CN115437384A

CN115437384A - Obstacle avoidance method, equipment and medium for mobile robot

Info

Publication number: CN115437384A
Application number: CN202211286236.7A
Authority: CN
Inventors: 李静; 马辰; 程瑶
Original assignee: Shandong New Generation Information Industry Technology Research Institute Co Ltd
Current assignee: Shandong New Generation Information Industry Technology Research Institute Co Ltd
Priority date: 2022-10-20
Filing date: 2022-10-20
Publication date: 2022-12-06

Abstract

The application discloses a method, equipment and medium for avoiding obstacles of a mobile robot, wherein the method comprises the following steps: in the moving process of the mobile robot, acquiring three-dimensional point cloud data of an obstacle through a plurality of obstacle detection devices arranged on the mobile robot; fusing the three-dimensional point cloud data respectively obtained by the plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data to a grid map to determine coordinates of obstacles in the grid map; acquiring a first moving path corresponding to the mobile robot, and judging whether the first moving path is overlapped with the coordinates of the obstacle or not; if so, fitting to obtain a two-dimensional contour line of the barrier, and determining the maximum span of the two-dimensional contour line; and determining an obstacle avoidance strategy of the mobile robot according to the maximum span and the vertical coordinate value of the obstacle.

Description

Obstacle avoidance method, equipment and medium for mobile robot

Technical Field

The application relates to the technical field of intelligent robots, in particular to an obstacle avoidance method, equipment and a medium for a mobile robot.

Background

The mobile robot is a complete or intermittent self-moving operation robot, is generally suitable for executing tasks such as self-inspection, explanation guidance, automatic cleaning and self-delivery, and is commonly a campus express delivery vehicle, a restaurant dish-serving robot and the like. However, when the robot performs a task, due to the influence of complex obstacles such as chairs and glass, the robot may collide due to the fact that no obstacle is detected, thereby causing task interruption and even damaging the machine.

However, the existing mobile robot can only recognize whether an obstacle exists in front of the robot, but cannot further recognize specific characteristics of the obstacle, so that it is difficult to perform corresponding obstacle avoidance operations for obstacles with different characteristics.

Disclosure of Invention

In order to solve the above problem, the present application provides an obstacle avoidance method for a mobile robot, including:

in the moving process of the mobile robot, acquiring three-dimensional point cloud data of an obstacle through a plurality of obstacle detection devices arranged on the mobile robot;

fusing the three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data to a grid map to determine coordinates of obstacles in the grid map;

acquiring a first moving path corresponding to the mobile robot, and judging whether the first moving path is overlapped with the coordinates of the obstacle;

if so, fitting to obtain a two-dimensional contour line of the obstacle, and determining the maximum span of the two-dimensional contour line;

and determining an obstacle avoidance strategy of the mobile robot according to the maximum span and the vertical coordinate value of the obstacle.

In one implementation manner of the present application, after determining whether the first moving path coincides with the coordinates of the obstacle, the method further includes:

determining a vertical coordinate value of the obstacle and a coordinate range where the vertical coordinate value is located; the vertical coordinate value is a vertical coordinate value corresponding to the highest point of the obstacle, and the coordinate range at least includes three ranges, namely a first range higher than a first preset threshold, a second range lower than the first preset threshold and higher than a second preset threshold, and a third range lower than the second preset threshold;

and under the condition that the vertical coordinate value is in the third range, determining that the moving obstacle does not need to be avoided.

In an implementation manner of the present application, determining the maximum span of the two-dimensional contour line specifically includes:

generating a convex polygon area corresponding to the two-dimensional contour line through a preset algorithm;

determining the moving direction and the current coordinate of the mobile robot, taking the current coordinate as a vertex, and making an extension line along the moving direction until the extension line is intersected with the convex polygon area;

generating a plurality of connecting lines parallel to the extension lines based on points on the convex polygon area;

and traversing the connecting lines in sequence, selecting the connecting line with the largest length from the connecting lines as a target connecting line, and taking the length of the target connecting line as the maximum span of the two-dimensional contour line.

