Positioning assessment method, device, electronic device and computer-readable storage medium

Abstract

The present application belongs to the field of positioning technology, and discloses a positioning evaluation method, device, electronic device and computer-readable storage medium, wherein the method comprises obtaining the current relative position information between the robot and the corresponding reference laser points, and obtaining the positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model. In this way, the accuracy of the current positioning information of the robot can be evaluated.

Description

Technical Field

[0001] The present application relates to the field of positioning technology, and more particularly to a method, device, electronic device and computer-readable storage medium for positioning evaluation. Background Art

[0002] With the gradual development of science and technology, robots have been widely used in people’s daily lives, such as service robots in restaurants or hotels, express delivery robots in logistics, sweeping robots in daily life, etc.

[0003] Under the existing technology, when the robot is positioning, it is usually due to the positioning sensor being blocked or the robot being moved, resulting in inaccurate positioning, which in turn causes errors in the robot's subsequent tasks. Therefore, it is usually necessary to evaluate the robot's positioning quality so as to perform corresponding remedial measures when the robot's positioning is inaccurate.

[0004] Therefore, how to evaluate the positioning quality of the robot is a problem that needs to be solved. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a positioning evaluation method, device, electronic device and computer-readable storage medium, which are used to solve the problem of how to evaluate the positioning quality of the robot when the robot is positioned.

[0006] In a first aspect, an embodiment of the present application provides a method for positioning assessment, which is applied to a robot and includes:

[0007] The relative position information between the current robot and the corresponding reference laser points is obtained, wherein each reference laser point is obtained by multiple laser beams currently emitted by the robot; the positioning evaluation result of the robot is obtained according to the current relative position information of the robot and a trained positioning evaluation model, wherein the positioning evaluation model is constructed based on a neural network and is used to evaluate the positioning accuracy of the robot based on the relative position information of each reference laser point and the robot.

[0008] In the above implementation process, the robot is positioned through the relative position information between the robot and each reference laser point, and the positioning evaluation model evaluates the positioning quality, so that when the positioning is inaccurate, operations such as repositioning or terminating the current task can be performed.

[0009] In conjunction with the first aspect, in one implementation, the relative position information includes:

[0010] Laser distance, mapping distance, and intersection distance;

[0011] The relative position information between the current robot and the corresponding reference laser points is obtained, including: determining the distance between the robot and each reference laser point as the corresponding laser distance; determining the shortest distance between each reference laser point and the edge line of the movable path of the robot as the corresponding mapping distance; determining the distance between each reference laser point and the corresponding laser intersection point as the intersection distance, wherein the laser intersection point corresponding to each reference laser point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot.

[0012] In the above implementation process, the position of the robot is represented by the relative position information between the current robot and the corresponding reference laser points.

[0013] In combination with the first aspect, in one embodiment, a positioning evaluation result of the robot is obtained based on the current relative position information of the robot and a trained positioning evaluation model, including: obtaining the current relative position matrix of the robot based on the current relative position information of the robot; inputting the current relative position matrix into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot.

[0014] In the above implementation process, the current relative position information of the robot is converted into a current relative position matrix that is convenient for inputting into the positioning evaluation model to realize data preprocessing.

[0015] In combination with the first aspect, in one embodiment, after obtaining the current positioning evaluation result of the robot, the method also includes: obtaining at least one historical positioning evaluation result of the robot; performing weighted summation of the current positioning evaluation result and at least one historical positioning evaluation result to obtain a new current positioning evaluation result.

[0016] In the above implementation process, the current positioning evaluation result estimated by the model is adjusted through the historical positioning evaluation results to improve the accuracy of the positioning quality evaluation.

[0017] In combination with the first aspect, in one implementation, before obtaining a positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model, the method further includes:

[0018] Acquire multiple position information samples of the robot and the actual positioning result of each position information sample in the multiple position information samples of the robot; obtain sample evaluation results of each position information sample according to the multiple position information samples of the robot and the obtained initial evaluation model; adjust the model parameters of the initial evaluation model according to the sample evaluation results and the actual positioning results of each position information sample to obtain a trained positioning evaluation model.

