WO2017088811A1

WO2017088811A1 - Self-moving robot and walking mode conversion method and walking method therefor

Info

Publication number: WO2017088811A1
Application number: PCT/CN2016/107247
Authority: WO
Inventors: 汤进举
Original assignee: 科沃斯机器人股份有限公司
Priority date: 2015-11-27
Filing date: 2016-11-25
Publication date: 2017-06-01
Also published as: CN106814732A

Abstract

A self-moving robot and a walking mode conversion method and a walking method therefor. The self-moving robot comprises a sensing unit and a control unit, and is further provided with planning navigation and non-planning navigation walking modes. The control unit comprises a processing module and a path planning module connected to the sensing unit. In the planning navigation walking mode, according to received information, the path planning module plans a moving path, and a mobile robot implements the walking of the moving path; and when the walking according to the pre-set moving path planning fails, a processing module converts the walking mode from the planning navigation walking mode to the non-planning navigation walking mode according to the self-moving robot. By means of free conversion between the planning navigation walking mode and the non-planning navigation walking mode, the self-moving robot can also remain in a normal operating state when in a complex environment which is difficult to recognize or cannot be recognized, and the structure is simple, the operation is convenient and free, and the operating efficiency is high.

Description

Self-moving robot and walking mode conversion method and walking method thereof

Technical field

The invention relates to a self-mobile robot, a walking mode conversion method and a walking method thereof, and belongs to the technical field of small household appliance manufacturing.

Background technique

Earlier self-mobile robots are only provided with collision sensors or side-view sensors, such as the self-moving robots disclosed in US 6,809,490, which perform cleaning on the ground mainly by one or a combination of random walking, welt walking or fixed-point walking. jobs. Since the self-moving robot has no ranging scanner, it is impossible to establish a global map of the indoor environment, and the working environment cannot be identified. Only a few walking modes can be used to perform the cleaning work, which results in many areas in the cleaning work that cannot be cleaned or repeatedly cleaned, and cleaned. low efficiency.

Another example is a self-mobile robot with a scanning range finder (or camera distance information extraction device), such as the self-mobile robot disclosed in CN101809461A, which preferentially uses a scanning range finder to measure the distance of indoor obstacles, thereby establishing a global indoor environment. Map, then perform path planning based on the indoor map. The self-mobile robot traverses the cleaning surface according to the map execution path plan, and can completely cover the indoor ground at a time, and rarely has a missing sweep area or a repeated coverage area. However, the cleaning of the self-moving robot with path planning capability has a fatal flaw. Once the positioning error occurs, such as the environment is incapable of recognizing the map for positioning or artificially moving the position and misalignment, the self-mobile robot cannot continue to work.

Therefore, the walking mode and path planning of mobile robots have always been a subject of research in the industry. The existing self-mobile robots mainly obtain environmental information through their own sensors, draw and update environmental maps, and navigate and plan optimal walking paths in real time. In most environments, self-traveling robots in real-time navigation can work unimpededly and efficiently according to their own algorithms, but they will make mistakes when they encounter certain complex environments or environments that are not recognized or difficult to identify from mobile robots. In this case, it is generally set that the mobile robot repeatedly performs an action until the error is exited, or if it is still unrecognizable after performing N times of work, it is directly shut down and stops working. Obviously, the above two methods are not ideal and Efficient countermeasures do not allow self-mobile robots to work efficiently and efficiently in a variety of environments.

Summary of the invention

The technical problem to be solved by the present invention is to provide a self-mobile robot, a walking mode conversion method thereof and a walking method, aiming at free conversion between a planned navigation walking mode and an unplanned navigation walking mode, aiming at the deficiencies of the prior art. Mobile robots can cope with complex environments that are difficult to identify or even recognize. If the ground is kept in a normal working state, the structure is simple, the operation is convenient and the work efficiency is high.

The technical problem to be solved by the present invention is achieved by the following technical solutions:

A self-moving robot includes a sensing unit and a control unit; the sensing unit is configured to sense information of a working area state of the self-mobile robot and a position of the self-mobile robot in a working area, and the information Transmitting to the control unit; the self-mobile robot is provided with a plurality of walking modes, including: a planned navigation walking mode and an unplanned navigation walking mode; the control unit includes a processing module and a path connected to the sensing unit a planning module; in the planning navigation walking mode, according to the received information, the path planning module plans a moving path, and the self-mobile robot performs walking of the moving path, and when walking according to a preset moving path plan Upon failure, the processing module controls the self-mobile robot to shift the walking mode from the planned navigation walking mode to the unplanned navigation walking mode. The unplanned navigation walking mode includes a random walking mode and/or a welt walking mode.