In an implementation manner of the present application, determining an obstacle avoidance policy of the mobile robot according to the maximum span and a vertical coordinate value of the obstacle specifically includes:

determining a span threshold that the mobile robot can span and comparing the span threshold to the maximum span;

when the span threshold is smaller than the maximum span or the vertical coordinate value is in the first range, determining that an obstacle avoidance strategy of the mobile robot is a circumvention strategy;

and determining that the obstacle avoidance strategy of the mobile robot is crossing under the condition that the span threshold is larger than the maximum span and the vertical coordinate value is in the second range.

In an implementation manner of the present application, the three-dimensional point cloud data respectively obtained by the plurality of obstacle detection devices are fused to obtain fused obstacle point cloud data, which specifically includes:

filtering the three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices, and respectively inputting the filtered three-dimensional point cloud data to a first feature extraction network to which the corresponding obstacle detection device belongs to so as to obtain first feature data respectively corresponding to the obstacle detection devices;

respectively inputting the first feature data corresponding to the obstacle detection devices to second feature extraction networks to which the obstacle detection devices belong so as to obtain second feature data corresponding to the obstacle detection devices;

and fusing the first characteristic data and the second characteristic data to obtain fused obstacle point cloud data.

In one implementation manner of the present application, after determining that the obstacle avoidance policy of the mobile robot is bypassing, the method further includes:

acquiring a current coordinate of the mobile robot and a target coordinate corresponding to a destination;

determining a connection path between the current coordinate and the destination coordinate and an obstacle in a rectangular area with the connection path as a diagonal line on the grid map;

acquiring a target obstacle which is intersected with the connection path and is closest to the mobile robot from the obstacles in the rectangular area;

and determining a boundary point set corresponding to the target obstacle, and generating a second moving path capable of bypassing the target obstacle according to the boundary point set.

In one implementation of the present application, after generating the second movement path that can bypass the target obstacle, the method further includes:

respectively determining the path lengths of a plurality of second moving paths under the condition that the plurality of second moving paths exist, and arranging the plurality of second moving paths according to the sequence of the path lengths from large to small to obtain a corresponding first moving path sequence;

determining the coordinate position of the mobile robot when encountering the obstacle again in the process of moving along the second movement paths, and determining the lengths of the movement paths respectively corresponding to the mobile robot moving along the second movement paths according to the coordinate position;

arranging the plurality of second moving paths according to the sequence of the lengths of the moving paths from small to large to obtain a corresponding second moving path sequence;

determining a first priority value corresponding to each second moving path in the first moving path sequence and a second priority value corresponding to each second moving path in the second moving path sequence;

and performing weighted summation on the first priority value and the second priority value to determine a final moving path of the mobile robot according to the result of the weighted summation.

In one implementation manner of the application, the obstacle detection device is sequentially provided with a depth camera, a laser radar and an ultrasonic sensor from high to low in detection height;

the depth camera is configured to detect obstacles in the first range, the lidar is configured to detect obstacles in the second range, and the ultrasonic sensor is configured to detect obstacles in the third range.

The embodiment of the application provides a mobile robot's obstacle avoidance equipment, its characterized in that, equipment includes: at least one processor;

and a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to cause the at least one processor to:

in the moving process of the mobile robot, acquiring three-dimensional point cloud data of obstacles through a plurality of obstacle detection devices arranged on the mobile robot;

An embodiment of the present application provides a non-volatile computer storage medium, which stores computer-executable instructions, and is characterized in that the computer-executable instructions are configured to:

fusing the three-dimensional point cloud data respectively obtained by the plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data onto a grid map to determine the coordinates of the obstacles in the grid map;