[0019] In the above implementation process, the model is trained by positioning sample data to obtain a trained positioning evaluation model.

[0020] In combination with the first aspect, in one implementation, obtaining multiple position information samples of the robot and corresponding positioning result samples includes:

[0021] Obtain historical relative position information and corresponding actual positioning results of multiple historical positions corresponding to when the robot's historical positioning is accurate;

[0022] For each historical relative position information, perform the following steps:

[0023] According to the historical relative position information of a historical position, a plurality of positive sample position information and a positioning result representing accurate positioning are obtained, wherein each positive sample position information is located within a first preset range of the historical position;

[0024] According to historical relative position information of a historical position, a plurality of negative sample position information and a positioning result representing inaccurate positioning are obtained, wherein each negative sample position information is located within a second preset range of the historical position;

[0025] Taking each historical relative position information, each positive sample position information and each negative sample position information as position information samples;

[0026] Each actual positioning result, each positioning result corresponding to the position information of the positive sample, and each positioning result corresponding to the position information of the negative sample are taken as positioning result samples.

[0027] In the above implementation process, new sample data is constructed based on historical sample data, thereby expanding the sample data.

[0028] In a second aspect, an embodiment of the present application provides a positioning assessment device, the device comprising:

[0029] An acquisition module is used to acquire the relative position information between the robot and the corresponding reference laser points, wherein each reference laser point is obtained by the laser currently emitted by the robot;

[0030] The processing module is used to obtain the positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model, wherein the positioning evaluation model is constructed based on a neural network and is used to evaluate the positioning accuracy of the robot based on the relative position information of each reference laser point and the robot.

[0031] In conjunction with the second aspect, in one implementation, the relative position information includes: laser distance, mapping distance, and intersection distance;

[0032] Obtain the relative position information between the robot and the corresponding reference laser points, including:

[0033] The distance between the robot and each reference laser point is determined as the corresponding laser distance;

[0034] The shortest distance between each reference laser point and the edge line of the movable path of the robot is determined as the corresponding mapping distance;

[0035] The distance between each reference laser point and the corresponding laser intersection point is respectively determined as the intersection distance, wherein the laser intersection point corresponding to each reference laser point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot.

[0036] In conjunction with the second aspect, in one implementation, the processing module is specifically configured to:

[0037] According to the current relative position information of the robot, the current relative position matrix of the robot is obtained;

[0038] The current relative position matrix is ​​input into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot.

[0039] In conjunction with the second aspect, in one implementation, the processing module is specifically configured to:

[0040] Obtain at least one historical positioning evaluation result of the robot;

[0041] A weighted sum is performed on the current positioning evaluation result and at least one historical positioning evaluation result to obtain a new current positioning evaluation result.

[0042] In conjunction with the second aspect, in one implementation, the positioning assessment device further includes:

[0043] A model training module, used for obtaining multiple position information samples of the robot and an actual positioning result of each position information sample in the multiple position information samples of the robot;

[0044] According to the multiple position information samples of the machine and the obtained initial evaluation model, a sample evaluation result of each position information sample is obtained respectively;

[0045] According to the sample evaluation results and actual positioning results of each location information sample, the model parameters of the initial evaluation model are adjusted to obtain a trained positioning evaluation model.

[0046] In conjunction with the second aspect, in one implementation, the model training module is used to obtain multiple position information samples and corresponding positioning result samples of the robot, specifically for:

[0047] Obtain historical relative position information and corresponding actual positioning results of multiple historical positions corresponding to when the robot's historical positioning is accurate;

[0048] For each historical relative position information, perform the following steps:

[0049] According to the historical relative position information of a historical position, a plurality of positive sample position information and a positioning result representing accurate positioning are obtained, wherein each positive sample position information is located within a first preset range of the historical position;

[0050] According to historical relative position information of a historical position, a plurality of negative sample position information and a positioning result representing inaccurate positioning are obtained, wherein each negative sample position information is located within a second preset range of the historical position;

[0051] Taking each historical relative position information, each positive sample position information and each negative sample position information as position information samples;

[0052] Each actual positioning result, each positioning result corresponding to the position information of the positive sample, and each positioning result corresponding to the position information of the negative sample are taken as positioning result samples.

