Robot anti-falling method, robot and storage medium

Abstract

The invention relates to the technical field of robots, in particular to a robot anti-falling method, a robot and a storage medium, the method is applied to the robot, and the method comprises the following steps: acquiring point cloud data corresponding to a ground environment in a driving direction of the robot, and determining whether a falling environment exists according to the point cloud data; if the falling environment exists, determining the boundary of the falling environment; and determining the advancing strategy of the robot according to the boundary of the falling environment. According to the invention, after the existence of the falling environment is determined according to the acquired ground point cloud data, the advancing strategy of the robot is determined according to the boundary of the falling environment. According to the method, different advancing strategies are flexibly adopted according to the boundary of the falling environment, and the influence of the falling environment on normal operation of the robot can be reduced, so that the influence on task execution of the robot is reduced, and the task execution efficiency is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of robots, and in particular to a robot anti-falling method, a robot and a storage medium. Background Art

[0002] With the vigorous development of artificial intelligence technology, robot intelligent unmanned delivery services have gradually become popular, and the market size continues to grow, making unmanned delivery services have very broad application prospects in many fields, such as express delivery, takeout, and B2C retail. In actual application scenarios, when robots perform tasks of delivering items in warehouses, hospitals, hotels, KTVs, industrial parks and other places, they usually encounter environments that are easy to fall or fall, such as puddles and stairs.

[0003] At present, after detecting a falling environment, the robot will sound an alarm and stop working, which affects the normal operation of the robot, thereby affecting the normal execution of the robot's tasks and reducing the efficiency of task execution. Summary of the invention

[0004] In view of this, an object of an embodiment of the present invention is to provide a robot anti-falling method, a robot and a storage medium, aiming to solve the technical problem in the prior art that the falling environment affects the normal operation of the robot and reduces the task execution efficiency.

[0005] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a robot anti-falling method, which is applied to the robot, and the method includes:

[0007] Acquire point cloud data corresponding to the ground environment in the direction of robot travel, and determine whether there is a falling environment according to the point cloud data;

[0008] If the falling environment exists, determining the boundary of the falling environment;

[0009] A moving strategy of the robot is determined according to the boundary of the falling environment.

[0010] In some embodiments, obtaining point cloud data corresponding to the ground environment in the direction of travel of the robot, and determining whether there is a falling environment according to the point cloud data, includes:

[0011] Determine the ground point cloud of the ground where the robot is located according to the point cloud data, and fit a first plane;

[0012] If, in the point cloud data, there are target point clouds whose height is lower than the preset height of the first plane, and the number of target point clouds is greater than or equal to the preset number, it is determined that a falling environment exists;

[0013] In the point cloud data, if there is no target point cloud whose height is lower than the preset height of the first plane, and / or the number of target point clouds is less than a preset number, it is determined that there is no falling environment.

[0014] In some embodiments, determining the boundary of the falling environment includes:

[0015] Determine a boundary area between the ground point cloud and the target point cloud as a boundary of a falling environment;

[0016] or,

[0017] Fitting a second plane according to the target point cloud;

[0018] Determining a preset value according to a height difference between the first plane and the second plane;

[0019] Each point cloud in the point cloud data is traversed, and a point cloud having a height difference with an adjacent point cloud greater than a preset value is determined as a point cloud corresponding to a boundary of the falling environment.

[0020] In some embodiments, determining the robot's travel strategy according to the boundary of the falling environment includes:

[0021] Determining whether the robot can travel around the boundary of the falling environment according to the boundary of the falling environment and the traversable area of ​​the electronic map;

[0022] If one side of the boundary of the falling environment is a passable area, and the minimum width of the passable area is greater than a preset width value, it is determined that the robot can travel around the boundary of the falling environment;

[0023] If the robot can travel around the boundary of the falling environment, then updating part of the grids of the traversable area of ​​the electronic map along the boundary of the falling environment as obstacle areas, so that the robot stays away from the boundary of the falling environment when planning a path;

[0024] A path is planned according to the updated electronic map to travel around the boundary of the falling environment.

[0025] In some embodiments, the method further comprises:

[0026] If one side of the boundary of the falling environment is an obstacle area, and / or the minimum width of the passable area on one side of the boundary of the falling environment is less than or equal to a preset width value, it is determined that the robot cannot travel around the boundary of the falling environment;

[0027] If the robot cannot travel around the boundary of the falling environment, and the distance between the robot and the boundary of the falling environment is greater than or equal to a first preset distance, reducing the speed of the robot to the first preset speed, and driving forward at the first preset speed;

[0028] If the distance between the robot and the boundary of the falling environment is less than the first preset distance, the robot stops moving forward.

[0029] In some embodiments, the method further comprises:

[0030] Determining the type of the falling environment according to the target point cloud, wherein the type of the falling environment includes a pit and a multi-step;

[0031] The moving strategy of the robot is determined according to the type of the falling environment.

[0032] In some embodiments, determining the robot's travel strategy according to the type of the falling environment includes:

[0033] If the type of the falling environment is a pit, detecting the slope of the connecting slope between the ground where the robot is located and the bottom of the pit;

[0034] When the slope of the connecting slope is less than or equal to a preset slope, it is determined that the robot can cross the pit and move forward;

[0035] When the slope of the connecting slope is greater than a preset slope, it is determined that the robot cannot cross the pit, and the path is replanned to avoid the pit.

[0036] In some embodiments, the method further comprises:

[0037] If the type of the falling environment is multiple steps, the multiple steps are marked as obstacles on the electronic map.

[0038] In a second aspect, an embodiment of the present invention provides a robot, comprising:

[0039] A processor and a memory communicatively connected to the processor;

[0040] The memory stores computer program instructions executable by the processor, and when the computer program instructions are called by the processor, the processor executes any one of the robot anti-falling methods proposed in the first aspect, or executes steps in any one of the possible implementation methods of any one of the robot anti-falling methods proposed in the first aspect.