The obstacle avoidance method for the mobile robot has the following beneficial effects that:

the method comprises the steps of obtaining obstacle point cloud data by fusing three-dimensional point cloud data obtained by a plurality of obstacle detection devices, and mapping the obstacle point cloud data to a grid map so as to display an obstacle in a two-dimensional mode, so that whether the mobile robot has a collision risk can be more intuitively identified; under the condition that the mobile robot has collision risks, the maximum span of the obstacle is determined through the two-dimensional contour line of the obstacle, and an obstacle avoidance strategy of the mobile robot is further determined according to the maximum span and the vertical coordinate value of the obstacle, so that the accuracy of obstacle avoidance operation is improved, and the method is more targeted.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is a schematic flowchart of an obstacle avoidance method for a mobile robot according to an embodiment of the present disclosure;

FIG. 2 is a schematic view of an obstacle according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of an obstacle avoidance device of a mobile robot according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

As shown in fig. 1, an obstacle avoidance method for a mobile robot according to an embodiment of the present application includes:

s101: in the moving process of the mobile robot, three-dimensional point cloud data of obstacles are acquired through a plurality of obstacle detection devices arranged on the mobile robot.

In this application embodiment, be equipped with a plurality of obstacle detection equipment on the mobile robot, and according to the detection height by high depth camera, laser radar and ultrasonic sensor in proper order to low order, regional height and the angle diverse that different obstacle detection equipment can be surveyed have guaranteed that the mobile robot can realize omnidirectional obstacle detection at the removal in-process.

In the moving process of the mobile robot, the obstacle information in front of the mobile robot can be acquired in real time through the plurality of obstacle detection devices. It should be noted that the obstacle information exists in the form of three-dimensional point cloud data, and when acquiring data, the distance measurement information acquired by the ultrasonic sensor needs to be further converted into the form of three-dimensional point cloud.

S102: and fusing the three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data to a grid map to determine the coordinates of the obstacles in the grid map.

After the three-dimensional point cloud data of the obstacle is acquired, the obstacle detected by a single obstacle detection device is limited by the detection height or the detection angle, and the point cloud data of the obstacle acquired by a plurality of obstacle detection devices respectively needs to be fused to acquire more comprehensive information of the obstacle.

Specifically, the three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices is filtered to remove obstacle information repeatedly detected by each obstacle detection device, so that the subsequent data processing pressure is reduced, and the obstacle avoidance efficiency is improved. And respectively inputting the three-dimensional point cloud data after filtering into a first feature extraction network to which the obstacle detection device corresponding to the three-dimensional point cloud data belongs to obtain first feature data corresponding to the obstacle detection device respectively, wherein the extracted first feature data are shallow features of the obstacle, and the first feature data also need to be output to a second feature network to which the obstacle detection device corresponding to the obstacle detection device belongs to obtain corresponding second feature data, and the extracted second feature data are deep features. After the first characteristic data and the second characteristic data of the obstacle are obtained respectively, the first characteristic data and the second characteristic data need to be fused to obtain fused obstacle point cloud data. And after the fused obstacle point cloud data are obtained, mapping the obstacle point cloud data to a preset grid map so as to further determine the coordinates corresponding to the obstacles.

S103: and acquiring a first moving path corresponding to the mobile robot, and judging whether the first moving path is overlapped with the coordinates of the obstacle.

After the obstacle in front of the mobile robot is detected, whether the obstacle can collide with the mobile robot needs to be further determined, and therefore whether the mobile robot needs to avoid the obstacle is judged. Based on the above, a first moving path corresponding to the mobile robot is obtained, and whether the first moving path is overlapped with the coordinates of the obstacle or not is judged, wherein the first moving path is an initial path planned according to the initial position and the destination to be reached before the mobile robot starts to work. If the coincidence exists, the mobile robot possibly collides with the obstacle if the mobile robot continues to travel according to the first movement path, and if the coincidence does not exist, the mobile robot indicates that the first movement path of the current mobile robot does not collide with the detected obstacle, and the mobile robot can ignore the obstacle and continue to travel along the original path.

