a virtual second lowermost horizontal line based on the second lowermost position point; recording an area surrounded by the second leftmost vertical line, the second rightmost vertical line, the second uppermost horizontal line and the second lowermost horizontal line as a sweeping range of the current global area. After the robot is charged, a map is searched from the current position, the sweeping range which is swept in the global area is determined through the extreme position points, the robot can directly search the position and the path in the determined area, the global area in the plan is not required to be searched completely, and therefore the map searching efficiency of the robot is improved. Moreover, by recording the determined area, the robot can not repeatedly clean when continuously sweeping, thereby improving the cleaning efficiency of the robot.

Preferably, the receiving a function instruction, and acting on the function instruction based on the recorded current position and cleaning state and the external navigation path, includes the steps of: receiving a functional instruction, and judging whether the functional instruction is a key cleaning instruction or not; if yes, performing key cleaning; if not, judging whether the functional instruction is a current point cleaning instruction or not; if so, cleaning is started from the current position point; if not, judging whether the functional instruction is an automatic recharging instruction or not; if yes, automatically returning to the seat for charging; if not, judging whether the functional instruction is a recording point continuous scanning instruction or not; if so, starting subsequent cleaning from the current position or returning to the recorded current position to start subsequent cleaning according to the recorded cleaning state; if not, the next functional instruction is waited for. The functional instruction can be customized, and different functional items can be added or deleted according to different requirements. And the robot receives different functional instructions and correspondingly executes different functions according to the recorded current position, the recorded cleaning state and the external navigation path.

Preferably, as shown in fig. 3, the performing of the focused cleaning includes the following steps: with the current position point as a first reference point P1, walking forward from the first reference point P1 in a first direction S1 for a set distance, then backing to the first reference point P1, then walking backward from the first reference point P1 in a second direction S2 opposite to the first direction S1 for a set distance, and then walking forward to the first reference point P1; walking a set width in a third direction S3 perpendicular to the first direction S1 from the first reference point P1, then walking forward a set distance in the first direction S1 from the second reference point P2 with the current position point as a second reference point P2, then backing to the second reference point P2, then backing a set distance in the second direction S2 from the second reference point P2, and then walking forward to the second reference point P2; walking twice the set width in a fourth direction S4 opposite to the third direction S3 from the second reference point P2, then walking forward a set distance in the first direction S1 from the third reference point P3 with the current position point being the third reference point P3, then backing to the third reference point P3, then backing a set distance in the second direction S2 from the third reference point P3, then walking forward to the third reference point P3; walking three times the set width in the third direction S3 from the third reference point P3, then walking forward a set distance in the first direction S1 from the fourth reference point P4 with the current position point as a fourth reference point P4, then backing to the fourth reference point P4, then backing a set distance in the second direction S2 from the fourth reference point P4, and then walking forward to the fourth reference point P4; walking four times the set width in a fourth direction S4 starting from the fourth reference point P4, then walking forward a set distance in a first direction S1 starting from the fifth reference point P5 with the current position point being a fifth reference point P5, then backing to the fifth reference point P5, then backing a set distance in the second direction S2 starting from the fifth reference point P5, then walking forward to the fifth reference point P5; by analogy, the key cleaning is finished until the distance from the first reference point P1 to the third direction S3 reaches N times of the set width. The set distance and the set width can be set differently according to different requirements, the set distance can be set between 0.5m and 1.5m, and the set width can be set between 0.1m and 0.3 m. The N is a natural number which is larger than 1, the N can be set according to actual requirements, the N is generally based on the ground of a conventional household, and the N can be set to be 3. When the area that needs to be cleaned with emphasis (i.e., the dirty area) is large, the user may manually input the values of the set distance and the set width and the value of N according to the actual area. In addition, the robot adopts the mode of advancing backward, can greatly improve and clean efficiency, avoids frequently turning to the time waste that brings, simultaneously, reciprocal walking cleans and can improve the cleanliness factor of cleaning. Of course, besides the above-mentioned method for cleaning the key points, a spiral method can be used for cleaning the key points, that is, the current point is used as the original point, the cleaning is gradually and spirally conducted outwards, and after the current point reaches a certain radius, the cleaning is gradually and spirally conducted inwards according to the original path to return to the original point, so that the cleaning is repeated, and the key cleaning area can be cleaned completely.

