Visual robot mapping methods, initial cleaning methods, chips and visual robots

Abstract

This application discloses a mapping method for a visual robot, an initial cleaning method, a chip, and a visual robot. The mapping method includes: controlling the visual robot to build a first local map based on a charging base station; after completing the construction of the first local map, controlling the visual robot to enter an edge-walking mode based on a seat avoidance signal from the charging base station; building a global map in the edge-walking mode; and confirming that the visual robot has completed the construction of the global map when it re-enters the area of ​​the first local map through edge-walking. This application improves the efficiency of the visual robot's initial mapping, achieving the technical effect of quickly completing the initial mapping, and avoids the visual robot's mistaken attempt to navigate along the edges of short obstacles such as table legs and chair legs.

Description

Technical Field

[0001] This application relates to the field of visual mapping, specifically to a visual robot mapping method, a visual robot initial cleaning method, a chip, and a visual robot. Background Technology

[0002] Currently, visual robots using visual sensors for mapping often suffer from difficulties and long initial mapping times, resulting in a poor user experience. Furthermore, during edge-based mapping, visual robots frequently encounter anomalies where they attempt to map along the legs of tables or chairs in a room. This edge-based mapping under such circumstances is useless for building a global map, severely impacting the mapping efficiency of the visual robot. Summary of the Invention

[0003] This application provides a mapping method for a visual robot, an initial cleaning method, a chip, and a visual robot. The specific technical solutions are as follows:

[0004] The visual robot mapping method specifically includes: Step 1: Control the visual robot to build a first local map based on the charging base station, and proceed to Step 2 after the first local map is built; Step 2: Control the visual robot to enter the edge walking mode based on the charging base station avoidance signal, and proceed to Step 3; Step 3: Control the visual robot to build a global map in the edge walking mode. When the visual robot walks along the edge again and enters the area where the first local map is located, it is confirmed that the visual robot has completed the construction of the global map.

[0005] Furthermore, the controlled vision robot constructs a first local map based on the charging base station, specifically including: using the location of the avoidance signal transmitter of the charging base station as the origin of the charging base station coordinate system, using the direction of the charging base station perpendicular to the wall behind it as the Y-axis of the charging base station coordinate system, using the direction of the wall pointing towards the avoidance signal transmitter of the charging base station as the positive direction of the Y-axis of the charging base station coordinate system, and using the direction of the charging base station parallel to the wall behind it as the X-axis of the charging base station coordinate system; controlling the vision robot to move horizontally a first distance from the charging base station in the first direction of the X-axis of the charging base station coordinate system, and recording the coordinates of the position of the vision robot after moving horizontally a first distance in the charging base station coordinate system as the first... Landmarks; the vision robot is controlled to move a first distance in the positive Y-axis direction of the charging base station coordinate system from the first landmark, and then obtains the coordinates of the point whose X-axis projection distance from the origin of the charging base station coordinate system is a first distance in the first X-axis direction and whose Y-axis projection distance from the origin of the charging base station coordinate system is a first distance in the positive Y-axis direction. This point is recorded as the second landmark. The vision robot is controlled to execute several sets of bow-shaped walking processes from the second landmark to construct a first local map with a projection length of the second distance in the X-axis direction and a projection length of the first distance in the Y-axis direction of the charging base station coordinate system. The second distance is twice the first distance.

[0006] Furthermore, the vision robot is controlled to start from the second waypoint and execute several sets of bow-shaped walking processes in a bow-shaped walking pattern to construct a first local map with a projection length of the second distance in the X-axis direction of the charging base station coordinate system and a projection length of the first distance in the Y-axis direction of the charging base station coordinate system. Specifically, this includes controlling the vision robot to detect its current position in real time when executing each set of bow-shaped walking processes. When the vision robot moves to a point where its X-axis projection distance from the origin of the charging base station coordinate system is a first distance along the second X-axis direction and its Y-axis projection distance from the origin of the charging base station coordinate system is also a first distance along the positive Y-axis direction, the current coordinate point of the vision robot is recorded as the third landmark point. When the vision robot moves to a point where its X-axis projection distance from the origin of the charging base station coordinate system is a first distance along the second X-axis direction and its Y-axis projection distance from the origin of the charging base station coordinate system is 0 along the positive Y-axis direction, the current coordinate point of the vision robot is recorded as the fourth landmark point. Once the vision robot has completed the construction of the third and fourth landmark points, the vision robot is controlled to... The robot completes several sets of bow-shaped walking processes in a bow-shaped walking pattern; based on the connection of the first, second, third, and fourth landmarks, a first local map is formed with a projection length of the second distance in the X-axis direction of the charging base station coordinate system and a projection length of the first distance in the Y-axis direction of the charging base station coordinate system; wherein, each set of bow-shaped walking processes includes: controlling the vision robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system, then controlling the vision robot to move a first distance in the negative Y-axis direction of the charging base station coordinate system, then controlling the vision robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system, and then controlling the vision robot to move a first distance in the positive Y-axis direction of the charging base station coordinate system.

[0007] Furthermore, the control of the visual robot to enter the edge-walking mode based on the charging base station avoidance signal specifically includes: controlling the visual robot to move to a position three distances away from the charging base station; controlling the visual robot to detect whether it has received a charging base station avoidance signal from the charging base station; when the visual robot receives the charging base station avoidance signal, controlling the visual robot to execute the edge-walking process based on the received charging base station avoidance signal, thereby enabling the visual robot to enter the edge-walking mode; when the visual robot does not receive the charging base station avoidance signal, controlling the visual robot to rotate in place, and during the rotation, detecting whether it has received the charging base station avoidance signal; when the visual robot receives the charging base station avoidance signal, controlling the visual robot to execute the edge-walking process based on the received charging base station avoidance signal, thereby enabling the visual robot to enter the edge-walking mode.

[0008] Furthermore, the control vision robot executes an edge-following process based on the received charging base station avoidance signal, enabling the vision robot to enter an edge-following walking mode. Specifically, this includes: controlling the left and right moving structures of the vision robot to operate at different speeds, so that the vision robot uses the location of the charging base station avoidance signal transmitter as the center of its movement, determines the moving radius based on the preset avoidance signal radiation distance, determines a circular movement trajectory based on the center of the movement and the moving radius, and controls the vision robot to move from its current position to the position closest to the circular movement trajectory, so that the vision robot moves in an arc along the circular movement trajectory, and enters the edge-following walking mode when the edge sensor detects the edge during the arc movement.

[0009] Furthermore, the control of the vision robot to execute the edge-following process based on the received charging base station avoidance signal, thereby enabling the vision robot to enter the edge-following walking mode, also includes: during the arc-shaped movement of the vision robot along the circular movement trajectory, at preset first time intervals, controlling the vision robot to execute the charging base station avoidance signal reception detection process; if the charging base station avoidance signal reception detection result is that the charging base station avoidance signal cannot be received, then the control robot continues to move along the circular movement trajectory; if the charging base station avoidance signal reception detection result is that the charging base station avoidance signal can be received, then the control robot re-plans the circular movement trajectory and moves according to the re-planned circular movement trajectory.

