Relocation Methods for Self-Moving Robots and Self-Moving Robots

Abstract

This application provides a relocalization method for a self-moving robot and a self-moving robot in the field of self-moving robot technology. The self-moving robot includes a lidar. The method includes: when relocalization is triggered, determining at least two candidate poses of the self-moving robot on a map using a particle filter localization algorithm; controlling the self-moving robot to move, so as to collect surrounding environmental information through the lidar during the movement; if, among the at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot's movement, then that candidate pose is taken as the relocalization result. Using this application can avoid errors in the relocalization result.

Description

Technical Field

[0001] This application relates to the field of self-moving robot technology, and in particular to a relocation method for a self-moving robot and the self-moving robot itself. Background Technology

[0002] When a self-moving robot is moved or slips, its pose on the map is usually lost. In this case, the self-moving robot needs to be relocalized.

[0003] In related technologies, self-moving robots typically use particle filter localization algorithms for relocalization. The specific steps are as follows: multiple particles are randomly scattered on the map, each particle representing the pose of the self-moving robot on the map. The self-moving robot collects environmental information around it using LiDAR at its current pose. The LiDAR observation results at the current pose are mapped onto the map with the pose corresponding to each particle to obtain a LiDAR point cloud. The matching degree between the LiDAR point cloud and obstacles on the map is calculated as the score for each particle. Particles with scores greater than a preset threshold are selected, and the pose of the particle on the map is used as the relocalization result.

[0004] However, the map may contain obstacle information for multiple poses, all of which match the observations of the robot's LiDAR at the current pose. This means that multiple particles have scores greater than a preset threshold, for example, see [link to relevant documentation]. Figure 1 , Figure 1 Particle P in the map shown i and P j The information about the surrounding obstacles is similar and matches the observations from the lidar. In this situation, the self-moving robot has difficulty selecting the correct particles, leading to incorrect relocation results. Summary of the Invention

[0005] The purpose of this application is to provide a relocation method for a self-moving robot and a self-moving robot, which can avoid errors in the relocation results. The specific technical solution is as follows:

[0006] In a first aspect, a relocation method for a self-moving robot is provided, the self-moving robot including a lidar, the method comprising:

[0007] When relocation is triggered, at least two candidate poses of the self-moving robot on the map are determined by a particle filter localization algorithm;

[0008] Control the self-moving robot's movements to collect surrounding environmental information via lidar during the movement;

[0009] If, among the at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot's action, then that candidate pose is taken as the relocalization result.

[0010] Optionally, controlling the actions of the self-moving robot includes:

[0011] Control the rotation and / or movement of the self-moving robot.

[0012] Optionally, controlling the rotation and / or movement of the self-moving robot includes:

[0013] Control the rotation of the self-moving robot;

[0014] If, after rotation, the surrounding map information of multiple candidate poses among the at least two candidate poses matches the environmental information collected after the self-moving robot's action, then the cleaning robot is controlled to move.

[0015] Optionally, controlling the rotation and / or movement of the self-moving robot includes:

[0016] Identify distinguishable objects from the surrounding map information of the at least two candidate poses;

[0017] Control the self-moving robot to rotate / move in the direction of detecting the distinguishing object.

[0018] Optionally, obtaining at least two candidate poses through the particle filter localization algorithm includes:

[0019] Multiple particles are generated, each representing the pose of the self-moving robot on the map.

[0020] The observation results of the lidar at the current pose are mapped to the map according to the pose of each particle to obtain the lidar point cloud. The matching degree between the lidar point cloud and the obstacles on the map is calculated as the score of each particle.

[0021] If at least two particles have scores greater than a preset score threshold, and the distance between the at least two particles is greater than a preset distance threshold, then the poses corresponding to the at least two particles are taken as candidate poses.

[0022] Optionally, generating multiple particles includes: generating multiple particles within a set range;

[0023] If there are no particles with a score greater than the preset score threshold, the set range is expanded, and the process returns to the step of generating multiple particles within the set range.

[0024] Optionally, the center of the set range is the pose of the self-moving robot on the map saved before triggering relocation.

[0025] Optionally, the method further includes:

[0026] If only one particle has a score greater than the preset score threshold, then the pose corresponding to that particle is determined as the relocalization result.

