A lawn mowing robot and its control method

Abstract

This application discloses a lawnmower robot and its control method, relating to the field of robotics. The robot comprises a drive module, an operation module, a navigation module, an energy module, and a frame. The navigation module comprises two RTK antennas, a laser radar, and a computing module. The computing module is configured to determine the position and orientation of the lawnmower robot as first positioning information using satellite signals received by the two RTK antennas and correction data sent by a base station. When the error of the first positioning information is greater than a first set value, a voxel map is constructed using the point cloud obtained by laser radar scanning, and the position and orientation of the lawnmower robot are then determined based on the voxel map as second positioning information. A global path is then planned based on the second positioning information. The drive module is controlled to drive the lawnmower robot to move along the global path, and the operation module is controlled to operate to perform mowing. This application improves positioning accuracy through the RTK antennas or laser radars, and the computing module enables adaptation to changing terrain.

Description

Technical Field

[0001] The present application relates to the field of robotics, and in particular to a lawn mowing robot and a control method thereof. Background Art

[0002] With the acceleration of urbanization and the expansion of green areas, traditional manual mowing methods can no longer meet the needs of rapid development. Especially in large lawns, gardens, parks, sports fields and other places, traditional mowing methods have problems such as low efficiency, high labor costs, and unsafe working environments.

[0003] Most of the application scenarios of some existing intelligent lawn mowing robots are limited to small lawns or home gardens. For the mowing needs of large lawns and large green areas, the existing technology still has problems such as low navigation system accuracy and limited ability to adapt to complex terrain. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to provide a lawn mowing robot and a control method thereof, so as to improve the local navigation accuracy of the lawn mowing robot and thereby improve its ability to adapt to complex terrain.

[0005] To achieve the above-mentioned object, one aspect of an embodiment of the present application provides a lawn mowing robot, the lawn mowing robot comprising: a driving module, an operation module, a navigation module, an energy module and a frame;

[0006] Wherein, the driving module, the operating module, the navigation module and the energy module are all installed on the vehicle frame;

[0007] The navigation module includes two RTK antennas, a laser radar and a computing module;

[0008] The calculation module includes:

[0009] a first positioning unit, configured to determine the position and orientation of the mowing robot as first positioning information using satellite signals received by the two RTK antennas and correction data sent by a reference station;

[0010] a second positioning unit, configured to construct a voxel map from the point cloud obtained by the laser radar scanning when an error in the first positioning information is greater than a first set value, and further determine the position and orientation of the mowing robot as second positioning information based on the voxel map;

[0011] a path planning unit, configured to plan a global path according to the first positioning information or the second positioning information;

[0012] a driving control unit, configured to control the driving module to drive the mowing robot to move along the global path;

[0013] The operation control unit is used to control the operation module to operate so as to mow the grass.

[0014] In some embodiments, the second positioning unit includes:

[0015] a pre-processing unit, configured to determine a dedistorted point cloud using IMU data measured by an IMU unit in the navigation module;

[0016] a state estimation unit for registering the point cloud using factor graph optimization and the dedistorted point cloud as a priori; and then dynamically registering the scanned point cloud to the voxel map;

[0017] a point cloud matching and optimization unit, configured to update the matching between the point cloud and the voxel map using an extended Kalman filter method to optimize the position and posture of the mowing robot; calculate the residual from the point cloud to the voxel map, and minimize the error from the point cloud to the voxel map under the constraints of the IMU unit;

[0018] A local map updating unit, configured to store the voxel map closest to the current moment and within a set time period in a sliding window manner;

[0019] The loop update unit is used to identify whether the mowing robot has reached the same location before, so as to eliminate accumulated errors, optimize pose estimation and relocalization.

[0020] In some embodiments, the loopback update unit includes:

[0021] A key frame management unit, configured to store the point cloud data and posture at the corresponding moment as a key frame after a set period or when the posture change of the mowing robot exceeds a preset threshold;

[0022] A frame search unit, using a KD-Tree to search the stored key frames for candidate key frames similar to the current key frame;

[0023] a point cloud matching unit, configured to calculate an ICP match between the candidate keyframe and the current keyframe to obtain a relative transformation matrix and an ICP error;

[0024] A factor graph optimization unit is configured to add a loop constraint to the factor graph if the ICP error is less than a second set value; and perform global optimization using iSAM2 incremental optimization to correct the drift error of the voxel map.

[0025] In some embodiments, the path planning unit includes:

[0026] a boundary generating unit, configured to generate a working boundary according to the first positioning information or the second positioning information;

[0027] The path generation unit is used to generate parallel lines with intervals of the radius of the cutter head in the operation module within the working boundary, and then connect the parallel lines end to end to form a continuous global path to completely cover the working boundary.

