Control method and device of intelligent work robot, intelligent work equipment and medium

Abstract

The application relates to a control method and device of an intelligent work robot, intelligent work equipment and a medium, and relates to the control field of intelligent robots. The method comprises the following steps: in the process of controlling the intelligent work robot to run in an edge work mode, in the case that a virtual detection frame arranged in the advancing direction of the intelligent work robot covers the boundary grid in a global cost map, determining a boundary tangent of a wall existing in the advancing direction based on the covered boundary grid; the boundary grid corresponding to the wall is marked on the global cost map; controlling the intelligent work robot to run in the current advancing direction until the distance between the intelligent work robot and the wall reaches a preset near-stop distance range; adjusting the advancing direction of the intelligent work robot to be parallel to the boundary tangent, and controlling the intelligent work robot to advance along the wall based on a preset range of a heading angle; the heading angle is the included angle between the advancing direction and the boundary tangent. The method can improve the work quality of the robot.

Description

Technical Field

[0001] This application relates to the field of intelligent robot control, and in particular to a control method, apparatus, intelligent operating equipment, computer-readable storage medium, and computer program product for an intelligent operating robot. Background Technology

[0002] Intelligent cleaning robots are widely used in home cleaning and personal care due to their convenience and automation. In practical applications, when working along solid walls, intelligent cleaning robots can use sensors to detect the distance between themselves and the wall to smoothly conform to the wall edge. However, when the robot passes through virtual partition boundaries without physical obstructions, the sensors cannot collect effective distance measurement signals, and the working effect of the intelligent cleaning robot will be significantly reduced.

[0003] For example, with robotic vacuum cleaners, users can customize the cleaning environment by zone or room to meet their individual needs. However, when a user sets the current room as the area to be cleaned, the robotic vacuum cleaner, while cleaning along the walls, may accidentally step outside the virtual boundary of the area to be cleaned if it reaches a doorway or other virtual boundary without a physical wall. In this case, the robot may not only enter unnecessary areas but also interrupt the normal cleaning task, and may even create large blind spots along the virtual boundary, affecting the overall cleaning quality. Summary of the Invention

[0004] Therefore, it is necessary to provide a control method, device, intelligent operation equipment, computer-readable storage medium, and computer program product for intelligent operation robots that can improve the quality of wall-following operations, in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a control method for an intelligent work robot, the method comprising:

[0006] During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the global cost map is marked with the boundary grid corresponding to the wall between preset partitions.

[0007] Control the intelligent robot to move along the current direction of travel until the distance between it and the wall reaches the preset near-stop distance range;

[0008] The forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0009] In one embodiment, determining the boundary tangent of the wall present in the forward direction based on the covered boundary grid includes:

[0010] Among the multiple boundary grids covered by the virtual detection frame, the neighborhood boundary grid with the smallest distance to the intelligent operation robot in the first direction is taken as the first reference point; the first direction is the forward direction of the intelligent operation robot.

[0011] From the neighboring boundary grids adjacent to the first reference point, determine the neighboring boundary grid with the smallest distance to the intelligent robot in the second direction as the second reference point; the second direction is the direction perpendicular to the moving direction of the intelligent robot.

[0012] The boundary tangent of the wall is determined by the line connecting the second reference point and the first reference point.

[0013] In one embodiment, the method further includes:

[0014] In a local coordinate system constructed based on the current pose of the intelligent robot, a virtual detection frame is set at a first preset distance from the intelligent robot along its forward direction; wherein, the current pose of the intelligent robot includes its current position and forward direction; the origin of the local coordinate system is the current position of the intelligent robot, the horizontal axis of the local coordinate system is the forward direction, and the vertical axis of the local coordinate system is perpendicular to the forward direction.

[0015] In one embodiment, the virtual detection box is rectangular, and the length direction of the virtual detection box is the same as the forward direction; the method further includes:

[0016] The length parameter of the virtual detection frame and / or the first preset distance are determined based on the running speed of the intelligent operation robot.

[0017] In one embodiment, controlling the intelligent work robot to move along the wall within a preset range based on the heading angle includes:

[0018] During the process of controlling the intelligent operation robot to move forward, the real-time heading angle between the forward direction of the intelligent operation robot and the boundary tangent of the wall is determined;

[0019] The intelligent robot's forward direction is adjusted based on the real-time heading angle so that the real-time heading angle is within a preset range.

[0020] In one embodiment, during the process of controlling the intelligent work robot to move forward, determining the real-time heading angle between the forward direction of the intelligent work robot and the boundary tangent of the wall includes:

[0021] During the process of controlling the intelligent work robot to move forward, if there is a boundary grid within a first preset distance range in the second direction, the boundary grid with the smallest distance to the intelligent work robot in the second direction is used as the third reference point; the second direction is the direction perpendicular to the forward direction of the intelligent work robot.

[0022] From the neighboring boundary grids adjacent to the third reference point, the neighborhood boundary grid with the smallest distance to the intelligent robot in the first direction is determined as the fourth reference point; the first direction is the forward direction of the intelligent robot.

[0023] Update the boundary tangent of the wall based on the line connecting the fourth reference point and the third reference point;

[0024] The angle between the forward direction of the intelligent robot and the updated boundary tangent is taken as the real-time heading angle.

[0025] In one embodiment, controlling the intelligent work robot to move along its current direction of travel until the distance between it and the wall reaches a preset near-stop distance range includes:

[0026] Control the intelligent robot to decelerate along the current direction of travel and stop at a position where the distance between it and the wall reaches a preset near-stop distance range.

[0027] Secondly, this application also provides a control device for an intelligent work robot, the device comprising:

[0028] The detection module is used to determine the boundary tangent of the wall existing in the forward direction based on the covered boundary grid when the virtual detection frame set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map during the operation of the intelligent operation robot in the edge operation mode; wherein the global cost map is marked with the boundary grid corresponding to the wall between preset partitions.

[0029] The near-stop trigger module is used to control the intelligent operation robot to run along the current forward direction until the distance between it and the wall reaches the preset near-stop distance range;

[0030] The edge control module is used to adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent, and to control the intelligent operation robot to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0031] Thirdly, this application also provides an intelligent operation device, which includes an intelligent operation robot and a controller. The controller is used to determine the boundary tangent of a wall existing in the forward direction based on the covered boundary grid when the virtual detection frame set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map during the operation of the intelligent operation robot in an edge operation mode. The global cost map is marked with boundary grids corresponding to walls between preset partitions.