After being transferred to the unplanned navigation walking mode, when the self-moving robot satisfies the recovery condition, it automatically returns to the planned navigation walking mode; or, according to the information received by the control unit, when the path planning module makes the When the mobile robot resumes the mobile path planning capability in the work area, the path planning module sends an instruction to the processing module, and the processing module controls the self-mobile robot to move the walking mode from the unplanned navigation walking mode. Automatically resumes to plan navigation walking mode.

The sensing unit includes: a camera assembly, a ranging sensor (such as a laser scanning range finder, an infrared ranging sensor, and an ultrasonic ranging sensor), a code wheel, an encoder, a collision sensor, a side view sensor, or a down view, as needed. One or a combination of sensors.

The present invention provides a walking mode conversion method of a self-moving robot, the walking mode of the self-moving robot includes a planned navigation walking mode and an unplanned navigation walking mode; the conversion method includes the following steps: Step S100: the self-mobile robot In the working area, the planning navigation walking mode is adopted, in which the self-moving robot plans to walk according to the preset moving path; step S200: when the walking failure according to the preset moving path planning occurs, the self-mobile robot automatically transfers to the non-planning Navigation walking mode walking.

Before the step S100, the method further includes: step S099: initial mapping is performed on the working area of the mobile robot, and the construction proceeds to step S100; otherwise, step S110 is performed: directly entering the unplanned navigation walking mode.

The step S110 further includes: the self-mobile robot walks on the area where the mobile station is located, and after the completion of the area, stores information that the area has completed walking; and performs step S099 on the remaining area.

In the step S200, when the travel failure is planned according to the preset movement path, the specific method includes: the self-mobile robot cannot continue to walk normally due to encountering an obstacle in the existing movement path.

The specific content of the inability to continue to walk normally is: the self-moving robot walks in the planned navigation mode in the work area, and after the number of encounters of obstacles in the unit time reaches the preset value, the walking cannot be performed according to the moving path.

The specific content of the inability to continue to walk normally is: the self-moving robot walks in the planned navigation mode in the work area, encounters obstacles and continues to encounter obstacles after the number of turns reaches the preset value, and cannot walk according to the moving path.

In an embodiment of the invention, the obstacle is annular.

The step of performing the walking failure according to the preset movement path described in the step S200 includes: the actual position of the self-moving robot does not match the position of the environment map established during the initial construction or planning walking.

The non-planned navigation walking mode further includes: a random walking mode and/or a welt walking mode, specifically including: Case 1: when the number of steering reaches a preset value and continues to encounter an obstacle, the random walking mode is entered; : When the number of obstacles encountered in the unit time reaches the preset value, it is transferred to the sticking walking mode; or, the third situation: after the mobile robot enters the unplanned navigation walking mode, the random walking mode is preferentially entered.

The step S200 further includes a step S300 of automatically restoring the planned navigation walking mode when the self-mobile robot walks in the unplanned navigation walking mode to satisfy the recovery condition.

The recovery condition in the step S300 includes: the walking time after the mobile robot enters the unplanned navigation walking mode reaches a preset interval time; or the number of random collisions reaches a preset value; or, the self-mobile robot can resume normal operation. Path planning to walk.

The invention also provides a walking method of a self-moving robot, wherein the walking mode of the self-moving robot comprises a planned navigation walking mode and an unplanned navigation walking mode; the walking method comprises the following steps:

Step S100': the self-mobile robot is in a planned navigation walking mode in the work area, in which the self-mobile robot plans to walk according to a preset moving path;

Step S200': when the walking failure according to the preset moving path planning occurs, the self-mobile robot automatically moves into the unplanned navigation walking mode to walk;

Step S400': The self-mobile robot completes walking in the work area.

The step S100' further includes:

Step S099': the initial mapping is performed by the mobile robot on the working area, and the construction proceeds to step S100';

Otherwise, step S110' is performed: directly entering the unplanned navigation walking mode walking.

The step S110' further includes: the self-mobile robot walks on the area where it is currently located, and after the area is completed, stores information that the area has completed walking; and the step S099' is re-executed for the remaining area.

After the step S200' and before the step S400', the method further includes the step S300': when the self-mobile robot is When the non-planned navigation walking mode meets the recovery condition, the planned navigation walking mode is automatically restored.