[0027] Optionally, the method further includes:

[0028] Relocation is triggered when one of the following conditions is detected:

[0029] The self-moving robot is tilted / lifted;

[0030] The self-moving robot slipped;

[0031] The self-moving robot was involved in an impact;

[0032] The distance actually traveled by the mobile robot is inconsistent with the distance calculated by the odometer.

[0033] In a second aspect, a self-moving robot is provided, including a processor and a machine-readable storage medium storing machine-executable instructions that can be executed by the processor, which in turn cause the processor to perform the aforementioned method steps.

[0034] This application provides a relocalization method for a self-moving robot. Upon triggering relocalization, a particle filter localization algorithm is used to determine at least two candidate poses of the self-moving robot on a map. The robot's movements are controlled to collect surrounding environmental information via LiDAR during these movements. If, among the at least two candidate poses, the surrounding map information of only one candidate pose matches the environmental information collected after the robot's movements, then that candidate pose is taken as the relocalization result. Compared to existing technologies that cannot select the correct pose, this application, even with multiple candidate poses, controls the self-moving robot's movements to allow the LiDAR to acquire richer environmental information. This acquired environmental information is then matched with the surrounding map information of the candidate poses to distinguish the correct pose and obtain the correct localization result.

[0035] Furthermore, this application generates multiple particles within a set range; if no particle has a score greater than a preset score threshold, the set range is expanded, and the process returns to generating multiple particles within the set range. Compared to the existing technology's relocation method of globally generating particles, this application generates particles by gradually increasing the set range, which reduces computational load and shortens the relocation time. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A schematic diagram of the pose of a particle provided in an embodiment of this application;

[0038] Figure 2 A flowchart illustrating a relocation method for a self-moving robot provided in an embodiment of this application;

[0039] Figure 3 A schematic diagram illustrating particle generation provided in an embodiment of this application;

[0040] Figure 4 This is a schematic diagram illustrating how a self-moving robot can rotate to select the correct pose, as provided in an embodiment of this application.

[0041] Figure 5 This is a schematic diagram illustrating how a self-moving robot can be used to filter out the correct pose in an embodiment of this application.

[0042] Figure 6 A flowchart illustrating another relocation method for a self-moving robot provided in an embodiment of this application;

[0043] Figure 7 This is a structural schematic diagram of a self-moving robot provided in an embodiment of this application. Detailed Implementation

[0044] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0045] This application provides a relocalization method for a self-moving robot. The self-moving robot includes a lidar. When relocalization is triggered, the self-moving robot can determine at least two candidate poses on a map using a particle filter localization algorithm. The self-moving robot is controlled to move, and during the movement, environmental information is collected from the surrounding environment using the lidar. If, among the at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot's movement, then that candidate pose is taken as the relocalization result.

[0046] The following will describe in detail, with reference to specific implementation methods, a relocation method for a self-moving robot provided in the embodiments of this application, such as... Figure 2 As shown, the specific steps are as follows:

[0047] Step 201: When relocalization is triggered, at least two candidate poses of the self-moving robot on the map are determined by the particle filter localization algorithm.

[0048] In this embodiment of the application, when the pose of the self-moving robot is lost, the self-moving robot triggers relocalization. When relocalization is triggered, at least two candidate poses of the self-moving robot on the map are determined by the particle filter localization algorithm, so as to select the correct candidate pose from the at least two candidate poses as the relocalization result.

[0049] The self-moving robot may lose its pose and trigger relocalization when it detects one of the following conditions: the self-moving robot tilts / is lifted; the self-moving robot slips; the self-moving robot collides; or the distance actually moved by the self-moving robot is inconsistent with the distance calculated by the odometry.

[0050] Optionally, the specific process of obtaining at least two candidate poses through the particle filter localization algorithm can be as follows: generate multiple particles, each representing the pose of the self-moving robot on the map; map the observation results of the current pose LiDAR to the map with the pose corresponding to each particle to obtain the LiDAR point cloud; calculate the matching degree between the LiDAR point cloud and the obstacles on the map as the score of each particle; if there are at least two particles whose scores are greater than a preset score threshold and the distance between at least two particles is greater than a preset distance threshold, then the poses corresponding to at least two particles are taken as candidate poses.