In one embodiment, after determining whether the coordinates of the obstacle overlap with the first movement path of the mobile robot, a z value, which is a vertical coordinate value of the obstacle, is determined, and it should be noted that there may be a height unevenness of the obstacle. The vertical coordinate value is used for preliminarily determining an obstacle avoidance strategy of the mobile robot, the mobile robot can directly ignore the obstacle with a lower vertical coordinate value without adopting any obstacle avoidance strategy, and for the obstacle with a certain height, the mobile robot needs to adopt a corresponding obstacle avoidance strategy to avoid collision.

Specifically, after the vertical coordinate value of the obstacle is determined, the coordinate range in which the vertical coordinate value is located is determined, where the coordinate range includes at least three ranges, namely a first range higher than a first preset threshold, a second range lower than the first preset threshold and higher than a second preset threshold, and a third range lower than the second preset threshold. For the obstacle with the vertical coordinate value in the third range, obstacle avoidance is not needed, and the mobile robot still can travel according to the original first moving path.

The depth camera adopts an MKC-11 binocular structure optical camera and is used for detecting obstacles in a first range and meeting the multi-dimensional obstacle avoidance requirement, the laser radar is used for detecting obstacles in a second range, the ultrasonic sensor adopts an HC-SR04 ultrasonic unit, the detection range is 0.04-3m, the detection angle is 15 degrees, and the ultrasonic sensor is annularly distributed around the mobile robot and is used for detecting obstacles in a third range.

S104: if so, fitting to obtain a two-dimensional contour line of the obstacle, and determining the maximum span of the two-dimensional contour line.

Under the condition that the first moving path and the obstacle have collision risks, at the moment, a two-dimensional contour line of the obstacle needs to be obtained through fitting according to obstacle point cloud data mapped on a grid map, under the normal condition, the two-dimensional contour line is an irregular graph, and after the moving robot touches the obstacle, whether the moving robot can avoid the obstacle in a crossing mode needs to be judged according to a span threshold value which can be crossed by the moving robot, so that the resource use brought by replanning the moving path is saved. And if the judgment process is to be realized, the maximum span of the two-dimensional contour line needs to be determined, wherein the maximum span refers to the length of the maximum connecting line corresponding to the area enclosed by the two-dimensional contour line along the advancing direction of the mobile robot.

Specifically, since the two-dimensional contour line is an irregular figure in general, a convex polygon region corresponding to the two-dimensional contour line is generated by a preset algorithm, which is more convenient for determining the maximum span, and the preset algorithm may be a convex hull algorithm. After the convex polygon area is generated, the moving direction and the current coordinate of the mobile robot are determined, the current coordinate is used as a vertex, and an extension line is made along the moving direction until the extension line is intersected with the convex polygon area. Based on the points on the convex polygon area, a plurality of connecting lines parallel to the extension line are generated, and it should be noted that the connecting lines at least need to pass through any two points on the convex polygon area and intersect with the convex polygon area. And after a plurality of connecting lines are obtained, traversing the connecting lines in sequence, selecting the connecting line with the largest length from the connecting lines as a target connecting line, and taking the length of the target connecting line as the maximum span of the two-dimensional contour line. The maximum span is consistent with the actual moving direction of the mobile robot, and if the mobile robot can span a distance not less than the maximum span, the mobile robot can avoid obstacles in a crossing manner when facing the obstacles.

S105: and determining an obstacle avoidance strategy of the mobile robot according to the maximum span and the vertical coordinate value of the obstacle.

After the maximum span of the two-dimensional contour line is determined, the obstacle avoidance strategy of the mobile robot can be determined according to the maximum span. A span threshold is determined that the mobile robot is able to span and the span threshold is compared to the maximum span. If the span threshold is smaller than the maximum span, it indicates that the maximum distance that the mobile robot can span is smaller than the maximum span of the obstacle, and the mobile robot has a certain distance limit when crossing the obstacle, and at this time, the obstacle-avoiding strategy for the mobile robot can be used to avoid the obstacle. In the embodiment of the application, the first preset threshold is the maximum height that the mobile robot can span, and for an obstacle whose vertical coordinate value is in the first range, the mobile robot also needs to adopt a circumvention strategy to avoid the obstacle. If the span threshold value of the mobile robot is larger than the maximum span and the vertical coordinate value of the obstacle is in the second range, at the moment, the transverse distance of the obstacle can enable the mobile robot to pass through, and the height of the obstacle does not exceed the maximum height which can be spanned by the mobile robot, and at the moment, the mobile robot can adopt a spanning strategy to avoid the obstacle.