The appearance described in the step S200' fails to travel according to the preset movement path. The specific content is that the actual position of the mobile robot does not match the position of the environment map established during the initial construction or planning walking.

The step of performing the walking failure according to the preset movement path described in the step S200' includes: the self-mobile robot cannot continue to walk normally due to encountering an obstacle in the existing moving path.

In summary, the present invention provides a self-mobile robot and a walking mode conversion method and a walking method thereof, which are difficult to recognize or even unrecognizable by planning a free transition between a navigation walking mode and an unplanned navigation walking mode. In the complex environment, it is also able to cope with the normal working state, with simple structure, convenient operation and high work efficiency.

The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

DRAWINGS

Figure 1 is a flow chart of a walking method of the present invention;

2 is a schematic view showing the environment position of the self-mobile robot according to the first case of the present invention;

3 is a schematic view showing the environment position of the self-mobile robot in the second case of the present invention;

4 is a schematic view showing the environment position of the self-mobile robot in the third case of the present invention;

Figure 5 is a schematic diagram showing the environmental position of the self-mobile robot in the fourth case of the present invention;

Figure 6 is a schematic view showing the detailed position of the self-moving robot in the fourth case of the present invention;

Figure 7 is a schematic diagram showing the environmental position of the self-mobile robot in the sixth case of the present invention;

Figure 8 is a schematic view showing the detailed position of the self-moving robot in the sixth case of the present invention;

9 and 10 are schematic views respectively showing two initial mapping methods of the present invention.

detailed description

The present invention provides a self-mobile robot including a sensing unit and a control unit; the sensing unit is configured to sense information of a working area state of the self-mobile robot and a position of the self-mobile robot in a working area, and Transmitting the information to the control unit; the self-mobile robot is provided with a plurality of walking modes, including: a planned navigation walking mode and an unplanned navigation walking mode; the control unit includes a processing module and the sensing unit a connected path planning module; in the planned navigation walking mode, according to the received information, the path planning module plans a moving path, and the self-mobile robot performs walking of the moving path, when a preset movement occurs When the path planning fails to travel, the processing module controls the self-mobile robot to follow the walking mode The navigation walking mode is changed to the unplanned navigation walking mode. After being transferred to the unplanned navigation walking mode, when the self-moving robot satisfies the recovery condition, it automatically returns to the planned navigation walking mode; or, according to the information received by the control unit, when the path planning module makes the When the mobile robot resumes the moving path planning capability in the work area, the processing module controls the self-mobile robot to automatically restore the walking mode from the unplanned navigation walking mode to the planned navigation walking mode.

Additionally, the unplanned navigation walking mode includes a random walking mode and/or a welt walking mode. Of course, in practical applications, the self-mobile robot is not limited to the above two walking modes in the unplanned navigation walking mode, and other walking modes, such as traversing walking or fixed-point walking mode, may also be adopted. When the mobile robot enters the unplanned navigation walking mode, the two walking modes may be selected according to preset conditions, or may be directly entered into the random walking mode.

It should be noted that, before planning the navigation walking mode, the self-mobile robot needs to establish an initial map of the working area. 9 and 10 are schematic views respectively showing two initial mapping methods of the present invention. The initial two maps can usually be established by the following two methods: Method 1. Walk the mobile robot to build a map. As shown in Fig. 9, the self-moving robot starts from the starting point s and walks until it returns to the starting point s, thereby establishing an initial map of the M-area. Method 2: Scanning ranging and mapping from a mobile robot. As shown in FIG. 10, the 360-degree rotation of the scanning range finder on the mobile robot directly obtains the distance from the mobile robot from the boundary of the room, thereby obtaining an initial map of the room. Of course, the combination of Method 1 and Method 2 will have a better effect. For details of the construction process, refer to the disclosure in the Chinese patent CN201010282605.6, and no further details are provided herein.

In addition, according to different needs, the sensing unit may include a plurality of types, such as: a camera assembly, a ranging sensor (a ranging sensor includes: a laser scanning range finder, an infrared ranging sensor, an ultrasonic ranging sensor, etc.), One or a combination of a code wheel, an encoder, a collision sensor, a side view sensor, or a down view sensor.

1 is a flow chart of a walking method of the present invention. As shown in FIG. 1 , the present invention provides a walking method of a self-moving robot, wherein the walking mode of the self-moving robot includes a planned navigation walking mode and an unplanned navigation walking mode; the walking method includes the following steps:

Step S400': The self-mobile robot completes walking in the work area.