[0051] In this embodiment, the self-moving robot can randomly scatter multiple particles on a map, with each particle representing the pose of the self-moving robot on the map. For example, the self-moving robot can generate multiple particles within a set range, or generate multiple particles globally.

[0052] The self-moving robot can observe the actual obstacle information around it using LiDAR at its current pose. The observation results are then mapped onto a map to obtain a LiDAR point cloud, with each particle corresponding to its pose. The matching degree between the LiDAR point cloud and the obstacles on the map is calculated (e.g., the number of overlaps between the LiDAR point cloud and obstacles on the map). This score is used as a rating for each particle; the higher the matching degree, the higher the score, and the greater the probability that the particle's pose on the map is the self-moving robot's current pose. The rating can be calculated using a likelihood field model or a beam model.

[0053] If only one particle has a score greater than the preset score threshold, then the pose corresponding to that particle is taken as the relocalization result; if at least two particles have scores greater than the preset score threshold, and the distance between the at least two particles is greater than the preset distance threshold, then the poses corresponding to the at least two particles are taken as candidate poses.

[0054] The self-moving robot can also use the poses corresponding to the top-scoring particles as candidate poses. If a randomly seeded particle enters an obstacle or is less than a set distance from an obstacle, the particle will be reseeded.

[0055] Optionally, the specific process for generating multiple particles can be as follows: generate multiple particles within a set range; if there are no particles with a score greater than a preset score threshold, expand the set range and return to the step of generating multiple particles within the set range.

[0056] The center of the set range can be the pose of the self-moving robot on the map saved before the relocation is triggered.

[0057] In this embodiment, the self-moving robot can disperse particles within a radius R centered on pose A, where pose A is the most recently saved correct pose (and the pose of the self-moving robot on the map saved before triggering relocation). Then, it is determined whether there are particles with a score greater than a preset score threshold; if not, the radius R is expanded, and the process of dispersing particles within a radius R centered on pose A is returned.

[0058] For example, see Figure 3 , Figure 3 The middle arrow represents a particle. The mobile robot disperses particles within a radius of R1. If there are no particles with a score greater than the preset score threshold within the radius of R1, particles are dispersed within the range of R1-R2. If there are no particles with a score greater than the preset score threshold within the range of R1-R2, particles are dispersed within the range of R2-Rmax. If there are no particles with a score greater than the preset score threshold within the range of R2-Rmax, particles are dispersed globally.

[0059] This application generates multiple particles within a set range; if no particle has a score greater than a preset score threshold, the set range is expanded, and the process returns to the step of generating multiple particles within the set range. Compared to the existing technology that uses a global particle generation relocation method, this application generates particles by gradually increasing the set range, which can reduce the amount of computation and shorten the relocation time.

[0060] Step 202: Control the self-moving robot to collect surrounding environmental information via lidar during the movement.

[0061] Step 203: If, among at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot's action, then that candidate pose is taken as the relocalization result.

[0062] In this embodiment, when at least two candidate poses are determined on the map, the self-moving robot can control its movements and collect surrounding environmental information via LiDAR during the movements to obtain more comprehensive environmental information. Then, the acquired environmental information is mapped onto the map and matched with the map information surrounding each candidate pose. It is determined whether only one of the at least two candidate poses has surrounding map information that matches the environmental information collected after the self-moving robot's movements. If so, this single candidate pose is used as the relocalization result; otherwise, the self-moving robot continues to be controlled, and surrounding environmental information is collected via LiDAR to obtain more comprehensive environmental information, until only one of the at least two candidate poses has surrounding map information that matches the environmental information collected after the self-moving robot's movements.

[0063] Compared to existing technologies that cannot select the correct pose, this application, when multiple candidate poses exist, controls the movement of a self-moving robot to enable the LiDAR to acquire richer environmental information, and matches the acquired environmental information with the surrounding map information of the candidate poses to distinguish the correct pose and obtain the correct localization result.

[0064] The specific process for controlling the movement of a self-moving robot can be: controlling the rotation and / or movement of the self-moving robot.