[0041] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions executable by a processor are stored, and when the computer program instructions are called by the processor, the processor executes any one of the robot anti-falling methods proposed in the first aspect, or executes the steps in any one of the possible implementation methods of any one of the robot anti-falling methods proposed in the first aspect.

[0042] Beneficial effects of the embodiments of the present invention: Different from the prior art, the robot anti-fall method provided by the embodiments of the present invention is applied to the robot, and the method includes: obtaining point cloud data corresponding to the ground environment in the direction of the robot's travel, and determining whether there is a falling environment based on the point cloud data; if there is a falling environment, determining the boundary of the falling environment; and determining the robot's travel strategy based on the boundary of the falling environment.

[0043] In the present invention, after determining the existence of a falling environment based on the acquired ground point cloud data, the robot's travel strategy is determined according to the boundary of the falling environment. The method flexibly adopts different travel strategies according to the boundary of the falling environment, which can reduce the impact of the falling environment on the normal operation of the robot, thereby reducing the impact on the robot's task execution and ensuring the efficiency of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for the description of the prior art or the embodiments. Obviously, the drawings described below only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope of protection. For those of ordinary skill in the art, other related drawings can also be obtained based on these drawings without creative work.

[0045] Figure 1 is a schematic diagram of an application scenario of a robot anti-fall method provided by some embodiments of the present invention;

[0046] Figure 2 is a schematic diagram of the structure of a robot provided by some embodiments of the present invention;

[0047] Figure 3 is a flowchart of a robot anti-fall method provided by some embodiments of the present invention;

[0048] Figure 4 yes Figure 3 A schematic diagram of a sub-process of step S100 in the robot anti-falling method shown in the embodiment;

[0049] Figure 5 yes Figure 3 A schematic diagram of a sub-process of step S300 in the robot anti-fall method shown in the embodiment;

[0050] Figure 6 Schematic diagram of application scenarios of robot anti-fall in other embodiments of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose and advantages of the embodiments of the present invention easier to understand, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The detailed description of the embodiments of the present invention in the drawings below does not limit the scope of protection claimed by the present invention, but only represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0052] It should be noted that, if no conflict is constituted, the various technical features involved in the embodiments of the present invention described below can be combined with each other, and are all within the scope of protection of the present invention. In addition, although the functional modules are divided in the device or structural schematic diagram, and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order than the module division in the device, or in a different order than in the flow chart. In addition, the "first", "second", "third" and other similar expressions used herein do not limit the data and execution order, but are only for the purpose of convenience of explanation and to distinguish between the same or similar items with basically the same functions and effects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features.

[0053] Unless otherwise defined, the technical terms and scientific terms used in this specification have the same meanings as those generally understood by those of ordinary skill in the art of the present invention. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. It should be understood that the term "and / or" used in this specification includes any and all combinations of one or more of the related listed items.

[0054] See also Figure 1 , Figure 1 A schematic diagram schematically illustrates an application scenario of a robot anti-fall method provided by some embodiments of the present invention.

[0055] like Figure 1 As shown, the application scenario includes: a robot 100 and a pit 200. The robot 100 moves along a first direction N. During the movement of the robot 100 along the first direction N, the robot 100 may encounter a falling environment (for example, Figure 1For this purpose, the robot 100 needs to use various sensors equipped on itself to detect whether there is a falling environment on the ground environment in the driving direction (ie, the first direction N).

[0056] In some embodiments, the robot 100 is equipped with a laser radar ( Figure 1 (not shown) detect the ground environment in the first direction N to obtain point cloud data about the ground environment in the first direction N. Then determine whether there is a falling environment in the ground environment in the first direction N according to the obtained point cloud data. Obviously, Figure 1 In the application scenario shown, there is a falling environment (ie, a pit 200 ) on the ground environment in the first direction N.

[0057] In order to ensure that the task can be performed smoothly and reduce the impact of the falling environment on the task, the robot 100 determines the boundary of the falling environment after detecting the existence of the falling environment (pit 200), and thus determines the movement strategy of the robot 100 according to the boundary of the falling environment. For example, in some embodiments, when the robot 100 is far away from the boundary of the falling environment (i.e., the target distance between the robot 100 and the boundary of the falling environment is greater than the first distance threshold), it is detected that there is a falling environment in the ground environment in the first direction N, then it can be considered that the robot 100 has a safe distance from the boundary of the falling environment, and the robot 100 has enough time and distance to re-plan a new driving path according to the ground environment in the first direction N to avoid the encountered falling environment, so that the task can be performed smoothly according to the new driving path.

[0058] In other embodiments, the robot 100 detects the existence of a falling environment in the ground environment in the first direction N only when it is close to the boundary of the falling environment (i.e., the target distance between the robot 100 and the boundary of the falling environment is less than or equal to the first distance threshold), then it can be considered that the robot 100 does not have a safe distance from the boundary of the falling environment, and the robot 100 does not have enough time and distance to replan a new driving path according to the ground environment in the first direction N. The robot 100 decelerates until the distance between the robot 100 and the boundary of the falling environment is less than or equal to the second distance threshold, stops driving, and sends an alarm signal to notify management personnel or users to perform rescue.

[0059] It should be understood that Figure 1The application scenario shown schematically only shows a situation where the robot 100 encounters a pit 200 while traveling along the first direction N, and does not impose any limitation on any situation such as the structure, type, and number of the robot 100 in other application scenarios or embodiments. It also does not impose any limitation on the ground environment in the direction of travel of the robot 100. For example, the ground environment in the direction of travel of the robot 100 may also contain other types of falling environments, or may not contain falling environments.

[0060] To facilitate understanding of the robot anti-fall method provided by the embodiment of the present invention, the robot provided by the embodiment of the present invention is first introduced in detail.

[0061] See also Figure 2 , Figure 2 The structural diagram of the robot 100 provided in some embodiments of the present invention is schematically shown.