The step S100' further includes:

Step S099': the initial mapping is performed by the mobile robot on the working area, and the drawing successfully enters the step. S100’;

After the step S200' and before the step S400', the step S300' is further included: when the mobile robot is walking in the unplanned navigation walking mode to satisfy the recovery condition, the planned navigation walking mode is automatically resumed.

The present invention also provides a walking mode conversion method of a self-mobile robot, the walking mode of the self-moving robot includes a planned navigation walking mode and an unplanned navigation walking mode; the conversion method includes the following steps:

Step S100: The self-mobile robot is in a planned navigation walking mode in the work area, in which the self-mobile robot plans to walk according to a preset moving path;

Step S200: When the walking failure is planned according to the preset movement path, the self-mobile robot automatically moves into the unplanned navigation walking mode.

In an embodiment of the invention, the obstacle is annular.

The step of performing the walking failure according to the preset movement path described in the step S200 includes: The actual position of the mobile robot does not match the location of the environment map established during the initial construction or planning walk.

The non-planned navigation walking mode further includes: a random walking mode and/or a welt walking mode, and specifically includes:

Case 1: When the number of turns reaches the preset value and continues to encounter obstacles, it shifts to the random walk mode;

Case 2: When the number of encounters with obstacles in the unit time reaches the preset value, it is transferred to the sticking walking mode;

Or, Case 3: After the mobile robot enters the unplanned navigation walking mode, the random walking mode is preferentially entered.

The recovery condition in the step S300 includes: the walking time after the mobile robot enters the unplanned navigation walking mode reaches a preset interval time; or, the number of random collisions reaches a preset value; or, the self-mobile robot can resume normal The path is planned to walk.

Hereinafter, the walking method of the above-described self-moving robot and the walking mode switching method applied during the walking process of the present invention will be described in detail in combination with various situations that occur during the normal operation of the mobile robot, and various methods as described below are described. In the case, the self-moving robots are all sweeping robots, and perform cleaning operations on the walking area during walking.

Situation one

2 is a schematic view showing the environmental position of the self-mobile robot according to the first case of the present invention. As shown in FIG. 2, the working area of the mobile robot 100 is a corridor having two left and right parallel walls 200, and the mobile robot 100 can recognize the distance between the left and right parallel walls 200, but cannot recognize the clip. How long is the distance between the front and rear open spaces between the walls on both sides. In this environment, the self-mobile robot 100 is likely to be unable to locate; or since the mobile robot 100 enters the environment from other environments, accurate positioning cannot be performed; or since the mobile robot 100 is manually moved to the work area The autonomous positioning navigation and planning algorithms cannot work normally in this environment, and thus lose their direction. That is to say, when the mobile robot enters the long corridor area, the initial map cannot be established, and the initial construction failure fails. The self-mobile robot automatically enters the unplanned navigation walking mode and walks while walking. The cleaning operation is performed on the walking area, for example, the random walking mode can be used to complete the work in the area in a random collision manner. Specifically, for the narrow area shown in FIG. 2, the self-mobile robot 100 can also divide the virtual robot 100 into a plurality of closed areas and clean them one by one by setting a virtual boundary during walking.

Situation two

3 is a schematic view showing the environmental position of the self-mobile robot in the second case of the present invention. In an indoor environment as shown in FIG. 3, a rectangular line is provided in the rectangular space, one end of the folding line is closed with one side wall of the rectangular space, and the other end forms a narrow opening A with the other side wall of the rectangular space. The self-moving robot 100 starts the sticking at the starting position S to establish an initial map of the indoor environment. The irregular line L in FIG. 3 indicates the path from the time when the mobile robot 100 is traveling on the side. Since somewhere in the indoor environment is a narrow mouth A, since the mobile robot is walking to the end point E, the self-moving robot 100 may not be able to get out of the narrow mouth A due to the angle problem, that is, the self-mobile robot is likely to be in the The folding line and the side wall surrounded by the rectangular space are looped and lapped, that is, the self-moving robot 100 walks in the planned navigation mode in the working area, and continues to encounter the obstacle after the number of times the steering reaches the preset value. Obstacle and cannot walk according to the moving path. In this environment, the self-mobile robot 100 automatically enters the unplanned navigation walking mode because of the initial mapping failure, and starts cleaning while walking. At this time, the control unit controls the self-mobile robot to shift the working mode. In the random walking mode, the self-moving robot walks in the random walking mode and first completes the cleaning in the area surrounded by the side walls of the fold line and the rectangular space, and stores the information that the area has been cleaned in the control unit. The mobile robot 100 can complete the narrow angle A by finding a suitable angle by randomly colliding, and construct the remaining part of the rectangular space. The process of drawing the map is to identify the boundary of the work area and complete the closed shape. The process of graphics. After the completion of the drawing, the self-moving robot plans a corresponding moving path according to the shape of the complete closed figure, and walks according to the moving path and completes the walking of the remaining part of the rectangular space, and completes the cleaning during the walking process.