When the mobile robot adopts an obstacle avoidance strategy of bypassing the obstacle, the corresponding second moving path needs to be re-planned, so that the mobile robot can bypass the obstacle along the second moving path. Because the appearance of barrier has uncertainty and real-time, this application only regards as keeping away the barrier standard with the nearest barrier of mobile robot, replans the second and moves the route, can avoid mobile robot to bump like this, can also update many times in the short time and move the route, and the real-time is stronger to further ensure mobile robot's the safety of marcing.

Specifically, the current coordinates of the mobile robot and the destination coordinates corresponding to the destination are acquired, and after the coordinates are acquired, a connection path between the current coordinates and the destination coordinates and an obstacle in a rectangular area with the connection path as a diagonal are determined on a grid map. And acquiring a target obstacle which is intersected with the connection path and is closest to the mobile robot from the obstacles in the rectangular area, wherein the target obstacle is an obstacle needing to be avoided and is determined according to the current position of the mobile robot. And determining a boundary point set of the target obstacle, and generating a second moving path capable of bypassing the target obstacle according to the boundary point set. For example, as shown in a schematic diagram of an obstacle shown in fig. 2, each polygon shown in the diagram represents an obstacle surrounded by a set of boundary points, point a represents the current position of the mobile robot, and point B represents the position of the destination of the mobile robot. And connecting the point A and the point B, establishing a rectangle by taking a connecting line between the two points as a diagonal line, wherein the rectangle comprises three polygons, namely an obstacle 1, an obstacle 2 and an obstacle 3, in total, the distance between the mobile robot and the obstacle 1 is the shortest, at the moment, the obstacle 1 is required to be used as a target obstacle, and a second moving path capable of bypassing the obstacle 1 is planned.

It should be noted that, the process of generating the second moving path has randomness, and therefore, the number of the finally generated second moving paths may be multiple, and at this time, the optimal second moving path needs to be selected as the final moving path of the mobile robot. When the final moving path is selected, the final moving path can be determined according to the length of the path and the probability of encountering an obstacle in the process of advancing along the path.

Firstly, the path lengths of a plurality of second moving paths are respectively determined, and the plurality of second moving paths are arranged according to the sequence of the path lengths from large to small so as to obtain a corresponding first moving path sequence. After the first moving path sequence is obtained, the coordinate position of the mobile robot when encountering the obstacle again is determined in the process of moving along the second moving paths, and the lengths of the moving paths respectively corresponding to the mobile robot when moving along the second moving paths are determined according to the coordinate position. And arranging the plurality of second moving paths according to the sequence of the lengths of the moving paths from small to large to obtain a corresponding second moving path sequence. And then, determining a first priority value corresponding to each second movement path in the first movement path sequence and a second priority value corresponding to each second movement path in the second movement path sequence, and performing weighted summation on the first priority value and the second priority value to determine a final movement path of the mobile robot according to a weighted summation result. It can be understood that the final moving path to be finally selected by the mobile robot needs to ensure that the path length is shortest and the possibility of encountering an obstacle again in the process of advancing along the path is low, so that the times of re-planning the path are reduced and the advancing efficiency is improved. For example, there is now a second movement path: and the path lengths of the path a, the path B and the path C are 5m, 10m and 8m, and the moving path lengths when the obstacle is encountered again are 3m, 1m and 4m, so that the first moving path sequence generated correspondingly is the path B, the path C and the path a, and the second moving path sequence is the path B, the path a and the path C. Assuming that the priority values corresponding to the first movement path sequence are 1, 2, and 3, the priority values corresponding to the second movement path sequence are 1, 2, and 3, and the weighting coefficients corresponding to the first movement path sequence and the second movement sequence are 0.4 and 0.6, respectively, the results corresponding to the path a, the path B, and the path C are 2.6, 1, and 1.6, respectively, by weighted summation, obviously, the smaller the result, the longer the path length, the shorter the movement path length, and then the path a is the final movement path of the mobile robot.