Wherein, after the key cleaning is finished, the method also comprises the following steps: and according to the recorded cleaning state, starting to perform subsequent cleaning from the current position, or returning to the recorded current position to perform subsequent cleaning. The specific implementation of this step is the same as the scanning mode after receiving the recording dot scanning command described later.

Wherein if the robot detects an obstacle or cliff (i.e. a place where the finger is hovering) while walking forward in the first direction S1, the robot directly moves backward to the corresponding reference point. If the robot detects an obstacle or cliff during the retreat in the second direction S2, it directly walks forward to the corresponding reference point. In the process of important cleaning, the robot cleans a specific range with a lot of garbage, and the range is generally small, so that the problem of whether to bypass an obstacle or not is not required to be considered like local area cleaning and global area cleaning, and the efficiency of important cleaning is further improved.

Preferably, the cleaning from the current position point includes the steps of: determining an initial cleaning angle from an initial position point when the robot initially starts cleaning, based on the initial position point; adjusting the current cleaning angle of the robot at the current position point to enable the current cleaning angle to be the same as the initial cleaning angle; and starting the arch-shaped cleaning, and stopping cleaning the current route and switching to the next route of the arch shape to continue cleaning if M continuous grid units are detected as cleaned units in the arch-shaped cleaning process. Wherein M is a natural number greater than or equal to 3. When the robot receives an instruction of starting to clean at the current position, the robot means to start to clean again from the position, however, a map of an area which is originally swept still needs to be kept, all azimuth information needs to be inherited, otherwise, after the robot restarts to clean, the robot cleans the area which is originally swept again, and the cleaning efficiency is low. Therefore, the method of the invention records the initial position point and the initial cleaning angle when the cleaning is started initially, so that the subsequent cleaning is carried out again according to the same cleaning angle, the integration of the current cleaning mode and the original cleaning mode can be realized, and the formed map can be continued, thereby effectively avoiding the problem of repeated cleaning and greatly improving the cleaning efficiency. For example, the starting point position when cleaning is started initially is (x0, y0, theta0), the starting point position when cleaning is started now is (x1, y1, theta1), and theta1 is set to theta0 in order to ensure inheritance of the map. Where theta represents the angle value detected by the gyroscope of the robot. When the zigzag type is cleaned, the robot judges whether 3 continuous grid units (3 is the value of M, and 3 can be set as other values according to different requirements) in front are swept or not, if the grid units are swept (the swept grid units are marked as the swept units in a map by the robot), the robot stops continuing the forward sweeping, and the next route shifted to the zigzag type is continuously swept to avoid repeated sweeping.

Preferably, the cleaning device further includes a cleaning device for performing subsequent cleaning from a current position or returning to the current position according to the recorded cleaning state, the cleaning device including: judging whether the recorded cleaning state is a local cleaning stage, if so, returning the robot to the recorded current position point according to the recorded XY coordinate information and angle information, judging whether the recorded current state is in a bow-shaped cleaning state, if so, continuing cleaning according to the recorded cleaning direction, otherwise, determining that the recorded current state is in an obstacle-detouring cleaning state, and continuing cleaning according to the recorded obstacle-detouring direction; if not, judging whether the recorded cleaning state is a global cleaning stage or not, if so, searching a global map by the robot, and then automatically navigating to the searched missed cleaning area from the current position to continue cleaning; if not, determining that the recorded cleaning state is a global edgewise stage, returning the robot to the recorded current position point according to the recorded XY coordinate information and angle information, and then continuing cleaning according to the recorded edgewise direction. Because the cleaning modes adopted by the robot are different in different cleaning stages, the method controls the robot to subsequently adopt corresponding cleaning actions through the recorded stages so as to achieve the high efficiency of the robot cleaning.

Specifically, if it is determined that local cleaning is currently required according to the recorded information, when the robot reaches a continuous-cleaning position point (i.e., the recorded current position point), based on the recorded cleaning range of the current local area, other non-cleaned places are continuously cleaned on the basis of the cleaning range, and after completion of the zigzag cleaning of the entire local area, the missed-cleaning places in the local area are subjected to supplementary cleaning. And based on the recorded cleaning range of the current global area, after the cleaning of the current local area is finished, other local areas which are not cleaned in the global area are continuously cleaned. If the cleaning range of the current local area and the cleaning range of the current global area are not recorded, the robot cannot judge which places are cleaned and which places are not cleaned in the current local area and the current global area, and thus subsequent cleaning work cannot be effectively carried out. In addition, whether the recorded current state is in a bow-shaped cleaning state or in a barrier-bypassing cleaning state is also determined; if the arc-shaped cleaning state is achieved, the cleaning direction in the arc-shaped cleaning state is recorded, and if the obstacle-detouring cleaning state is achieved, the obstacle-detouring direction in the obstacle-detouring cleaning state is recorded. If these specific states and directions are not recorded, the robot cannot determine how to go and where to go next. After the record is made, the robot will not be blindly cleaned. For example, if the cleaning direction is from left to right, the robot turns right when turning around during the zigzag cleaning, and turns left when turning around if the cleaning direction is from right to left. If the obstacle detouring direction is from the left side of the obstacle, the robot continues to clean along the left side of the obstacle; if the obstacle detouring direction is from the right side of the obstacle, the robot continues to sweep along the right side of the obstacle.

Specifically, if it is determined that global sweeping is currently required according to the recorded information, the global map is searched, based on the recorded sweeping range of the current global area, the navigation is automatically performed to the searched missed-sweeping area from the current position, and other areas (i.e., missed-sweeping areas) in the global area are continuously swept. If the cleaning range of the current global area is not recorded, the robot cannot judge which places are cleaned and which places are not cleaned in the current global area, and therefore subsequent cleaning work cannot be effectively carried out. In the global cleaning stage, the robot does not need to return to the recorded position point, but directly starts to perform subsequent cleaning from the current position, namely directly performs map search at the current position and then navigates to the searched missed-cleaning area to perform cleaning. The cleaning efficiency of the robot can be further improved through the control mode, and in the global cleaning stage, the main task of the robot is to search out the area which is not cleaned and clean, so that the area which is not cleaned is finally searched out and cleaned no matter where the area is searched, the area which is not cleaned is directly searched at the current position and searched again compared with the area which is searched back to the recorded position, the link of walking from the current position to the recorded position is reduced, the cleaning time is shortened, and the cleaning efficiency is improved.

Specifically, if it is determined that global edgewise operation is currently required according to the recorded information, when the robot reaches a continuous scanning position point (i.e., the recorded current position point), it is determined how long the robot is edgewise based on the recorded starting position point of the edgewise operation and the already-edgewise path, and it is accurately determined which direction the robot should continue to edgewise next according to the recorded current edgewise direction. And in the process of continuing the edge, judging the difference value between the current angle of the robot and the angle according to the angle at the initial position point, and taking the difference value as one of the conditions for judging whether to finish the edge. If the starting position point of the edge and the path which is already along the edge are not recorded, the length of the edge of the robot cannot be known; if the direction along the edge is not recorded, the direction along the edge cannot be known; if the angle at the starting position point is not recorded, whether the robot has a circle along the edge cannot be judged. Therefore, according to the recorded information, the robot can be helped to better complete the subsequent edge work.

As shown in fig. 4, the line with an arrow represents the travel locus of the robot. And a maximum rectangle formed by four edges at the outermost side represents the boundary of the global map. The position point B is the current position of the robot, and the arrow-shaped line is the cleaning track of the robot. The zigzag cleaning means that when the robot travels to a turning point along a straight line moving path, the robot travels by a certain width after turning 90 degrees, then turns 90 degrees again to enable the current traveling direction to be opposite to the original straight line moving path, and then continues to travel to the next turning point. The trajectory of the robot travelling in this way is similar to a Chinese character bow, and is called Chinese character bow cleaning.

Preferably, the robot searches a global map, and then automatically navigates to the searched missed-scan area from the current position to continue to perform cleaning, comprising the following steps:

in a first step, based on the global map searched, the entry boundaries of the cleaned area and the uncleaned area (i.e., the missed-sweep area, the same applies below) are determined. In fig. 4, the area occupied by the zigzag line is a cleaned area, and the area within the dotted line frame is an uncleaned area. After the conventional cleaning is completed, the non-cleaned area needs to be cleaned again, so the entrance boundaries of the cleaned area and the non-cleaned area (i.e. the line segment from d1 to d2, the line segment from c1 to c2, the line segment from c3 to c4, etc.) are determined first.

And secondly, respectively taking the end points of the two ends of each inlet boundary as two inlet reference points. As shown in fig. 4, a point d1, a point d2, a point c1, a point c2, a point c3, a point c4, and the like are taken as inlet reference points.

Thirdly, establishing an XY axis coordinate system by taking the current position as a coordinate origin; analyzing the coordinate positions of two inlet reference points of the same inlet boundary; when the X-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the Y-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the X-axis value and the Y-axis value of the two inlet reference points are different, judging the position relation between the non-cleaned area and the cleaned area; selecting the entry reference point having a large X-axis value as an optimal entry point if the unswept area is to the left of the swept area; selecting said entry reference point with a small X-axis value as the optimal entry point if the unswept area is to the right of the swept area. As shown in fig. 4, the current position is point B, an XY axis coordinate system is established with point B as an origin, and coordinate positions of a1 point and a3 point, a4 point and a5 point of two entry boundaries a1-a3 point and a4-a5 point (which may be an entry boundary because a1-a2 boundary and a2-a3 boundary are connected with each other) of area a in the uncleaned area are analyzed, and since X axis values and Y axis values of a1 point and a3 point with respect to the origin of coordinates (point B) are different and area a is on the left side of the cleaned area, a1 point with a large X axis value is selected as the best entry point (since X axis values of a1 point and a3 point are both negative, X axis values are larger, the closer to the origin in the X axis direction are larger). Since the Y-axis values of the a4 point and the a5 point are the same, the a5 point closest to the current position point is selected as another optimal entry point of the a-region. Similarly, point b2, point b4 and point b8 are selected as the optimal entry points of the b-region; c1 point, c5 point and c4 point were selected as the best entry points for the c region; the point d1 is selected as the best entry point for the d-field. By the aid of the entry selection conditions, each entry boundary of each uncleaned area can be analyzed, and an optimal entry point is selected as a selection object of a subsequent priority cleaning reference point, so that even if the shape of the uncleaned area is complex, which position point is more appropriate to be used as the optimal entry point can be comprehensively and effectively analyzed, and a more objective and more accurate object is provided for selection of the priority cleaning reference point.

And fourthly, selecting the optimal entry point closest to the current position in an unswept area as a priority sweeping reference point. Selecting two priority cleaning reference points which are closest to the current position; judging whether the difference value between the current position and the two priority cleaning reference points is smaller than a preset distance value; if so, selecting the priority cleaning reference point corresponding to the inlet boundary with longer length as a cleaning starting point; if not, selecting the priority cleaning reference point closest to the current position as the cleaning starting point. By analyzing the two preferential cleaning reference points which are closest to the current position and combining the length of the entrance boundary, the area which is more suitable for preferential cleaning can be determined more comprehensively and more accurately, the problem that the cleaning efficiency is not high due to the fact that the prior preferential cleaning area is determined only by the distance is solved, and therefore the cleaning efficiency of the robot is improved. For example, there are two regions that are relatively close to the current location point, where the entry boundary of the nearest region is short in length, typically indicating that the area of the region is small, and the entry boundary of the other region is long, typically indicating that the area of the region is large. If the robot only considers the distance, the robot firstly goes to the area with short entrance boundary length, so that the cleaning efficiency is low due to the fact that the large-area is not cleaned in time, the product use experience of a user is affected, and when the robot firstly goes to the entrance boundary with short entrance boundary length, the cleaning starting point needs to be found in a long time due to map errors or walking errors and the like. Therefore, the most efficient method is to preferentially clean a region having a long entrance boundary when determining two regions having relatively close distances, and to clean the region having the closest distance only when the distance between the two regions and the current position is significantly different. If the robot does not clean the area with a relatively close side, the robot runs a long distance to clean the area with a relatively large area, the cleaning efficiency of the robot is reduced because too much time is spent on the walking route, and meanwhile, a user feels that the robot is relatively clumsy and not intelligent enough, so that the use experience of products is reduced. As shown in fig. 4, after the optimal entry points of each of the unswept areas are determined, a1 point closest to the B point among the optimal entry points of the a area (a1 point and a5 point) is selected as a preferential sweeping reference point; selecting a point b4 in the area b as a preferential sweeping reference point; selecting a point c1 in the zone c as a preferential sweeping reference point; the point d1 is selected as the preferential sweeping reference point in the d zone. Then two priority cleaning reference points, namely a1 point and a B4 point, which are closest to the point B are selected, and since the difference between the distance from the point a1 to the point B and the distance from the point B4 to the point B is greater than a preset distance value, the point a1 closest to the point B is selected as a cleaning starting point. The preset distance value is a settable value, different parameter values can be set according to specific requirements, and the body widths of two robots are selected as the preset distance value in the embodiment.

Fifthly, walking from the current position to the cleaning starting point, and judging the position relation of two end points of the inlet boundary where the cleaning starting point is located; when the X-axis values of the two end points are the same, directly walking towards the other end point of the inlet boundary from the cleaning starting point, and cleaning an uncleaned area according to a zigzag track form; when the Y-axis values of the two endpoints are the same; if the non-cleaned area is positioned above the inlet boundary, walking along the positive direction of the Y axis from the cleaning starting point, and cleaning the non-cleaned area according to the track form of the Chinese character 'gong'; if the non-cleaning area is positioned below the inlet boundary, walking along the Y-axis negative direction from the cleaning starting point, and cleaning the non-cleaning area in a zigzag track form; when the X-axis value and the Y-axis value of the two end points are different, the cleaning machine walks along the direction vertical to the X-axis of the other end point from the cleaning starting point, and cleans the non-cleaned area in a bow-shaped track mode. As shown in fig. 4, when the a zone is cleaned, it is determined that the positional relationship between the a1 point and the a3 point is different, and the X-axis value and the Y-axis value are different between the a1 point and the a3 point, so that the a zone is cleaned from right to left in a manner of a bow-shaped trajectory by traveling in the X-axis direction (i.e., the a2 point) in which the a1 point is perpendicular to the a3 point. By judging the position relationship of two points on the boundary of the entrance and correspondingly adopting different cleaning modes, the problem that the corners of an uncleaned area with a complex shape are missed to be cleaned can be effectively avoided, so that the cleaning comprehensiveness and integrity of the robot are improved.

And sixthly, after the sweeping is finished, determining the next optimal entry point as the sweeping starting point of the next preferential sweeping area, and repeating the steps until all the unscreened areas meeting the sweeping conditions of the unscreened areas are swept, and ending the global sweeping stage. As shown in fig. 4, when the a-zone is cleaned, the robot walks to the cleaning end point (i.e., a5 point), at this time, the a5 point is used as the current position point (i.e., corresponds to a new current position), the d1 point, the c1 point, the c4 point, the c5 point, the b1 point, the b3 point, and the b8 point are selected as the optimal entry points according to the entry selection conditions, and the d1 point is determined as the cleaning start point according to the priority cleaning conditions. When the robot travels from the point a5 to the point d1, the cleaning of the area is completed when the area d is cleaned from right to left to the end point of the cleaning (point d 4) in the form of a bow-shaped trajectory in the direction of the point d 2. And taking the point d4 as the current position point, continuously determining the point c4 as the cleaning starting point according to the entrance selection condition and the priority cleaning condition, and after the robot moves from the point d4 to the point c4, cleaning the c area from left to right to the cleaning end point (point c 1) in a bow-shaped track form towards the point c3, and then cleaning the area. And taking the point c1 as the current position point, continuously determining the point b8 as the cleaning starting point according to the entrance selection condition and the priority cleaning condition, after the robot travels from the point c1 to the point b8, the robot travels along the direction vertical to the X axis of the point b6, and cleans the b area from left to right to the cleaning end point (point b 4) in a zigzag track form, and at the moment, the global cleaning phase is ended.

Wherein, judging whether the cleaning condition of the non-cleaning area is met comprises the following steps: judging whether the lengths of all the inlet boundaries of one uncleaned area are smaller than a preset inlet length; if so, the uncleaned area corresponding to the entrance boundary does not meet the uncleaned area cleaning condition; if not, the uncleaned area corresponding to the entrance boundary meets the uncleaned area cleaning condition. The preset inlet length is also a settable value, and can be set according to specific requirements, in this embodiment, the preset inlet length is set to be 1.2 times of the width of the robot body, and as long as the length of an inlet boundary is greater than or equal to 1.2 times of the width of the robot body, the area can be cleaned through the inlet boundary. If all the entry boundaries are less than 1.2 times the width of the robot body, the robot cannot or is difficult to access the area for cleaning.

Preferably, after the step of continuing the cleaning in the recorded edgewise direction, the method further includes the steps of: judging whether the sum of the length of the path traveled when the edgewise cleaning is continued and the length of the path already along is greater than or equal to the total perimeter; judging whether the difference between the current angle of the robot and the recorded angle at the initial position point is larger than 360 degrees or not in the process of continuing the edgewise cleaning; judging whether the current position point of the robot is within the preset range of the recorded initial position point of the edge in the process of continuing the edge cleaning; if the judgment results are all yes, the edge is ended; if the judgment result is negative, continuing the edgewise cleaning, and if the difference between the current angle of the robot and the recorded angle at the initial position point is more than 540 degrees in the process of continuing the edgewise cleaning, or the sum of the path length traveled when continuing the edgewise cleaning and the path length already along the edge is more than 2.5 times of the total circumference, finishing the edgewise cleaning. The robot can generate walking errors due to the influence of various self or external environments in the process of walking along the edge. In this case, if a single condition such as whether the total circumference is reached or whether the rotation is 360 ° is simply used as a basis for the determination, erroneous determination is likely to occur. The method provided by the invention can improve the accuracy of judgment by combining three conditions for judgment. Namely, the sum of the length of the path which is traveled when the edgewise sweeping is continued and the length of the path which is already edgewise is more than or equal to the total circumference; and in the process of continuing the edgewise cleaning, the difference between the current angle of the robot and the recorded angle at the initial position point is more than 360 degrees; in the process of continuing the edgewise cleaning, the current position point of the robot is within the preset range of the recorded initial position point of the edgewise cleaning; it can be determined that the robot has finished along the edge. If during the edge-following process, some accident (for example, the gyroscope is broken) occurs, and one of the three conditions cannot be obtained accurately, a forced judgment condition needs to be added, namely, as long as the difference between the current angle of the robot and the recorded angle at the starting position point is greater than 540 °, or the sum of the path length traveled when the edge-following cleaning is continued and the path length along the edge is greater than 2.5 times of the total circumference, the robot edge-following end can be determined. By adding the forced judgment condition, the condition that the robot is not limited along the edge indefinitely due to some unexpected conditions can be effectively avoided, and the practicability of the robot is further improved.

In the above embodiments, the sweeping includes sweeping, sucking and/or wiping. The map is a grid map with grid cells as basic units.

The above embodiments are merely provided for full disclosure and not for limitation, and any replacement of equivalent technical features based on the creative work of the invention should be regarded as the scope of the disclosure of the present application.

Claims (10)

1. A control method of an intelligent cleaning robot is characterized by comprising the following steps:

receiving an external navigation instruction, wherein the external navigation instruction is a navigation instruction generated by a direction key on a robot remote controller or a navigation instruction generated by a smart phone;

recording the current position and the cleaning state;

walking according to the external navigation instruction, and recording the external navigation path of walking;

receiving a functional instruction, and acting according to the functional instruction based on the recorded current position, the cleaning state and the external navigation path;

the recording of the current position comprises the following steps:

recording XY coordinate information of the current position point;

recording angle information of a current position point;

if the functional instruction is a recorded point continuous sweeping instruction, then according to the recorded sweeping state, starting subsequent sweeping from the current position, or returning to the recorded current position to start subsequent sweeping;

the cleaning method comprises the following steps of starting subsequent cleaning from the current position or returning to the recorded current position to start subsequent cleaning according to the recorded cleaning state:

judging whether the recorded cleaning state is a local cleaning stage, if so, returning the robot to the recorded current position according to the recorded XY coordinate information and angle information, judging whether the recorded current state is in a bow-shaped cleaning state, if so, continuing cleaning according to the recorded cleaning direction, otherwise, determining that the recorded current state is in an obstacle-detouring cleaning state, and continuing cleaning according to the recorded obstacle-detouring direction;

if not, judging whether the recorded cleaning state is a global cleaning stage or not, if so, searching a global map by the robot, and then automatically navigating to the searched missed cleaning area from the current position to continue cleaning;

if not, determining that the recorded cleaning state is a global edgewise stage, returning the robot to the recorded current position according to the recorded XY coordinate information and angle information, and then continuing cleaning according to the recorded edgewise direction.

2. The method of claim 1, wherein recording the purge condition comprises the steps of:

judging whether the current sweeping stage belongs to a local sweeping stage;

if so, recording a local cleaning stage, recording a cleaning range of a current local area and a cleaning range of a current global area, recording whether the current state is in a bow-shaped cleaning state or a barrier-bypassing cleaning state, and recording a cleaning direction in the bow-shaped cleaning state or a barrier-bypassing direction in the barrier-bypassing cleaning state;

if not, judging whether the current sweeping stage belongs to a global sweeping stage or not;

if so, recording as a global cleaning stage, and recording the cleaning range of the current global area;

if not, recording as a global edge stage, and recording the starting position point of the edge, the angle at the starting position point, the path along the edge and the current edge direction.

3. The method of claim 2, wherein:

the method for recording the cleaning range of the current local area comprises the following steps:

determining a first leftmost position point positioned at the leftmost side, a first rightmost position point positioned at the rightmost side, a first uppermost position point positioned at the uppermost end and a first lowermost position point positioned at the lowermost end in the walking path based on the walking path of the robot in the current local area;

establishing a virtual first leftmost vertical line based on the first leftmost position point, establishing a virtual first rightmost vertical line based on the first rightmost position point, establishing a virtual first uppermost horizontal line based on the first uppermost position point, and establishing a virtual first lowermost horizontal line based on the first lowermost position point;

recording an area surrounded by the first leftmost vertical line, the first rightmost vertical line, the first uppermost horizontal line and the first lowermost horizontal line as a sweeping range of the current local area;

and/or the presence of a gas in the gas,

the method for recording the sweeping range of the current global area comprises the following steps:

determining a second leftmost position point located at the leftmost side, a second rightmost position point located at the rightmost side, a second uppermost position point located at the uppermost end, and a second lowermost position point located at the lowermost end in the walking path based on the walking path of the robot in the current global area;

establishing a virtual second leftmost vertical line based on the second leftmost position point, establishing a virtual second rightmost vertical line based on the second rightmost position point, establishing a virtual second uppermost horizontal line based on the second uppermost position point, and establishing a virtual second lowermost horizontal line based on the second lowermost position point;

recording an area surrounded by the second leftmost vertical line, the second rightmost vertical line, the second uppermost horizontal line and the second lowermost horizontal line as a sweeping range of the current global area.

4. The method of claim 1, wherein: the receiving function instruction, based on the recorded current position and cleaning state and external navigation path, acts according to the function instruction, comprising the following steps:

receiving a functional instruction, and judging whether the functional instruction is a key cleaning instruction or not;

if yes, performing key cleaning;

if not, judging whether the functional instruction is a current point cleaning instruction or not;

if so, cleaning is started from the current position point;

if not, judging whether the functional instruction is an automatic recharging instruction or not;

if yes, automatically returning to the seat for charging;

if not, judging whether the functional instruction is a recording point continuous scanning instruction or not;

if not, the next functional instruction is waited for.

5. The method of claim 4, wherein: the key cleaning comprises the following steps:

taking a current position point as a first reference point, walking forward from the first reference point along a first direction for a set distance, then backing to the first reference point, then backing from the first reference point along a second direction opposite to the first direction for a set distance, and then walking forward to the first reference point;

walking from the first reference point along a third direction perpendicular to the first direction for a set width, then taking the current position point as a second reference point, walking from the second reference point forward along the first direction for a set distance, then backing to the second reference point, then backing from the second reference point along the second direction for a set distance, and then walking forward to the second reference point;

walking twice the set width in a fourth direction opposite to the third direction from the second reference point, then walking forward for a set distance in the first direction from the third reference point by taking the current position point as a third reference point, then backing to the third reference point, then backing for a set distance in the second direction from the third reference point, and then walking forward to the third reference point;

walking the set width three times along the third direction from the third reference point, then walking the set distance forward along the first direction from the fourth reference point by taking the current position point as the fourth reference point, then backing to the fourth reference point, backing the set distance along the second direction from the fourth reference point, and then walking the set distance forward to the fourth reference point;

walking the set width four times along the fourth direction from the fourth reference point, then walking the set distance forward along the first direction from the fifth reference point by taking the current position point as the fifth reference point, then backing to the second direction from the fifth reference point by the set distance, and then walking the set distance forward to the fifth reference point;

in this way, until the distance from the first reference point to the third direction reaches N times of the set width, the key cleaning is finished;

wherein N is a natural number greater than 1.

6. The method of claim 5, wherein: after the key cleaning is finished, the method also comprises the following steps:

and according to the recorded cleaning state, starting to perform subsequent cleaning from the current position, or returning to the recorded current position to perform subsequent cleaning.

7. The method of claim 5, wherein:

if the robot detects an obstacle or a cliff in the process of walking forward along the first direction, directly backing to a corresponding reference point;

if the robot detects an obstacle or cliff during the backward movement in said second direction, it walks straight forward to the corresponding reference point.

8. The method of claim 4, wherein: the cleaning from the current position point comprises the following steps:

determining an initial cleaning angle from an initial position point when the robot initially starts cleaning, based on the initial position point;

adjusting the current cleaning angle of the robot at the current position point to enable the current cleaning angle to be the same as the initial cleaning angle;

starting the arch-shaped cleaning, and stopping cleaning the current route and switching to the next route of the arch-shaped to continue cleaning if M continuous grid units are detected as cleaned units in the arch-shaped cleaning process;

wherein M is a natural number greater than or equal to 3.

9. The method according to claim 4 or 6, characterized in that: the robot searches a global map, automatically navigates to a searched missed-scanning area from the current position and continues to clean, and the method comprises the following steps:

determining entry boundaries of the cleaned area and the uncleaned area based on the searched global map;

respectively taking end points at two ends of each entry boundary as two entry reference points;

establishing an XY axis coordinate system by taking the current position as a coordinate origin; analyzing the coordinate positions of two inlet reference points of the same inlet boundary; when the X-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the Y-axis values of the two entrance reference points are the same, selecting the entrance reference point closest to the current position as the optimal entrance point; when the X-axis value and the Y-axis value of the two inlet reference points are different, judging the position relation between the non-cleaned area and the cleaned area; selecting the entry reference point having a large X-axis value as an optimal entry point if the unswept area is to the left of the swept area; selecting said entry reference point having a small X-axis value as an optimal entry point if the unswept area is to the right of the swept area;

selecting one optimal entry point closest to the current position as a priority cleaning reference point in an unswept area, and selecting two priority cleaning reference points closest to the current position; judging whether the difference value between the current position and the two priority cleaning reference points is smaller than a preset distance value; if so, selecting the priority cleaning reference point corresponding to the inlet boundary with longer length as a cleaning starting point; if not, selecting the priority cleaning reference point closest to the current position as a cleaning starting point;

walking from the current position to the cleaning starting point, and judging the position relation of two end points of the inlet boundary where the cleaning starting point is located; when the X-axis values of the two end points are the same, directly walking towards the other end point of the inlet boundary from the cleaning starting point, and cleaning an uncleaned area according to a zigzag track form; when the Y-axis values of the two endpoints are the same; if the non-cleaned area is positioned above the inlet boundary, walking along the positive direction of the Y axis from the cleaning starting point, and cleaning the non-cleaned area according to the track form of the Chinese character 'gong'; if the non-cleaning area is positioned below the inlet boundary, walking along the Y-axis negative direction from the cleaning starting point, and cleaning the non-cleaning area in a zigzag track form; when the X-axis value and the Y-axis value of the two end points are different, walking along the direction vertical to the X-axis of the other end point from the cleaning starting point, and cleaning an uncleaned area according to a track form of a Chinese character 'gong';

and after the sweeping is finished, determining the next optimal entry point as the sweeping starting point of the next preferential sweeping area, and repeating the steps until all the non-sweeping areas meeting the sweeping conditions of the non-sweeping areas are completely swept, and finishing the global sweeping stage.

10. The method according to claim 4 or 6, characterized in that: after the step of continuing to clean in the recorded edgewise direction, the method further comprises the following steps:

judging whether the sum of the length of the path traveled when the edgewise cleaning is continued and the length of the path already along is greater than or equal to the total perimeter; judging whether the difference between the current angle of the robot and the recorded angle at the initial position point is larger than 360 degrees or not in the process of continuing the edgewise cleaning; judging whether the current position point of the robot is within the preset range of the recorded initial position point of the edge in the process of continuing the edge cleaning;

if the judgment results are all yes, the edge is ended;

if the judgment result is negative, continuing the edgewise cleaning, and if the difference between the current angle of the robot and the recorded angle at the initial position point is more than 540 degrees in the process of continuing the edgewise cleaning, or the sum of the path length traveled when continuing the edgewise cleaning and the path length already along the edge is more than 2.5 times of the total circumference, finishing the edgewise cleaning.

Images