[0032] Control the intelligent robot to move along the current direction of travel until the distance between it and the wall reaches the preset near-stop distance range;

[0033] The forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0035] During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the global cost map is marked with the boundary grid corresponding to the wall between preset partitions.

[0036] Control the intelligent robot to move along the current direction of travel until the distance between it and the wall reaches the preset near-stop distance range;

[0037] The forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0038] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0039] During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the global cost map is marked with the boundary grid corresponding to the wall between preset partitions.

[0040] Control the intelligent robot to move along the current direction of travel until the distance between it and the wall reaches the preset near-stop distance range;

[0041] The forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0042] The aforementioned control method, device, intelligent operation equipment, computer-readable storage medium, and computer program product for intelligent operation robots, when controlling the intelligent operation robot to operate in an edge-following operation mode, determine the boundary tangent of the wall existing in the forward direction based on the covered boundary grid when a virtual detection frame set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map; wherein, the boundary grid corresponding to the wall between preset zones is marked on the global cost map; the intelligent operation robot is controlled to run along the current forward direction until the distance between it and the wall reaches a preset near-stop distance range; the forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent. Therefore, in the edge-following operation mode, when a wall is detected in the forward direction of the intelligent operation robot, the intelligent operation robot is controlled to run until the distance between it and the wall reaches the preset near-stop distance range, then its forward direction is adjusted to be parallel to the boundary tangent of the wall, and then it is controlled to move along the wall boundary, with the heading angle within the preset range. In this way, the intelligent robot does not make large turns while moving along the wall, and can work stably along the edge without deviating from the boundary and missing work or running out of the preset area, which can effectively improve the quality of work. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a structural diagram of an intelligent operation device according to one embodiment;

[0045] Figure 2 This is a flowchart illustrating a control method for an intelligent work robot in one embodiment;

[0046] Figure 3 This is a flowchart illustrating the process of determining the boundary tangent of a wall existing in the forward direction based on a covered boundary grid in one embodiment.

[0047] Figure 4 This is a schematic diagram of the boundary tangent of the wall in the forward direction in one embodiment;

[0048] Figure 5 This is a flowchart illustrating the control method for an intelligent work robot in another embodiment;

[0049] Figure 6 This is a schematic diagram of the process for controlling an intelligent operation robot to run along a wall boundary within a preset range based on the heading angle in one embodiment;

[0050] Figure 7 This is a flowchart illustrating the process of determining the real-time heading angle between the intelligent operation robot's forward direction and the boundary tangent of the wall during the control of the intelligent operation robot's forward movement in one embodiment.

[0051] Figure 8 This is a schematic diagram of the boundary tangent of the wall in the direction perpendicular to the forward direction in one embodiment;

[0052] Figure 9 This is a flowchart illustrating the control method for an intelligent work robot in yet another embodiment;

[0053] Figure 10 This is a structural block diagram of the control device for an intelligent work robot in one embodiment;

[0054] Figure 11 This is an internal structural diagram of an intelligent operating device in one embodiment. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] The control method for intelligent work robots provided in this application can be applied to various intelligent work equipment. Intelligent work equipment includes, but is not limited to, intelligent cleaning equipment, intelligent lawnmowers, and intelligent window cleaning equipment. Correspondingly, the intelligent work robot is adapted to the corresponding type of equipment; for example, when the intelligent work equipment is an intelligent cleaning equipment, the intelligent work robot is a sweeping robot.

[0057] In some embodiments, such as Figure 1 As shown, the intelligent operation equipment 100 includes a controller 102 and a connected intelligent operation robot 104, the intelligent operation robot 104 being adapted to a corresponding type of equipment. The controller 102, while controlling the intelligent operation robot to operate in an edge-operation mode, determines the boundary tangent of a wall existing in the forward direction based on the covered boundary grid in the global cost map, provided that a virtual detection frame set along the robot's forward direction covers the boundary grid in the forward map; wherein the global cost map marks the boundary grid corresponding to the walls between preset zones; controls the intelligent operation robot to move along the current forward direction until the distance to the wall reaches a preset near-stop distance range; adjusts the robot's forward direction to be parallel to the boundary tangent, and controls the robot to move along the wall within a preset range based on a heading angle; the heading angle is the angle between the robot's forward direction and the boundary tangent.

[0058] It is understandable that the control method for this intelligent work robot can also be applied to servers, and to systems that include both intelligent work equipment and servers, and implemented through the interaction between the intelligent work equipment and the server. Specifically, the control method for this intelligent work robot can be implemented through the controller built into the intelligent work equipment, or through other external control devices.

[0059] In one exemplary embodiment, such as Figure 2 As shown, a control method for an intelligent work robot is provided, which is applied to... Figure 1 Taking controller 102 as an example, the explanation includes the following steps 202 to 206. Wherein:

[0060] Step 202: During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid.

[0061] The global cost map is marked with boundary grids corresponding to the walls between preset zones. These boundary grids are the grids in the global cost map that map the walls. Walls can be physical walls or virtual walls. Physical walls refer to real, physical obstacles that can be detected and identified by sensors, such as building walls or furniture. Virtual walls refer to the boundaries of zones formed by user partitioning, room division, or delineation of work areas; they have no actual physical obstructions and cannot be detected and identified by sensors.

[0062] The wall is mapped onto the global cost map as a strip-shaped area with width. For example, the strip-shaped area occupies 2 to 3 grids in the thickness direction to simulate the thickness characteristics of a real wall and ensure the continuity of the boundary between the physical wall and the virtual wall.

[0063] In the global cost map, different cost values ​​can be set for grids mapping different regions to distinguish region types. For example, the cost value of a grid in an open area to be worked on is set to 0, the cost value of a grid for a regular obstacle is set to 100, the cost value of a boundary grid is set to 110, and the cost value of an unknown area is set to -1. When a grid with a cost value of 110 is detected within the virtual detection box, it can be determined that the virtual detection box covers the boundary grid in the global cost map, thus confirming the presence of a wall in the robot's direction of movement.