The above is the method embodiment proposed by the present application. Based on the same idea, one or more embodiments of the present specification further provide a device and a medium corresponding to the above method.

Fig. 3 is a schematic structural diagram of an obstacle avoidance device of a mobile robot according to an embodiment of the present application, where the obstacle avoidance device includes:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

fusing the three-dimensional point cloud data respectively obtained by the plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data to a grid map to determine coordinates of obstacles in the grid map;

acquiring a first moving path corresponding to the mobile robot, and judging whether the first moving path is overlapped with the coordinates of the obstacle or not;

if so, fitting to obtain a two-dimensional contour line of the barrier, and determining the maximum span of the two-dimensional contour line;

An embodiment of the present application provides a non-volatile computer storage medium, in which computer-executable instructions are stored, and the computer-executable instructions are set to:

fusing three-dimensional point cloud data respectively obtained by a plurality of obstacle detection devices to obtain fused obstacle point cloud data, and mapping the obstacle point cloud data onto a grid map to determine coordinates of obstacles in the grid map;

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the device and media embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference may be made to some descriptions of the method embodiments for relevant points.

The device and the medium provided by the embodiment of the application correspond to the method one by one, so the device and the medium also have the beneficial technical effects similar to the corresponding method.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art to which the present application pertains. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An obstacle avoidance method for a mobile robot, the method comprising:

2. An obstacle avoidance method for a mobile robot according to claim 1, wherein after determining whether the first moving path coincides with the coordinates of the obstacle, the method further comprises:

3. The obstacle avoidance method for a mobile robot according to claim 1, wherein determining the maximum span of the two-dimensional contour line specifically includes:

4. The obstacle avoidance method of a mobile robot according to claim 2, wherein determining the obstacle avoidance strategy of the mobile robot according to the maximum span and the vertical coordinate value of the obstacle specifically comprises:

determining that the obstacle avoidance strategy of the mobile robot is bypassing when the span threshold is smaller than the maximum span or the vertical coordinate value is in the first range;

5. The obstacle avoidance method for the mobile robot according to claim 1, wherein three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices are fused to obtain fused obstacle point cloud data, and the method specifically includes:

filtering the three-dimensional point cloud data respectively acquired by the plurality of obstacle detection devices, and respectively inputting the filtered three-dimensional point cloud data to a first feature extraction network to which the corresponding obstacle detection devices belong, so as to obtain first feature data respectively corresponding to the obstacle detection devices;

inputting the first feature data respectively corresponding to the obstacle detection devices to a second feature extraction network to which the obstacle detection devices belong respectively so as to obtain second feature data respectively corresponding to the obstacle detection devices;

6. An obstacle avoidance method for a mobile robot according to claim 4, wherein after determining that the obstacle avoidance policy of the mobile robot is circumvention, the method further comprises:

7. The obstacle avoidance method for a mobile robot according to claim 6, wherein after generating a second moving path that can bypass the target obstacle, the method further comprises:

and performing weighted summation on the first priority value and the second priority value so as to determine a final moving path of the mobile robot according to the result of the weighted summation.

8. An obstacle avoidance method for a mobile robot according to claim 2, wherein the obstacle detection device is provided with a depth camera, a laser radar and an ultrasonic sensor in sequence from high to low in detection height;

the depth camera is used for detecting obstacles in the first range, the laser radar is used for detecting obstacles in the second range, and the ultrasonic sensor is used for detecting obstacles in the third range.

9. An obstacle avoidance apparatus of a mobile robot, the apparatus comprising: at least one processor;

the memory stores instructions executable by the at least one processor to enable the at least one processor to:

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions configured to: