A method for identifying a robot entering a narrow lane, a chip and a robot

Abstract

The application discloses a method for recognizing that a robot enters a narrow lane, a chip and the robot. An execution subject of the method is a robot equipped with a ranging sensor. The method comprises the following steps: S1, when the robot detects an obstacle by using the ranging sensor, the robot plans an obstacle-avoiding path leading to a predetermined target position, walks along the obstacle-avoiding path, and obtains a reference contour line segment from point cloud data collected by the ranging sensor; S2, when two obstacle-avoiding points are extracted in a search direction of the reference contour line segment, if it is detected that a walking direction of the robot changes from a reference obstacle-avoiding direction to a direction from a current position to the predetermined target position, and it is detected that a distance between the two obstacle-avoiding points is within a preset distance range, the robot recognizes that it enters the narrow lane.

Description

Technical Field

[0001] This invention relates to the field of mobile robot technology, and in particular to a method, chip, and robot for identifying robot entry into narrow passages based on robot obstacle avoidance. Background Technology

[0002] In the working environment of a robotic vacuum cleaner, there is a working area defined by walls, and a non-working area between two walls. This non-working area generally refers to a narrow passageway, or simply a narrow passage. The entrance and exit of this narrow passageway are both designated as "narrow passage openings." Thus, within the robotic vacuum cleaner's working environment, wall obstacles divide the space into narrow passageways or multiple room areas with narrow passageways. The narrow passageway is flanked by the outlines of two obstacles. When setting the movement path, the robotic vacuum cleaner is generally considered as a single point. To facilitate the robotic vacuum cleaner's movement to a target location within another room area, the movement path may need to pass through a narrow passageway.

[0003] When a robot uses point cloud data detected by line laser detection for real-time localization and synchronous navigation, it usually plans its navigation path autonomously according to the set task. When the robot detects obstacles or narrow passages during its journey, it tries to avoid them by following the outline of the obstacles. However, since the calculated outline of the obstacles may contain errors, the robot may be unable to pass through some narrow passages, thus reducing the navigation success rate. Summary of the Invention

[0004] To address the problem of accurate obstacle avoidance for robots, this invention provides a method, chip, and robot for identifying obstacles. During obstacle avoidance maneuvers, the robot uses laser data to obtain contour lines and extracts relevant obstacle avoidance points through fitting, identifying narrow passages slightly wider than the robot's body width and determining whether it can enter those passages. The specific technical solution is as follows:

[0005] A method for identifying a robot entering a narrow passage, wherein the robot is equipped with a ranging sensor and the identification method includes: Step S1, when the robot detects an obstacle using the ranging sensor, the robot plans an obstacle avoidance path to a predetermined target position, walks along the obstacle avoidance path, and obtains a reference contour line segment through point cloud data collected by the ranging sensor; Step S2, if the robot extracts two obstacle avoidance points in the search direction of the reference contour line segment, and if it detects that the robot's walking direction changes from the reference obstacle avoidance direction to the direction from the current position to the predetermined target position, and detects that the distance between the two obstacle avoidance points is within a preset distance range, then the robot identifies its behavior as entering a narrow passage.

[0006] Furthermore, step S2 also includes: if the robot extracts two obstacle avoidance points in the search direction of the reference contour line segment, and if the distance between the two obstacle avoidance points is detected to be within a preset distance range, then the channel where the two obstacle avoidance points are located is identified as a narrow passage; the direction from the current position to the predetermined target position in step S2 is the passable area inside the channel where the two obstacle avoidance points are located.

[0007] Furthermore, the robot configures the reference obstacle avoidance direction as the direction of the outer contour line formed by the obstacles where the two obstacle avoidance points are located, so as to form the extension direction of the obstacle avoidance path before the robot enters the narrow passage; the robot sets the search direction of the reference contour line segment to the vertical direction of the reference contour line segment, forming the width direction of the passage where the two obstacle avoidance points are located; the reference contour line segment corresponds to the obstacle where one of the obstacle avoidance points is located, and forms one of the boundary lines of the passage where the two obstacle avoidance points are located.

[0008] Furthermore, during the process of the robot walking along the obstacle avoidance path, if the robot adjusts its walking direction to point from the current position to the predetermined target position, and detects that the distance between the two obstacle avoidance points is within the preset distance range, the robot determines that it has begun to enter a narrow passage. The narrow passage is a gap formed between the obstacles where the two obstacle avoidance points are located. The gap is located in the passable area. The robot also configures the direction from the current position to the predetermined target position to be different from the reference obstacle avoidance direction.

[0009] Furthermore, between steps S1 and S2, the method further includes: after the robot detects an obstacle and before extracting two obstacle avoidance points in the search direction of the reference contour line segment of another obstacle, the robot continues to walk along the obstacle avoidance path and sets the current extension direction of the obstacle avoidance path as the historical obstacle avoidance direction; during the robot's movement along the obstacle avoidance path, after the robot obtains the reference contour line segment, if the robot detects that the extension direction of the obstacle avoidance path starting from the current position is not the direction from the current position to the predetermined target position, the robot stops continuing to walk along the obstacle avoidance path and adjusts its walking direction to the direction from the current position to the predetermined target position so that the robot begins to enter the channel where the two obstacle avoidance points are located.

[0010] Furthermore, when the robot first detects the first obstacle using a ranging sensor, it plans an obstacle avoidance path to the predetermined target position. The robot then walks along this path to follow the outline of the first obstacle until it reaches the gap between the first and second obstacles and / or collides with the second obstacle. The robot then marks the direction of its movement as the historical obstacle avoidance direction. The reference outline of the second obstacle has a first endpoint and a second endpoint. The robot then acquires the reference outline of the second obstacle, marks the direction from the first endpoint to the second endpoint as the direction from the current position to the predetermined target position, and marks the direction from the second endpoint to the first endpoint as the extension direction of the historical obstacle avoidance direction. The robot configures the extension direction of the historical obstacle avoidance direction as the reference obstacle avoidance direction.

[0011] Furthermore, the extension direction of the passable area inside the channel where the two obstacle bypass points are located is parallel to the reference contour line segment of the second obstacle, the reference contour line segment of the second obstacle is perpendicular to the search direction, the passable area inside the channel where the two obstacle bypass points are located is connected to the predetermined target position, and one of the obstacle bypass points is located on the reference contour line segment of the second obstacle.

[0012] Furthermore, the robot collects point cloud data through a ranging sensor. The point cloud data is configured to reflect the position information of obstacles detected by the ranging sensor. The robot then fits the collected point cloud data into the outline of the obstacle to represent the local outline of the detected obstacle or the obstacle envelope. The reference outline segment belongs to the fitted outline. When the robot marks the extension direction of the passable area inside the channel where the two obstacle avoidance points are located as the preset channel direction, the two obstacle avoidance points are respectively located on the outline of the first obstacle in the preset channel direction and the outline of the second obstacle in the preset channel direction, or the two obstacle avoidance points are respectively located on the first obstacle and the second obstacle.

[0013] Furthermore, when the robot detects the first obstacle, the position it has traversed along the obstacle avoidance path and the predetermined target position are located on opposite sides of the first obstacle. The robot then marks the side where it has traversed along the obstacle avoidance path as the first side and the side where the predetermined target position is located as the second side. The robot also configures the first endpoint of the reference contour line segment of the second obstacle to be located on the first side of the second obstacle or the first side of the first obstacle, and configures the second endpoint of the reference contour line segment of the first obstacle to be located on the second side of the second obstacle or the second side of the first obstacle. Then, the robot marks the direction of the contour line formed by connecting the contour line of the first side of the first obstacle and the contour line of the first side of the second obstacle as the direction of the outer contour line composed of the obstacles where the two obstacle avoidance points are located.

[0014] Furthermore, when the robot first detects the first obstacle using the ranging sensor, the robot has already collided with the first obstacle; when the robot walks to the gap formed between the first obstacle and the second obstacle, the robot has already collided with the second obstacle; the robot sets the preset distance range to be greater than or equal to the robot's body width, and the upper limit of the preset distance range is the sum of the robot's body width and the preset redundancy.

[0015] A chip storing program code, which, when executed, implements the steps of the robot entry into a narrow passage recognition method as described above.

[0016] A robot equipped with a ranging sensor, wherein the robot is configured with the aforementioned chip, which controls the robot to use the ranging sensor to detect obstacles and obtain their corresponding outlines and obstacle avoidance points, so as to facilitate the identification of the robot entering a narrow passage.

[0017] The beneficial technical effect of this invention is that, when the robot walks along the pre-planned obstacle avoidance path, it is easy for the robot to misjudge a narrow passage formed between two obstacles that is just larger than the width of the robot body by a small gap distance (1cm) as impassable. The invention obtains the contour line segment associated with the passage factor of the gap formed between the two obstacles, which guides the robot to adjust its obstacle avoidance direction before and after entering the gap. When the obstacle avoidance direction changes, the distance between the two obstacle avoidance points searched in the search direction of the contour line segment is used to identify the passage where the two obstacle avoidance points are located as a narrow passage, and it is determined that the robot has started to enter a narrow passage but not through the pre-planned obstacle avoidance path. Therefore, it can accurately distinguish narrow passages that robots can pass through in the environment where robot vacuums, lawnmowers, or mobile toys are located. This overcomes the problem that the width of narrow passages is so small that the grid area inside the narrow passage or its opening is easily misjudged as an obstacle blocking the robot's passage. It also enables a smooth transition from the robot's action of bypassing obstacles to entering the narrow passage formed between two obstacles. In this way, the robot can accurately pass through the narrow passage or the narrow passage opening based on laser data, thereby improving the smoothness and success rate of passing through narrow areas under laser navigation conditions.

[0018] No matter how frequently the robot collides with the obstacles it is trying to avoid before entering the passage containing the two obstacle avoidance points, as long as the robot's walking direction changes from the reference obstacle avoidance direction to the direction pointing from the current position to the predetermined target position (i.e., the previous obstacle avoidance direction is different from the current obstacle avoidance direction), and the distance between the two obstacle avoidance points is within the preset distance range, the robot will recognize the passage containing the two obstacle avoidance points as a narrow passage. This makes the robot's navigation from the current position to the predetermined target position smoother or the navigation path shorter, thereby improving the robot's navigation efficiency. Attached Figure Description

[0019] Figure 1 This is a flowchart of a narrow passage identification method based on robot obstacle avoidance, disclosed in one embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the robot's movement trajectory before and after entering and exiting a narrow passage, as disclosed in another embodiment of the present invention. In the diagram, the black rectangle at the top represents the second obstacle, and the black rectangle at the bottom represents the first obstacle.

[0021] Figure 3 This is a schematic diagram of a robot changing its obstacle-avoiding direction according to another embodiment of the present invention, wherein the black rectangle at the top of the diagram represents the second obstacle. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. To further illustrate the embodiments, the present invention provides accompanying drawings. These drawings are part of the disclosure of the present invention, mainly used to illustrate the embodiments, and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementations and the advantages of the present invention. The flowchart depicts a process or method. Although the flowchart describes the steps as sequential processes, many of the steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. The process can be terminated when its operation is complete, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0023] For narrow passages, the point cloud data (radar points) collected by the laser sensor disclosed in this embodiment are all represented in discrete coordinates. These discrete coordinates are specifically used to represent the location of obstacles. Most of them need to be filtered and grouped, and then the grouped point cloud data is fitted to obtain a fitted straight line or fitted curve. Then, based on the ratio between world coordinates and image coordinates, the obstacle envelope is obtained through the coordinate points of the fitted curve. The obstacle envelope is displayed in real time on the grid map (the point cloud data is discretely distributed around the obstacle). This not only corresponds to the laser detection distance from the obstacle to the edge of the robot, but can also be used to represent the obstacle position and the obstacle outline. When using a line laser sensor to collect point cloud data (radar points), since the line laser sensor projects a straight line, it can measure the distance of low obstacles in front of the robot and convert it into the obstacle's point cloud. Point cloud data can be fitted into straight line segments. In some embodiments, the two endpoints of these straight line segments are configured as reference points for the robot to bypass obstacles, which can be regarded as obstacle-avoidance points. Due to the accumulated errors of the robot's sensors, different obstacle materials, and interference from different lighting environments, the obstacle envelope (the outer contour of the obstacle, a fitted curve or a connection of multiple fitted curves) obtained from the point cloud data collected by the laser sensor cannot match the actual obstacle position. As a result, the robot excludes narrow passages from its passable area when searching for a path. Even if the grid side length is set appropriately, the grid occupied by narrow passages with smaller widths will be marked as obstacle grids. Therefore, the error of the obstacle envelope will cause the robot to misjudge narrow passages. The robot cannot find a path to enter the narrow passage, resulting in the robot being unable to pass through the narrow passage. For example, when a robot vacuum cleaner is cleaning, if it travels to a narrow passage formed by two walls or two low obstacles that is slightly wider than the robot's body, it will actively avoid the obstacle by making a sharp turn. If the robot vacuum cleaner's collision sensor detects a collision, it will turn to the opposite side and move around the obstacle, but it will not attempt to enter the narrow passage. This will prevent it from achieving full coverage cleaning, reduce the success rate of robot navigation, and make it easy for it to get stuck between four table or chair legs and unable to extricate itself.

[0024] To enable robots to navigate smoothly around obstacles and enter narrow passages, this invention discloses a narrow passage identification method based on adjusting the obstacle avoidance trend direction. The main implementer of the narrow passage identification method is a robot equipped with a ranging sensor. The ranging sensor can be a line laser sensor, which acquires the position information reflected back from obstacles by emitting a line laser beam externally. Line laser sensors include multi-line lidar and single-line lidar. Single-line lidar refers to a radar where the laser source emits a single-line beam. It is widely used in robotics, primarily in service robots, to help robots avoid obstacles. It features fast scanning speed, high resolution, and high reliability. Compared to multi-line lidar, single-line lidar reacts faster in terms of angular frequency and sensitivity, thus providing greater accuracy in measuring the distance to surrounding obstacles. This method is particularly suitable for cleaning robots that walk on ground surfaces, lawnmowers that navigate lawns with narrow passages defined by boundary lines, floor scrubbers, and security patrol robots. This embodiment does not limit the types of entities to which the narrow passage identification method is applicable. The robot can also be equipped with inertial sensors (including but not limited to odometers for measuring walking distance, collision sensors for detecting collisions with obstacles, and gyroscopes for measuring the rotation angle of the robot body) or visual sensors (which can be any type of depth information acquisition device, including but not limited to monocular cameras, binocular cameras, etc.) to detect two-dimensional point cloud data of the surrounding environment and build a two-dimensional point cloud map in a timely manner. The number of sensors installed on the robot body can be one or more.

[0025] See Figure 1 It is understood that the narrow passage recognition method includes: Step S1, when the robot detects an obstacle using a ranging sensor, the robot plans an obstacle-avoidance path to a predetermined target position, then walks along the obstacle-avoidance path, and uses the point cloud data collected by the ranging sensor to obtain a reference contour line segment. Before the robot executes step S1, the robot uses a heuristic search algorithm to plan a navigation path, including a navigation path searched using the A* algorithm on the map, with the predetermined target position being the endpoint of the navigation path; after planning the navigation path, the robot walks along the navigation path. When the robot detects an obstacle using the ranging sensor, such as when it detects an obstacle for the first time, or even when it collides with an obstacle, the robot starts from its current position or a preset starting point and uses a heuristic search algorithm to plan a navigation path to bypass the currently detected obstacle, marking the navigation path as an obstacle-avoidance path, with the predetermined target position being the endpoint of the obstacle-avoidance path. Specifically, before the robot starts walking along the obstacle-avoidance path, regardless of whether it collides with an obstacle, its current position or preset starting position is used as the search starting point, and the direction from the search starting point to the predetermined target position corresponds to... Figure 1 and Figure 2In the search algorithm, the starting point (start) points towards the predetermined target location (target). During the path planning process using a heuristic search algorithm, the robot searches for free grids (grid areas representing passable areas in a grid map) in the neighborhood of the starting point as path nodes to ensure that the path nodes are not located in areas occupied by obstacles. The robot finds multiple path nodes in the neighborhood of the starting point; these path nodes are all free grids, and each path node corresponds to an initial search direction. Then, each path node is updated as a new starting point, and free grids are searched in the neighborhood of each starting point in the direction pointing to the same predetermined target location. This process is repeated until the predetermined target location is found in the grid map. The searched grids are then connected in the order of the search to form multiple navigation paths or obstacle avoidance paths. The robot finds one navigation path or obstacle avoidance path corresponding to each initial search direction. The width of the grid occupied by each navigation path or obstacle avoidance path in the path width direction is greater than the robot's body width, and each navigation path or obstacle avoidance path is a passable path extending from the robot's current position to the predetermined target location. Preferably, the neighborhood can be a grid area consisting of 4, 8, or 12 neighborhoods centered on the grid where the robot's current position (the center position of the robot's body) is located, or it can be a range of 10, 15, 20, or 30 grids in the four directions of front, back, left, and right of the robot's center position. The number of grids of 10, 15, 20, or 30 is only an example.

[0026] In step S1, the robot typically uses the location of its first collision with an obstacle as the starting point for its search. Then, as the robot moves along the planned obstacle-avoidance path from this starting point, if it collides with another obstacle, it can be considered to have detected that obstacle. The robot then obtains a reference contour segment from the point cloud data collected by the ranging sensor. This reference contour segment is a fitting result of the point cloud data reflected from the other obstacle (a partial segment of the other obstacle's envelope, which can overlap with the robot's current position). This facilitates the exploration of gaps or channels formed between the two detected obstacles. In some embodiments, the reference contour segment can be a fitted line segment calculated based on the point cloud data pre-collected by the robot (e.g., derived from the contour of the first obstacle collision) or parallel to the fitted line segment. The robot may not have detected or collided with the other obstacle, but it can predict the possible gap between the two different obstacles. Therefore, the robot considers the reference contour segment as the boundary line where one end of the gap or one side of the channel is located.

[0027] Step S2: If the robot extracts two obstacle avoidance points in the search direction of the reference contour line segment, and if it detects that the robot's walking direction changes from the reference obstacle avoidance direction to the direction from the current position to the predetermined target position, and detects that the distance between the two obstacle avoidance points is within a preset distance range, then the robot recognizes its behavior as entering a narrow passage. At this time, the robot enters the narrow passage from the outside of the narrow passage at the instant, or the robot is in front of the entrance of the passage where the two obstacle avoidance points are located and its walking direction is to the passable area inside the passage where the two obstacle avoidance points are located, and the robot may twist left and right. Specifically, after acquiring the reference contour segment in step S1, the robot extracts two obstacle avoidance points along the search direction of the reference contour segment. This can be done by selecting one obstacle avoidance point and then extracting another along the search direction. These obstacle avoidance points are the endpoints of the fitted contour segment and are the two closest endpoints located on the two reference contour segments, forming obstacle avoidance points on the left and right sides of the robot. Then, the robot can continue searching for a new obstacle avoidance path from its current position, and then bypass the obstacle where one of the obstacle avoidance points is located along this new path, or adjust its walking direction to pass through the gap formed between the two extracted obstacle avoidance points in a straight line. The obstacle avoidance direction formed by the robot then differs from the obstacle avoidance direction formed in step S1, thus deviating from the original obstacle avoidance trend. If the distance between the two extracted obstacle avoidance points is still within a preset distance range, then it is determined that the robot is now entering a narrow passage. Preferably, the preset distance range is greater than the robot's body width. When the robot's shape is circular, the robot's body width is its diameter.

[0028] It should be noted that when the robot detects that the distance between the two obstacle avoidance points is within a preset distance range, the robot identifies the passage where the two obstacle avoidance points are located as a narrow passage. Furthermore, when the robot detects that its walking direction changes from the reference obstacle avoidance direction to a direction pointing from its current position to the predetermined target position, the robot determines that it has entered the narrow passage. Specifically, the robot configures the direction from its current position to the predetermined target position to point towards a passable area inside the passage where the two obstacle avoidance points are located. The current position is a path node traversed by the obstacle avoidance path planned in step S1, at which the robot's obstacle avoidance trend changes. Preferably, the preset distance range is greater than or equal to the robot's body width. When the width of the entrance to the passage where the two obstacle avoidance points are located (i.e., the gap formed between the two obstacle avoidance points) is greater than or equal to the robot's body width, it is determined that the passage where the two obstacle avoidance points are located allows the robot to enter. Then, when the robot walks to one of the obstacle avoidance points or its vicinity to collide with a new obstacle, it can adjust its walking direction to point from the current position to the predetermined target position. The adjusted direction can be a straight line pointing to the predetermined target position, or a zigzag path through the passage where the two obstacle avoidance points are located to extend to the predetermined target position. Then, when it is detected that the robot's walking direction changes from the reference obstacle avoidance direction to the direction from the current position to the predetermined target position, and when it is detected that the distance between the two obstacle avoidance points is within the preset distance range, the robot recognizes its behavior as entering a narrow passage, so that the robot can smoothly enter and pass through the narrow passage. On the other hand, if it is detected that the width of the entrance to the channel where the two obstacle avoidance points are located (i.e., the gap formed between the two obstacle avoidance points) is less than the width of the robot's body, and it is determined that the robot is not allowed to enter the channel where the two obstacle avoidance points are located, the robot may resume using the obstacle avoidance path in step S1, or avoid the obstacle, or adjust its walking direction to the opposite direction from the current position to the predetermined target position, or adjust its walking direction to extend from the current position to the outline of the most recently collided obstacle, so that the robot does not enter the channel where the two obstacle avoidance points are located, plans to form a new obstacle avoidance path, keeps walking along the outline of the obstacle, and avoids the new obstacle.

[0029] In summary, when the robot is walking along the pre-planned obstacle avoidance path, it is prone to misjudging narrow passages formed between two obstacles that are just a small gap (1cm) larger than the width of the robot body as impassable. To address this issue, the robot obtains contour lines associated with the passage factors of the gap formed between the two obstacles. These contour lines guide the robot's adjustment of its obstacle avoidance direction before and after entering the gap. When the obstacle avoidance direction changes, the robot combines the distance between two obstacle avoidance points extracted from the search direction of the contour lines to identify the passage where the two obstacle avoidance points are located as a narrow passage. This determines that the robot is entering a narrow passage but not through the pre-planned obstacle avoidance path. Therefore, it can accurately distinguish narrow passages that robots can pass through in the environment where robot vacuums, lawnmowers, or mobile toys are located. This overcomes the problem that the width of narrow passages is so small that the grid area inside the narrow passage or its opening is easily misjudged as an obstacle blocking the robot's passage. It also enables a smooth transition from the robot's action of bypassing obstacles to entering the narrow passage formed between two obstacles. In this way, the robot can accurately pass through the narrow passage or the narrow passage opening based on laser data, thereby improving the smoothness and success rate of passing through narrow areas under laser navigation conditions.

[0030] Based on the above embodiments, during the process of searching for each path leading to the predetermined target location within the grid map, the robot can, according to the direction from its current position to the predetermined target location, systematically mark the index numbers of the grids corresponding to the searched paths, and set the extension direction of the grids corresponding to the index numbers in ascending order as the robot's walking direction on the searched path, thereby determining the robot's walking direction on the searched path (including the obstacle avoidance path). Preferably, when the current position needs to pass through the narrow passage during the process of pointing to the predetermined target location, the direction from the current position to the predetermined target location in step S2 can be the direction from the entrance of the narrow passage to the exit of the narrow passage, ensuring that it points to the passable area inside the passage where the two obstacle avoidance points are located. Then, the robot's walking direction into the narrow passage can be determined according to the index number of the grid closest to the entrance of the narrow passage. For example... Figure 2As shown, before entering the gap between the two obstacles above and below, the robot collides with the obstacle below at the start position. This start position is marked as the search starting point and labeled with index number ①. Then, along the boundary line of the obstacle below, index numbers ② and ③ are marked at the searched path nodes. At the position corresponding to index number ③, the robot's walking direction changes to face the inside of the gap and towards the target position. The position corresponding to index number ③ is the closest position to the gap and is also the position where the robot changes its obstacle-avoiding direction. After determining that it has entered the gap, the robot marks index numbers ④, ⑤, and ⑥ at the searched path nodes according to the principle of passing through the passage of the gap, until the target position is found. The path search algorithm here includes, but is not limited to, heuristic search algorithms such as the A* algorithm and the D* algorithm, so that the searched path can avoid the obstacles shown in the figure.

[0031] As one embodiment, the robot configures the reference obstacle avoidance direction as the outline of the obstacles located at each of the two obstacle avoidance points, thus forming the extension direction of the obstacle avoidance path before the robot enters the narrow passage; wherein, each obstacle avoidance point corresponds to an obstacle, and the robot passing through an obstacle avoidance point can be regarded as bypassing the obstacle located at that obstacle, and the reference obstacle avoidance direction can be shown as along Figure 2 The direction in which the line connecting the outer contours of the two rectangular obstacles shown extends can be understood as enclosing... Figure 2 The curves on the outer sides of the two rectangular obstacles shown follow a trend close to the outlines of the two obstacles. When the robot enters the gap (opening) between the two obstacle-avoidance points, the robot simultaneously bypasses the obstacles at each of the two obstacle-avoidance points or passes through the gap formed between the obstacles at each of the two obstacle-avoidance points. On the other hand, in this embodiment, the robot sets the search direction of the reference contour line segment to the vertical direction of the reference contour line segment, forming the width direction of the channel where the two obstacle-avoidance points are located; the reference contour line segment corresponds to the obstacle at one of the obstacle-avoidance points, corresponding to... Figure 2 The obstacle above, line segment AB is the reference contour line segment, which can pass through... Figure 2 The fitted line segment at the lower left corner of the obstacle above may not necessarily belong to the actual outline of the obstacle, but it can form one of the boundary lines of the channel where the two obstacle-around points are located, that is, the boundary line of the narrow passage identified in the aforementioned embodiment. Figure 2 If line segment CD is considered perpendicular to line segment AB, then points C and D are two points around the obstacle. Point C can be located at... Figure 2A corner point of the obstacle (black rectangle) shown can also be a point on the outline of the obstacle in the direction perpendicular to AB, to fit the outline and distribution of the obstacle. The two obstacle avoidance points are points on the reference outline segments of the obstacles closest to the robot's left and right sides during the execution of the recognition method. They can also be understood as the two endpoints with the smallest distance between them, with a gap between them. In some embodiments, the distance between the two obstacle avoidance points can also be the minimum distance between the endpoints of the reference outline segments where each obstacle avoidance point is located, used to determine the passability of the gap between the two obstacle avoidance points.

[0032] As one embodiment, during the robot's movement along the obstacle avoidance path, if the robot adjusts its direction of travel to point from its current position to a predetermined target position, and detects that the distance between the two obstacle avoidance points is within the preset distance range, the robot determines that it is entering a narrow passage. This narrow passage is a gap formed between the obstacles where the two obstacle avoidance points are located. This gap is located within a passable area, and the boundary line of the passage containing the gap can be the outline of the obstacle or a fitted outline segment. The spacing between the boundary lines is within the preset distance range to improve judgment accuracy. Preferably, the width of the gap is the distance between the grids containing the two endpoints of the gap on the robot's travel plane, and the two endpoints of the gap are respectively located on the outline of the corresponding obstacle or the fitted outline segment. Additionally, the robot configures the direction from its current position to the predetermined target position to be different from the reference obstacle avoidance direction. Optionally, the reference obstacle avoidance direction corresponds to... Figure 2 The arrow pointing from the last position (which can be marked as the previous obstacle avoidance direction) corresponds to the direction from the current position to the predetermined target position. Figure 2 The arrows originating from the next position point (which can be marked as the current obstacle avoidance direction) all point to passable areas to facilitate the robot's avoidance of surrounding obstacles. Specifically, when the robot chooses to collide with an obstacle at one of the obstacle avoidance points or walks to the point where it collides with an obstacle (not the obstacle detected in step S1), the robot adjusts its walking direction to point from the current position to the predetermined target position. Then, when the distance between the two obstacle avoidance points is detected to be within the preset distance range, the gap formed between the currently collided obstacle and the obstacle detected in step S1 is considered a narrow passage, and the robot begins to enter this narrow passage. The reference contour line segment is set to belong to the currently collided obstacle. The obstacle avoidance path can belong to the historical obstacle avoidance path. The robot will then determine the path based on the obstacle avoidance path, the predetermined target position, and the predetermined target position. Figure 1 and Figure 2The search starting point (start) determines the robot's obstacle avoidance direction, allowing it to choose to bypass the corresponding obstacle from an obstacle avoidance point on its left or right. This can involve twisting left and right and colliding with two obstacle avoidance points sequentially. Regardless of how frequently the robot collides with the obstacle before entering the passage containing the two obstacle avoidance points, as long as the robot's direction changes from the reference obstacle avoidance direction to a direction pointing from the current position to the predetermined target position (i.e., the previous obstacle avoidance direction is different from the current one), and the distance between the two obstacle avoidance points is within the preset distance range, the robot will recognize the passage containing the two obstacle avoidance points as a narrow passage. This makes the robot's navigation from the current position to the predetermined target position smoother, the navigation path shorter, and improves the robot's navigation efficiency.

[0033] In some embodiments, the robot's working area may include a working area defined by boundary lines, and non-working areas may be formed between obstacles, or between working areas defined by boundary lines. This non-working area could be formed between corresponding boundary lines of two working areas, or between gaps or passages between two wall obstacles. Here, a non-working area generally refers to a narrow passage, or simply a narrow passage. The entrance and exit of the narrow passage are both designated as narrow passage openings. The narrow passage is flanked by the outlines of two obstacles. Furthermore, when setting the movement path, the robot is generally considered as a single point. To facilitate the robot's movement to the target location, the movement path may pass through... Narrow passages; generally, gaps between two or more obstacles (such as passages between two walls) are displayed as free passages in the grid map. A free passage refers to a path that connects two different work areas and can be a passageway for the robot, provided that its width is greater than the robot's body width. Specifically, the narrow passages that allow the robot to pass are identified by detecting the entrance, exit, length, and width of the passage, and the location information of the narrow passage is determined. For example, the grid area of ​​a doorway under a wall, a passage between two walls, or a passage surrounded by three walls with only one gap to be identified. Passages in the actual working scenario of the robot have certain characteristics, including three-dimensional shape features and size features.

[0034] As one embodiment, the obstacle avoidance path described in step S1 is a path planned by the robot using a heuristic search algorithm. The predetermined target position is the endpoint of the obstacle avoidance path. The specific search starting point can be the location where the robot first collides with the obstacle, a location selected near the obstacle, or the obstacle avoidance point or its vicinity. Between steps S1 and S2, the robot continues to move along the obstacle avoidance path, and before extracting two obstacle avoidance points along the search direction of the reference contour line segment of another obstacle (a newly detected obstacle, which can be located close to a previously detected obstacle and considered adjacent), and sets the current extension direction of the obstacle avoidance path as the historical obstacle avoidance direction, corresponding to... Figure 2 The position of index number ② points in the direction of index number ③, or from... Figure 3 The arrow pointing from the last position.

[0035] Between steps S1 and S2, the method further includes: during the robot's movement along the obstacle avoidance path, after the robot acquires the reference contour segment, if the robot detects that the extension direction of the obstacle avoidance path from the current position is not the direction pointing from the current position to the predetermined target position, i.e. Figure 2 The arrow at index number ② does not point to the intended target location or from... Figure 3 When the arrow originating from the last position does not point to the predetermined target position, the robot stops moving along the obstacle avoidance path and adjusts its direction of movement to point from the current position to the predetermined target position so that the robot can begin entering the channel containing the two obstacle avoidance points. Specifically, the adjustment is as follows: Figure 3 The arrow pointing to the position of index number ③, or Figure 2 The arrow pointing to the position of index number ③ (corresponding to) Figure 2 The cur position points to Figure 2 (The direction of the next position), so that the robot begins to enter the passage where the two obstacle avoidance points are located, then after the robot recognizes that its behavior is entering the narrow passage or after recognizing the narrow passage, it continues to... Figure 2 Go straight through the gap between the obstacles above and below (e.g., along the gap between the obstacles above and below). Figure 2 (The direction of index number ③ points to the direction of index number ④), or crosses... Figure 3 The passable area located at line segment CD is along Figure 3The robot moves in the direction of the next position, pointing from index number ③. This allows the robot to find narrow, easily misjudged passages between obstacles during obstacle avoidance and adjust its direction to enter them, instead of continuing to circle around them. This shortens the robot's overall navigation path and speeds up navigation to the predetermined target location.

[0036] As one example, combined with Figure 2 and Figure 3 It can be seen that this embodiment will Figure 2 and Figure 3 The obstacle above is marked as the second obstacle. Figure 2 The obstacle below is marked as the first obstacle; in Figure 3 In the middle, the reference contour line segment AB of the second obstacle has a first endpoint A and a second endpoint B, which are located at the second obstacle ( Figure 3 The left and right sides of the lower left corner of the rectangle filled with black shown. In this embodiment, when the robot walks along the path pre-planned by the heuristic search algorithm, when the robot first detects the first obstacle at the start position using a ranging sensor (such as a laser sensor), the robot may collide with the first obstacle. At the same time, the robot plans an obstacle-avoidance path to the predetermined target position. Specifically, it uses the collision position as the search starting point and the predetermined target position as the search ending point to plan an obstacle-avoidance path using the heuristic search algorithm. Then the robot walks along this obstacle-avoidance path to achieve walking along the outline of the first obstacle until it reaches the gap formed between the first obstacle and the second obstacle, corresponding to... Figure 2 In the diagram, the robot starts from the "start" position and moves along the arrow pointing from index ① to index ②. Then, it moves along the arrow pointing from index ② to index ③. The robot then reaches the gap between the first and second obstacles. Figure 3 At the position of cur, the robot can be positioned between the two obstacle avoidance points (which can correspond to...). Figure 3 When the distance between points C and D is within a preset distance range, the passage where the two obstacle-avoidance points are located is identified as a narrow passage. Simultaneously, the direction formed by the robot's movement is marked as the historical obstacle-avoidance direction, corresponding to the arrow pointing at index number ②; for example... Figure 3As shown, the robot then obtains the reference contour line segment AB of the second obstacle, and marks the direction from the first endpoint A to the second endpoint B as the direction from the current position to the predetermined target position (corresponding to the arrow pointing at the position of index number ③) to pass through the channel where the two obstacle bypass points are located. The direction from the second endpoint B to the first endpoint A is marked as the extension direction of the historical obstacle bypass direction. Then, as the robot walks along the extension direction of the historical obstacle bypass direction, it can bypass the upper contour of the second obstacle, thereby bypassing the top of the obstacle and navigating to the predetermined target position target. This achieves the goal of walking along the contour line of the second obstacle after bypassing the first obstacle, and maintaining the obstacle bypass operation. Corresponding to the aforementioned embodiment, the robot configures the extension direction of the historical obstacle bypass direction as the reference obstacle bypass direction to represent the general direction of the upper left contour lines of the first obstacle and the second obstacle.

[0037] It should be added that, for the aforementioned narrow passages, narrow passages, or gaps, in the area where the robot is located, there are doorways on the horizontal ground that penetrate two adjacent room areas, or gaps between two parallel walls. The entrance of the doorway or the entrance of the gap can be set as an opening formed between obstacles, which is an opening formed between the outlines of at least two obstacles. These obstacles can exist in isolation. When the robot uses a laser sensor to scan the surrounding environment, the two endpoints and the width of the opening are scanned by the robot's laser sensor and converted into point cloud coordinates in the corresponding coordinate system. Then, the two endpoints of the opening are scanned into the corresponding point cloud and converted into the corresponding grid in the grid map. In some embodiments, when the robot senses its surroundings through collisions, whenever the robot's collision sensor contacts the two endpoints of the opening, it marks the outline point of the obstacle it collided with or the reference outline segment at the corresponding grid cell in the grid map. In some embodiments, an evaluation quantity is also introduced to represent the passability of the robot at the gap formed between obstacles or at its corresponding grid cell (the number of free grid cells between the two endpoints or their proportion within the opening width range). This quantity can represent the passability probability in the corresponding grid area. Generally, the confidence level assigned when the robot scans the gap using a ranging sensor is higher than the confidence level assigned when the robot detects the gap through collision, because the positioning accuracy of the ranging sensor is higher than the positioning accuracy generated by the robot's physical contact.

[0038] Preferably, the extension direction of the passable area inside the channel where the two obstacle bypass points are located is parallel to the reference contour line segment of the second obstacle (corresponding to...). Figure 3 The line segment AB) is parallel to the line segment of the second obstacle, and the reference contour line segment of the second obstacle is parallel to the search direction (corresponding to the line segment AB). Figure 3The direction from point D to point C is perpendicular. The passable area inside the passage where the two obstacle bypass points are located is connected to the predetermined target position. One of the obstacle bypass points is located on the reference contour line segment of the second obstacle. For example, points D and C are marked as the obstacle bypass point of the upper second obstacle and the obstacle bypass point of the lower first obstacle, respectively. Point D is set to be located on the reference contour line segment AB, and the area where the line segment CD is located is a passable area. The line segment CD can be marked as the narrow passage.

[0039] In the aforementioned embodiment, the robot collects point cloud data through a ranging sensor. The point cloud data is configured to reflect the position information of obstacles detected by the ranging sensor. It is a collection of discrete points and carries environmental noise (feedback of ambient light interference or the influence of the material on the surface of the obstacle). The robot then fits the collected point cloud data into the contour line of the obstacle to represent the local contour or envelope of the detected obstacle. The fitting process here specifically involves sorting, grouping (to distinguish different types of obstacles), filtering, piecewise interpolation fitting, and then connecting the coordinate points of each fitted curve in each group to obtain the obstacle envelope or contour line. Specifically, it consists of fitted line segments, fitted curves, and their combinations. The aforementioned reference contour line segment belongs to the contour line obtained through the fitting process. Preferably, to identify a narrow passage with passable significance, when the robot marks the extension direction of the passable area inside the passage where the two obstacle avoidance points are located as the preset passage direction, the two obstacle avoidance points are respectively located on the outlines of the first obstacle and the second obstacle in the preset passage direction, or the two obstacle avoidance points are respectively located on the first obstacle and the second obstacle. The distance between the two obstacle avoidance points is set to the preset distance range to facilitate identification of the robot entering the narrow passage, or the width of the passage where the two obstacle avoidance points are located is within the preset distance range to determine that the robot can pass through the narrow passage; the preset passage direction is... Figure 3 The index number ③ is located in the direction of the next position, or Figure 2 The position of index number ③ points in the direction of index number ④, forming the entrance to the narrow passage that the robot has explored during its obstacle-avoidance journey and the way to enter the narrow passage.

[0040] As one example, combined with Figure 2 and Figure 3 It can be seen that when the robot detects the first obstacle, the position it traverses along the obstacle avoidance path is located on either side of the predetermined target position, for example... Figure 2 The start and target positions are located on the left and right sides of the first obstacle, respectively; then the robot marks the side where it has traveled along the obstacle-avoiding path as the first side, corresponding to... Figure 2 and Figure 3The left side of the first obstacle, and the side where the predetermined target position is located is marked as the second side, corresponding to... Figure 2 and Figure 3 The robot configures the first endpoint A of the reference contour segment of the second obstacle to be located on the first side of the second obstacle or the first side of the first obstacle, and configures the second endpoint B of the reference contour segment of the first obstacle to be located on the second side of the second obstacle or the second side of the first obstacle; this distinguishes the inner and outer sides surrounded by the first and second obstacles, with the inner side being the right side and the outer side being the left side. The extension direction of the narrow passage corresponds to the right side, and the entrance of the narrow passage corresponds to the left side. Then, the robot marks the contour line formed by connecting the contour segments of the first side of the first obstacle and the contour segments of the first side of the second obstacle as the outer contour line formed by the obstacles where the two obstacle bypass points are located, corresponding to the reference obstacle bypass direction disclosed in the aforementioned embodiment, to represent the extension direction of the contour line planned under the heuristic search algorithm that is close to the upper left of the first obstacle and the upper left of the second obstacle.

[0041] Preferably, when the robot detects the first obstacle using a laser sensor, the robot has already collided with the first obstacle; when the robot walks to the gap formed between the first obstacle and the second obstacle, the robot has already collided with the second obstacle; the robot sets the preset distance range to be greater than or equal to the robot's body width, so that when the distance between the two obstacle avoidance points is greater than or equal to the robot's body width, the gap formed between the obstacles where the two obstacle avoidance points are located is determined to be a narrow passage allowing the robot to enter. The upper limit of the preset distance range is the sum of the robot's body width and a preset redundancy. Since the width of the narrow passage is only slightly greater than the robot's body width, a preset distance range is set for judgment to reduce the error interference of the calculated contour line during the narrow passage identification process; specifically, the minimum value (lower limit) of the preset distance range is greater than the robot's body width, and the maximum value (upper limit) of the preset distance range is only slightly greater than the robot's body width, not exceeding twice the body width. The preset distance range can be determined based on the robot's body width and a preset fitting error, and the preset redundancy setting allows passages that are close to the robot's body width to be called the narrow passage. For example, if the width of the robotic vacuum cleaner is 30 centimeters, the preset distance range can be 32 centimeters to 35 centimeters, meaning the center of the robot is located at the entrance (or gap) of the passage where the two obstacle avoidance points are located. Figure 3 When the midpoint of line segment CD is at the center of its width direction, the preset redundancy setting is manifested as the gap between the left and right sides of the robot and the channel being between 1 cm and 2 cm.

[0042] In summary, the aforementioned embodiments utilize the change in the obstacle-avoiding direction before entering the gap formed between the two obstacles, as well as the distance between the obstacle-avoiding points corresponding to the two obstacles, to identify whether the robot has begun to enter the narrow passage. This overcomes the error caused by the contour line fitting calculation of the obstacles in identifying the narrow passage, and improves the accuracy of narrow passage identification and the success rate of the robot entering the narrow passage.

[0043] Preferably, the narrow passage is a channel formed by the outlines (fitted results) of the two obstacles as its boundary lines; the minimum distance between the obstacle avoidance points extracted from the outlines of the two obstacles is greater than or equal to the minimum value of the preset distance range; the minimum value of the preset distance range is greater than the width of the robot's body; the width of the narrow passage is within the preset distance range, and the entrance of the narrow passage is also within the preset distance range. The entrance of the narrow passage can be a small doorway in a room, and the obstacles on both sides of the doorway are the four walls of the same room, which are continuous and integral. In addition, the walls on the left and right sides of the passage where the obstacle avoidance point is located can also be approximately parallel.

[0044] Preferably, when the robot walks inside the narrow passage, the shortest distance between the two boundary lines of the passage's entrance and the corresponding side of the robot is equal to half of the preset redundancy. For example, when the robot enters the narrow passage along its center line, the vertical distance between the robot's left side and its left obstacle avoidance point is equal to half of the preset redundancy, and the vertical distance between the robot's right side and its right obstacle avoidance point is also equal to half of the preset redundancy. In this embodiment, half of the preset redundancy is preferably 1 to 2 centimeters, thereby avoiding errors in contour line fitting calculations.

[0045] Based on the foregoing embodiments, the present invention also discloses a chip storing program code. When the program code is executed, it implements the steps of the robot entering a narrow passage identification method as described above. When the program code corresponding to the steps of the robot entering a narrow passage identification method is stored in a chip, it is treated as a computer program product. The computer program is operable to cause a computer to execute some or all of the steps of any of the methods described in the above embodiments of the robot entering a narrow passage identification method. During obstacle avoidance maneuvers, a robot equipped with the aforementioned chip uses laser data to obtain contour lines and extract relevant obstacle avoidance points, identifies narrow passages with a width slightly larger than the robot's body width, and determines that it can enter the narrow passage. This also facilitates the selection of a smoother navigation path.

[0046] This invention also discloses a robot equipped with a ranging sensor. The robot is equipped with the chip disclosed in the aforementioned embodiments to control the robot to use the ranging sensor to detect obstacles and obtain corresponding contour lines and obstacle avoidance points, facilitating the identification of the robot entering a narrow passage. When the robot is walking along a pre-planned obstacle avoidance path, the robot is prone to misjudging narrow passages formed between two obstacles that are just larger than the width of the robot body by a small gap distance (1 cm) as impassable. The invention obtains contour line segments associated with the passage factors of the gap formed between two obstacles to guide the robot in adjusting its obstacle avoidance direction before and after entering the gap. When the obstacle avoidance direction changes, the distance between two obstacle avoidance points extracted from the search direction of the contour line segments is used to identify the passage where the two obstacle avoidance points are located as a narrow passage, and it is determined that the robot has begun to enter a narrow passage but not through the pre-planned obstacle avoidance path. Therefore, it can accurately distinguish narrow passages that robots can pass through in the environment where robot vacuums, lawnmowers, or mobile toys are located. This overcomes the problem that the width of narrow passages is so small that the grid area inside the narrow passage or its opening is easily misjudged as an obstacle blocking the robot's passage. It also enables a smooth transition from the robot's action of bypassing obstacles to entering the narrow passage formed between two obstacles. In this way, the robot can accurately pass through the narrow passage or the narrow passage opening based on laser data, thereby improving the smoothness and success rate of passing through narrow areas under laser navigation conditions.

[0047] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0048] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0049] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0050] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0051] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0052] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0053] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0054] The above provides a detailed description of an embodiment of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for identifying a robot entering a narrow passage, wherein the robot equipped with a ranging sensor is the main implementer of the identification method; characterized in that, The identification method includes: Step S1: When the robot detects an obstacle using the ranging sensor, the robot plans an obstacle avoidance path to the predetermined target location, then walks along the obstacle avoidance path, and obtains the reference contour line segment through the point cloud data collected by the ranging sensor. Step S2: If the robot extracts two obstacle avoidance points in the search direction of the reference contour line segment, and if it detects that the robot's walking direction changes from the reference obstacle avoidance direction to the direction from the current position to the predetermined target position, and detects that the distance between the two obstacle avoidance points is within the preset distance range, then the robot recognizes its behavior as entering the narrow passage. Step S2 further includes: If the robot extracts two obstacle avoidance points in the search direction of the reference contour line segment, and if the distance between the two obstacle avoidance points is detected to be within a preset distance range, then the channel where the two obstacle avoidance points are located is identified as a narrow channel. The direction from the current position to the predetermined target position mentioned in step S2 is the passable area inside the channel where the two obstacle bypass points are located; The robot will be configured with reference obstacle avoidance direction as the outer contour line formed by the obstacles where the two obstacle avoidance points are located, so as to form the extension direction of the obstacle avoidance path before the robot enters the narrow passage; The robot sets the search direction of the reference contour line segment to be perpendicular to the reference contour line segment, forming the width direction of the channel where the two obstacle bypass points are located; the reference contour line segment corresponds to the obstacle where one of the obstacle bypass points is located, and forms one of the boundary lines of the channel where the two obstacle bypass points are located. The robot sets the preset distance range to be greater than or equal to the robot's body width, and the upper limit of the preset distance range is the sum of the robot's body width and the preset redundancy.

2. The identification method according to claim 1, characterized in that, As the robot moves along the obstacle avoidance path, if it adjusts its direction of movement to point from its current position to a predetermined target position, and detects that the distance between the two obstacle avoidance points is within the preset distance range, the robot determines that it is entering a narrow passage. This narrow passage is a gap formed between the obstacles where the two obstacle avoidance points are located. This gap is located within a passable area. The robot also configures the direction from its current position to the predetermined target position to be different from the reference obstacle avoidance direction.

3. The identification method according to claim 1, characterized in that, Between step S1 and step S2, the following is also included: After the robot detects an obstacle and before extracting two obstacle avoidance points in the search direction of the reference contour line segment of another obstacle, it continues to walk along the obstacle avoidance path and sets the current extension direction of the obstacle avoidance path as the historical obstacle avoidance direction. During the process of the robot walking along the obstacle avoidance path, after the robot obtains the reference contour line segment, if the robot detects that the extension direction of the obstacle avoidance path starting from the current position is not the direction from the current position to the predetermined target position, then the robot stops walking along the obstacle avoidance path and adjusts its walking direction to the direction from the current position to the predetermined target position so that the robot can start entering the channel where the two obstacle avoidance points are located.

4. The identification method according to claim 3, characterized in that, When the robot detects a first obstacle using a ranging sensor, it plans an obstacle avoidance path to a predetermined target location. The robot then walks along this path, following the outline of the first obstacle, until it reaches the gap between the first and second obstacles and / or collides with the second obstacle. At this point, the robot marks the direction of its movement as the historical obstacle avoidance direction. The reference outline of the second obstacle has a first endpoint and a second endpoint. The robot then acquires the reference outline of the second obstacle, marks the direction from the first endpoint to the second endpoint as the direction from the current location to the predetermined target location, and marks the direction from the second endpoint to the first endpoint as the extension direction of the historical obstacle avoidance direction. The robot configures the extension direction of the historical obstacle avoidance direction as the reference obstacle avoidance direction.

5. The identification method according to claim 4, characterized in that, The passageway inside the two obstacle bypass points extends in a direction parallel to the reference contour line segment of the second obstacle. The reference contour line segment of the second obstacle is perpendicular to the search direction. The passageway inside the two obstacle bypass points is connected to the predetermined target location. One of the obstacle bypass points is located on the reference contour line segment of the second obstacle.

6. The identification method according to claim 5, characterized in that, The robot collects point cloud data through a ranging sensor. The point cloud data is configured to reflect the position information of obstacles detected by the ranging sensor. The robot then fits and processes the collected point cloud data into the outline of the obstacle to represent the local outline of the detected obstacle or the envelope of the obstacle. The reference contour line segment is a contour line obtained through fitting; When the robot marks the extension direction of the passable area inside the channel where the two obstacle avoidance points are located as the preset channel direction, the two obstacle avoidance points are respectively located on the outline of the first obstacle in the preset channel direction and the outline of the second obstacle in the preset channel direction, or the two obstacle avoidance points are respectively located on the first obstacle and the second obstacle.

7. The identification method according to claim 6, characterized in that, When the robot detects the first obstacle, the position it has traversed along the obstacle avoidance path is located on either side of the predetermined target position. The robot then marks the side where it has traversed along the obstacle avoidance path as the first side and the side where the predetermined target position is located as the second side. The robot also configures the first endpoint of the reference contour line segment of the second obstacle to be located on either the first side of the second obstacle or the first side of the first obstacle, and configures the second endpoint of the reference contour line segment of the first obstacle to be located on either the second side of the second obstacle or the second side of the first obstacle. The robot then marks the contour line formed by connecting the contour line of the first side of the first obstacle and the contour line of the first side of the second obstacle as the direction of the outer contour line formed by the obstacles where the two obstacle bypass points are located.

8. The identification method according to claim 4, characterized in that, When the robot first detects the first obstacle using its ranging sensor, it has already collided with the first obstacle; when the robot walks to the gap formed between the first obstacle and the second obstacle, it has already collided with the second obstacle.

9. A chip storing program code, characterized in that, When the program code is executed, it implements the steps of the robot entry into a narrow passage identification method as described in any one of claims 1 to 8.

10. A robot equipped with a ranging sensor, characterized in that, The robot is equipped with the chip described in claim 9, which is used to control the robot to use a ranging sensor to detect obstacles and obtain the corresponding outline and obstacle avoidance points, so as to facilitate the identification that the robot has begun to enter the narrow passage.

Images