[0064] In some embodiments, the controller stores a global environmental map, which is a map built based on the entire working environment. After receiving a partitioning setting command from the user, the controller calls a preset partitioning algorithm to extract the wall boundaries separating multiple preset partitions based on the global environmental map, and maps the walls onto the global cost map to generate corresponding boundary grids. These multiple preset partitions can be multiple target working areas, or they can include target working areas and non-working areas. The working parameters of the multiple target working areas can be the same or different; this embodiment does not limit this. The partitioning algorithm does not need to be limited; those skilled in the art can set it according to actual needs. It is understood that the number of grids occupied by the walls in the thickness direction may fluctuate due to the influence of the preset partitioning algorithm and grid resolution.

[0065] The boundary of the wall refers to the edge contour of the wall near the intelligent robot. In some embodiments, the controller determines the tangent line of the wall boundary near the intelligent robot based on the boundary grid covered by the virtual detection frame, which serves as the boundary tangent line. The tangent line is a straight line that fits the edge contour and reflects the extension direction and slope of the wall.

[0066] Edge-following operation mode refers to the working mode in which the intelligent robot keeps close to the boundary of the wall and performs tasks simultaneously (such as cleaning). When the intelligent robot is running in edge-following operation mode, it continuously identifies the boundaries of the surrounding physical or virtual walls and performs tasks at the same time (such as cleaning the gaps in the boundary).

[0067] In some embodiments, when the controller detects the existence of a boundary grid in the global cost map and the robot reaches a preset range around the boundary, it controls the intelligent operation robot to operate in an edge-operation mode. In some embodiments, the controller controls the intelligent operation robot to operate in an edge-operation mode according to the edge-operation mode control command received from the user.

[0068] Step 204: Control the intelligent operation robot to run along the current forward direction until the distance between it and the wall reaches the preset near-stop distance range.

[0069] The preset near-stop distance range can be set according to specific circumstances, such as the type of intelligent robot, control precision, and type of operation.

[0070] In this embodiment, when a wall is detected in the current direction of travel, the controller controls the intelligent robot to continue moving forward until it stops when it is within a preset stopping distance from the wall boundary. This allows for a safe stopping at a location close to the wall, effectively preventing the intelligent robot from crossing the virtual wall and entering non-target work areas (such as non-work areas or unknown areas).

[0071] Step 206: Adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent, and control the intelligent operation robot to move along the wall within a preset range based on the heading angle.

[0072] Here, the heading angle is the angle between the forward direction of the intelligent robot and the tangent of the boundary. In this embodiment, the intelligent robot is controlled to move along the wall, and the heading angle is kept within a preset range. Thus, the angle between the forward direction of the intelligent robot and the tangent of the boundary is maintained within the preset range. Once the heading angle exceeds the preset range, the controller will adjust the forward direction of the intelligent robot to reduce the angular deviation between the forward direction and the tangent of the boundary. The preset range can be determined based on factors such as control accuracy and forward speed; this embodiment does not limit this. As an example, the allowable deviation range of the heading angle is set to -0.2 rad (radians) to 0.2 rad.

[0073] After the intelligent robot adjusts its forward direction using the tangential direction as a reference, it can reduce the risk of deviation. During the robot's movement along the wall boundary, the controller monitors its heading angle in real time, keeping it stably within a preset range to prevent significant fluctuations and frequent deviations in its posture. This control method ensures the robot remains close to the physical or virtual wall boundary, maintaining a continuous and smooth trajectory, effectively improving the efficiency of edge-based operations.

[0074] The aforementioned control method for the intelligent work robot, during its operation in an edge-following mode, involves determining the boundary tangent of a wall existing in the forward direction based on the covered boundary grid when a virtual detection frame along the robot's forward direction covers the boundary grid in the global cost map. The global cost map marks the boundary grids corresponding to walls between preset zones. The robot is then controlled to move along its current forward direction until the distance to the wall reaches a preset near-stop distance. The robot's forward direction is then adjusted to be parallel to the boundary tangent, and the robot is controlled to move along the wall within a preset range based on its heading angle. The heading angle is the angle between the robot's forward direction and the boundary tangent. Therefore, in edge-following mode, when a wall is detected in the robot's forward direction, the robot is controlled to move until the distance to the wall reaches a preset near-stop distance, its forward direction is adjusted to be parallel to the wall's boundary tangent, and it is then controlled to move along the wall boundary while maintaining its heading angle within a preset range. In this way, the intelligent robot does not make large turns while moving along the wall, and can work stably along the edge without deviating from the boundary and missing work or running out of the preset area, which can effectively improve the quality of work.

[0075] It should be noted that in related technologies, when intelligent robots operate along edges, they often use coordinate boundary judgment logic to determine whether the robot has crossed the boundary. That is, if the robot's coordinates are detected to be outside the virtual wall area, it will turn around and turn back. The driving trajectory generated by this method has obvious jerks and is also prone to missing work at the boundary. Taking a robot vacuum cleaner as an example, when the robot vacuum cleaner is cleaning along the right side of the physical wall to the door position, the physical wall is interrupted and switches to a virtual partition boundary. At this time, physical sensors such as LiDAR and infrared have no detection feedback. Because there is no physical ranging feedback, the controller will control the robot vacuum cleaner to turn right directly. After crossing the virtual boundary, it will detect that the robot's actual coordinates are outside the virtual wall coordinates and trigger the turn-back logic. After turning back, it still cannot receive physical ranging feedback, and will repeat the turning, boundary crossing, and turn-back actions. The driving trajectory is jerky, and the boundary area is prone to forming cleaning blind spots.

[0076] This embodiment effectively prevents the intelligent robot from crossing virtual walls and entering non-target work zones, ensuring continuous and complete zone cleaning tasks and improving the problem of missed edge cleaning. At the same time, high-precision, smooth edge cleaning can be achieved through software, without relying on physical ranging sensors such as lidar and infrared to collect boundary distances, thus reducing dependence on physical sensors.

[0077] In one exemplary embodiment, such as Figure 3 As shown, the steps, based on the covered boundary grid, determine the boundary tangent of the wall existing in the forward direction, including steps 302 to 306. Wherein:

[0078] Step 302: Among the multiple boundary grids covered by the virtual detection box, the neighborhood boundary grid with the smallest distance to the intelligent operation robot in the first direction is taken as the first reference point.

[0079] Step 304: From the neighboring boundary grids adjacent to the first reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the second direction as the second reference point.

[0080] Step 306: Determine the boundary tangent of the wall based on the line connecting the second reference point and the first reference point.

[0081] Wherein, the first direction is the forward direction of the intelligent work robot, and the second direction is the direction perpendicular to the forward direction of the intelligent work robot. For example, the second direction is perpendicular to the forward direction of the intelligent work robot and points towards the side of the robot that is working along the edge. For instance, when the intelligent work robot is working along the right side of the wall in its forward direction, the second direction is perpendicular to the forward direction and points to the right.

[0082] In some embodiments, the neighborhood boundary grid adjacent to the first reference point is an eight-neighbor grid of the first reference point, taken from a 3×3 eight-neighbor search window centered on the first reference point.

[0083] When controlling the movement of the intelligent operation robot, the controller will determine whether the wall with a fixed grid thickness covers the virtual detection frame. If the virtual detection frame is detected to cover the boundary grid, the controller will execute the near stop control logic in step 204 on the one hand, and determine the boundary tangent of the wall on the other hand.

[0084] When determining the boundary tangent of the wall, the controller filters all boundary grids covered by the virtual detection frame, selecting the boundary grid closest to the intelligent robot along the first direction, and records it as the first reference point. Then, using the first reference point as the center, the valid boundary grids within its eight neighboring grids are extracted as candidate grids. Among the candidate grids, the grid closest to the intelligent robot along the second direction is selected as the second reference point. Connecting the first reference point and the second reference point forms a vector line segment, which is used as the boundary tangent of the wall.

[0085] The following is combined Figure 4 The following example illustrates how the boundary tangent of the wall is determined. The grid (black dots) where the wall (shown as dashed lines in the diagram) intersects with the virtual detection frame in the direction the intelligent robot is moving is the closest boundary grid to the intelligent robot along the moving direction, and is set as the first reference point. A 3×3 eight-neighborhood search window is constructed using this first reference point. The grids represented by green dots, red dots, and the blue dots between the two dashed lines within the window are all valid candidate wall grids with a value equal to a preset value (e.g., 110). The controller selects the grid closest to the robot's center (shown as red dots in the diagram) from the candidate grids that meet the criteria as the second reference point. The line vector connecting the first and second reference points is the boundary tangent corresponding to the actual extension direction of the boundary of the wall closer to the intelligent robot.

[0086] In this embodiment, the boundary grid closest to the intelligent robot's forward direction is first selected as the first reference point. Then, a second reference point perpendicular to the forward direction is selected through a 3×3 eight-neighbor window. The tangent line of the wall boundary is accurately determined by the vector connecting the two reference points. Based on this boundary tangent line, the robot's posture can be accurately calibrated, improving issues such as trajectory stuttering, missed boundary scans, and zone crossings. Furthermore, this method only performs local calculations on the valid boundary grids within the virtual detection frame, resulting in low computational load and fast response speed, thus meeting the control requirements of the intelligent robot's high-speed dynamic movement.

[0087] In some embodiments, such as Figure 5 As shown, the control method of the intelligent operation robot also includes the following step 502.

[0088] Step 502: In a local coordinate system constructed based on the current pose of the intelligent robot, a virtual detection frame is set at a first preset distance from the intelligent robot along its forward direction.

[0089] The current pose of the intelligent robot includes its current position and direction of travel. The origin of the local coordinate system is the current position of the intelligent robot, such as the center of its body. The horizontal axis of the local coordinate system (…) Figure 4The direction shown in the diagram (x-axis) is the forward direction, and the fuselage longitudinal axis of the local coordinate system (…). Figure 4 (The diagram shows the Y-axis) The direction is perpendicular to the direction of movement.

[0090] First preset distance ( Figure 4 The distance d) can be set according to the robot's travel speed, control response speed, etc., and is not limited in this embodiment. For example, the first preset distance d is 15 centimeters.

[0091] The virtual detection box can be set up in a flexible way. As an example, the center point of the border of the virtual detection box near the origin of the local coordinate system is spaced from the origin of the local coordinate system by a first preset distance d. For example, the virtual detection box is a rectangle, and the border of the rectangle extends along the horizontal and vertical axes of the fuselage, respectively. The center point of the border near the origin of the local coordinate system is spaced from the origin of the local coordinate system by a first preset distance d.

[0092] During the operation of the intelligent robot, its working position is continuously calibrated in the global coordinate system corresponding to the global map of the environment. Simultaneously, the controller constructs a local coordinate system for the robot based on its current pose, and builds a virtual detection box within this local coordinate system along the robot's forward direction. The global cost map stores the absolute coordinates of all grids in the global coordinate system. The controller can perform coordinate conversion between the global and local coordinate systems, converting the coordinates of the virtual detection box in the local coordinate system to global map coordinates. By iterating through the cost values ​​of all grids within the converted detection box, the controller determines whether the virtual detection box covers the boundary grids in the global cost map.

[0093] When selecting the first reference point and the second reference point, the boundary grid with the smallest absolute value of the horizontal axis coordinate of the fuselage among the multiple boundary grids covered by the virtual detection box can be used as the first reference point. Then, the boundary grid with the smallest absolute value of the vertical axis coordinate of the fuselage among the neighboring boundary grids adjacent to the first reference point can be used as the second reference point. Finally, the tangent of the wall boundary can be determined in the local coordinate system based on the second reference point and the first reference point.

[0094] In this embodiment, as the intelligent robot continues to operate, the controller synchronously follows the robot and dynamically refreshes the coordinate position of the virtual detection frame in real time, realizing uninterrupted detection of the boundary grid in front of the robot, which can avoid detection blind spots and thus improve the cleaning coverage of edge operations.

[0095] The virtual detection frame is rectangular, with its length parallel to and parallel to the forward direction, and its width perpendicular to the forward direction. The length parameter of the virtual detection frame can be set to a preset fixed value or adjusted flexibly according to actual working conditions.

[0096] In some embodiments, the control method for the intelligent work robot further includes the following steps:

[0097] The length parameter of the virtual detection frame is determined based on the running speed of the intelligent operation robot; and / or, a first preset distance is determined based on the running speed of the intelligent operation robot.

[0098] When the linear velocity of the intelligent robot is high, the controller linearly increases the first preset distance and / or increases the length of the rectangular detection frame proportionally. This extends the prediction time window for high-frequency traversal of the boundary grid, providing sufficient safety buffer distance for deceleration and near-stop operations at high speeds. When the linear velocity of the intelligent robot decreases, the controller linearly decreases the first preset distance and / or decreases the length of the rectangular detection frame proportionally. This prevents distant, irrelevant boundary grids from being included in the detection range, thus avoiding misjudgments and improving the robot's motion sensitivity when turning or adjusting its posture in narrow areas.

[0099] In some embodiments, such as Figure 6 As shown, the steps control the intelligent operation robot to run along the wall boundary within a preset range based on the heading angle, including the following steps 602 and 604.

[0100] Step 602: During the process of controlling the intelligent operation robot to move forward, determine the real-time heading angle between the forward direction of the intelligent operation robot and the boundary tangent of the wall.

[0101] Step 604: Adjust the forward direction of the intelligent operation robot based on the real-time heading angle so that the real-time heading angle is within the preset range of the heading angle.

[0102] Specifically, when the controller controls the intelligent operation robot to move along the wall boundary, it will continuously determine the real-time heading angle between the robot's forward direction and the tangent of the wall boundary, and calculate the corresponding heading angle deviation.

[0103] To ensure the intelligent robot's trajectory remains parallel to the wall boundary, the controller sets the target expectation value of its PID (Proportional-Integral-Derivative) controller to 0, inputs the real-time heading angle deviation into the PID controller, and dynamically adjusts the output control quantity based on the derivative term corresponding to the error change rate, while performing safety limiting on the output angular velocity. For example, the angular velocity compensation value is limited to the range of -0.2 rad / s (radians per second) to 0.2 rad / s, ultimately outputting an adapted angular velocity compensation command. The intelligent robot receives the angular velocity compensation command and continuously reduces and eliminates the heading angle deviation by frequently making small adjustments to its body posture, ensuring the body remains parallel to the tangent of the wall boundary. This closed-loop adjustment method enables edge-to-edge travel in scenarios seamlessly switching between physical and virtual wall boundaries. Even when traveling to virtual boundary positions without physical obstructions, such as doorways, there will be no significant turning, thus improving issues such as trajectory jerking and unstable work quality.

[0104] In some embodiments, such as Figure 7 As shown, step 602 includes the following steps 702-708.

[0105] Step 702: During the process of controlling the intelligent operation robot to move forward, if there is a boundary grid within the first preset distance range in the second direction, the boundary grid with the smallest distance to the intelligent operation robot in the second direction is used as the third reference point.

[0106] Step 704: From the neighboring boundary grids adjacent to the third reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the first direction as the fourth reference point.

[0107] Step 706: Update the boundary tangent of the wall based on the line connecting the fourth reference point and the third reference point.

[0108] Step 708: The angle between the forward direction of the intelligent robot and the updated boundary tangent is taken as the real-time heading angle.

[0109] The first preset distance range can be flexibly set according to the robot's body size, grid resolution, and response speed. For example, this first preset distance range is no more than 25 centimeters from the origin of the local coordinate system. The first preset distance range in the second direction corresponds to the interval along the Y-axis in the local coordinate system where the distance from the robot's center point (the origin of the local coordinate system) is within the first preset distance range. That is, during the process of controlling the intelligent robot to move forward, the controller continuously determines whether there are boundary grids within the first preset distance range from the center point of the intelligent robot along the second direction. For ease of understanding, the following will illustrate this further. Figure 8This document provides an example of how to determine the boundary tangent of a wall during the movement of an intelligent work robot. Assuming the intelligent work robot is working along the right-side wall, the controller effectively generates a lateral detection ray with a range of 25cm, covering the second direction area on the right side of the robot. When this detection ray identifies the boundary grid corresponding to the wall, the intersection grid of the detection ray becomes the third reference point (corresponding to...). Figure 8 (Black dot in the middle). The controller constructs a 3×3 eight-neighborhood search window using the third reference point, traverses all valid boundary candidate grids within the window whose cost value is equal to the preset value, compares the absolute value of the X-axis coordinate of each candidate point in the local coordinate system, and selects the point with the smallest displacement in the forward direction as the fourth reference point. Figure 8 (The red dot in the center).

[0110] Furthermore, the controller updates the current boundary tangent of the wall in real time based on the coordinate connection of the third and fourth reference points. Specifically, the local coordinate system coordinates of the third and fourth reference points are input into a bivariate arctangent function, and the slope of the connecting line is calculated by the difference between the coordinates of the two points to accurately calculate the actual extension angle of the wall boundary relative to the fuselage. This extension angle is used as the reference direction of the boundary tangent for calculating the real-time heading angle.

[0111] In this embodiment, during the operation of the intelligent robot along the edge, the boundary tangent of the wall boundary relative to the robot body can be calculated and updated in real time. This allows for the calculation of the real-time heading angle between the robot's forward direction and the wall boundary tangent, enabling timely adjustment of the forward direction and ensuring that the robot always travels parallel to the wall boundary, thereby improving the quality of the operation.

[0112] In some embodiments, step 204 includes the following steps:

[0113] Control the intelligent operation robot to decelerate along the current direction of travel and stop at a position where the distance between it and the wall boundary reaches the preset near-stop distance range.

[0114] In this embodiment, controlling the intelligent robot to decelerate along the current forward direction and stop within a preset near-stop distance range can constrain the robot's travel distance in advance, reserve space for adjusting direction, and avoid insufficient space for posture adjustment due to the robot body being too close to the wall boundary.

[0115] In some embodiments, stepped deceleration control can also be implemented based on the real-time distance between the intelligent robot and the wall boundary. For example, a three-level control strategy can be set: long-distance pre-deceleration, close-distance strong braking, and critical-point safe near-stop. When the distance between the intelligent robot and the wall boundary is within the long-distance threshold range, long-distance pre-deceleration is implemented, specifically by gradually reducing the robot's straight-line travel speed to weaken its inertia. When the distance between the intelligent robot and the wall boundary shrinks to the close-distance threshold range, close-distance strong braking is implemented, significantly reducing the travel speed and shortening the braking stroke. When the intelligent robot reaches the preset near-stop distance critical point, safe near-stop is executed, completely halting forward movement.

[0116] By employing stepped deceleration control, the stability and robustness of the system's boundary-crossing control under different travel speeds can be significantly improved. For example, it effectively counteracts the motion inertia generated when the intelligent robot is moving at high speed, preventing the robot from crossing the boundary of a virtual wall without physical obstructions; at the same time, the deceleration process is smooth at low speeds, without sudden braking jolting, ensuring the robot's stable operation.

[0117] To better understand the above embodiments, the following detailed explanation is provided with reference to an optional embodiment. In one embodiment, taking a robotic vacuum cleaner as an example, such as... Figure 9 As shown, the control method includes the following steps:

[0118] During the operation of the sweeping robot controlled by the controller, when the controller detects that the sweeping robot has reached the perimeter of the preset zone boundary, it controls the intelligent operation robot to operate in the edge operation mode.

[0119] The controller obtains the global cost map corresponding to the operating environment, identifies and extracts the virtual contour boundaries of each preset area through a preset partitioning algorithm, maps the extracted virtual contour boundaries to the underlying grid map, and assigns a unique special value of 110 to the grids that constitute the virtual boundary.

[0120] During the robot's boundary cleaning operation, the controller uses the robot's real-time pose (global coordinates X, Y, and orientation angle) as a reference and configures a 15cm forward preset offset distance (i.e., the first preset distance). It dynamically generates a rectangular virtual detection frame matching the robot's width in front of it; for example, the virtual detection frame is set to 15cm in length and 20cm in width. The controller frequently traverses the value of all grids within the coverage area of ​​the virtual detection frame. Once a boundary grid with a value equal to 110 is encountered, it determines that a wall exists in the robot's forward direction. The controller immediately terminates straight-line control and initiates graded deceleration control, gradually reducing the straight-line speed based on the real-time distance between the robot and the virtual boundary until the distance between the robot and the virtual boundary reaches a preset stopping distance of 3cm, at which point the robot completely stops, effectively preventing it from running out of the current cleaning zone.

[0121] After completing the deceleration and near-stop, the controller transforms the global coordinates of all boundary grids with a cost value of 110 within the virtual detection box to a local coordinate system with the robot's body center as the origin. It then compares the absolute values ​​of the X-axis (forward direction) coordinates of each boundary grid in the local coordinate system and selects the grid with the smallest value as the first reference point for solving the wall tangent.

[0122] A 3×3 eight-neighbor search window is constructed by converting the planar coordinates of the first reference point into the row and column index of the raster map. A 3×3 eight-neighbor raster search window is constructed with this index as the center, covering a total of 8 adjacent rasters above, below, left, right and diagonally of the center raster.

[0123] Traverse all grids within the eight-neighborhood search window, filtering candidate boundary grids with a cost value equal to 110. Ensure that the first and second reference points belong to the same continuous wall boundary to avoid angle calculation distortion caused by cross-boundary sampling. Transform the coordinates of all candidate grids that meet the cost value condition to the robot's local coordinate system. If multiple candidate grids have the same absolute value of their grid axis coordinates, compare the straight-line physical distances between each candidate grid and the robot's center, prioritizing the grid closer to the center as the second reference point.

[0124] The robot calculates the tangent of the wall boundary and adjusts its parallel posture. Based on the coordinate difference between the first and second reference points in the local coordinate system, the bivariate arctangent function (Atan2) is used to solve for the angle between the vectors connecting the two points, accurately obtaining the direction of the tangent of the wall boundary. The controller issues rotation control commands based on the calculated boundary direction angle, driving the robot chassis to adaptively steer, making the robot's forward direction parallel to the tangent of the wall boundary, and smoothly switching to the edge-hugging wall operation posture.

[0125] The controller generates a virtual detection ray with a maximum detection range of 25cm, starting from the center of the fuselage and extending to the right side of the fuselage. This detection method enables seamless switching between physical walls and virtual boundaries. When the ray hits a boundary grid with a cost value of 110, the grid coordinates are recorded as the third reference point for real-time sampling. A 3×3 grid search window is constructed centered on the third reference point. Candidate grids with a cost value of 110 within the window are filtered and transformed to the local coordinate system. The absolute values ​​of the Y-axis coordinates of the candidate grids are compared, and the grid corresponding to the minimum value is selected as the fourth reference point, ensuring that the two sampling points are closely attached to the same side wall boundary.

[0126] The vector angle between the two points is calculated using the bivariate arctangent function (Atan2) to obtain the real-time heading angle of the current wall boundary; the real-time heading angle deviation is solved by comparing the robot's real-time forward direction with the wall heading angle.

[0127] The real-time heading angle deviation is input into the PID controller, and the differential term of the error change rate is used to dynamically output the adjustment amount. The output angular velocity is also subject to a safety limit, with the angular velocity compensation value constrained within the range of -0.2 rad / s to 0.2 rad / s. By using PID control to make high-frequency, small-amplitude corrections to the robot's orientation, the heading angle deviation is continuously reduced, allowing the robot to travel smoothly along the wall throughout the entire process. There is no trajectory jitter, no boundary crossing, and no blind spots during the switching between physical walls and virtual invisible boundaries.

[0128] The aforementioned control method for intelligent work robots can uniformly and accurately identify virtual walls at the software level, and enable intelligent work robots to smoothly fit virtual walls and perform high-quality edge operations without the assistance of physical sensors.

[0129] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0130] Based on the same inventive concept, this application also provides a control device for an intelligent work robot to implement the control method of the intelligent work robot described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the control device for the intelligent work robot provided below can be found in the limitations of the control method for the intelligent work robot described above, and will not be repeated here.

[0131] In one exemplary embodiment, such as Figure 10 As shown, a control device for an intelligent work robot is provided, including: a detection module 802, a near-stop triggering module 804, and an edge control module 806, wherein:

[0132] The detection module 802 is used to determine the boundary tangent of the wall existing in the forward direction based on the covered boundary grid when the virtual detection frame set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map during the operation of the intelligent operation robot in the edge operation mode; wherein the global cost map is marked with the boundary grid corresponding to the wall between the preset partitions.

[0133] The near-stop trigger module 804 is used to control the intelligent operation robot to run along the current forward direction until the distance between it and the wall reaches the preset near-stop distance range.

[0134] The edge control module 806 is used to adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent and control the intelligent operation robot to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0135] In some embodiments, the detection module 802 is further configured to: use the neighborhood boundary grid with the smallest distance to the intelligent operation robot in a first direction among the multiple boundary grids covered by the virtual detection frame as a first reference point; the first direction is the forward direction of the intelligent operation robot; determine the neighborhood boundary grid with the smallest distance to the intelligent operation robot in a second direction from the neighboring neighborhood boundary grids adjacent to the first reference point as a second reference point; the second direction is a direction perpendicular to the forward direction of the intelligent operation robot; and determine the boundary tangent of the wall based on the line connecting the second reference point and the first reference point.

[0136] In some embodiments, the detection module 802 is further configured to set a virtual detection box at a first preset distance from the intelligent operation robot along the forward direction of the intelligent operation robot in a local coordinate system constructed based on the current pose of the intelligent operation robot; wherein, the current pose of the intelligent operation robot includes the current position and the forward direction; the origin of the local coordinate system is the current position of the intelligent operation robot, the horizontal axis of the local coordinate system is the forward direction, and the vertical axis of the local coordinate system is the direction perpendicular to the forward direction.

[0137] In some embodiments, the detection module 802 is further configured to determine the length parameter of the virtual detection frame and / or a first preset distance based on the running speed of the intelligent operation robot.

[0138] In some embodiments, the edge control module 806 is further configured to determine the real-time heading angle between the forward direction of the intelligent operation robot and the boundary tangent of the wall during the forward movement of the intelligent operation robot; and adjust the forward direction of the intelligent operation robot based on the real-time heading angle so that the real-time heading angle is within a preset range of the heading angle.

[0139] In some embodiments, the edge control module 806 is further configured to, during the movement of the intelligent work robot, if there are boundary grids within a first preset distance range in the second direction, use the boundary grid with the smallest distance to the intelligent work robot in the second direction as a third reference point; the second direction is a direction perpendicular to the moving direction of the intelligent work robot; from the neighboring boundary grids adjacent to the third reference point, determine the neighboring boundary grid with the smallest distance to the intelligent work robot in the first direction as a fourth reference point; the first direction is the moving direction of the intelligent work robot; update the boundary tangent of the wall according to the line connecting the fourth reference point and the third reference point; and use the angle between the moving direction of the intelligent work robot and the updated boundary tangent as the real-time heading angle.

[0140] In some embodiments, the near stop triggering module 804 is further configured to control the intelligent operation robot to decelerate along the current forward direction and stop at a position where the distance between it and the wall boundary reaches a preset near stop distance range.

[0141] The various modules in the control device of the aforementioned intelligent work robot can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the intelligent work device in hardware form or independent of it, or stored in the memory of the intelligent work device in software form, so that the processor can call and execute the operations corresponding to each module.

[0142] In one exemplary embodiment, an intelligent operation device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 11As shown, this intelligent work device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores XX data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a control method for an intelligent work robot.

[0143] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the intelligent operating equipment to which the present application is applied. Specific intelligent operating equipment may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0144] In one exemplary embodiment, an intelligent work device is provided, including an intelligent work robot and a controller, the controller being used to control the operation of the intelligent work robot according to the steps in any of the above method embodiments.

[0145] In some embodiments, the controller includes a memory and a processor, the memory storing a computer program, which the processor executes to implement the steps in the above method embodiments.

[0146] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0147] During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the boundary grid corresponding to the wall between the preset partitions is marked on the global cost map.

[0148] Control the intelligent operation robot to move along the current direction of travel until the distance to the wall reaches the preset near-stop distance range;

[0149] Adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent, and control the intelligent operation robot to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0150] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: taking the neighborhood boundary grid with the smallest distance to the intelligent operation robot in a first direction among the multiple boundary grids covered by the virtual detection box as a first reference point; the first direction is the forward direction of the intelligent operation robot;

[0151] From the neighboring boundary grids adjacent to the first reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the second direction as the second reference point; the second direction is the direction perpendicular to the moving direction of the intelligent robot.

[0152] The boundary tangent of the wall is determined by the line connecting the second reference point and the first reference point.

[0153] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: in a local coordinate system constructed based on the current pose of the intelligent robot, a virtual detection box is set at a first preset distance from the intelligent robot along the forward direction of the intelligent robot; wherein, the current pose of the intelligent robot includes the current position and the forward direction; the origin of the local coordinate system is the current position of the intelligent robot, the horizontal axis of the local coordinate system is the forward direction, and the vertical axis of the local coordinate system is the direction perpendicular to the forward direction.

[0154] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the length parameter of the virtual detection frame and / or a first preset distance based on the running speed of the intelligent robot.

[0155] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: during the process of controlling the intelligent work robot to move forward, determining the real-time heading angle between the forward direction of the intelligent work robot and the boundary tangent of the wall;

[0156] The forward direction of the intelligent robot is adjusted based on the real-time heading angle so that the real-time heading angle is within the preset range.

[0157] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: during the process of controlling the intelligent work robot to move forward, if there is a boundary grid within a first preset distance range in the second direction, the boundary grid with the smallest distance to the intelligent work robot in the second direction is used as a third reference point; the second direction is a direction perpendicular to the forward direction of the intelligent work robot.

[0158] From the neighboring boundary grids adjacent to the third reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the first direction as the fourth reference point; the first direction is the forward direction of the intelligent robot.

[0159] Update the boundary tangent of the wall based on the line connecting the fourth reference point and the third reference point;

[0160] The angle between the direction of travel of the intelligent robot and the updated boundary tangent is used as the real-time heading angle.

[0161] In one embodiment, when the computer program is executed by the processor, it also performs the following steps: controlling the intelligent work robot to decelerate along the current direction of travel and stop at a position where the distance between it and the wall boundary reaches a preset near-stop distance range.

[0162] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0163] During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the boundary grid corresponding to the wall between the preset partitions is marked on the global cost map.

[0164] Control the intelligent operation robot to move along the current direction of travel until the distance to the wall reaches the preset near-stop distance range;

[0165] Adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent, and control the intelligent operation robot to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

[0166] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: taking the neighborhood boundary grid with the smallest distance to the intelligent operation robot in a first direction among the multiple boundary grids covered by the virtual detection box as a first reference point; the first direction is the forward direction of the intelligent operation robot;

[0167] From the neighboring boundary grids adjacent to the first reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the second direction as the second reference point; the second direction is the direction perpendicular to the moving direction of the intelligent robot.

[0168] The boundary tangent of the wall is determined by the line connecting the second reference point and the first reference point.

[0169] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: in a local coordinate system constructed based on the current pose of the intelligent robot, a virtual detection box is set at a first preset distance from the intelligent robot along the forward direction of the intelligent robot; wherein, the current pose of the intelligent robot includes the current position and the forward direction; the origin of the local coordinate system is the current position of the intelligent robot, the horizontal axis of the local coordinate system is the forward direction, and the vertical axis of the local coordinate system is the direction perpendicular to the forward direction.

[0170] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the length parameter of the virtual detection frame and / or a first preset distance based on the running speed of the intelligent robot.

[0171] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: during the process of controlling the intelligent work robot to move forward, determining the real-time heading angle between the forward direction of the intelligent work robot and the boundary tangent of the wall;

[0172] The forward direction of the intelligent robot is adjusted based on the real-time heading angle so that the real-time heading angle is within the preset range.

[0173] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: during the process of controlling the intelligent work robot to move forward, if there is a boundary grid within a first preset distance range in the second direction, the boundary grid with the smallest distance to the intelligent work robot in the second direction is used as a third reference point; the second direction is a direction perpendicular to the forward direction of the intelligent work robot.

[0174] From the neighboring boundary grids adjacent to the third reference point, determine the neighborhood boundary grid with the smallest distance to the intelligent robot in the first direction as the fourth reference point; the first direction is the forward direction of the intelligent robot.

[0175] Update the boundary tangent of the wall based on the line connecting the fourth reference point and the third reference point;

[0176] The angle between the direction of travel of the intelligent robot and the updated boundary tangent is used as the real-time heading angle.

[0177] In one embodiment, when the computer program is executed by the processor, it also performs the following steps: controlling the intelligent work robot to decelerate along the current direction of travel and stop at a position where the distance between it and the wall boundary reaches a preset near-stop distance range.

[0178] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0179] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0180] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A control method for an intelligent work robot, characterized in that, The method includes: During the process of controlling the intelligent operation robot to operate in the edge operation mode, when the virtual detection box set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map, the boundary tangent of the wall existing in the forward direction is determined based on the covered boundary grid; wherein, the global cost map is marked with the boundary grid corresponding to the wall between preset partitions. Control the intelligent robot to move along the current direction of travel until the distance between it and the wall reaches the preset near-stop distance range; The forward direction of the intelligent operation robot is adjusted to be parallel to the boundary tangent, and the intelligent operation robot is controlled to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

2. The method according to claim 1, characterized in that, The determination of the boundary tangent of the wall existing in the forward direction based on the covered boundary grid includes: Among the multiple boundary grids covered by the virtual detection frame, the neighborhood boundary grid with the smallest distance to the intelligent operation robot in the first direction is taken as the first reference point; the first direction is the forward direction of the intelligent operation robot. From the neighboring boundary grids adjacent to the first reference point, determine the neighboring boundary grid with the smallest distance to the intelligent robot in the second direction as the second reference point; the second direction is the direction perpendicular to the moving direction of the intelligent robot. The boundary tangent of the wall is determined by the line connecting the second reference point and the first reference point.

3. The method according to claim 2, characterized in that, The method further includes: In a local coordinate system constructed based on the current pose of the intelligent robot, a virtual detection frame is set at a first preset distance from the intelligent robot along its forward direction; wherein, the current pose of the intelligent robot includes its current position and forward direction; the origin of the local coordinate system is the current position of the intelligent robot, the horizontal axis of the local coordinate system is the forward direction, and the vertical axis of the local coordinate system is perpendicular to the forward direction.

4. The method according to claim 3, characterized in that, The virtual detection box is rectangular, and the length direction of the virtual detection box is the same as the forward direction; the method further includes: The length parameter of the virtual detection frame and / or the first preset distance are determined based on the running speed of the intelligent operation robot.

5. The method according to claim 1, characterized in that, The control of the intelligent operation robot to move along the wall within a preset range based on the heading angle includes: During the process of controlling the intelligent operation robot to move forward, the real-time heading angle between the forward direction of the intelligent operation robot and the boundary tangent of the wall is determined; The intelligent robot's forward direction is adjusted based on the real-time heading angle so that the real-time heading angle is within a preset range.

6. The method according to claim 5, characterized in that, During the process of controlling the intelligent work robot to move forward, determining the real-time heading angle between the intelligent work robot's forward direction and the boundary tangent of the wall includes: During the process of controlling the intelligent work robot to move forward, if there is a boundary grid within a first preset distance range in the second direction, the boundary grid with the smallest distance to the intelligent work robot in the second direction is used as the third reference point; the second direction is the direction perpendicular to the forward direction of the intelligent work robot. From the neighboring boundary grids adjacent to the third reference point, the neighborhood boundary grid with the smallest distance to the intelligent robot in the first direction is determined as the fourth reference point; the first direction is the forward direction of the intelligent robot. Update the boundary tangent of the wall based on the line connecting the fourth reference point and the third reference point; The angle between the forward direction of the intelligent robot and the updated boundary tangent is taken as the real-time heading angle.

7. The method according to claim 1, characterized in that, Controlling the intelligent robot to move along its current direction of travel until the distance between it and the wall reaches a preset near-stop distance range includes: Control the intelligent robot to decelerate along the current direction of travel and stop at a position where the distance between it and the wall reaches a preset near-stop distance range.

8. A control device for an intelligent work robot, characterized in that, The device includes: The detection module is used to determine the boundary tangent of the wall existing in the forward direction based on the covered boundary grid when the virtual detection frame set along the forward direction of the intelligent operation robot covers the boundary grid in the global cost map during the operation of the intelligent operation robot in the edge operation mode; wherein the global cost map is marked with the boundary grid corresponding to the wall between preset partitions. The near-stop trigger module is used to control the intelligent operation robot to run along the current forward direction until the distance between it and the wall reaches the preset near-stop distance range; The edge control module is used to adjust the forward direction of the intelligent operation robot to be parallel to the boundary tangent, and to control the intelligent operation robot to move along the wall within a preset range based on the heading angle; the heading angle is the angle between the forward direction of the intelligent operation robot and the boundary tangent.

9. An intelligent operating device, characterized in that, It includes an intelligent work robot and a controller, the controller being used to control the intelligent work robot according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images