[0062] like Figure 2 As shown, the robot 100 includes at least one processor 110 and a memory 120 that are communicatively connected. Figure 2 In the example, the bus system 130 is connected to a processor. The various components in the robot 100 are coupled together through the bus system 130, and the bus system 130 is used to realize the connection and communication between these components. It is easy to understand that the bus system 130 includes a power bus, a control bus, and a status signal bus in addition to a data bus. However, for the sake of clarity and brevity, Figure 2 In the figure, various buses are labeled as bus system 130. It can be understood that Figure 2 The structure shown in the embodiment is only for illustration and does not impose any limitation on the structure of the robot 100. For example, the robot 100 may also include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations shown.

[0063] Specifically, the processor 110 is used to provide computing and control capabilities to control the robot 100 to perform corresponding tasks. For example, the robot 100 is controlled to perform any one of the robot anti-fall methods provided in the embodiments of the present invention, or to perform any one of the steps in any possible implementation of any one of the robot anti-fall methods provided in the embodiments of the present invention. Those skilled in the art will appreciate that the processor 110 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0064] The memory 120, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs, instructions and modules, for example, programs, instructions and modules corresponding to the robot anti-fall method in the embodiment of the present invention. In some embodiments, the memory 120 may include a program storage area and a data storage area, and the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the processor 110, etc. The processor 110 executes various functional applications and data processing of the robot 100 by running the non-transitory software programs, instructions and modules stored in the memory 120 to implement any robot anti-fall method provided in the embodiment of the present invention, or execute any possible implementation of any robot anti-fall method provided in the embodiment of the present invention. Among them, the memory 120 may include a high-speed random access memory, and may also include a non-transitory memory. For example, at least one disk storage device, a flash memory device or other non-transitory solid-state storage device. In some embodiments, the memory 120 may also include a memory remotely arranged relative to the processor 110, and these remotely arranged memories may be connected to the processor 110 through a communication network. It is understandable that examples of the above-mentioned communication network include but are not limited to the Internet, corporate intranet, local area network, mobile communication network and combinations thereof.

[0065] It is easy to understand that the implementation subject of any robot anti-falling method provided in the embodiments of the present invention can be any suitable type of robot with certain computing and control capabilities. In some possible implementations, any robot anti-falling method provided in the embodiments of the present invention can be implemented by a processor calling a computer program instruction stored in a memory.

[0066] It can be understood from the above that the robot anti-fall method provided in the embodiment of the present invention can be implemented by various suitable types of robots with certain computing and control capabilities, for example, it can be implemented and executed by the above robot 100. The robot anti-fall method provided in the embodiment of the present invention will be described in detail below in conjunction with the exemplary application and implementation of the robot provided in the embodiment of the present invention.

[0067] See also Figure 3 , Figure 3 The flowchart of the robot anti-fall method provided by some embodiments of the present invention is schematically shown.

[0068] Those skilled in the art will appreciate that the robot anti-fall method provided in the embodiment of the present invention can be applied to the above-mentioned robot (eg, robot 100). Specifically, the robot anti-fall method is executed by one or at least two processors of the robot.

[0069] like Figure 3 As shown, the robot anti-fall method includes but is not limited to the following steps S100-S300:

[0070] S100: Acquire point cloud data corresponding to the ground environment in the direction of the robot's travel, and determine whether there is a falling environment based on the point cloud data.

[0071] Specifically, the robot acquires point cloud data of the ground environment in the driving direction in real time through a laser radar, depth camera or stereo vision system equipped on itself and facing the driving direction. After acquiring the point cloud data, the point cloud data is filtered and denoised using methods such as voxel filtering and statistical filtering to remove noise and errors in the point cloud data acquisition process.

[0072] In order to determine whether there is a falling environment on the ground environment in the driving direction, it is necessary to extract ground environment information from the point cloud data. In some possible implementations, a plane fitting algorithm (such as RANSAC (random sampling consensus algorithm) is used to identify flat ground, or the ground, obstacles, and pits can be distinguished by the height difference of the point cloud data.

[0073] The robot detects the falling environment by analyzing the height change of the ground point cloud in the driving direction. If a significant height drop (such as a cliff or the edge of a staircase) is detected in the ground point cloud in the ground environment information in a certain area, it can be considered that there is a risk of falling in the area, and the area is marked as a falling environment.

[0074] In some embodiments, the falling environment can also be identified by detecting the ground edge. For example, a place where the ground point cloud decreases sharply is found in the point cloud as a mark of the falling environment boundary, thereby determining the falling environment.

[0075] See also Figure 4 , Figure 4 A sub-flow diagram of step S100 in the robot anti-falling method provided in some embodiments of the present invention is schematically shown.

[0076] like Figure 4 As shown, in some embodiments, point cloud data corresponding to the ground environment in the direction of the robot's travel is obtained, and whether there is a falling environment is determined according to the point cloud data, specifically including but not limited to the following steps S110-S130:

[0077] S110: Determine the ground point cloud of the ground where the robot is located according to the point cloud data, and fit a first plane.

[0078] Ground point cloud extraction is to determine the point cloud belonging to the ground where the robot is located from the point cloud data.

[0079] In some embodiments, the points in the point cloud data are screened according to the height (Z axis), that is, by comparing the height of each point in the point cloud data with a preset height threshold range, the ground point cloud and non-ground point cloud (such as obstacles, walls, etc.) of the ground where the robot is located are determined. Points whose heights are within the preset height threshold range are ground point clouds, and points whose heights are not within the preset height threshold range are non-ground point clouds.

[0080] After determining the ground point cloud where the robot is located, the least squares method is used to fit the ground point cloud, and the normal vector and other parameters of the fitting plane are calculated to obtain the ground plane where the robot is currently located, that is, the first plane is fitted.

[0081] In some embodiments, a plane fitting algorithm (for example, RANSAC (random sampling consensus algorithm)) may be used to fit a plane. That is, a plane model is selected, three points are randomly selected in the point cloud data to define a plane, a first distance from each point in the point cloud data to the plane is calculated, and points whose first distance is within a preset distance threshold range are taken as inliers, until a plane with the most inliers is found as the optimal fitting plane, that is, the first plane is fitted.

[0082] S120: In the point cloud data, if there are target point clouds whose height is lower than the preset height of the first plane, and the number of target point clouds is greater than or equal to the preset number, it is determined that a falling environment exists.

[0083] S130: In the point cloud data, if there is no target point cloud whose height is lower than the preset height of the first plane, and / or the number of target point clouds is less than a preset number, it is determined that there is no falling environment.

[0084] Specifically, after fitting the first plane, each point in the point cloud data is compared with the first plane, that is, each point in the point cloud data is traversed to calculate the height difference between each point and the first plane. If there is a target point cloud in the point cloud data whose height is lower than the preset height of the first plane (that is, the height is lower than the first plane, and the height difference with the first plane is greater than the preset height), and the number of target point clouds is greater than or equal to the preset number, it means that there is a falling risk in the area where the target point cloud is located, and the area where the target point cloud is located is a falling area, then it is determined that there is a falling environment in the ground environment in the driving direction.

[0085] In some embodiments, if there is no target point cloud in the point cloud data whose height is lower than the preset height of the first plane (that is, the height is lower than the first plane, and the height difference with the first plane is less than the preset height), or there is a target point cloud whose height is lower than the preset height of the first plane, but the number of target point clouds is less than the preset number, it means that there is no falling risk in the area where the target point cloud is located, and the area where the target point cloud is located is not a falling area, then it is determined that there is no falling environment in the ground environment in the driving direction.

[0086] It should be understood that those skilled in the art can set and determine the preset height and preset number according to engineering design requirements and historical experience data. For example, in some embodiments, the preset height can be 8 cm and the preset number can be 1000. In addition, the preset height and the preset number can be adjusted and changed according to actual application conditions, and the embodiments of the present invention do not impose any limitation on this.

[0087] S200: If a falling environment exists, determining a boundary of the falling environment.

[0088] After determining that there is a falling environment on the ground in the direction of the robot's travel, in order for the robot to accurately identify the dangerous falling area and make an appropriate travel strategy, it is necessary to first determine the boundary of the falling environment.

[0089] Specifically, the point cloud belonging to the falling environment is extracted from the point cloud data and used as a new point cloud set to represent the potential falling area. Then the convex hull algorithm is used to calculate the minimum convex hull of the falling environment point cloud, that is, the minimum polygonal boundary that can surround the falling environment point cloud, and the edge of the convex hull is used as the boundary of the falling environment. The convex hull algorithm is simple and efficient, and is suitable for situations where the boundary shape is regular.

[0090] In some embodiments, an α-shape algorithm may be used, that is, a suitable α parameter is selected to generate the boundary of the falling environment point cloud. The boundary generated by the α-shape algorithm is more in line with the actual environment contour and is suitable for concave and complex shapes. The α parameter determines the tightness of the boundary, and a smaller α value can generate a finer boundary.

[0091] In some embodiments, for situations where the point cloud density is high, edge detection technology can be used. For each point cloud, the distribution of the surrounding point clouds is analyzed. If there is a height difference exceeding a threshold in the neighborhood of a point cloud, the point cloud is located at the edge of the falling environment. All point clouds located at the edge of the falling environment are extracted to form a boundary point cloud. The edge detection method is suitable for more complex environments and can accurately identify irregular boundaries.

[0092] In some embodiments, determining the boundary of the falling environment specifically includes but is not limited to the following steps S210:

[0093] S210: Determine a boundary area between the ground point cloud and the target point cloud as a boundary of the falling environment.

[0094] Specifically, a plane fitting method is used to extract the ground point cloud where the robot is currently located, and a target point cloud below a preset height of the first plane is extracted through height threshold screening.

[0095] In some implementations, a convex hull algorithm is used to calculate the boundary of the target point cloud to obtain a polygon that can enclose the target point cloud, and the convex hull boundary represents the edge of the ground.

[0096] In some implementations, a convex hull algorithm or an α-shape algorithm is used to generate a boundary of the target point cloud to obtain an outer contour of the falling environment.

[0097] After obtaining the boundary area of ​​the ground point cloud and the boundary area of ​​the target point cloud, the boundary of the ground point cloud and the boundary of the target point cloud are merged to form a polygonal boundary area containing both, and the boundary of the falling environment is obtained. Among them, the merged boundary area represents a dangerous falling environment that the robot should avoid.

[0098] In some embodiments, for complex environments, it is necessary to compare the ground point cloud boundary and the target point cloud boundary and adjust the boundary according to the actual terrain conditions. For example, if the ground point cloud boundary intersects with the target point cloud boundary, the intersection area can be used as part of the fall boundary.

[0099] In some other embodiments, determining the boundary of the falling environment specifically includes but is not limited to the following steps S220-S240:

[0100] S220: Fitting a second plane according to the target point cloud.

[0101] S230: Determine a preset value according to a height difference between the first plane and the second plane.

[0102] S240: Traverse each point cloud in the point cloud data, and determine a point cloud whose height difference with an adjacent point cloud is greater than a preset value as a point cloud corresponding to the boundary of the falling environment.

[0103] Exemplarily, after determining that there is a target point cloud that meets the requirements (i.e., a target point cloud whose height is lower than a preset height of the first plane, and the number of target point clouds is greater than or equal to a preset number), the corresponding target point cloud is extracted. Then a plane fitting algorithm (e.g., RANSAC (Random Sampling Consensus Algorithm) is used to fit the plane. That is, a plane model is selected, three points are randomly selected from the target point cloud to define a plane, and a second distance from each point in the target point cloud to the plane is calculated, and points whose second distance is within a preset distance threshold range are taken as inliers, until a plane with the most inliers is found as the optimal fitting plane, that is, the second plane is fitted.

[0104] The preset value is determined according to the height difference between the first plane and the second plane, so that the robot can effectively identify the danger level of the falling environment and make corresponding travel decisions. Optionally, the preset value is the height difference between the first plane and the second plane. Alternatively, the preset value can be a value that is smaller than the height difference between the first plane and the second plane by a preset multiple. After fitting the second plane, the height difference between the first plane and the second plane is calculated using the point clouds in the first plane and the second plane. The height difference is used to measure the relative height difference between the two planes.

[0105] After determining the preset value, traverse each point cloud in the point cloud data, and calculate the height difference between each point cloud and the adjacent point cloud. If the height difference between a point cloud and the adjacent point cloud is greater than the preset value, the point cloud is the boundary point cloud of the falling environment, and the point cloud is determined as the point cloud corresponding to the boundary of the falling environment. The boundary point clouds of all falling environments are extracted as the point clouds corresponding to the boundary of the falling environment, thereby determining the boundary of the falling environment. It can be understood that adjacent point clouds refer to the point clouds with the closest distance in three-dimensional space. In some embodiments, the adjacent point clouds of each point cloud can be determined using the K-Nearest Neighbors (KNN) or Fixed Radius Neighbors method.

[0106] S300: Determine the robot's moving strategy according to the boundary of the falling environment.

[0107] Specifically, the boundary information of the falling environment is analyzed to determine the shape and distribution of the boundary, the geometric characteristics of the boundary (such as the continuity, size and direction of the boundary), and whether the falling environment belongs to a larger falling area (such as a cliff, steps, etc.) or a smaller falling area (such as a small step or ramp). Then the distance between the current position of the robot and the boundary is calculated to determine the relative position of the robot and the boundary.

[0108] In some embodiments, if the boundary of the falling environment is located on the robot's driving path and the depth of the boundary exceeds the threshold range that the robot can safely cross, the robot uses a path planning algorithm (such as an A* algorithm, a Dijkstra algorithm, or a graph-based path search algorithm, etc.) to re-plan a driving path that avoids the falling environment based on existing map information and boundary data.

[0109] In some embodiments, if there is no feasible safe path around the boundary of the falling environment, or the robot cannot bypass the boundary of the falling environment, the robot stops moving and sends an alarm signal to notify the operator or trigger safety measures, such as waiting for new instructions or taking automatic recovery measures.

[0110] In some embodiments, if the boundary of the falling environment belongs to a smaller falling area (such as a small step or a gentle slope), and the height difference is within the threshold range that the robot can safely cross, the robot reduces its driving speed and carefully drives through the boundary area to avoid the robot overturning or falling due to the height difference.

[0111] See also Figure 5 , Figure 5 A sub-flow diagram of step S300 in the robot anti-fall method provided in some embodiments of the present invention is schematically shown.

[0112] like Figure 5 As shown, in some embodiments, the robot's travel strategy is determined according to the boundary of the falling environment, specifically including but not limited to the following steps S310-S340:

[0113] S310: Determine whether the robot can travel around the boundary of the falling environment according to the boundary of the falling environment and the traversable area of ​​the electronic map.

[0114] It is understandable that to determine whether the robot can travel around the boundary of the falling environment, it is necessary to combine the detected boundary information of the falling environment and the traversable area in the electronic map, and comprehensively consider the robot's own travel ability, the position and shape of the falling environment boundary, and the traversability of the surrounding environment.

[0115] Among them, the electronic map is a map or a predefined map generated in real time based on environmental scanning, which is used for robot navigation and positioning, path planning and environmental perception. The electronic map contains obstacle information, the passability of different areas (such as roads, narrow passages, etc.) and the location of other objects in the environment. The electronic map marks the places where the robot can travel safely, that is, the passable areas of the electronic map. It is easy to understand that the electronic map in the embodiment of the present invention can be any suitable type of map such as a grid map, a point cloud map, a semantic map, and a landmark map.

[0116] Specifically, the detected boundary of the falling environment is analyzed, for example, the geometric shape, length, width, detour space, distance from the robot, etc. of the boundary of the falling environment, and based on the passable area around the robot obtained from the electronic map, the boundary point cloud of the falling environment is projected onto the electronic map, and the relative position of the boundary area and the passable area is determined, thereby determining whether the robot can travel around the boundary of the falling environment. If the passable area is wide enough and there are no other obstacles blocking it, the robot may be able to bypass the boundary of the falling environment and continue to move. In some embodiments, at least one driving path connecting the current position of the robot and the target position for performing the task can be planned in the electronic map, and the at least one driving path will not pass through the falling environment and its boundary, then it is considered that the robot can travel around the boundary of the falling environment.

[0117] S320: If one side of the boundary of the falling environment is a passable area, and the minimum width of the passable area is greater than a preset width value, it is determined that the robot can travel around the boundary of the falling environment.

[0118] In some embodiments, after the boundary point cloud of the falling environment is projected onto the electronic map, if one side of the boundary of the falling environment is a passable area of ​​the electronic map, the narrowest part of the passable area, i.e., the minimum width, is measured and obtained. If one side of the boundary of the falling environment is a passable area, and the minimum width of the passable area is greater than a preset width value, indicating that the robot can bypass the boundary of the falling environment, it is determined that the robot can travel around the boundary of the falling environment.

[0119] It is understandable that the preset width value should be at least greater than the actual width of the robot, and should also be greater than the actual width of the robot by a certain safety width to ensure that the robot can safely bypass the boundary of the falling environment. Those skilled in the art can set and determine the preset width value based on factors such as the actual width of the robot, the complexity of the falling environment, and the driving ability of the robot.

[0120] S330: If the robot can travel around the boundary of the falling environment, then update part of the grids in the traversable area of ​​the electronic map along the boundary of the falling environment as obstacle areas, so that the robot stays away from the boundary of the falling environment when planning the path.

[0121] Specifically, if the robot can travel around the boundary of the falling environment, that is, the robot can bypass the boundary of the falling environment, then the vicinity of the boundary of the falling environment is marked on the electronic map along the boundary of the falling environment as a boundary area that needs to be avoided during path planning.

[0122] The boundary point cloud of the falling environment is projected onto the grid of the electronic map, and each boundary point cloud corresponds to one or more grids in the electronic map. In the vicinity of the grid corresponding to the boundary point cloud, the grid within the preset distance range is selected for marking, that is, the grid within the preset distance range corresponding to the boundary point cloud is selected in the passable area of ​​the electronic map. Obviously, the preset range can be determined according to factors such as the size of the robot, the degree of danger of the falling environment, and the required safety distance.

[0123] Then, part of the grids in the marked passable area are updated to obstacle grids as obstacle areas of the electronic map, so that when planning a path, the obstacle grids in the updated electronic map are regarded as inaccessible areas to plan a driving path away from the boundary of the falling environment.

[0124] S340: Planning a path according to the updated electronic map to travel around the boundary of the falling environment.

[0125] Specifically, after updating the grid information marked as obstacles to the electronic map to obtain an updated electronic map, in the updated electronic map, the robot plans a safe bypass path through the boundary of the falling environment based on the current position, the target position of the task execution, the traversable area and the boundary information of the falling environment, etc., using a path planning algorithm (for example, based on the RRT (rapidly explored random trees) algorithm, the Dijkstra algorithm and the A* algorithm, etc.), so as to bypass the boundary of the falling environment according to the bypass path and prevent the robot from entering the falling environment.

[0126] In some embodiments, the robot anti-fall method further includes but is not limited to the following steps S350-S370:

[0127] S350: If one side of the boundary of the falling environment is an obstacle area, and / or the minimum width of the passable area on one side of the boundary of the falling environment is less than or equal to a preset width value, it is determined that the robot cannot travel around the boundary of the falling environment.

[0128] In some embodiments, after projecting the boundary point cloud of the falling environment onto the electronic map, if one side of the boundary of the falling environment is an obstacle area of ​​the electronic map, or if one side of the boundary of the falling environment is a passable area, but the minimum width of the passable area is less than or equal to a preset width value, indicating that the robot cannot bypass the boundary of the falling environment, it is determined that the robot cannot travel around the boundary of the falling environment.

[0129] S360: If the robot cannot travel around the boundary of the falling environment, and the distance between the robot and the boundary of the falling environment is greater than or equal to the first preset distance, the speed of the robot is reduced to the first preset speed, and the robot moves forward at the first preset speed.

[0130] Specifically, if the robot cannot go around the boundary of the falling environment, that is, the robot cannot go around the boundary of the falling environment, then the distance between the robot and the boundary of the falling environment is measured by a sensor (for example, a laser radar, an ultrasonic sensor, or a camera, etc.). If the distance between the robot and the boundary of the falling environment is greater than or equal to the first preset distance, it means that the robot is far from the boundary of the falling environment and does not need to stop moving forward. The speed of the robot is reduced to the first preset speed, and the robot continues to move forward along the current driving direction at the first preset speed.

[0131] It should be understood that those skilled in the art can set and determine the first preset distance and the first preset speed according to factors such as the driving speed and driving ability of the robot, the complexity of the falling environment, etc. For example, in some embodiments, the first preset distance can be 30 cm, 50 cm, or any other suitable distance value, and the first preset speed can be 0.2 m / s, 0.3 m / s, or any other suitable speed value.

[0132] S370: If the distance between the robot and the boundary of the falling environment is less than the first preset distance, the robot stops moving forward.

[0133] Specifically, if the distance between the robot and the boundary of the falling environment is less than the first preset distance, it means that the robot is very close to the boundary of the falling environment and there is a risk of driving into the falling environment, and it is necessary to stop driving immediately to avoid falling or entering a dangerous falling environment. After receiving the stop driving instruction, the robot control system turns off the driving device, motor and other working components and stops driving.

[0134] In some embodiments, the drop environment includes pits and multiple steps, for example, Figure 1 The pit 200 and Figure 6 The multi-step stairs 300 shown, that is, the types of the falling environment include pits and multi-step stairs, etc. In some embodiments of the present invention, the type of the falling environment can also be determined according to the target point cloud, and then the robot's moving strategy can be determined according to the type of the falling environment.

[0135] In some embodiments, the robot anti-fall method further includes but is not limited to the following steps S400-S500:

[0136] S400: Determine the type of the falling environment according to the target point cloud, wherein the type of the falling environment includes a pit and a multi-step.

[0137] Among them, the characteristic of a pit is that the height of the point cloud in a region suddenly drops relative to the ground point cloud, forming a continuous concave area that is partially lower than the surrounding plane. The height variation of the concave area is small, uniform and continuous. The boundary shape of the pit is relatively smooth or elliptical, and the whole is relatively continuous. The characteristic of a multi-step is that there are multiple height drops in the point cloud height in the region, and there is a relatively flat area between each height drop area, forming a stepped structure. The boundary shape of a multi-step is manifested as multiple parallel straight lines, or a stepped structure, with multiple obvious dividing lines.

[0138] Exemplarily, after determining that there is a target point cloud whose height is lower than the preset height of the first plane in the point cloud data, the height of each point of the target point cloud is calculated to obtain the height difference of different points. And the boundary of the falling environment is determined by using a convex hull algorithm, an α-shape algorithm, etc. to determine the geometric shape of the falling environment.

[0139] Then, the type of the falling environment is determined based on the height difference and height change trend of each point of the target point cloud, the geometric shape of the falling environment, etc. The types of falling environments include pits and multiple steps. In some embodiments, if it is detected that the height of the point cloud in a certain area suddenly drops relative to the ground point cloud, and there is no further sudden drop in height after the change (that is, the height in the area remains on a lower plane), and the boundary contour of the area is smooth or irregular, and the whole appears as a continuous concave area, then the area can be determined to be a pit.

[0140] In some embodiments, if there are multiple obvious height drops in the target point cloud, and there are small flat areas between these height drops (that is, there are flat point cloud areas connecting the height drop areas), and the boundary of the area appears as multiple parallel straight lines or a stepped structure with clear dividing lines and multiple planes at different heights, then the area can be determined to be a multi-step step.

[0141] S500: Determine the robot's travel strategy according to the type of the falling environment.

[0142] In some embodiments, if the type of falling environment is detected to be a pit, since entering the pit may cause the robot to fall or be damaged, the robot avoids entering the pit area and needs to re-plan the first driving path that avoids the pit area, so as to travel around the edge of the pit according to the first driving path. If the path planning cannot bypass the pit, or the pit area is too large to be avoided, the robot stops driving forward. In some embodiments, for some pits with a small slope that the robot will not fall when driving through and will not damage the robot, the robot may not re-plan the driving path to avoid the pit area, but instead travel across the pit area according to the original driving path, thereby not affecting the execution of the task and ensuring the efficiency of task execution.

[0143] In some embodiments, if the type of the falling environment is detected to be multi-step stairs, since multi-step stairs may be an area that the robot can pass through, depending on the height of the stairs and the robot's climbing ability, therefore, for multi-step stairs, if the robot has the ability to go up and down stairs, and the height and number of the stairs are within the robot's climbing ability, then the robot can climb through the multi-step stairs under safe conditions.

[0144] If the steps are too high or too many, that is, the height and / or number of the steps exceed the robot's climbing ability, especially for wheeled mobile robots, the robot needs to re-plan the path to bypass the multi-step area. When the steps are too high or too many and a safe bypass path cannot be planned, the robot stops moving forward and waits for further instructions or operations.

[0145] In some embodiments, according to the type of the falling environment, the robot's travel strategy is determined, which specifically includes but is not limited to the following steps S510-S530:

[0146] S510: If the type of the falling environment is a pit, detect the slope of the connecting slope between the ground where the robot is located and the bottom of the pit.

[0147] S520: When the slope of the connecting slope is less than or equal to the preset slope, it is determined that the robot can cross the pit and move forward.

[0148] S530: When the slope of the connecting slope is greater than the preset slope, it is determined that the robot cannot cross the pit, and the path is replanned to avoid the pit.

[0149] In some embodiments, if the type of drop environment is a pit (e.g., Figure 1The pit 200 shown in the figure is then acquired through a laser radar, a depth camera or other sensors to collect and obtain the point cloud data of the slope area from the current position of the robot to the bottom of the pit, and extract the point cloud of the slope area, that is, extract the point cloud between the plane where the robot is located and the bottom of the pit. According to the extracted slope point cloud data, the slope plane is fitted using the least squares method or other geometric fitting algorithms, and the slope of the slope is calculated using the normal vector of the fitted slope plane, that is, the slope of the connecting slope between the ground where the robot is located and the bottom of the pit is obtained. Obviously, the slope of the connecting slope refers to the ratio of the vertical height of the slope to the horizontal width. The greater the slope, the steeper the slope. On the contrary, the smaller the slope, the gentler the slope. Let the angle between the normal vector of the slope plane and the horizontal plane be θ, then the slope of the connecting slope is tan(π / 2-).

[0150] Then compare the slope of the connecting slope with the preset slope. If the slope of the connecting slope is less than or equal to the preset slope, it means that the slope of the pit is relatively gentle and the robot can safely drive through the pit. Then it is determined that the robot can cross the pit. The robot does not need to replan the first driving path to avoid the pit. It continues to move forward according to the original driving path and drives smoothly across the pit area at a suitable speed. If there is also a slope on the other side of the pit, the robot should adopt a climbing strategy to safely climb the edge of the pit through appropriate speed and power distribution to complete the crossing.

[0151] In some embodiments, if the slope of the connecting slope is greater than the preset slope, it means that the slope of the pit is too steep and the robot cannot safely drive through the pit, then it is determined that the robot cannot cross the pit. In this case, the robot should choose to bypass, that is, the robot needs to use the path planning algorithm to re-plan a new driving path to bypass the pit, and bypass the pit according to the new driving path to avoid the pit area that cannot be crossed. If bypassing the pit is not feasible, for example, the pit area is too large to plan a new driving path to bypass the pit, the robot stops driving and waits for further instructions or operations.

[0152] It is easy to understand that those skilled in the art can set and determine the preset slope based on factors such as the robot's maximum climbing ability and the complexity of the slope. For example, the preset slope can be any suitable slope value, and the embodiments of the present invention do not impose any limitation on this.

[0153] In some embodiments, since the robot does not have the ability to go up and down stairs, when encountering a multi-step environment, the multi-step stairs are marked as obstacles in the electronic map.

[0154] In some embodiments, the robot anti-fall method further includes but is not limited to the following steps S540:

[0155] S540: If the type of the falling environment is multiple steps, the multiple steps are marked as obstacles on the electronic map.

[0156] Specifically, when it is detected that the type of falling environment is a multi-step staircase (for example Figure 6 When the robot encounters a multi-step staircase 300 shown in FIG. 1 , the robot regards the multi-step staircase as an obstacle that cannot be safely crossed. To ensure that the robot does not choose to cross the multi-step staircase area during path planning, the multi-step staircase needs to be marked as an obstacle and updated to the electronic map. Specifically, the boundary points of the multi-step staircase are determined, and according to the boundary points of the multi-step staircase, the entire multi-step staircase area is marked as an obstacle area on the electronic map through the boundary points, that is, the grid occupied by the multi-step staircase area is updated to an obstacle grid in the electronic map, so that the area marked as an obstacle can be identified and avoided during subsequent path planning.

[0157] To summarize, the robot anti-fall method provided in an embodiment of the present invention is applied to a robot, and the method includes: obtaining point cloud data corresponding to the ground environment in the direction of the robot's travel, and determining whether there is a falling environment based on the point cloud data; if there is a falling environment, determining the boundary of the falling environment; and determining the robot's moving strategy based on the boundary of the falling environment.

[0158] In the present invention, after determining the existence of a falling environment based on the acquired ground point cloud data, the robot's travel strategy is determined according to the boundary of the falling environment. The method flexibly adopts different travel strategies according to the boundary of the falling environment, which can reduce the impact of the falling environment on the normal operation of the robot, thereby reducing the impact on the robot's task execution and ensuring the efficiency of task execution.

[0159] An embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions executable by a processor are stored. When the computer program instructions are called by the processor, the processor executes any one of the robot anti-falling methods provided by the embodiments of the present invention, or executes the steps in any one of the possible implementation methods of any one of the robot anti-falling methods provided by the embodiments of the present invention.

[0160] In some embodiments, the storage medium may be a flash memory, a hard disk, an optical disk, a register, a magnetic surface storage, a removable disk, a CD-ROM, a random access memory (RAM), a read-only memory (ROM), an electrically programmable ROM, and an electrically erasable programmable ROM, or any other form of storage medium known in the technical field, or various devices including one or any combination of the above storage media.

[0161] In some embodiments, computer program instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0162] As an example, computer program instructions may, but do not necessarily, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0163] As an example, the computer program instructions may be deployed to be executed on a computing device (including devices such as smart terminals and servers), or on multiple computing devices located at one location, or on multiple computing devices distributed at multiple locations and interconnected by a communication network. It is easy to understand that all or part of the steps of the method described in the embodiments of the present invention may be directly implemented using electronic hardware or processor executable computer program instructions, or a combination of the two.

[0164] Those skilled in the art will appreciate that the embodiments provided by the present invention are only illustrative, and the order of writing the steps in the method of the embodiment does not mean a strict execution order and does not constitute any limitation on the implementation process. The order can be adjusted, merged and deleted according to actual needs, and the modules or sub-modules, units or sub-units in the device or system of the embodiment can be merged, divided and deleted according to actual needs. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0165] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general hardware platform, and of course can also be implemented by hardware. Those skilled in the art can understand that all or part of the processes in the above embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0166] It should be noted that the above embodiments are intended to illustrate the technical concept and features of the present invention, and their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it accordingly, and cannot be used to limit the scope of protection of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiments can be implemented according to the technical solutions recorded in the embodiments of the present invention, or some of the technical features can be equivalently replaced. It can be understood that these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and should be regarded as equal changes and modifications based on the embodiments of the present invention, and should all fall within the scope of the claims of the present invention.

Claims

1. A robot anti-fall method, characterized in that: Applied to a robot, the method comprises: Acquire point cloud data corresponding to the ground environment in the direction of robot travel, and determine whether there is a falling environment according to the point cloud data; If the falling environment exists, determining the boundary of the falling environment; A moving strategy of the robot is determined according to the boundary of the falling environment.

2. The method according to claim 1, characterized in that The step of obtaining point cloud data corresponding to the ground environment in the direction of the robot's travel, and determining whether there is a falling environment according to the point cloud data, comprises: Determine the ground point cloud of the ground where the robot is located according to the point cloud data, and fit a first plane; If, in the point cloud data, there are target point clouds whose height is lower than the preset height of the first plane, and the number of target point clouds is greater than or equal to the preset number, it is determined that a falling environment exists; In the point cloud data, if there is no target point cloud whose height is lower than the preset height of the first plane, and / or the number of target point clouds is less than a preset number, it is determined that there is no falling environment.

3. The method according to claim 2, characterized in that Determining the boundary of the falling environment includes: Determine a boundary area between the ground point cloud and the target point cloud as a boundary of a falling environment; or, Fitting a second plane according to the target point cloud; Determining a preset value according to a height difference between the first plane and the second plane; Each point cloud in the point cloud data is traversed, and a point cloud having a height difference with an adjacent point cloud greater than a preset value is determined as a point cloud corresponding to a boundary of the falling environment.

4. The method according to claim 2 or 3, characterized in that: Determining the robot's travel strategy according to the boundary of the falling environment includes: Determining whether the robot can travel around the boundary of the falling environment according to the boundary of the falling environment and the traversable area of ​​the electronic map; If one side of the boundary of the falling environment is a passable area, and the minimum width of the passable area is greater than a preset width value, it is determined that the robot can travel around the boundary of the falling environment; If the robot can travel around the boundary of the falling environment, then updating part of the grids of the traversable area of ​​the electronic map along the boundary of the falling environment as obstacle areas, so that the robot stays away from the boundary of the falling environment when planning a path; A path is planned according to the updated electronic map to travel around the boundary of the falling environment.

5. The method according to claim 4, characterized in that The method further comprises: If one side of the boundary of the falling environment is an obstacle area, and / or the minimum width of the passable area on one side of the boundary of the falling environment is less than or equal to a preset width value, it is determined that the robot cannot travel around the boundary of the falling environment; If the robot cannot travel around the boundary of the falling environment, and the distance between the robot and the boundary of the falling environment is greater than or equal to a first preset distance, reducing the speed of the robot to the first preset speed, and driving forward at the first preset speed; If the distance between the robot and the boundary of the falling environment is less than the first preset distance, the robot stops moving forward.

6. The method according to claim 5, characterized in that The method further comprises: Determining the type of the falling environment according to the target point cloud, wherein the type of the falling environment includes a pit and a multi-step; The moving strategy of the robot is determined according to the type of the falling environment.

7. The method according to claim 6, characterized in that Determining the robot's travel strategy according to the type of the falling environment includes: If the type of the falling environment is a pit, detecting the slope of the connecting slope between the ground where the robot is located and the bottom of the pit; When the slope of the connecting slope is less than or equal to a preset slope, it is determined that the robot can cross the pit and move forward; When the slope of the connecting slope is greater than a preset slope, it is determined that the robot cannot cross the pit, and the path is replanned to avoid the pit.

8. The method according to claim 6, characterized in that The method further comprises: If the type of the falling environment is multiple steps, the multiple steps are marked as obstacles on the electronic map.

9. A robot, characterized in that: include: A processor and a memory communicatively connected to the processor; The memory stores computer program instructions executable by the processor, and when the computer program instructions are called by the processor, the processor executes the robot anti-fall method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions executable by a processor, and when the computer program instructions are called by the processor, the processor executes the robot anti-fall method according to any one of claims 1 to 8.

Images