Situation three

4 is a schematic view showing the environmental position of the self-mobile robot in the third case of the present invention. As shown in FIG. 4, since the mobile robot 100 constructs a picture in the environment to be operated, it is possible to encounter an infinite loop condition, such as the L1 path in FIG. 4, since the mobile robot 100 starts the sticking from the starting position S1. Until the end of E1, since the welt envelope of the whole environment is not completed, the mobile robot thinks that the welt walking fails, that is, since the mobile robot is unable to return to the original starting position S1, it will start the next posting from the beginning. While walking, such as the L2 path, the mobile robot 100 starts to walk from the starting position S2 until the end of the E2, and the self-moving robot still believes that the welt fails, so that it will re-walk again, thus circulating all the time. That is to say, since the mobile robot 100 can not complete the welt mode by the outer edge and the inner edge, the loop is always in this mode, and the map cannot be built. At this time, the number of dead loops can be automatically recognized by the self-moving robot. For example, when the number of dead loops exceeds three times, the self-mobile robot will shift to the random walking mode, thereby completing the walking of the entire environment and during walking. Perform cleaning work.

In the above three cases, after the initial mapping failure, the self-moving robot 100 enters the random walking mode to complete the walking in the area where the drawing fails, and performs cleaning during the walking process, and stores the information that the area has already traveled. In the control unit. Then, the remaining space of the work area is treated and then walked and walked, and the complete walking of the work area is finally completed. Specifically, as shown in FIG. 2, FIG. 3 or FIG. 4, initially, the self-moving robot 100 usually walks along the indoor circle to determine a closed area and then performs a planned navigation mode walk. In FIG. 1, the self-moving robot 100 cannot determine the front and rear boundaries of the long corridor, and cannot initially construct the long corridor, and then moves into the random mode; in FIG. 2, after the mobile robot 100 enters the narrow-port A area, If you can't come out, you can go to the random mode and walk on the area first. In Figure 3, when the mobile robot 100 performs the pasting construction, it cannot return to the starting point, and also enters the random mode to complete the walking of a part of the area first, and then Other parts continue to build maps and walk, and perform sweeps while walking. In the actual operation, the self-moving robot may also preferentially perform a short segment of the welt path before it is transferred to the random mode, which is convenient for the self-moving robot to get out of trouble.

Situation four

FIG. 5 is a schematic diagram showing the environment position of the self-mobile robot in the fourth case of the present invention; FIG. 6 is a schematic diagram showing the detailed position of the self-mobile robot in the fourth case of the present invention. As shown in FIG. 5, the self-moving robot 100 has completed the original construction of the room to be cleaned, forming a complete closed boundary in which the obstacle C is removed in the rectangular space. Then, the mobile robot 100 can start to perform the normal traversal planning walking of the pending work according to the preset moving path in the planned navigation walking mode, for example, the bow-shaped walking, that is, the moving path is X. The area that the walking robot 100 has traveled before moving to point A is saved in the control unit. When the mobile robot 100 walks to the A point, an unexpected situation occurs, for example, the child will move from the mobile robot 100 to the point B. At this time, the positioning error occurs from the mobile robot 100, and the self-mobile robot 100 still follows the point A. When the method continues to travel in the original direction, it will always hit the obstacle C and become stuck. That is to say, since the mobile robot cannot continue to walk normally due to encountering an obstacle in the existing moving path, the self-mobile robot 100 cannot Continue to work in navigation mode. At this time, the self-moving robot 100 is turned into the random walking mode, thereby being separated from the obstacle C. More specifically, as shown in FIG. 5 and with reference to FIG. 6 , if the positioning error occurs in the mobile robot 100, S2 is the current position of the mobile robot 100. Due to an algorithm problem, the positioning error is determined from the mobile robot 100, and the current location is considered to be S1. The correct planning path is S1E1. Since the actual walking path of the mobile robot 100 becomes S2E2 after the positioning error, if there is an obstacle 300 on the S2E2 path, the self-mobile robot 100 cannot pass, and tries several times, for example: After three failures, he switched to a random walking mode. If the obstacle is 300, the mobile robot 100 can pass, walk to E2, and then continue to position and navigate, and plan to walk according to a fixed walking mode. That is to say, in this embodiment, since the self-moving robot is manually moved, it is of course possible to manually move the obstacle to the guided path. As a result, the corresponding positioning error occurred, and in the stored path map, an obstacle was encountered when walking again in the previously traveled path. At this time, it is necessary to convert from the planned navigation walking mode to the unplanned navigation walking mode, especially the random walking mode, so that the walking of the mobile robot 100 in the working area can be successfully completed and the operation is performed during the walking.

Further, when the mobile robot 100 walks in the unplanned navigation walking mode to satisfy the recovery condition, the planned navigation walking mode is automatically resumed. Specifically, since the area that has passed since the mobile robot 100 moved to the point A is stored in the control unit, after the mobile robot 100 moves into the random walking mode, it may move to the previously preset moving path X. Going up, returning to the known or stored map, for example, the partial graphic returning to position A matches the local graphic corresponding to the established environmental map A position, that is, the self-mobile robot can resume the normal path planning walking. At this point, it will automatically resume to the previous planning navigation walking mode.

Of course, in addition to the recovery conditions described above, it is also possible to return the self-mobile robot to the previous planned navigation walking mode by satisfying other recovery conditions, for example, an interval time can be preset in the control unit. The mobile robot counts after entering the unplanned navigation walking mode, and automatically resumes planning the navigation walking mode when the walking time reaches the preset interval time. In addition, the number of random collisions can also be set as a condition for restoring the planned navigation walking mode. Similarly, the preset value of the collision number can be set in the control unit before, and the recording is started after the mobile robot enters the unplanned navigation walking mode. The actual number of collisions with the obstacle during the walking process automatically resumes the planned navigation walking mode when the actual number of collisions reaches the preset value. The above-described setting conditions for restoring the walking mode of the mobile robot can be adaptively selected according to different needs and walking conditions of the environment to be operated.

Situation five

As can also be seen in FIG. 5, if there is no obstacle C in FIG. 5, the mobile robot 100 obtains an unconventional rectangular work space map through the original construction, and then starts to follow the preset movement path in the planned navigation walking mode. The waiting work is performed in a normal traversal planning, such as: bow-shaped walking, that is, the moving path is X. The area that the walking robot 100 has traveled before moving to point A is saved in the control unit. When the mobile robot 100 walks to point A, it moves to the point B. At this time, the actual position of the mobile robot 100 does not match the position of the environment map established in the initial map, such as after the mobile robot moves. The pattern around the point B is completely mismatched with the pattern around the previous point A. Even if there is no obstacle C, it is impossible to continue to follow the preset bow shape, and then move to the unplanned navigation walking mode. In this case, since the shape of the space to be worked is relatively regular, it can be preset by the time interval in the control unit, when the self-moving machine When the walking time after the person enters the unplanned navigation walking mode reaches the preset interval time, the planned navigation walking mode is automatically resumed.

Situation six

FIG. 7 is a schematic diagram showing the environment position of the self-mobile robot according to the sixth case of the present invention; FIG. 8 is a schematic diagram showing the detailed position of the self-mobile robot in the sixth case of the present invention. As shown in FIG. 7, when the self-mobile robot 100 is located in the complex area M, the area M is a complex environment composed of the table 400 and the plurality of chairs 500. The construction work has also been carried out on the area to be worked before the cleaning operation is started. At this time, since the mobile robot 100 traverses and performs the cleaning operation in the M area, since the number of the table 400 and the chair 500 is large, it is difficult to perform path planning, as shown in FIG. 7 and in conjunction with FIG. 8, the obstacle here. Mainly for the chair 500, the arrangement of the obstacles in the area to be worked is a ring shape. That is to say, the self-moving robot 100 may repeatedly collide with the table 400 or the chair 500 during the cleaning process, that is, the number of encounters of obstacles per unit time reaches a preset value and cannot follow the moving path. It is also possible that after encountering an obstacle and the number of turns reaches a preset value, the obstacle continues to be encountered and cannot follow the moving path. The preset value is previously preset in the control unit, and the number thereof can be determined as needed. In the present example, the preset value has a value of 3. At this time, since the mobile robot 100 has been unable to walk in the normal bow-shaped or zigzag walking mode, it is necessary to shift to the random walking mode. Specifically, as shown in FIG. 8, when there are a plurality of obstacles 300 in the walking environment, the plurality of obstacles 300 form a closed area from which the mobile robot 100 plans a path to the destination M, such as A path, but since the mobile robot 100 may hit the obstacle 300 when entering or working in the area, or manually place other obstacles on the path A, the self-moving robot 100 may consider the path to be unable to walk, It is identified as an inaccessible area, and is re-planned as a B, C, or D path. However, if there is a situation such as the A path, the mobile robot 100 will not find a feasible path and cannot get out of the closed area. At this time, the self-mobile robot 100 enters the random walking mode after planning the walking failure, and completes the work in the area. That is to say, in the present embodiment, the self-mobile robot 100 encounters a closed obstacle during the walking process, and can walk to the target location in a preset manner. At this time, it is necessary to transfer from the planned path walking mode to the random walking mode to complete the task. Similarly, after the mobile robot 100 is transferred to the unplanned walking mode, the walking operation may be performed on the vicinity of the to-be-worked area and the cleaning operation may be performed while walking, and the information of the already-running area is saved in the control unit. Then build maps, walk and clean other areas. In the above process, when the self-mobile robot 100 is in the unplanned walking mode, once the recovery condition is satisfied, it is still possible to resume walking and work in the planned walking mode.

It can be seen from the above that the situation 1 to 3 is the failure of the initial mapping to make the self-mobile robot directly enter the non-planning guide. The voyage mode, while the four to six cases are obstacles encountered during the operation of the mobile robot or do not match the initial mapping (or during walking), resulting in the inability to continue walking and into the unplanned navigation walking mode. It should be specially stated that, during the initial construction process or during the walking operation after the completion of the construction, when the self-mobile robot moves from the planned navigation walking mode to the unplanned navigation walking mode, There is no particular limitation on the walking of the welt, and it can be selected according to some corresponding conditions set in advance. Of course, from the viewpoint of facilitating the self-moving robot to enter the working state in the shortest time, in the above various cases, Self-mobile robots are preferentially converted directly into a random walk mode.

In summary, the present invention provides a self-mobile robot, a walking mode conversion method thereof, and a walking method. When the path planning walking mode is difficult to perform, that is, when a walking failure occurs according to a preset moving path plan, the automatic conversion is performed. In the non-planned walking mode, including the failure of mapping, map mismatch and encounter obstacles can not continue to walk in three main situations. When the unplanned walking mode is operated for a period of time or is separated from the complex environment, it can be converted back to the path planning walking mode; that is, the present invention enables self-transformation between the planned navigation walking mode and the unplanned navigation walking mode. In a complex environment that is difficult to identify or even unrecognizable, the mobile robot can cope with the normal working state with ease, simple structure, convenient operation and high work efficiency.

Claims

A self-moving robot includes a sensing unit and a control unit; the sensing unit is configured to sense information of a working area state of the self-mobile robot and a position of the self-mobile robot in a working area, and the information Transmitted to the control unit; characterized in that

The self-mobile robot is provided with a plurality of walking modes, including: a planned navigation walking mode and an unplanned navigation walking mode;

The control unit includes a processing module and a path planning module connected to the sensing unit;

In the planned navigation walking mode, according to the received information, the path planning module plans a moving path, and the self-mobile robot performs walking of the moving path, and when a walking failure is planned according to a preset moving path, The processing module controls the self-mobile robot to switch the walking mode from the planned navigation walking mode to the unplanned navigation walking mode.
The self-mobile robot according to claim 1, wherein after the transition to the unplanned navigation walking mode, when the self-motive robot satisfies the recovery condition, the self-mobile robot automatically returns to the planned navigation walking mode;

Or, according to the information received by the control unit, when the path planning module causes the self-mobile robot to restore the moving path planning capability in the working area, the path planning module sends an instruction to the processing The module is controlled by the processing module to automatically restore the walking mode from the unplanned navigation walking mode to the planned navigation walking mode.
The self-mobile robot according to claim 1 or 2, wherein the sensing unit comprises: one of a camera assembly, a distance measuring sensor, a code wheel, an encoder, a collision sensor, a side view sensor or a down view sensor or Its combination.
The self-mobile robot of claim 3, wherein the unplanned navigation walking mode comprises a random walking mode and/or a welt walking mode.
A walking mode conversion method of a self-moving robot, characterized in that the walking mode of the self-moving robot comprises a planned navigation walking mode and an unplanned navigation walking mode; the conversion method comprises the following steps:

Step S100: The self-mobile robot is in a planned navigation walking mode in the work area, in which the self-mobile robot plans to walk according to a preset moving path;

Step S200: When the walking failure is planned according to the preset moving path, the self-mobile robot automatically transfers to Non-planned navigation walking mode walking.
The walking mode conversion method according to claim 5, further comprising: before the step S100:

Step S099: the initial mapping is performed on the working area of the mobile robot, and the mapping proceeds to step S100;

Otherwise, step S110 is performed: directly entering the unplanned navigation walking mode walking.
The walking mode conversion method according to claim 6, wherein the step S110 further comprises: the self-mobile robot performs walking on an area in which it is located, and stores information that the area has completed walking after the area is completed; Re-execute step S099 for the remaining area.
The walking mode conversion method according to claim 5, wherein the step of walking in the step S200 is as follows: the self-mobile robot encounters an obstacle in the existing moving path. Things can't continue to walk normally.
The walking mode conversion method according to claim 8, wherein the specific content of the non-continuous normal walking is: the self-moving robot walks in a planned navigation mode in the work area, and the number of encounters of obstacles in a unit time reaches a pre-preparation. After setting the value, you still cannot follow the moving path.
The walking mode conversion method according to claim 8, wherein the specific content of the non-continuous normal walking is: the self-moving robot walks in the planned navigation mode in the work area, encounters an obstacle, and the number of turns reaches a preset value. Still continue to encounter obstacles and cannot walk according to the moving path.
The walking mode conversion method according to claim 10, wherein the obstacle is a ring shape.
The walking mode conversion method according to claim 6, wherein the step of performing the walking failure according to the preset movement path in the step S200 includes: the actual position of the self-moving robot and the initial mapping or The location of the environment map established during the planning walk does not match.
The walking mode conversion method according to claim 5, wherein the non-planned navigation walking mode further comprises: a random walking mode and/or a welt walking mode, specifically comprising:

Case 1: When the number of turns reaches the preset value and continues to encounter obstacles, it shifts to the random walk mode;

Case 2: When the number of encounters with obstacles in the unit time reaches the preset value, it is transferred to the sticking walking mode;

Or, Case 3: After the mobile robot enters the unplanned navigation walking mode, the random walking mode is preferentially entered.
The walking mode conversion method according to claim 5, wherein the step S200 further comprises a step S300 of automatically restoring the planned navigation walking mode when the self-moving robot is traveling in the unplanned navigation walking mode to satisfy the recovery condition.
The walking mode conversion method according to claim 14, wherein the recovery condition in the step S300 comprises:

The walking time after the mobile robot enters the unplanned navigation walking mode reaches a preset interval time;

Or, set the number of random collisions to a preset value;

Alternatively, the self-mobile robot can resume normal path planning walking.
A walking method of a self-moving robot, characterized in that the walking mode of the self-moving robot comprises a planned navigation walking mode and an unplanned navigation walking mode; the walking method comprises the following steps:

Step S100': the self-mobile robot is in a planned navigation walking mode in the work area, in which the self-mobile robot plans to walk according to a preset moving path;

Step S200': when the walking failure according to the preset moving path planning occurs, the self-mobile robot automatically moves into the unplanned navigation walking mode to walk;

Step S400': The self-mobile robot completes walking in the work area.
The walking method according to claim 16, wherein the step S100' further includes:

Step S099': the initial mapping is performed by the mobile robot on the working area, and the construction proceeds to step S100';

Otherwise, step S110' is performed: directly entering the unplanned navigation walking mode walking.
The walking method according to claim 17, wherein said step S110' further comprises: said self-mobile robot walking on an area in which it is currently located, and storing the area after the area is completed is completed The information of walking is performed; step S099' is re-executed for the remaining area.
The walking method according to claim 16, wherein after the step S200' and before the step S400', the method further includes the step S300': when the self-mobile robot walks in the unplanned navigation walking mode to satisfy the recovery condition, automatically Resume planning navigation walking mode.
The walking method according to claim 17, wherein the occurrence of the step S200' occurs in accordance with a preset movement path, and the specific content is: the actual position of the mobile robot and the initial construction or planning. The location of the environment map established during walking does not match.
The walking method according to claim 16, wherein the step of walking in the step S200′ when the travel failure is planned according to the preset movement path includes: the self-mobile robot encounters the existing movement path. Obstacle and unable to continue to walk normally.