[0028] In some embodiments, the drive control unit includes:

[0029] a forward-looking point selection unit, configured to dynamically select a forward-looking point on the global path so that the forward-looking point maintains a set distance from the current position of the mowing robot until the forward-looking point coincides with an end point;

[0030] The direction adjustment unit is used to dynamically adjust the direction of the mowing robot and control the driving module to drive the mowing robot to move toward the forward point.

[0031] In some embodiments, the drive control unit further includes an obstacle avoidance unit, and the obstacle avoidance unit includes:

[0032] An obstacle recognition unit is configured to determine the surface of an object based on the return time of the laser radar beam or the angle between the incident beam and the return beam, project the object surface onto a global map, and thereby identify whether the object is an obstacle; or alternatively, utilize the camera module in the navigation module to acquire an image, segment the image using an AI model, and project the segmented object onto the global map to thereby determine whether the object is an obstacle;

[0033] A temporary path generating unit, configured to generate a temporary path when an obstacle is identified;

[0034] The detour unit is used to expand the boundary of the obstacle by a set range when an obstacle is identified, thereby controlling the mowing robot to move according to the temporary path to detour the obstacle after the boundary expansion, and then return to the global path.

[0035] In some embodiments, the temporary path generation unit includes:

[0036] A feasible speed space generating unit, configured to calculate and generate a feasible speed space of the lawn mowing robot within a dynamic time window according to the kinematic constraints of the lawn mowing robot;

[0037] A trajectory prediction unit, configured to predict a predicted trajectory corresponding to each speed combination within a preset time period in the feasible speed space; wherein the speed combination includes a linear velocity and an angular velocity;

[0038] a trajectory scoring unit, configured to score the predicted trajectory according to the distance between the end point of the predicted trajectory and the target point, the forward speed of the predicted trajectory, the length of the predicted trajectory, and the distance between the predicted trajectory and the obstacle;

[0039] The trajectory selection unit is configured to determine the predicted trajectory with the highest score and then combine the corresponding speeds to generate the temporary path.

[0040] In some embodiments, the obstacle avoidance unit further includes:

[0041] The obstacle boundary judgment unit is used to judge whether the distance between the obstacle and the working boundary in the global map is greater than the width of the lawn mowing robot, so as to judge whether the circumvention of the obstacle will exceed the working boundary; if the circumvention of the obstacle will exceed the working boundary, the current path will be interrupted, the starting point of the current path will be returned, and the navigation path of the remaining unworked area will be replanned after subtracting the worked area.

[0042] In some embodiments, the energy module includes multiple batteries, and the battery compartment of each battery adopts a quick-release structure;

[0043] The calculation module also includes:

[0044] The energy management unit is used to switch the current battery to the remaining batteries if the current of the current battery is lower than the preset power threshold, and send a message to the management terminal of the lawn mowing robot to remind the user to replace the battery.

[0045] To achieve the above object, another aspect of the present invention provides a control method, which includes the following steps:

[0046] The position and orientation of the mowing robot are determined as first positioning information using satellite signals received by the two RTK antennas and correction data sent by the base station;

[0047] When the error of the first positioning information is greater than a first set value, a voxel map is constructed using the point cloud obtained by laser radar scanning, and the position and direction of the lawn mowing robot are determined as second positioning information according to the voxel map;

[0048] Planning and obtaining a global path according to the first positioning information or the second positioning information;

[0049] A control driving module drives the mowing robot to move according to the global path;

[0050] Control the operation module to work to mow the grass.

[0051] The embodiments of the present application include at least the following beneficial effects:

[0052] The lawn mower robot of the present application includes: a drive module, an operation module, a navigation module, an energy module and a frame; the navigation module includes two RTK antennas, a laser radar and a computing module; the computing module includes: a first positioning unit for determining the position and direction of the lawn mower robot as first positioning information using satellite signals received by the two RTK antennas and correction data sent by a base station; a second positioning unit for constructing a voxel map from the point cloud obtained by laser radar scanning when the error of the first positioning information is greater than a first set value, and then determining the position and direction of the lawn mower robot based on the voxel map as second positioning information; a path planning unit for planning a global path based on the first positioning information or the second positioning information; a drive control unit for controlling the drive module to drive the lawn mower robot to move along the global path; and an operation control unit for controlling the operation module to operate in order to mow. The present application improves positioning accuracy by using RTK antennas or laser radars for positioning, thereby providing accurate position information for navigation; planning a global path by the computing module can adapt to changing terrain and improve the adaptability to complex terrain. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0054] Figure 1 This is an example structural diagram of a lawn mowing robot provided in an embodiment of the present application;

[0055] Figure 2 A flow chart of a control method provided in an embodiment of the present application;

[0056] FIG3( a ) is a top view of a lawn mowing robot provided in an embodiment of the present application;

[0057] FIG3( b ) is an oblique view of a lawn mowing robot provided in an embodiment of the present application;

[0058] FIG3( c ) is a front view of a lawn mowing robot provided in an embodiment of the present application;

[0059] FIG3( d ) is a side view of a lawn mowing robot provided in an embodiment of the present application.

[0060] Reference numerals: 101 is a cutting knife, 102 is a lifting chassis, 103 is a crawler track, 104 is an RTK antenna hole, 105 is a battery compartment, 106 is a video transmission camera, 107 is a laser radar, and 108 is a driving wheel. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are merely examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0062] It will be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0063] The terms "at least one", "plurality", "each", "any", etc. used in this application include "at least one", "two" or more, "plurality" or "each", "any" or "any one", "each" or "any one" as used herein.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0065] Before describing the embodiments of the present application in detail, some of the related technologies involved in the embodiments of the present application are first described as follows:

[0066] RTK: Real-time kinematic real-time dynamic carrier phase differential technology.

[0067] LiDAR: LiDAR (Light Laser Detection and Ranging) is the abbreviation of laser detection and ranging system.

[0068] SLAM: Simultaneous Localization and Mapping (SLAM).

[0069] Loop detection: also known as closed-loop detection, refers to the robot's ability to recognize that it has visited a certain scene, thus closing the map loop.

[0070] PID control algorithm: PID is the abbreviation of Proportional, Integral, and Differential. The PID control algorithm is a control algorithm that combines proportional, integral, and differential.

[0071] Voxel Map: A voxel is a volume pixel in three-dimensional space, similar to a pixel in a two-dimensional image, but extended to three dimensions. A voxel map divides the environment into a uniform grid of cubes, with each cube (voxel) storing information within that space (such as occupancy probability, color, semantic label, etc.).

[0072] ESKF, or Extended Kalman Filter, is an improved Kalman filter algorithm that focuses on estimating state errors (such as position and attitude errors) rather than directly estimating system states (such as position and attitude). It solves the state estimation problem for nonlinear systems by linearizing the error dynamics model.

[0073] KD-Tree: A binary tree data structure used to efficiently organize points in k-dimensional space. It accelerates nearest neighbor searches by recursively partitioning the space (selecting the median point in different dimensions).

[0074] ICP matching: An iterative algorithm for aligning two point clouds (or a point cloud and a map) to minimize the distance error between them. The core steps include:

[0075] 1. Association: Find the nearest neighbor points in the target point cloud.

[0076] 2. Solve the transformation: Calculate the rotation and translation matrices through SVD decomposition.

[0077] 3. Iterative optimization: repeat until convergence.

[0078] iSAM2: A factor graph-based incremental optimization framework for SLAM backend optimization. It efficiently updates pose and map variables via a Bayesian tree.

[0079] Robot dynamics model: A mathematical model that describes the relationship between robot motion and forces / torques, usually based on Newton-Euler equations or Lagrangian mechanics.

[0080] In response to the problems existing in the prior art, some embodiments of the present application aim to solve the following technical problems:

[0081] Large, specialized robots must be capable of all-weather, all-terrain travel and operation, which inevitably leads to overall structural size and weight challenges. While an excessively long body and ground contact length may provide excellent maneuverability, the load and friction from turning can cause soil damage and compaction. This can be mitigated by adjusting motor output and torque, while also reducing machine weight.

[0082] Based on this, this application proposes a lawn mowing robot and its control method, which aims to achieve efficient and precise mowing operations on large areas of lawns through advanced intelligent control technology, an efficient battery system and the collaborative work of multiple sensors, with strong adaptability, long endurance and environmental protection.

[0083] Reference Figure 1 , an embodiment of the present application provides a lawn mowing robot, the lawn mowing robot comprising: a driving module, an operation module, a navigation module, an energy module and a frame;

[0084] Wherein, the driving module, the operating module, the navigation module and the energy module are all installed on the vehicle frame;

[0085] The navigation module includes two RTK antennas, a laser radar and a computing module;

[0086] The calculation module includes:

[0087] a first positioning unit, configured to determine the position and orientation of the mowing robot as first positioning information using satellite signals received by the two RTK antennas and correction data sent by a reference station;

[0088] a second positioning unit, configured to construct a voxel map from the point cloud obtained by the laser radar scanning when an error in the first positioning information is greater than a first set value, and further determine the position and orientation of the mowing robot as second positioning information based on the voxel map;

[0089] a path planning unit, configured to plan a global path according to the first positioning information or the second positioning information;

[0090] a driving control unit, configured to control the driving module to drive the mowing robot to move along the global path;

[0091] The operation control unit is used to control the operation module to operate so as to mow the grass.

[0092] As an optional implementation manner, the second positioning unit includes:

[0093] a pre-processing unit, configured to determine a dedistorted point cloud using IMU data measured by an IMU unit in the navigation module;

[0094] a state estimation unit for registering the point cloud using factor graph optimization and the dedistorted point cloud as a priori; and then dynamically registering the scanned point cloud to the voxel map;

[0095] a point cloud matching and optimization unit, configured to update the matching between the point cloud and the voxel map using an extended Kalman filter method to optimize the position and posture of the mowing robot; calculate the residual from the point cloud to the voxel map, and minimize the error from the point cloud to the voxel map under the constraints of the IMU unit;

[0096] A local map updating unit, configured to store the voxel map closest to the current moment and within a set time period in a sliding window manner;

[0097] The loop update unit is used to identify whether the mowing robot has reached the same location before, so as to eliminate accumulated errors, optimize pose estimation and relocalization.

[0098] Furthermore, the loop update unit includes:

[0099] A key frame management unit, configured to store the point cloud data and posture at the corresponding moment as a key frame after a set period or when the posture change of the mowing robot exceeds a preset threshold;

[0100] A frame search unit, using a KD-Tree to search the stored key frames for candidate key frames similar to the current key frame;

[0101] a point cloud matching unit, configured to calculate an ICP match between the candidate keyframe and the current keyframe to obtain a relative transformation matrix and an ICP error;

[0102] A factor graph optimization unit is configured to add a loop constraint to the factor graph if the ICP error is less than a second set value; and perform global optimization using iSAM2 incremental optimization to correct the drift error of the voxel map.

[0103] As an optional implementation, the path planning unit includes:

[0104] a boundary generating unit, configured to generate a working boundary according to the first positioning information or the second positioning information;

[0105] The path generation unit is used to generate parallel lines with intervals of the radius of the cutter head in the operation module within the working boundary, and then connect the parallel lines end to end to form a continuous global path to completely cover the working boundary.

[0106] As an optional implementation, the drive control unit includes:

[0107] a forward-looking point selection unit, configured to dynamically select a forward-looking point on the global path so that the forward-looking point maintains a set distance from the current position of the mowing robot until the forward-looking point coincides with an end point;

[0108] The direction adjustment unit is used to dynamically adjust the direction of the mowing robot and control the driving module to drive the mowing robot to move toward the forward point.

[0109] As an optional embodiment, the drive control unit further includes an obstacle avoidance unit, and the obstacle avoidance unit includes:

[0110] An obstacle recognition unit is configured to determine the surface of an object based on the return time of the laser radar beam or the angle between the incident beam and the return beam, project the object surface onto a global map, and thereby identify whether the object is an obstacle; or alternatively, utilize the camera module in the navigation module to acquire an image, segment the image using an AI model, and project the segmented object onto the global map to thereby determine whether the object is an obstacle;

[0111] A temporary path generating unit, configured to generate a temporary path when an obstacle is identified;

[0112] The detour unit is used to expand the boundary of the obstacle by a set range when an obstacle is identified, thereby controlling the mowing robot to move according to the temporary path to detour the obstacle after the boundary expansion, and then return to the global path.

[0113] Furthermore, the temporary path generating unit includes:

[0114] A feasible speed space generating unit, configured to calculate and generate a feasible speed space of the lawn mowing robot within a dynamic time window according to the kinematic constraints of the lawn mowing robot;

[0115] A trajectory prediction unit, configured to predict a predicted trajectory corresponding to each speed combination within a preset time period in the feasible speed space; wherein the speed combination includes a linear velocity and an angular velocity;

[0116] a trajectory scoring unit, configured to score the predicted trajectory according to the distance between the end point of the predicted trajectory and the target point, the forward speed of the predicted trajectory, the length of the predicted trajectory, and the distance between the predicted trajectory and the obstacle;

[0117] The trajectory selection unit is configured to determine the predicted trajectory with the highest score and then combine the corresponding speeds to generate the temporary path.

[0118] As another further embodiment, the obstacle avoidance unit further includes:

[0119] The obstacle boundary judgment unit is used to judge whether the distance between the obstacle and the working boundary in the global map is greater than the width of the lawn mowing robot, so as to judge whether the circumvention of the obstacle will exceed the working boundary; if the circumvention of the obstacle will exceed the working boundary, the current path will be interrupted, the starting point of the current path will be returned, and the navigation path of the remaining unworked area will be replanned after subtracting the worked area.

[0120] As an optional embodiment, the energy module includes multiple batteries, and the battery compartment of each battery adopts a quick-release structure;

[0121] The calculation module also includes:

[0122] The energy management unit is used to switch the current battery to the remaining batteries if the current of the current battery is lower than the preset power threshold, and send a message to the management terminal of the lawn mowing robot to remind the user to replace the battery.

[0123] Reference Figure 2 , the present application also provides a control method, the control method comprising the following steps S1 to S5:

[0124] S1: Determine the position and orientation of the mowing robot as the first positioning information using the satellite signals received by the two RTK antennas and the correction data sent by the base station;

[0125] S2: When the error of the first positioning information is greater than a first set value, constructing a voxel map using the point cloud obtained by laser radar scanning, and then determining the position and direction of the lawn mowing robot as second positioning information based on the voxel map;

[0126] S3: Plan and obtain a global path according to the first positioning information or the second positioning information;

[0127] S4: controlling the driving module to drive the mowing robot to move according to the global path;

[0128] S5: Control the operation module to perform mowing.

[0129] Next, the solution of the embodiment of the present application will be introduced and explained in detail with reference to specific application examples.

[0130] Referring to Figures 3(a), 3(b), 3(c), and 3(d), where Figure 3(a) is a top view of the mowing robot, Figure 3(b) is an oblique view, Figure 3(c) is a front view, and Figure 3(d) is a side view, the mowing robot of this embodiment may include a drive module, an operation module, a navigation module, and an energy module. These four modules are all constructed on the vehicle frame. The drive module utilizes dual upright motors directly connected to a horizontally positioned dual reducer, with drive wheels driving tracks located on both sides of the machine. The operation module, located at the bottom of the vehicle frame, utilizes dual asymmetrical motors. A powered push-rod lift platform allows for movement 2.5-13.5 cm above the ground. The navigation module consists of two components: a sensor and a computing module. The sensors are located on the top and head of the vehicle, respectively, and consist of an RTK antenna, a lidar, and a camera module. The RTK antenna receives satellite signals for positioning and navigation. The lidar provides enhanced positioning when near trees or other obstructing environments, and works together with the camera module to identify and avoid obstacles. The computing module, located beneath the vehicle's shell and mounted on the vehicle frame, processes satellite signals, identifies obstacles, and performs local navigation. The energy module, located at the front of the vehicle and shielded by a hatch, contains two lithium iron phosphate batteries. If one battery is insufficient, it automatically switches to another, allowing a single person to disassemble and replace the batteries.

[0131] 1. Manual remote control: Connect the robot via the remote control. The left joystick can be used to control the robot's movement and steering. The right scroll wheel on the top of the remote control can be used to manually adjust the cutter head height.

[0132] 2. Automatic operation:

[0133] 1) Move the lever on the right to turn on the automatic operation mode.

[0134] 2) Connect the robot and wait for it to automatically receive the RTK signal and analyze and calibrate its real-time positioning based on the measurement station.

[0135] 3) Click on the device details and select the planned path method (U-shaped or Bow-shaped).

[0136] 4) Move the robot and click on the newly added points to mark multiple corners of the work area.

[0137] 5) If there is a restricted area within the area, click on the restricted area and repeat the steps in 4) to add a new point in the restricted area.

[0138] 6) After marking the work area and prohibited area, click the return point, move the robot to the specified location, and click Next to confirm the starting point and return point.

[0139] 7) Click Finish to complete the map drawing.

[0140] 8) Click Start to start the automation operation on the selected map.

[0141] Specifically, this embodiment provides an intelligent electric lawn mower robot for large-area lawn maintenance, which integrates high-precision RTK positioning technology to achieve autonomous navigation, obstacle avoidance, and precise mowing.

[0142] 1. Robot startup and environmental perception:

[0143] 1. The RTK antenna receives carrier phase signals from satellite systems such as GPS and Beidou. The base station (with known precise coordinates) compares the received satellite signals with its own position, generates differential correction data, and transmits it to the rover (robot) in real time via radio or network. The rover combines the received satellite signals with the base station's correction data to eliminate errors such as atmospheric delay and satellite clock error, resolve the carrier phase ambiguity, and ultimately achieve centimeter-level positioning (±2cm). The use of dual antennas allows the robot's position and orientation to be calculated based on their mounting positions. Simultaneously, the robot continuously transmits positioning data (latitude, longitude, and elevation) in real time to the navigation module for path planning adjustments and position calibration.

[0144] Second, when there are many buildings or trees around the robot, resulting in poor RTK data quality, laser simultaneous localization and mapping (SLAM) technology can be used to provide mapping and positioning information.

[0145] 1) Preprocessing stage:

[0146] IMU pre-integration: Use IMU data to predict the initial pose of the current frame.

[0147] Point cloud dedistortion: The point cloud distortion generated during the scanning process is compensated according to the IMU motion, so that the point cloud returns to the unified time coordinate.

[0148] 2) Recursive state estimation:

[0149] Factor graph optimization is used to efficiently register point clouds using IMU odometry predictions as priors.

[0150] Directly register new scan points to the voxel map to avoid building a complete point cloud frame, thereby reducing the amount of calculation.

[0151] 3) Point cloud matching and optimization:

[0152] Compared to frame-to-frame matching, this embodiment can directly match laser points with points in the local map:

[0153] ESKF (Extended Kalman Filter) is used to update the state and optimize the posture.

[0154] Calculate the residual from the laser point to the map and minimize the error under the IMU constraints.

[0155] 4) Recursive local map update:

[0156] Using voxel map, only key point clouds are retained instead of complete point clouds.

[0157] The local map is managed by a sliding window method, and only the point cloud of the most recent period is stored to prevent map expansion and reduce the storage requirements of the mowing robot.

[0158] 5) Loopback update:

[0159] The above process can only guarantee the accuracy of the local map. When the site is large, the accumulated errors of the local map will lead to a large error in the final global map. To solve the above problem, this embodiment can perform loop detection and loop update. The loop update can eliminate the accumulated errors caused by the local map update, optimize the pose estimation and relocalization. The steps of loop detection are as follows:

[0160] a. Keyframe management:

[0161] Only keyframes (point cloud data and poses at critical moments) are stored instead of all point cloud frames to reduce computational effort.

[0162] Keyframes are usually added at a certain interval or when there is a large enough change in pose.

[0163] b. Approximate Nearest Neighbor Search (ANN):

[0164] Use KD-Tree to search for similar point cloud frames in the historical keyframe database to find possible candidate keyframes.

[0165] c. Point Cloud Matching (ICP):

[0166] Calculate the ICP matching between the candidate keyframe and the current keyframe to obtain the relative transformation matrix (discard if the error is large).

[0167] d. Factor graph optimization:

[0168] If the ICP error is small, the loop closure constraint is added to the factor graph (the pose constraint between loop closure frames).

[0169] Use iSAM2 incremental optimization to perform global optimization and correct the drift error of the entire map.

[0170] Third, while gradually building the map, the robot analyzes the single-frame point cloud and converts it into the robot's coordinate system. When the point cloud height exceeds the set grass height, it is considered an obstacle. The detected obstacle location is then mapped to the global map and marked as an avoidance zone.

[0171] 2. Autonomous path planning and mowing operations:

[0172] First, accept positioning information provided by RTK or laser SLAM to generate a working boundary. Within this boundary, parallel lines are generated, separated by the cutter radius. These parallel lines are connected end to end to form a continuous path (global path) that completely covers the polygon enclosed by the working boundary. This ensures that the robot's cutting blade covers all areas within the boundary during movement. Then, on the pre-planned trajectory, a target point within a certain range from the current vehicle is selected, called a look-ahead point. The robot's direction is continuously adjusted to move toward the look-ahead point. The look-ahead point continuously slides along the predetermined trajectory, maintaining a certain distance from the robot, until it coincides with the end point, allowing the robot to move along the path.

[0173] 2. During mowing operations, the lidar emits a laser beam that returns after reaching the surface of an object. The position of the object's surface (a point in space) is determined based on the time it takes for the laser to return or the angle between the incident and returning beams. The radar also emits a series of light beams, which generate a series of point clouds in space. These point clouds correspond to the contours of the object's surface. The camera module also identifies obstacles, which are generally low (the radar cannot distinguish between the ground and low obstacles). The AI ​​model is used to perform image segmentation, and after the obstacle is segmented, it is projected onto the ground. When the robot finds an object blocking its walking path, it generates a temporary path to avoid the blocking object. This path ensures that there is no collision between the robot's outline and the object's point cloud during walking, and it stays as close to the global path as possible. After obstacle avoidance is completed, it returns to the global path as soon as possible.

[0174] This embodiment generates a temporary path, so that the mowing robot can detour when encountering an obstacle while moving along the global path, and return to the global path as soon as possible after completing the obstacle avoidance, avoiding the need to plan a new global path again due to not moving along the global path due to the detour.

[0175] The steps to generate a temporary path include:

[0176] 1) Generate feasible speed space:

[0177] The robot has different linear and angular velocity combinations to choose from at each moment.

[0178] Calculate the feasible velocity space:

[0179] Robot kinematic constraints (maximum acceleration, maximum velocity, maximum angular velocity, maximum angular acceleration).

[0180] The range of speeds that the robot can reach within a dynamic window (limited time).

[0181] 2) Predicted trajectory:

[0182] In the feasible speed space, predict the trajectory corresponding to each speed combination.

[0183] The duration of trajectory simulation is usually 1-2 seconds to evaluate the movement effect in the short term.

[0184] 3) Evaluate trajectory scores:

[0185] A comprehensive score is calculated for each trajectory, mainly considering:

[0186] Heading: The distance between the trajectory endpoint and the target point (the closer the distance to the target, the higher the score. The target point is a point on the path that has not yet been executed and is not occupied by obstacles. If all the paths that have not yet been executed are occupied, the next path will be selected in turn).

[0187] Velocity: The forward speed of the trajectory (faster speeds mean higher scores).

[0188] Path: The shorter the detour path, the higher the score.

[0189] Clearance: The distance between the trajectory and obstacles (safer trajectories have higher scores).

[0190] The final score is obtained through weighted calculation.

[0191] 4) Select the optimal trajectory:

[0192] The trajectory with the highest score is selected as the motion command (linear velocity and angular velocity) at the current moment.

[0193] The robot executes the instruction and enters the next cycle.

[0194] Third, the machine analyzes its real-time position and adjusts the cutterhead's lift motor to ensure the machine maintains close contact with the ground, allowing it to mow at the desired height. When the LiDAR and camera modules detect and identify obstacles, they avoid them while reducing the blade's rotation speed, preventing repeated mowing in previously mowed areas and resulting in inconsistent results.

[0195] 3. Obstacle avoidance and dynamic adjustment:

[0196] During mowing, a lidar emits a laser beam that strikes the surface and returns. The laser's return time or the angle between the incident and returning beams determines the surface's location (a point in space). The lidar also emits a series of beams simultaneously, generating a series of point clouds in space that correspond to the surface contours.

[0197] Second, the obstacle information identified and segmented by the LiDAR and camera modules is projected onto the global map. The obstacle boundary is expanded by 10-15 cm to ensure the robot maintains a safe distance from the obstacle at all times (10 cm for stationary obstacles, 15 cm for moving obstacles). The robot also determines whether the distance from the obstacle boundary in the global map is greater than the machine width, and whether detours will exceed the working boundary. If the obstacle is sufficiently far from the boundary to allow the robot to detour, the navigation algorithm will plan a detour route based on the obstacle boundary, avoiding obstacles while also preventing missed cuts. If the obstacle is less than the machine width from the working boundary, detours are not possible, and the path is interrupted. The robot returns to the starting point of the path, subtracting the already worked area and replanning the navigation path for the remaining unworked area.

[0198] 4. Battery life management and recharging:

[0199] 1. When the power of a single battery is less than 5%, the robot automatically switches to the backup battery, and the APP reminds the staff to replace the battery.

[0200] 2. It can power multiple batteries at the same time. The robot's battery compartment adopts a quick-release structure, which facilitates battery replacement and achieves longer working time.

[0201] The beneficial effects of this embodiment include:

[0202] Using high-precision RTK positioning technology, accurate positioning can be achieved even in very open environments (such as dams).

[0203] Intelligent path planning is realized to avoid repeated coverage and missed areas, thus improving work efficiency.

[0204] Obstacle avoidance through multi-sensor fusion improves safety and can work stably in complex environments.

[0205] The mowing height can be adjusted to meet the needs of different types of lawns and improve the mowing quality.

[0206] Remote management and data analysis can monitor the working status in real time through the APP and perform intelligent scheduling.

[0207] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0208] Those skilled in the art will understand that the technical solutions shown in the drawings do not constitute a limitation on the embodiments of the present application, and may include more or fewer technical solutions than shown in the drawings, or a combination of certain technical solutions, or different technical solutions.

[0209] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0210] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0211] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A lawn mowing robot, characterized in that: The lawn mowing robot comprises: a driving module, an operation module, a navigation module, an energy module and a frame; Wherein, the driving module, the operating module, the navigation module and the energy module are all installed on the vehicle frame; The navigation module includes two RTK antennas, a laser radar and a computing module; The calculation module includes: a first positioning unit, configured to determine the position and orientation of the mowing robot as first positioning information using satellite signals received by the two RTK antennas and correction data sent by a reference station; a second positioning unit, configured to construct a voxel map from the point cloud obtained by the laser radar scanning when an error in the first positioning information is greater than a first set value, and further determine the position and orientation of the mowing robot as second positioning information based on the voxel map; a path planning unit, configured to plan a global path according to the first positioning information or the second positioning information; a driving control unit, configured to control the driving module to drive the mowing robot to move along the global path; An operation control unit, used for controlling the operation module to operate so as to mow the grass; The driving control unit further includes an obstacle avoidance unit, which includes: An obstacle recognition unit is configured to determine the surface of an object based on the return time of the laser radar beam or the angle between the incident beam and the return beam, project the object surface onto a global map, and thereby identify whether the object is an obstacle; or alternatively, utilize the camera module in the navigation module to acquire an image, segment the image using an AI model, and project the segmented object onto the global map to thereby determine whether the object is an obstacle; A temporary path generating unit, configured to generate a temporary path when an obstacle is identified; a detour unit, configured to, when an obstacle is identified, expand the boundary of the obstacle by a set range, thereby controlling the mowing robot to move according to the temporary path to detour the obstacle after the boundary expansion, and then return to the global path; The temporary path generating unit includes: A feasible speed space generating unit, configured to calculate and generate a feasible speed space of the lawn mowing robot within a dynamic time window according to the kinematic constraints of the lawn mowing robot; A trajectory prediction unit, configured to predict a predicted trajectory corresponding to each speed combination within a preset time period in the feasible speed space; wherein the speed combination includes a linear velocity and an angular velocity; a trajectory scoring unit, configured to score the predicted trajectory according to the distance between the end point of the predicted trajectory and the target point, the forward speed of the predicted trajectory, the length of the predicted trajectory, and the distance between the predicted trajectory and the obstacle; The trajectory selection unit is configured to determine the predicted trajectory with the highest score and then combine the corresponding speeds to generate the temporary path.

2. The lawn mowing robot according to claim 1, characterized in that: The second positioning unit includes: a pre-processing unit, configured to determine a dedistorted point cloud using IMU data measured by an IMU unit in the navigation module; a state estimation unit for registering the point cloud using factor graph optimization and the dedistorted point cloud as a priori; and then dynamically registering the scanned point cloud to the voxel map; a point cloud matching and optimization unit, configured to update the matching between the point cloud and the voxel map using an extended Kalman filter method to optimize the position and posture of the mowing robot; calculate the residual from the point cloud to the voxel map, and minimize the error from the point cloud to the voxel map under the constraints of the IMU unit; A local map updating unit, configured to store the voxel map closest to the current moment and within a set time period in a sliding window manner; The loop update unit is used to identify whether the mowing robot has reached the same location before, so as to eliminate accumulated errors, optimize pose estimation and relocalization.

3. The lawn mowing robot according to claim 2, characterized in that: The loopback update unit includes: A key frame management unit, configured to store the point cloud data and posture at the corresponding moment as a key frame after a set period or when the posture change of the mowing robot exceeds a preset threshold; A frame search unit, using a KD-Tree to search the stored key frames for candidate key frames similar to the current key frame; a point cloud matching unit, configured to calculate an ICP match between the candidate keyframe and the current keyframe to obtain a relative transformation matrix and an ICP error; A factor graph optimization unit is configured to add a loop constraint to the factor graph if the ICP error is less than a second set value; and perform global optimization using iSAM2 incremental optimization to correct the drift error of the voxel map.

4. The lawn mowing robot according to claim 1, characterized in that: The path planning unit includes: a boundary generating unit, configured to generate a working boundary according to the first positioning information or the second positioning information; The path generation unit is used to generate parallel lines with intervals of the radius of the cutter head in the operation module within the working boundary, and then connect the parallel lines end to end to form a continuous global path to completely cover the working boundary.

5. The lawn mowing robot according to claim 1, characterized in that: The drive control unit includes: a forward-looking point selection unit, configured to dynamically select a forward-looking point on the global path so that the forward-looking point maintains a set distance from the current position of the mowing robot until the forward-looking point coincides with an end point; The direction adjustment unit is used to dynamically adjust the direction of the mowing robot and control the driving module to drive the mowing robot to move toward the forward point.

6. The lawn mowing robot according to claim 1, characterized in that: The obstacle avoidance unit further includes: The obstacle boundary judgment unit is used to judge whether the distance between the obstacle and the working boundary in the global map is greater than the width of the lawn mowing robot, so as to judge whether the circumvention of the obstacle will exceed the working boundary; if the circumvention of the obstacle will exceed the working boundary, the current path will be interrupted, the starting point of the current path will be returned, and the navigation path of the remaining unworked area will be replanned after subtracting the worked area.

7. The lawn mowing robot according to any one of claims 1 to 6, characterized in that: The energy module includes multiple batteries, and the battery compartment of each battery adopts a quick-release structure; The calculation module also includes: The energy management unit is used to switch the current battery to the remaining batteries if the current of the current battery is lower than the preset power threshold, and send a message to the management terminal of the lawn mowing robot to remind the user to replace the battery.

8. A control method for a lawn mowing robot, characterized in that: The control method is used to control the lawn mowing robot according to claim 1, and the control method comprises the following steps: The position and orientation of the mowing robot are determined as first positioning information using satellite signals received by the two RTK antennas and correction data sent by the base station; When the error of the first positioning information is greater than a first set value, a voxel map is constructed using the point cloud obtained by laser radar scanning, and the position and direction of the lawn mowing robot are determined as second positioning information according to the voxel map; Planning and obtaining a global path according to the first positioning information or the second positioning information; A control driving module drives the mowing robot to move according to the global path; Control the operation module to work to mow the grass.

Images