[0010] Furthermore, during the arc-shaped movement, at preset first time intervals, the visual robot is controlled to execute a charging base station avoidance signal reception and detection process. Specifically, this includes: during the arc-shaped movement of the visual robot along a circular trajectory, at preset first time intervals, the visual robot stops moving, and the orientation of the visual robot when it stops is recorded as a marked orientation; the visual robot is then controlled to rotate by a preset angle; the charging base station avoidance signal reception and detection device of the visual robot continuously searches for and receives the charging base station avoidance signal during the preset angle rotation; if the charging base station avoidance signal reception and detection device detects the charging base station avoidance signal during the preset angle rotation, the result of the charging base station avoidance signal reception and detection is that the visual robot can receive the charging base station avoidance signal, and the visual robot is controlled to rotate so that the robot's orientation... Once the orientation is restored to the marked orientation, the vision robot terminates the current charging base station avoidance signal reception and detection process. The vision robot then replans its circular movement trajectory and continues moving along it. During this movement, the vision robot executes the charging base station avoidance signal reception and detection process at preset intervals. If the vision robot has rotated forward by a preset angle and the charging base station avoidance signal reception and detection device continues to fail to detect the charging base station avoidance signal, the result is that the vision robot has not received the signal. The vision robot then terminates the current charging base station avoidance signal reception and detection process, rotates in the opposite direction to restore its orientation to the marked orientation, and continues moving along the circular movement trajectory.

[0011] Furthermore, the control vision robot replans its circular movement trajectory, specifically including: extending the movement radius of the circular movement trajectory by a preset distance to obtain a replanned movement radius; taking the location of the charging base station's avoidance signal transmitter as the center of the movement circle, and taking the opposite direction of the charging base station relative to the side where the vision robot is located as the arc movement direction, and obtaining the replanned circular movement trajectory based on the replanned movement radius.

[0012] Furthermore, the control of the vision robot to execute the edge-following process based on the received charging base station avoidance signal, so that the vision robot enters the edge-following walking mode, also includes: during the arc movement of the vision robot according to the circular movement trajectory, the control of the edge sensors on the vision robot body to continuously perform edge detection; when the edge sensors detect the existence of an edge, the vision robot receives the edge detection success signal output by the edge sensors, determines that the vision robot has found the wall, controls the vision robot to stop the arc movement and controls the vision robot to enter the edge-following walking mode based on the edge sensors.

[0013] Furthermore, the step of entering the edge-walking mode based on the edge sensor specifically includes: controlling the visual robot to rotate in the opposite direction to the charging base station relative to the side where the visual robot is located, so that the forward direction of the visual robot's moving structure is parallel to the edge detected by the edge sensor; controlling the edge sensor to perform edge detection in real time, and adjusting the forward direction of the visual robot in real time based on the edge detected by the edge sensor, so as to realize the visual robot entering the edge-walking mode.

[0014] Furthermore, the visual robot mapping method further includes: during the process of constructing a global map based on the edge-walking mode, controlling the visual robot to use a line laser module to identify low obstacles in the environment, and performing edge-walking around the identified low obstacles.

[0015] Furthermore, the visual robot mapping method further includes: during the process of constructing a global map based on the edge-walking mode, controlling the visual robot to use a visual sensor to collect visual images, identifying whether the area near the visual robot meets the conditions for the possible existence of a door area based on the visual images, and if the area near the visual robot meets the conditions for the possible existence of a door area, controlling the visual robot to execute a door area detection process to determine whether a door area exists.

[0016] Furthermore, the control vision robot executes a door area detection process to determine whether a door area exists. Specifically, this includes: when the vision robot is identified as having a door area that meets the conditions for potential existence, the vision robot stops moving, records its current coordinates as re-edge coordinates, and records its current orientation as re-edge orientation; based on the re-edge coordinates and re-edge orientation, a first door area where a door may exist is defined; the vision robot switches from an edge-walking mode to a bow-shaped walking mode; the vision robot moves in a bow shape within the first door area based on the bow-shaped walking mode, and uses a vision sensor to collect visual images, and detects the presence of doors in the first door area in real time based on the visual images.

[0017] Furthermore, the process of controlling the vision robot to perform the door area detection process to determine whether the door area exists also includes: when the vision robot determines that there is a door in the first door area based on real-time detection of the door in the first door area based on the visual image, the vision robot switches from the bow-shaped walking mode to the edge-walking mode, navigates back to the edge coordinates, ends the door area detection process, and continues to execute the global map construction based on the edge-walking mode.

[0018] Furthermore, the process of controlling the vision robot to perform a door area detection procedure to determine whether a door area exists also includes: determining the existence of a door when the vision robot moves in a bow-shaped pattern within the first door area for a preset second time without detecting the existence of a door; or determining the existence of a door when the number of frames of visual images acquired by the vision robot in the door area detection procedure reaches a preset number of image frames without detecting the existence of a door; or determining the existence of a door when the vision robot has traversed the first door area in a bow-shaped pattern without detecting the existence of a door. In this case, the vision robot switches from the bow-shaped walking mode to the edge-walking mode, navigates back to the edge coordinates, ends the door area detection procedure, and continues to execute the global map construction based on the edge-walking mode.

[0019] Furthermore, the visual robot mapping method also includes: if it is determined that there is a door in the first door area, then controlling the visual robot to record the first door area information of the existing door; after the visual robot completes the construction of the global map, the global map is updated based on the first door area information of all existing doors recorded by the visual robot, so as to obtain a global map with door information.

[0020] This application also discloses a method for initial cleaning of a visual robot. This method is based on a global map constructed using the aforementioned visual robot mapping method. Specifically, the initial cleaning method includes: controlling the visual robot to receive room cleaning instructions transmitted from a user terminal, and determining whether the received instructions are whole-house cleaning instructions or specific room cleaning instructions; when the received instructions are whole-house cleaning instructions, controlling the visual robot to start from the charging base station and traverse and clean each room according to the global map; when the received instructions are specific room cleaning instructions, controlling the visual robot to plan a target navigation path from the charging base station to the specific room based on historical edge trajectories during the global map construction process; and controlling the visual robot to move from the charging base station to the specific room according to the target navigation path and perform traversal cleaning of the specific room.

[0021] This application also discloses a chip that stores a computer program internally. When the computer program stored internally in the chip is run by a processor, it executes the visual robot mapping method as described above, or runs the initial cleaning method of the visual robot as described above.

[0022] This application also discloses a visual robot, comprising: a visual sensor for acquiring visual images, enabling the visual robot to detect the presence of doors based on the visual images; a charging base station avoidance signal receiving and detection device for receiving charging base station avoidance signals; an edge sensor for performing edge detection; a line laser sensor for detecting low obstacles; a cleaning device for performing cleaning work; a moving device including a left moving structure and a right moving structure, the left moving structure being disposed on the left side of the bottom of the visual robot body, and the right moving structure being disposed on the right side of the bottom of the visual robot body, for realizing the moving function of the visual robot, wherein the visual robot achieves arc-shaped movement when the left moving robot and the right moving structure move at different speeds; a chip internally storing a computer program, the computer program stored internally in the chip being executed by a processor to perform the visual robot mapping method as described above, or to run the initial cleaning method of the visual robot as described above; and a processor for running the computer program stored internally in the chip.

[0023] The visual robot mapping method, initial cleaning method, chip, and visual robot described in this application first establish a small-area first local map based on the charging base station before constructing the global map. This enables the visual robot to have accurate positioning capabilities near the charging base station. Then, based on the first local map, the installation characteristics of the charging base station being set against a wall, combined with the charging base station's avoidance signal, guide the visual robot to quickly and correctly find the wall, improving the efficiency of the visual robot's initial edge mapping and achieving the technical effect of quickly completing the initial mapping. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a visual robot mapping method according to one embodiment of this application.

[0025] Figure 2 This is a flowchart illustrating the initial cleaning method of a vision robot according to one embodiment of this application. Implementation

[0026] The embodiments of this application will now be described in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described below are for illustrative purposes only and are not intended to limit the scope of this application.

[0027] Currently, visual robots using visual sensors for mapping often suffer from difficulties and long initial mapping times, resulting in a poor user experience. Furthermore, during edge-based mapping, visual robots frequently encounter anomalies where they attempt to map along the legs of tables or chairs in a room. This edge-based mapping under such circumstances is useless for building a global map, severely impacting the mapping efficiency of the visual robot.

[0028] To address the aforementioned challenges, this application provides a visual robot mapping method, aiming to improve the efficiency of initial mapping by the visual robot and achieve the technical effect of quickly completing the initial mapping. Furthermore, it guides the visual robot to map along walls based on a charging base station, avoiding the unnecessary edge-mapping work by the visual robot when encountering obstacles such as furniture. It should be noted that the concept of "wall" in this application includes not only walls in the conventional sense, but also obstacles of a certain length, such as the edges of cabinets and sofas, which allow the visual robot to map along a certain distance in a straight line. Specifically, for example... Figure 1 As shown, the visual robot mapping method specifically includes:

[0029] Step 1: Control the vision robot to build a first local map based on the charging base station. After the first local map is built, proceed to Step 2. In this step, the vision robot is controlled to build a first local map based on the charging base station. Since the charging base station is usually set up in a fixed location in the home, the vision robot uses the charging base station as a reference point to build a small-scale local map. This allows the vision robot to use the first local map as a reference when building the global map, thereby realizing the basic positioning function of the vision robot and improving the accuracy of the global map building.

[0030] Step 2: Control the visual robot to enter edge-walking mode based on the charging base station avoidance signal, and proceed to Step 3; Specifically, in order to ensure a sufficiently wide recharging range for the visual robot, the charging base station is usually placed against a wall. Therefore, this step, by controlling the visual robot to use the charging base station avoidance signal emitted by the charging base station, can efficiently guide the visual robot to walk along the wall, effectively solving the problem of the visual robot's ineffective edge-walking on furniture. It should be noted that the visual robot mapping method provided in this application is implemented for a home environment where the charging base station is placed against a wall.

[0031] Step 3: Control the visual robot to build a global map in edge-walking mode. When the visual robot enters the area of ​​the first local map again along the edge, it is confirmed that the visual robot has completed the construction of the global map. The visual robot mapping method provided in this embodiment first establishes a small first local map based on the charging base station before building the global map. This enables the visual robot to have accurate positioning capabilities near the charging base station. Then, based on the first local map, the installation characteristics of the charging base station being set against the wall guide the visual robot to quickly and correctly find the wall. This avoids the visual robot making mistakes when building maps along the edge of short-distance obstacles such as furniture legs, improves the efficiency of the visual robot's first edge-walking mapping, and achieves the technical effect of quickly completing the first mapping.

[0032] In one implementation, the controlled visual robot constructs a first local map based on the charging base station, specifically including: using the location of the avoidance signal transmitter of the charging base station as the origin of the charging base station coordinate system, using the direction perpendicular to the wall behind the charging base station as the Y-axis of the charging base station coordinate system, using the direction from the wall behind the charging base station to the location of the avoidance signal transmitter as the positive direction of the Y-axis of the charging base station coordinate system, and using the direction parallel to the wall behind the charging base station as the X-axis of the charging base station coordinate system; controlling the visual robot to move horizontally a first distance from the charging base station in the first direction of the X-axis of the charging base station coordinate system, and recording the coordinates of the position of the visual robot after moving horizontally a first distance in the charging base station coordinate system as a first landmark point; controlling the visual robot to move horizontally from the first landmark point towards the charging base station... After moving a first distance along the positive Y-axis of the coordinate system, the coordinates of the points whose X-axis projection distance from the origin of the charging base station coordinate system is a first distance along the first X-axis and whose Y-axis projection distance from the origin of the charging base station coordinate system is a first distance along the positive Y-axis of the charging base station coordinate system are recorded as second landmarks. The vision robot is controlled to start from the second landmark and execute several sets of bow-shaped walking processes in a bow-shaped walking mode to construct a first local map with a projection length of the second distance along the X-axis of the charging base station coordinate system and a projection length of the first distance along the Y-axis of the charging base station coordinate system. The second distance is twice the first distance. The first distance is a pre-set distance value used to limit the width of the first local map, and the second distance is used to limit the length of the first local map. The first distance can be, but is not limited to, 1 meter, 2 meters, etc. It should be noted that this embodiment does not limit the positive and negative directions of the X-axis. The first direction and the second direction of the X-axis are a pair of opposite directions. When the first direction of the X-axis is the positive direction, the second direction is the negative direction, and vice versa. In this embodiment, the second distance is configured to be twice the first distance, thereby constructing a first local map with the location of the charging base station as the first local map. Figure 1 The midpoint of the side edge is used to construct the first local map, ensuring the success rate of the visual robot returning to the charging base station when it first constructs the global map.

[0033] In one implementation, the vision robot is controlled to start from the second waypoint and execute several sets of bow-shaped walking processes in a bow-shaped walking pattern to construct a first local map with a projection length of a second distance on the X-axis of the charging base station coordinate system and a projection length of a first distance on the Y-axis of the charging base station coordinate system. Specifically, this includes: controlling the vision robot to detect its current position in real time while executing each set of bow-shaped walking processes; and when the vision robot moves to a coordinate point where its X-axis projection distance from the origin of the charging base station coordinate system in the second direction of the X-axis is the first distance and its Y-axis projection distance from the origin of the charging base station coordinate system in the positive Y-axis direction is also the first distance, recording the current coordinate point of the vision robot as... The third landmark point; when the visual robot moves to a coordinate point whose X-axis projection distance from the origin of the charging base station coordinate system is a first distance in the second direction of the X-axis and whose Y-axis projection distance from the origin of the charging base station coordinate system is 0 in the positive direction of the Y-axis, the current coordinate point of the visual robot is constructed and recorded as the fourth landmark point; when the visual robot completes the construction of the third and fourth landmark points, the visual robot is controlled to end the execution of several sets of bow-shaped walking processes in the bow-shaped walking mode; the lines connecting the first, second, third, and fourth landmark points form a first local map with a projection length of a second distance in the X-axis direction of the charging base station coordinate system and a projection length of a first distance in the Y-axis direction of the charging base station coordinate system. This embodiment, by pre-defining the positions of the four landmark points of the first local map relative to the charging base station coordinate system, enables the visual robot to construct landmark points when walking in a bow shape, allowing the visual robot to obtain information about the area near the charging base station based on the first local map and achieve positioning and navigation.

[0034] Each bow-shaped walking sequence includes: controlling the visual robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system; then controlling the visual robot to move a first distance in the negative Y-axis direction of the charging base station coordinate system; then controlling the visual robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system; and finally controlling the visual robot to move a first distance in the positive Y-axis direction of the charging base station coordinate system. Specifically, the preset X-axis distance is used to limit the distance the visual robot moves in the X-axis direction each time during the bow-shaped walking sequence. It is set based on the second distance, and the specific setting condition requires that the second distance be controlled to be an integer multiple of the preset X-axis distance so that the visual robot can successfully reach the third waypoint during the bow-shaped walking sequence, and from the third waypoint, move in the negative Y-axis direction so that the visual robot can reach the fourth waypoint.

[0035] As one implementation method, the controlled vision robot enters an edge-walking mode based on a charging base station avoidance signal, specifically including:

[0036] The vision robot is controlled to move to a position three distances from the charging base station. Specifically, the third distance position refers to the position three distances from the center point of the charging base station. It should be noted that the third distance is comprehensively set based on the radiation range of the charging base station avoidance signal and the detection range of the charging base avoidance signal of the vision robot, so that the vision robot can usually receive the charging base station avoidance signal sent by the charging base station when it moves to the third distance position in front of the charging base station.

[0037] The system controls the visual robot to detect whether it receives a charging base station avoidance signal. If the visual robot receives this signal, it executes an edge-following procedure based on the received signal, entering an edge-following walking mode. If the visual robot does not receive this signal, it rotates in place, checking for the signal during rotation. If it does receive the signal, it executes an edge-following procedure based on the received signal, entering an edge-following walking mode. Specifically, the edge-following procedure refers to the visual robot searching for a wall to facilitate entering edge-following walking mode. Since the visual robot can usually receive the charging base station avoidance signal when it reaches the third distance before the charging base station, rotating in place assists in receiving the signal when it fails to do so. Since charging base stations are typically positioned against walls, and the purpose of the charging base station avoidance signal is to prevent the visual robot from being obstructed during movement, guiding the visual robot to avoid the charging base station, the avoidance signal typically radiates as a semi-circular area centered on the center point of the charging base station with a preset avoidance signal radiation distance as its radius. This semi-circular area connects to the wall against which the charging base station is located; the preset avoidance signal radiation distance is the longest distance that the avoidance signal can radiate. Therefore, this embodiment controls the visual robot to detect the charging base station avoidance signal. The semi-circular area radiated by the avoidance signal assists the visual robot in quickly finding walls, thereby efficiently entering the edge-walking mode and avoiding accidental edge-walking due to furniture in the environment, effectively improving edge-walking efficiency.

[0038] In one implementation, the controlled vision robot executes an edge-following process based on the received charging base station avoidance signal, causing the vision robot to enter an edge-following walking mode. Specifically, this includes controlling the left and right moving structures of the vision robot to operate at different speeds, so that the vision robot uses the location of the charging base station's avoidance signal transmitter as the center of its movement, determines the moving radius based on a preset avoidance signal radiation distance, determines a circular movement trajectory based on the center and radius, and controls the vision robot to move from its current position to the position closest to the circular movement trajectory, allowing the vision robot to move in an arc along the circular movement trajectory. During the arc movement, the edge sensor detects an edge, and the vision robot enters the edge-following walking mode. Determining a circle based on the center and radius is a common mathematical method; therefore, how to determine the circular movement trajectory based on the center and radius will not be explained here.

[0039] Specifically, the left-side moving structure of the vision robot refers to at least one set of moving structures located on the left side of the bottom of the vision robot body, and the right-side moving structure of the vision robot refers to at least one set of moving structures located on the right side of the bottom of the vision robot body. The moving structures can be, but are not limited to, wheeled moving structures or tracked moving structures, or other structures that assist the movement of the vision robot.

[0040] Specifically, the principle behind determining the movement radius based on the preset avoidance signal radiation distance is as follows: The distance between the visual robot and the charging base station avoidance signal transmitter is adjusted based on the movement radius. Since the visual robot faces the risk of collision with the charging base station when it can receive the avoidance signal, to ensure the visual robot avoids collisions while still enabling it to quickly find walls and edges guided by the avoidance signal, the movement radius can be, but is not limited to, set to a value greater than or equal to the preset avoidance signal radiation distance. When the movement radius is set equal to the preset avoidance signal radiation distance, the visual robot's current position is the starting point of the circular movement trajectory. Conversely, when the movement radius is set greater than the preset avoidance signal radiation distance, the planned circular movement trajectory does not overlap with the visual robot's current position, requiring the visual robot to move onto the circular movement trajectory. To ensure edge detection on the circular movement trajectory, the visual robot is controlled to move from its current position to the position closest to the circular movement trajectory.

[0041] Preferably, the direction of movement of the visual robot when making arc-shaped movements on the circular movement trajectory can be, but is not limited to, the opposite direction of the charging base station relative to the side where the visual robot is located. This is designed to minimize the distance the visual robot travels to the wall on the circular movement trajectory. When the charging base station is on the right side of the visual robot, considering the location of the charging base station's avoidance signal transmitter, it can be seen that the visual robot can find the wall via a shorter path by moving in an arc to the left. This embodiment selects a shorter circular movement trajectory while ensuring a safe distance between the visual robot and the charging base station, and quickly guides the visual robot along the edge by combining the characteristics of the charging base station's placement, thereby improving edge detection speed and efficiency.

[0042] In one implementation, the control of the vision robot to execute an edge-following process based on the received charging base station avoidance signal, thereby enabling the vision robot to enter an edge-following walking mode, further includes: during the visual robot's arc-shaped movement along a circular trajectory, at preset first time intervals, controlling the vision robot to execute a charging base station avoidance signal reception detection process; if the charging base station avoidance signal reception detection result is that the charging base station avoidance signal cannot be received, then the control robot continues to move along the circular trajectory; if the charging base station avoidance signal reception detection result is that the charging base station avoidance signal can be received, then the control robot replans the circular trajectory and moves according to the replanned circular trajectory. Specifically, the preset first time interval is a pre-designed time length for periodically calibrating the accuracy of the vision robot's edge-following process, which can be, but is not limited to, 3s, 5s, 10s, 30s, etc., adjusted considering various factors such as the vision robot's moving speed and the area of ​​the charging base station avoidance signal radiation region. This implementation determines whether to continue executing the edge-following process by controlling the vision robot to detect whether the charging base station avoidance signal can be received, so that the vision robot can correct errors in the circular trajectory in a timely manner, avoid collisions between the vision robot and the charging base station, and optimize the accuracy of the edge-following process.

[0043] In one implementation, during the arc-shaped movement, the visual robot is controlled to execute a charging base station avoidance signal reception and detection process at preset first time intervals. Specifically, this includes: during the arc-shaped movement of the visual robot along a circular trajectory, the visual robot is controlled to stop moving at preset first time intervals, and the orientation of the visual robot when it stops is recorded as a marked orientation; the visual robot is then controlled to rotate by a preset angle; the charging base station avoidance signal reception and detection device of the visual robot continuously searches for and receives the charging base station avoidance signal during the preset angle rotation; if the charging base station avoidance signal reception and detection device detects the charging base station avoidance signal during the preset angle rotation, the result of the charging base station avoidance signal reception and detection is that the visual robot can receive the charging base station avoidance signal, and the visual robot is controlled to rotate to such that… The robot's orientation is restored to the marked orientation. The vision robot terminates the current charging base station avoidance signal reception and detection process, replans a circular movement trajectory, and continues to move along the replanned circular movement trajectory. During the movement, the vision robot executes the charging base station avoidance signal reception and detection process at preset first time intervals. When the vision robot has rotated to a preset angle and the charging base station avoidance signal reception and detection device has not found the charging base station avoidance signal, the charging base station avoidance signal reception and detection result is that the vision robot has not received the charging base station avoidance signal. The vision robot terminates the current charging base station avoidance signal reception and detection process, rotates until its orientation is restored to the marked orientation, and continues to move along the circular movement trajectory. The preset angle is used to limit the maximum angle that the charging base station avoidance signal reception and detection device needs to rotate to search for the charging base station avoidance signal if the vision robot is within the radiation range of the charging base station avoidance signal. The preset angle is set based on the location and number of charging base station avoidance signal reception and detection devices on the vision robot body. The preset first duration is a pre-designed time length for periodically calibrating the wall-finding accuracy of the visual robot. It can be, but is not limited to, 3s, 5s, 10s, 30s, etc., and is adjusted considering various factors such as the visual robot's moving speed and the area of ​​the charging base station's avoidance signal radiation zone. This embodiment determines whether the current circular movement trajectory can maintain a safe distance between the robot and the charging base station by detecting whether the visual robot can receive the charging base station's avoidance signal, thus avoiding accidental collisions during the robot's arc-shaped movement. Furthermore, it can adjust the circular movement trajectory based on the radiation range of the charging base station's avoidance signal, enabling the robot to successfully find walls within the area radiated by the signal.

[0044] In one implementation, the controlled robot replans its circular movement trajectory, specifically including: extending the radius of the circular movement trajectory by a preset distance to obtain a replanned radius; using the location of the charging base station avoidance signal transmitter as the center of the movement circle, and obtaining a replanned circular movement trajectory based on the replanned radius and the center of the movement circle. The preset distance is a pre-set distance used to gradually increase the arc-shaped movement radius. This allows the robot to avoid the radiation range of the charging base station avoidance signal within a small area when it needs to replan its arc-shaped movement radius after receiving the signal, while still maintaining a certain distance between the robot and the radiation range, enabling the robot to find walls while remaining close to the radiation range of the charging base station avoidance signal.

[0045] In one implementation, the control robot's replanning of the circular movement trajectory further includes: after the vision robot acquires the replanned circular movement trajectory, it performs a charging base station avoidance signal reception detection process on the replanned circular movement trajectory to determine whether the circular movement trajectory needs to be replanned again. This implementation, by controlling the vision robot to perform a charging base station avoidance signal reception detection process on the replanned circular movement trajectory, ensures that the replanned circular movement trajectory maintains a distance between the robot and the radiation range of the charging base station avoidance signal before the robot officially moves along the replanned circular movement trajectory.

[0046] In one implementation, the charging base station avoidance signal reception detection process for the replanned circular movement trajectory, to determine whether the circular movement trajectory needs to be replanned again, specifically includes: controlling the vision robot to move onto the replanned circular movement trajectory; adjusting the orientation of the vision robot on the circular movement trajectory to the marked orientation during the last execution of the charging base station avoidance signal reception detection process; controlling the vision robot to rotate by a preset angle, and controlling the charging base station avoidance signal reception detection device mounted on the vision robot to continuously search for charging base station avoidance signals during the rotation of the preset angle; if the charging base station avoidance signal reception detection device finds a charging base station avoidance signal during the rotation of the preset angle, the charging base station avoidance signal reception detection result is that the robot can receive charging base station avoidance signals. If the charging base station avoidance signal is detected, the visual robot is controlled to rotate until its orientation returns to the marked orientation. The current charging base station avoidance signal reception and detection process ends, indicating that the visual robot needs to replan its circular movement trajectory. If the charging base station avoidance signal reception and detection device fails to detect the signal during the preset rotation angle, the result is that the visual robot has not received the signal. The current charging base station avoidance signal reception and detection process ends, and the visual robot is controlled to rotate until its orientation returns to the marked orientation. This indicates that the visual robot does not need to replan its circular movement trajectory, and the visual robot moves according to the circular movement trajectory.

[0047] In one implementation, the control vision robot executes an edge-following process based on the received charging base station avoidance signal, enabling the vision robot to enter an edge-following walking mode. This further includes: during arc-shaped movement, the edge sensors on the vision robot continuously perform edge detection; when the edge sensors detect an edge, the vision robot receives an edge detection success signal output by the edge sensors, determining that the vision robot has found a wall, controlling the vision robot to stop arc-shaped movement, and controlling the vision robot to enter an edge-following walking mode based on the edge sensors. Specifically, the edge detection performed by the edge sensors can be, but is not limited to, contact detection or non-contact detection. Contact detection refers to the edge sensors detecting the presence of an edge when the vision robot collides with an elastic probe. Non-contact detection can be, but is not limited to, scanning with photoelectric or ultrasonic transmitters and receivers, detecting the presence of an edge based on the reflected information. Since charging base stations are typically set against walls and there are usually no obstacles near them to ensure normal recharging of the vision robot, in this application, when the edge sensors detect an edge while the vision robot is moving along a circular trajectory, it is assumed that the vision robot has found a wall. This implementation method controls the vision robot to continuously perform edge detection using edge sensors during arc-shaped movement, in order to determine in real time whether the vision robot has successfully found a wall. When a wall is successfully found, the vision robot stops the edge-following process, that is, stops the arc-shaped movement and directly enters the edge-following walking mode, thereby improving the edge-following mapping efficiency of the vision robot.

[0048] In one implementation, the step of entering the edge-walking mode based on the edge sensor specifically includes: controlling the vision robot to rotate in a direction away from the center of the circular movement trajectory, so that the forward direction of the vision robot's moving structure is parallel to the edge detected by the edge sensor; controlling the edge sensor to perform edge detection in real time, and adjusting the forward direction of the vision robot in real time based on the edge detected by the edge sensor, so as to realize the vision robot entering the edge-walking mode. This implementation determines the edge-walking direction of the vision robot based on the direction of the center of the circular movement trajectory, avoiding collisions with the charging base station when the vision robot is walking along the edge, and improving edge-walking safety.

[0049] As one implementation method, the visual robot mapping method further includes: during the process of constructing a global map based on the visual robot's edge-walking mode, controlling the visual robot to use a line laser module to identify low obstacles in the environment, and to navigate around the identified low obstacles along the edges. Currently, existing technologies typically rely on infrared sensors to enable robots to walk along edges and walls. However, infrared sensors have the limitation of not being able to assist the robot in avoiding low obstacles, causing the robot to easily get stuck and slip when walking along edges, thus affecting mapping. The line laser module emits a linear laser beam at a fixed angle. When the laser beam illuminates an object, the visual robot's camera can capture it. By utilizing the structural relationship between the line laser module and the camera, combined with the principle of structured light ranging, the information of low obstacles in front of the visual robot can be accurately identified. Compared to existing technologies, this implementation method, by using a line laser module to identify low obstacles in the environment, can improve the success rate of identifying low obstacles in the environment and effectively solve the problem of low obstacles affecting mapping during the visual robot mapping process.

[0050] As one implementation, the visual robot mapping method further includes: during the process of constructing a global map based on the visual robot's edge-walking mode, controlling the visual robot to acquire visual images using a visual sensor, identifying whether the vicinity of the visual robot meets the conditions for the possible existence of a door area based on the visual images, and if the vicinity of the visual robot meets the conditions for the possible existence of a door area, controlling the visual robot to execute a door area detection process to determine whether a door area exists. The method for determining whether the vicinity of the visual robot meets the conditions for the possible existence of a door area based on visual images may be, but is not limited to, matching the line features and point features identified in the acquired visual images with the line features and point features in the preset door area standard image, and determining whether the vicinity of the visual robot meets the preset conditions for the possible existence of a door area based on the matching degree. In current technologies, when robots construct a global map by traversing edges, doorways are often left unmarked because they lack walls to traverse. However, due to the opening and closing characteristics of doors, existing technologies cannot accurately determine whether an area without walls is a doorway. This implementation method accurately identifies doorways using visual images, thereby providing precise feedback on doorways in the constructed global map and improving mapping accuracy. Furthermore, by acquiring visual images during the edge-traversing process to determine whether the vicinity of the visual robot meets the conditions for the possible existence of a doorway, the visual robot can perform a doorway detection process at both ends of the doorway. For example, when the visual robot enters a room by traversing the edge, it performs a doorway detection process. If this doorway detection process fails to detect the door, it can perform another doorway detection process when the visual robot leaves the room by traversing the edge, achieving dual doorway detection and improving the accuracy of doorway detection.

[0051] In one implementation, the control of the vision robot to perform a door area detection process to determine whether a door area exists includes: when the vision robot is identified as having a potential door area, it stops moving, records its current coordinates as re-edge coordinates, and records its current direction of movement as re-edge direction; based on the re-edge coordinates and re-edge direction, it delineates a first door area where a door may exist; it controls the vision robot to switch from an edge-walking mode to a bow-shaped walking mode; it controls the vision robot to move in a bow shape within the first door area based on the bow-shaped walking mode, and uses a vision sensor to acquire visual images, detecting the presence of doors in the first door area in real time based on the visual images. Specifically, by recording the coordinates when the vision robot switches from an edge-walking mode to a bow-shaped walking mode, it ensures that when the vision robot re-enters the edge area after switching from the bow-shaped walking mode, it can restart edge-walking at the same position, guaranteeing that edge mapping is not affected by the door area detection process. This implementation method performs a door area detection process when the conditions for a possible door area are met near the vision robot. By controlling the vision robot to move back and forth in the first defined door area based on a bow-shaped walking pattern, the vision robot can acquire visual images from multiple angles. Based on the multi-angle visual images, the existence of the door area can be more accurately identified, thus improving the detection accuracy of the door.

[0052] Specifically, the step of dividing the first door area, which may contain doors, based on the re-edge coordinates and re-edge orientation includes: using the re-edge coordinates as the origin of the marked coordinate system and the re-edge orientation as the positive Y-axis direction of the marked coordinate system; using the re-edge coordinates as the midpoint of the first area edge of the first door area, the first area edge is parallel to the X-axis of the marked coordinate system, and the length of the first area edge is equal to twice the preset door length; extending the first area edge along the positive Y-axis of the marked coordinate system by twice the preset door length to obtain a second area edge of the first door area, the second area edge is parallel to the first area edge, and the vertical distance between the second area edge and the first area edge is equal to twice the preset door length; connecting the two ends of the first area edge to the two ends of the second area edge respectively to form a third area edge and a fourth area edge; wherein, the first area edge, the second area edge, the third area edge, and the fourth area edge constitute the first door area, and the lengths of the four area edges are equal; the preset door length is a value set in advance based on the door length in the environment.

[0053] As one implementation method, the process of controlling the vision robot to perform a door area detection procedure to determine whether a door area exists also includes: when the vision robot determines the presence of a door in the first door area based on real-time detection of visual images, the vision robot switches from a bow-shaped walking mode to an edge-walking mode, navigates back to the edge coordinates, ends the door area detection procedure, and continues to build a global map based on the edge-walking mode. Specifically, when the vision robot moves back and forth in a bow shape in the first door area to acquire visual images from multiple angles, if the matching degree of line features and point features in a visual image with the preset door area standard image reaches a preset matching degree, then the presence of a door in the first door area is confirmed. At this time, there is no need for the vision robot to continue to acquire visual images from multiple angles for the potentially existing door area. The vision robot can be controlled to switch from a bow-shaped walking mode to an edge-walking mode. By navigating the vision robot to the edge coordinates, there is no need for the vision robot to search for walls to enter the edge, which makes the edge-walking mapping of the vision robot coherent and improves the edge-walking efficiency of the vision robot.

[0054] In one implementation, controlling the vision robot to perform a door area detection process to determine whether a door area exists further includes: if the vision robot moves in a bow-shaped pattern within the first door area for a preset second time period without detecting a door, indicating the existence of a door, then the vision robot switches from a bow-shaped walking mode to an edge-walking mode, navigates back to the edge coordinates, ends the door area detection process, and continues to build a global map based on the edge-walking mode. Specifically, this implementation limits the time the vision robot moves in a bow-shaped pattern within a potentially existing door area. When the vision robot fails to detect a door within the first door area for a preset second time period, the vision robot ends the door area detection process. This avoids the vision robot performing the door area detection process for an extended period, which could affect the efficiency of global map construction, and ensures that the vision robot performs the door area detection process within a reasonable timeframe.

[0055] In one implementation, if the visual robot fails to detect a door after collecting a preset number of visual image frames during the door area detection process, it is considered that a door exists. The robot then switches from a bow-shaped walking mode to an edge-walking mode, navigates back to the edge coordinates, ends the door area detection process, and continues building a global map based on the edge-walking mode. This implementation limits the number of visual image frames collected by the visual robot's visual sensor during the door area detection process. By ending the process when the preset number of frames is reached, it avoids the visual robot continuously collecting visual images for door detection even when no door is detected, thus consuming computational resources.

[0056] In one implementation, if the visual robot has traversed the first door area in a bow-shaped pattern and has not detected a door, thus confirming the existence of a door, the robot switches from bow-shaped walking mode to edge-walking mode, navigates back to the edge coordinates, and ends the door area detection process. It then continues to build a global map based on the edge-walking mode. This implementation limits the traversal of the first door area as the condition for ending the door area detection process, ensuring the visual robot performs only one traversal within that area. If no door is detected after one traversal, the robot terminates the door area detection process for that potentially existing door area.

[0057] As one implementation, the visual robot mapping method further includes: if it is determined that a door exists in the first door area, controlling the visual robot to record the first door area information of the existing door; after the visual robot completes the construction of the global map, updating the global map based on the first door area information of all existing doors recorded by the visual robot to obtain a global map with door information. Specifically, the first door area information may include, but is not limited to, the area range information of the first door area relative to the coordinate system of the charging base station, the visual image frame with the highest matching degree to the preset door area standard image corresponding to the first door area, etc. This implementation realizes the updating of door areas to the global map by recording door area information, so that the global map can contain door information, improving the navigation accuracy and path planning logic of the visual robot based on the global map.

[0058] One embodiment of this application also provides a method for the initial cleaning of a vision robot. The cleaning control method of the vision robot is implemented based on a global map constructed using the aforementioned vision robot mapping method. Specifically, as shown... Figure 2 The initial cleaning method of the vision robot specifically includes:

[0059] The control system receives room cleaning commands from the user terminal and determines whether the received command is for whole-house cleaning or a specific room cleaning. This step involves the vision robot determining whether the command is for whole-house cleaning or a specific room cleaning. Since the global map built by the vision robot is constructed along edges, it lacks information about all areas of the house. Therefore, to clean a specific room, historical edge-following paths need to be considered for accurate navigation path planning. By distinguishing the type of cleaning command, the vision robot can perform more precise and efficient path planning based on the attributes of the global map.

[0060] When the received room cleaning instruction is a whole-house cleaning instruction, the vision robot is controlled to start from the charging base station and traverse and clean each room according to the global map. Since the global map does not contain information on all areas of the house, but has a first partial map, the vision robot is controlled to start from the charging base station and clean each room according to the room partitions contained in the global map.

[0061] When the received room cleaning instruction is for a specific room, the visual robot is controlled to plan a target navigation path from the charging base station to the specified room based on the historical edge-trajectory during the global map building process. The visual robot then moves from the charging base station to the specified room according to the target navigation path and performs a thorough cleaning of the room. Since the global map does not contain information about all areas of the house, directly navigating the visual robot from the charging base station to the door of the specified room would not guarantee the presence of obstacles. Therefore, this step uses the historical edge-trajectory during the global map building process to control the target navigation path from the charging base station to the specified door. This target navigation path may not be the shortest or optimal path, but it ensures the visual robot can successfully move to the specified room, especially when the global map lacks information about each area after the initial mapping.

[0062] In one embodiment of this application, a chip is also provided, which internally stores a computer program. When the computer program stored in the chip is run by a processor, it executes the visual robot mapping method as described above, or runs the initial cleaning method of the visual robot as described above.

[0063] One embodiment of this application also provides a visual robot, including: a visual sensor for acquiring visual images, allowing the visual robot to detect the presence of a door based on the visual images; a charging base station avoidance signal receiving and detection device for receiving charging base station avoidance signals; an edge sensor for performing edge detection; a line laser sensor for detecting low obstacles; a cleaning device for performing cleaning work; a moving device including a left moving structure and a right moving structure, the left moving structure being disposed on the left side of the bottom of the visual robot body, and the right moving structure being disposed on the right side of the bottom of the visual robot body, for realizing the moving function of the visual robot, wherein the visual robot achieves arc-shaped movement when the left moving robot and the right moving structure move at different speeds; a chip internally storing a computer program, the computer program stored internally in the chip being executed by a processor to perform the visual robot mapping method as described above, or to run the initial cleaning method of the visual robot as described above; and a processor for running the computer program stored internally in the chip. This implementation utilizes the wall-mounted installation characteristics of the charging base station, combined with the charging base station's obstacle avoidance signal, to guide the vision robot to quickly and accurately find the wall. This avoids the inconsistencies that occur when the vision robot performs edge mapping along obstacles such as furniture, thus improving the efficiency of the vision robot's initial edge mapping and achieving the technical effect of quickly completing the initial mapping.

[0064] Obviously, the above embodiments are only some embodiments of the present invention, and not all embodiments. The technical solutions of various embodiments can be combined with each other. If terms such as "first," "second," and "third" appear in the embodiments, they are for the purpose of distinguishing related features and should not be construed as indicating or implying their relative importance, order, or number of technical features.

[0065] Those skilled in the art will understand that all or part of the steps in the methods described above can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a microcontroller, chip, or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0066] It should be noted that any process or method description in the flowchart or otherwise described herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order described or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which the embodiments of the invention pertain.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A visual robot mapping method, characterized in that, The visual robot mapping method specifically includes: Step 1: Control the vision robot to build a first local map based on the charging base station. After the first local map is built, proceed to Step 2. Step 2: Control the vision robot to enter the edge-walking mode based on the charging base station avoidance signal, and proceed to Step 3; Step 3: Control the vision robot to build a global map in edge-walking mode. When the vision robot walks along the edge again and enters the area where the first local map is located, it is confirmed that the vision robot has completed the construction of the global map.

2. The visual robot mapping method according to claim 1, characterized in that, The controlled vision robot constructs a first local map based on the charging base station, specifically including: The location of the signal transmitter of the charging base station is taken as the origin of the charging base station coordinate system. The direction of the charging base station perpendicular to the wall behind it is taken as the Y-axis of the charging base station coordinate system. The direction of the charging base station parallel to the wall behind it is taken as the X-axis of the charging base station coordinate system. Control the vision robot to move horizontally a first distance in the first direction of the X-axis of the charging base station coordinate system, starting from the charging base station, and record the coordinates of the position of the vision robot after moving horizontally a first distance in the charging base station coordinate system as the first landmark point; After controlling the vision robot to move a first distance in the positive Y-axis direction of the charging base station coordinate system from the first landmark point, the coordinate point whose X-axis projection distance from the origin of the charging base station coordinate system is a first distance in the first X-axis direction of the charging base station coordinate system and whose Y-axis projection distance from the origin of the charging base station coordinate system is a first distance in the positive Y-axis direction of the charging base station coordinate system is a first distance is constructed and recorded as the second landmark point. The vision robot is controlled to start from the second waypoint and execute several sets of bow-shaped walking processes in a bow-shaped walking pattern to construct a first local map with a projection length of the second distance in the X-axis direction of the charging base station coordinate system and a projection length of the first distance in the Y-axis direction of the charging base station coordinate system; wherein, the second distance is twice the first distance.

3. The visual robot mapping method according to claim 2, characterized in that, The vision robot is controlled to start from the second waypoint and execute several sets of bow-shaped walking processes in a bow-shaped walking pattern to construct a first local map with a projection length of the second distance along the X-axis of the charging base station coordinate system and a projection length of the first distance along the Y-axis of the charging base station coordinate system. Specifically, this includes: When the vision robot performs each bow-shaped walking process, its current position is detected in real time. When the vision robot moves to a coordinate point where the X-axis projection distance from the origin of the charging base station coordinate system in the second direction of the X-axis is the first distance and the Y-axis projection distance from the origin of the charging base station coordinate system in the positive direction of the Y-axis is the first distance, the current coordinate point of the vision robot is recorded as the third landmark point. When the visual robot moves to a coordinate point where the X-axis projection distance from the origin of the charging base station coordinate system in the second direction of the X-axis is the first distance and the Y-axis projection distance from the origin of the charging base station coordinate system in the positive direction of the Y-axis is 0, the current coordinate point of the visual robot is constructed and recorded as the fourth landmark point. Once the vision robot completes the construction of the third and fourth waypoints, control the vision robot to end the execution of several sets of bow-shaped walking processes in the bow-shaped walking mode; The lines connecting the first, second, third, and fourth landmarks form a first local map with a projection length of the second distance in the X-axis direction of the charging base station coordinate system and a projection length of the first distance in the Y-axis direction of the charging base station coordinate system. Each bow-shaped walking process includes: controlling the vision robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system, then controlling the vision robot to move a first distance in the negative Y-axis direction of the charging base station coordinate system, then controlling the vision robot to move a preset X-axis distance in the second direction of the X-axis of the charging base station coordinate system, and then controlling the vision robot to move a first distance in the positive Y-axis direction of the charging base station coordinate system.

4. The visual robot mapping method according to claim 3, characterized in that, The controlled vision robot enters the edge-walking mode based on the charging base station avoidance signal, specifically including: Control the vision robot to move to a position three distances away from the charging base station; Control the vision robot to detect whether it has received a charging base station avoidance signal from the charging base station; When the visual robot receives a charging base station avoidance signal from the charging base station, the visual robot is controlled to execute the edge-walking process based on the received charging base station avoidance signal, so that the visual robot enters the edge-walking mode. If the visual robot does not receive a charging base station avoidance signal from the charging base station, it will control the visual robot to rotate in place and detect whether it receives a charging base station avoidance signal during the rotation. If the visual robot receives a charging base station avoidance signal from the charging base station, it will control the visual robot to execute the edge-walking process based on the received charging base station avoidance signal, so that the visual robot enters the edge-walking mode.

5. The visual robot mapping method according to claim 4, characterized in that, The control vision robot executes an edge-following process based on the received charging base station avoidance signal, enabling the vision robot to enter an edge-following walking mode. Specifically, this includes controlling the left and right moving structures of the vision robot to operate at different speeds, so that the vision robot uses the location of the charging base station avoidance signal transmitter as the center of movement, determines the moving radius based on the preset avoidance signal radiation distance, determines a circular moving trajectory based on the moving center and moving radius, and controls the vision robot to move from its current position to the position closest to the circular moving trajectory, so that the vision robot moves in an arc according to the circular moving trajectory, and enters the edge-following walking mode when the edge sensor detects the edge during the arc movement.

6. The visual robot mapping method according to claim 5, characterized in that, The control vision robot executes an edge-following process based on the received charging base station avoidance signal, enabling the vision robot to enter an edge-following walking mode, and also includes: During the arc-shaped movement of the visual robot following a circular trajectory, the visual robot is controlled to perform the charging base station avoidance signal reception and detection process at preset first time intervals. If the detection result of the charging base station avoidance signal is that the charging base station avoidance signal cannot be received, the robot will continue to move along the circular movement trajectory. If the detection result of the charging base station avoidance signal is that the charging base station avoidance signal can be received, then the robot is controlled to replan its circular movement trajectory and move according to the replanned circular movement trajectory.

7. The visual robot mapping method according to claim 6, characterized in that, During the arc-shaped movement, at preset first time intervals, the vision robot is controlled to execute the charging base station avoidance signal reception and detection process, specifically including: During the arc-shaped movement of the visual robot along a circular trajectory, the visual robot is controlled to stop moving at a preset first time interval, and the orientation of the visual robot when it stops is recorded as the marked orientation. The visual robot is then controlled to rotate by a preset angle. The charging base station avoidance signal receiving and detection device controlling the vision robot continuously searches for and receives the charging base station avoidance signal during the rotation of a preset angle. If the charging base station avoidance signal receiving and detection device detects the charging base station avoidance signal during the rotation of the preset angle, the result of the charging base station avoidance signal receiving and detection is that the visual robot can receive the charging base station avoidance signal. The visual robot is then controlled to rotate until its orientation is restored to the marked orientation. The current charging base station avoidance signal receiving and detection process is then terminated. The visual robot is then controlled to replan its circular movement trajectory and continue to move according to the replanned circular movement trajectory. During the movement, the visual robot is controlled to execute the charging base station avoidance signal receiving and detection process at preset first time intervals. If the charging base station avoidance signal receiving and detection device fails to detect the charging base station avoidance signal after the vision robot has rotated to the preset angle, the result of the charging base station avoidance signal receiving and detection is that the vision robot has failed to receive the charging base station avoidance signal. The vision robot is then controlled to end the current charging base station avoidance signal receiving and detection process, rotated until the vision robot's orientation is restored to the marked orientation, and continues to move along the circular movement trajectory.

8. The visual robot mapping method according to claim 7, characterized in that, The control vision robot to replan its circular movement trajectory specifically includes: extending the movement radius of the circular movement trajectory by a preset distance to obtain a replanned movement radius; taking the location of the charging base station's avoidance signal transmitter as the center of the movement circle, and taking the opposite direction of the charging base station relative to the side where the vision robot is located as the arc movement direction, and obtaining the replanned circular movement trajectory based on the replanned movement radius.

9. The visual robot mapping method according to claim 8, characterized in that, The control vision machine executes an edge-following process based on the received charging base station avoidance signal, enabling the vision robot to enter an edge-following walking mode, and also includes: During the arc-shaped movement of the visual robot following a circular trajectory, the edge sensors on the robot body continuously perform edge detection. When the edge sensor detects the presence of an edge, the vision robot receives the edge detection success signal output by the edge sensor, confirming that the vision robot has found the wall. The vision robot is then controlled to stop its arc-shaped movement and enter the edge-walking mode based on the edge sensor.

10. The visual robot mapping method according to claim 9, characterized in that, The entry into edge-walking mode based on edge sensors specifically includes: Control the vision robot to rotate in a direction away from the center of the circular movement trajectory, so that the forward direction of the vision robot's moving structure is parallel to the edge detected by the edge sensor; The edge sensors are controlled to perform edge detection in real time. Based on the edge detection, the movement direction of the vision robot is adjusted in real time to enable the vision robot to enter the edge walking mode.

11. The visual robot mapping method according to claim 1, characterized in that, The visual robot mapping method further includes: during the process of constructing a global map based on the edge-walking mode, controlling the visual robot to use a line laser module to identify low obstacles in the environment, and performing edge-walking around the identified low obstacles.

12. The visual robot mapping method according to claim 1, characterized in that, The visual robot mapping method further includes: during the process of constructing a global map based on the edge-walking mode, controlling the visual robot to use a visual sensor to collect visual images, identifying whether the area near the visual robot meets the conditions for the possible existence of a door area based on the visual images, and if the area near the visual robot meets the conditions for the possible existence of a door area, controlling the visual robot to execute a door area detection process to determine whether a door area exists.

13. The visual robot mapping method according to claim 12, characterized in that, The controlled vision robot performs a door area detection process to determine whether a door area exists, specifically including: When the vision robot is identified as being in an area where a door may be located, the vision robot is controlled to stop moving. The current coordinates of the vision robot are recorded as the coordinates for re-edge movement, and the current orientation of the vision robot is recorded as the orientation for re-edge movement. Based on the re-alignment and re-orientation along the edge, the first door zone where doors may exist is delineated. Control the vision robot to switch from edge-walking mode to bow-shaped walking mode; The vision robot is controlled to move in a bow-shaped pattern within the first door area, and visual sensors are used to collect visual images to detect the presence of doors in the first door area in real time based on the visual images.

14. The visual robot mapping method according to claim 13, characterized in that, The process of controlling the vision robot to perform a door area detection procedure to determine whether a door area exists also includes: When the vision robot determines that a door exists in the first door area based on real-time detection of the visual image, it controls the vision robot to switch from the bow-shaped walking mode to the edge-walking mode, navigates back to the edge coordinates, ends the door area detection process, and continues to build a global map based on the edge-walking mode.

15. The visual robot mapping method according to claim 13, characterized in that, The process of controlling the vision robot to perform a door area detection procedure to determine whether a door area exists also includes: If the visual robot moves in a bow-shaped pattern within the first door area for a preset second time period without detecting the existence of a door, it is determined that a door exists. Alternatively, if the visual robot collects a preset number of visual image frames during the door area detection process without detecting the existence of a door, it is determined that a door exists. Or, if the visual robot has traversed the first door area in a bow-shaped pattern without detecting the existence of a door, it is determined that a door exists. Then, the visual robot is controlled to switch from the bow-shaped walking mode to the edge-walking mode, navigates back to the edge coordinates, ends the door area detection process, and continues to build a global map based on the edge-walking mode.

16. The visual robot mapping method according to claim 13, characterized in that, The visual robot mapping method further includes: if it is determined that there is a door in the first door area, then controlling the visual robot to record the first door area information of the existing door; after the visual robot completes the construction of the global map, the global map is updated based on the first door area information of all existing doors recorded by the visual robot to obtain a global map with door information.

17. A method for initial cleaning by a vision robot, characterized in that, The cleaning control method of the vision robot is implemented based on a global map constructed by the vision robot mapping method according to any one of claims 1 to 16, and the initial cleaning method of the vision robot specifically includes: The vision robot receives room cleaning instructions transmitted from the user terminal and determines whether the received room cleaning instructions are whole-house cleaning instructions or specific room cleaning instructions. When the received room cleaning instruction is a whole house cleaning instruction, the vision robot is controlled to start from the charging base station and traverse and clean each room according to the global map. When the received room cleaning instruction is a designated room cleaning instruction, the vision robot is controlled to plan the target navigation path from the charging base station to the designated room based on the historical edge trajectory during the global map construction process; the vision robot is controlled to move from the charging base station to the designated room according to the target navigation path and traverse and clean the designated room.

18. A chip internally storing a computer program, characterized in that, The computer program stored inside the chip is executed by the processor to perform the visual robot mapping method as described in any one of claims 1 to 16, or to run the visual robot initial cleaning method as described in claim 17.

19. A visual robot, characterized in that, The visual robot includes: Visual sensors are used to acquire visual images, which are then used by visual robots to detect the presence of doors. A charging base station avoidance signal receiving and detection device; used to receive charging base station avoidance signals. Edge sensors are used to perform edge detection; Line laser sensors are used to detect low-lying obstacles; A cleaning device used to perform cleaning tasks; The mobile device includes a left-side mobile structure and a right-side mobile structure. The left-side mobile structure is located on the left side of the bottom of the visual robot body, and the right-side mobile structure is located on the right side of the bottom of the visual robot body. It is used to realize the mobile function of the visual robot. When the left-side mobile robot and the right-side mobile structure move at different speeds, the visual robot can move in an arc. The chip has a computer program stored inside it. When the computer program stored inside the chip is run by the processor, it executes the visual robot mapping method as described in any one of claims 1 to 16, or runs the visual robot initial cleaning method as described in claim 17. A processor is used to run computer programs stored inside a chip.

Images