[0065] In one implementation, controlling the rotation and / or movement of the self-moving robot can be: controlling the rotation of the self-moving robot.

[0066] For example, see Figure 4 Assume the candidate pose on the map is P. i and P j Then the pose of the self-moving robot may be a candidate pose P. i It may also be a candidate pose P j To select the correct candidate pose, the self-moving robot rotates (e.g., counterclockwise by an angle θ) and collects surrounding environmental information using LiDAR. Figure 4 As shown in the map, assuming the pose of the self-moving robot is P j After its rotation, the circular obstacle in the upper right corner can be detected by lidar. Assuming the self-moving robot's position is P... iEven after rotation, the circular obstacle in the upper right corner cannot be detected by the lidar. Therefore, the self-moving robot detects the surrounding environment information using lidar during rotation and matches the surrounding map information of the candidate pose with the environmental information collected after the self-moving robot's action. Assuming the self-moving robot detects the circular obstacle using lidar, the lidar detection result only matches the pose P. j By matching the surrounding map information, the self-moving robot can determine its pose. j As a result of localization, the pose of the self-moving robot can be P. j+1 pose P j Add the rotation angle from before.

[0067] In another implementation, controlling the rotation and / or movement of the self-moving robot can be: controlling the movement of the self-moving robot.

[0068] For example, see Figure 5 Assume the candidate pose on the map is P. i and P j Then the pose of the self-moving robot may be a candidate pose P. i It may also be a candidate pose P j To select the correct candidate pose, the self-propelled robot moves (e.g., forward) and collects surrounding environmental information using LiDAR. Figure 5 As shown in the map, assuming the pose of the self-moving robot is P j After it moves, it can detect the circular obstacle in the upper right corner using lidar. Let's assume the self-moving robot's position is P. i Furthermore, even after moving, the robot cannot detect the circular obstacle in the upper right corner using lidar. Therefore, the self-moving robot detects surrounding environmental information using lidar during its movement and matches the surrounding map information of the candidate pose with the environmental information collected after the self-moving robot's actions. Assuming the self-moving robot detects the circular obstacle using lidar, the lidar detection result only matches the pose P. j By matching the surrounding map information, the self-moving robot can determine its pose. j As a result of localization, the pose of the self-moving robot can be P. j+1 pose P j Add the movement path we just described.

[0069] In another implementation, controlling the rotation and / or movement of the self-moving robot can be: controlling the self-moving robot to rotate; if, after rotation, there are multiple candidate poses whose surrounding map information matches the environmental information collected after the self-moving robot's action, then controlling the cleaning robot to move.

[0070] In another implementation, controlling the rotation and / or movement of the self-moving robot can be done by: identifying distinguishable objects from surrounding map information of at least two candidate poses; and controlling the self-moving robot to rotate / move in the direction of detecting the distinguishable objects.

[0071] For example, see Figure 4 Assume the candidate pose on the map is P. i and P j If the distinguishing object in the surrounding map information is the circular obstacle in the upper right corner, then the self-moving robot can rotate counterclockwise in order to rotate in the direction of the detected distinguishing object.

[0072] For example, see Figure 5 Assume the candidate pose on the map is P. i and P j If the distinguishing object in the surrounding map information is the circular obstacle in the upper right corner, then the self-moving robot can move upwards in order to move in the direction of the detected distinguishing object.

[0073] in, Figure 1 , Figure 4 and Figure 5 The dashed line in the diagram represents the detection beam of the lidar.

[0074] See Figure 6 This application also provides an example of a relocation method for a self-moving robot, the specific steps of which are as follows:

[0075] Step 601: Generate multiple particles within a set range.

[0076] Step 602: Map the observation results of the current pose lidar to the map with the pose corresponding to each particle to obtain the lidar point cloud, calculate the matching degree between the lidar point cloud and the obstacles on the map, and use it as the score for each particle.

[0077] Step 603: Determine whether there are particles with scores greater than a preset score threshold.

[0078] If there are no particles with scores greater than the preset score threshold, proceed to step 604; if there is only one particle with a score greater than the preset score threshold, proceed to step 605; if there are at least two particles with scores greater than the preset score threshold, proceed to step 606.

[0079] Step 604: Expand the setting range and return to step 601.

[0080] Step 605: If only one particle has a score greater than the preset score threshold, then the pose corresponding to that particle is taken as the relocalization result.

[0081] Step 606: If at least two particles have scores greater than a preset score threshold and the distance between the at least two particles is greater than a preset distance threshold, then the poses corresponding to the at least two particles are taken as candidate poses.

[0082] Step 607: Control the self-moving robot to collect surrounding environmental information via LiDAR during the movement.

[0083] Step 608: Determine whether, among at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot's action.

[0084] If yes, proceed to step 609; otherwise, return to step 607.

[0085] Step 609: The candidate pose that matches the surrounding map information with the environmental information collected after the self-moving robot's actions is used as the relocalization result.

[0086] The specific implementation process of steps 601-609 can be referred to steps 201-203, and will not be repeated here in the embodiments of this application.

[0087] This application also provides an electronic device, such as... Figure 7 As shown, the system includes a processor 701, a memory 703, and a communication bus 702. The processor 701 and the memory 703 communicate with each other through the communication bus 702. The memory 703 is used to store computer programs. When the processor 701 executes the program stored in the memory 703, it implements the steps of the above-mentioned relocation method.

[0088] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0089] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0090] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0091] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above relocation methods.

[0092] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the steps of any of the relocation methods described above.

[0093] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0094] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0095] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for self-moving robots, computer-readable storage media, and computer program products are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0096] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A relocation method for a self-moving robot, the self-moving robot comprising a lidar, characterized in that, The method includes: When relocation is triggered, at least two candidate poses of the self-moving robot on the map are determined by a particle filter localization algorithm; Control the self-moving robot to rotate and / or move, so as to collect surrounding environmental information via lidar during rotation and / or movement; If, among the at least two candidate poses, only one candidate pose has surrounding map information that matches the environmental information collected after the self-moving robot rotates or moves, then that candidate pose is taken as the relocalization result. The control of the self-moving robot's rotation and / or movement includes: determining a distinguishable object from the surrounding map information of the at least two candidate poses; and controlling the self-moving robot to rotate / move in the direction of detecting the distinguishable object. The step of obtaining at least two candidate poses through the particle filter localization algorithm includes: Multiple particles are generated, each representing the pose of the self-moving robot on the map. The observation results of the lidar at the current pose are mapped to the map according to the pose of each particle to obtain the lidar point cloud. The matching degree between the lidar point cloud and the obstacles on the map is calculated as the score of each particle. If at least two particles have scores greater than a preset score threshold, and the distance between the at least two particles is greater than a preset distance threshold, then the poses corresponding to the at least two particles are taken as candidate poses.

2. The method according to claim 1, characterized in that, The control of the self-moving robot's rotation and / or movement includes: Control the rotation of the self-moving robot; If, after rotation, the surrounding map information of multiple candidate poses matches the environmental information collected after the self-moving robot's action, then the cleaning robot is controlled to move until the surrounding map information of only one candidate pose matches the environment collected after the self-moving robot rotates or moves.

3. The method according to claim 1, characterized in that, The generation of multiple particles includes: generating multiple particles within a set range; The method further includes: If there are no particles with a score greater than the preset score threshold, the set range is expanded, and the process returns to the step of generating multiple particles within the set range.

4. The method according to claim 3, characterized in that, The center of the set range is the pose of the self-moving robot on the map saved before the relocation is triggered.

5. The method according to claim 1, characterized in that, The method further includes: If only one particle has a score greater than the preset score threshold, then the pose corresponding to that particle is determined as the relocalization result.

6. The method according to claim 1, characterized in that, The method further includes: Relocation is triggered when one of the following conditions is detected: The self-moving robot is tilted / lifted; The self-moving robot slipped; The self-moving robot was involved in an impact; The distance actually traveled by the mobile robot is inconsistent with the distance calculated by the odometer.

7. A self-moving robot, characterized in that, The method includes a processor and a machine-readable storage medium storing machine-executable instructions that can be executed by the processor, the processor being prompted by the machine-executable instructions to perform the steps of the method according to any one of claims 1-6.

Images