[0053] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0054] A processor, a memory and a bus, wherein the processor is connected to the memory via the bus, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, they are used to implement the method as described in the first aspect and any one of the embodiments of the first aspect.

[0055] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a server, the steps in the method provided in the first aspect and any one of the embodiments of the first aspect are implemented.

[0056] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by practicing the embodiments of the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A flowchart of a model training method provided in an embodiment of the present application;

[0059] Figure 2 A flowchart of a method for positioning evaluation provided in an embodiment of the present application;

[0060] Figure 3 A schematic diagram of a positioning scenario provided in an embodiment of the present application;

[0061] Figure 4 A schematic diagram of a positioning assessment model structure provided in an embodiment of the present application;

[0062] Figure 5 A schematic diagram of the structure of a positioning assessment device provided in an embodiment of the present application;

[0063] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0065] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0066] It is a necessary condition for a mobile robot to obtain its own positioning in the surrounding environment before it can carry out subsequent tasks. When the robot's positioning sensor is blocked or the robot is moved, the robot's positioning may shift. Therefore, the robot needs to evaluate whether the current positioning is accurate to determine whether to start subsequent tasks.

[0067] In the embodiment of the present application, the execution subject may be a robot. In the embodiment of the present application, before positioning evaluation is performed on the robot, model training is first performed using ... sample data to obtain a trained ...

[0068] See also Figure 1 As shown, Figure 1 The embodiment of the present application provides a flow chart of an implementation method of a model training method, and the specific implementation process of the method is as follows:

[0069] Step 100: Acquire multiple position information samples of the robot and corresponding positioning result samples.

[0070] Specifically, when executing step 100, any one or combination of the following methods may be used:

[0071] Method 1: Obtain historical relative position information and corresponding actual positioning results of multiple historical positions corresponding to the accurate historical positioning of the robot, and use each historical relative position information as a position information sample and each actual positioning result as a positioning result sample.

[0072] Method 2: construct a new position information sample and a corresponding positioning result sample based on the historical relative position information of multiple historical positions corresponding to the accurate historical positioning of the robot.

[0073] When executing method 2, the following steps can be used:

[0074] For each historical relative position information, perform the following steps:

[0075] S1001: According to historical relative position information of a historical position, multiple positive sample position information and positioning results representing accurate positioning are obtained.

[0076] Wherein, each positive sample position information is located within a first preset range of the one historical position.

[0077] S1002: Obtaining multiple negative sample position information and positioning results representing inaccurate positioning based on historical relative position information of a historical position.

[0078] Wherein, each negative sample position information is located within a second preset range of the one historical position.

[0079] S1003: taking each positive sample position information and each negative sample position information as position information samples.

[0080] S1004: taking the positioning results corresponding to the position information of each positive sample and the positioning results corresponding to the position information of each negative sample as positioning result samples.

[0081] As an embodiment, when the historical positioning of the robot is accurate, the corresponding multiple historical position data sets are [P0], and one of the historical position data (x, y, θ) is selected.

[0082] Where x represents the value of the robot's X-axis in the current robot coordinate system, y represents the value of the robot's Y-axis in the current robot coordinate system, and θ represents the current robot's deflection angle;

[0083] Construct a sample position data (x′, y′, θ′) of a robot;

[0084] Among them, |xx′|=Δx1, |yy′|=Δy1, |θ-θ′|=Δθ1.

[0085] When the sample position data (x′, y′, θ′) satisfies the first preset range [(Δx1) 2 +(Δy1) 2 ] 0.5 <0.5m and Δθ1<6°), the sample data (x′, y′, θ′) constructed at this time is positive sample data, and the positioning result corresponding to the positive sample data is accurate positioning.

[0086] According to this method, multiple positive sample position data are constructed to form a positive sample position data set [P1].

[0087] Construct a sample position data (x″, y″, θ″) of a robot;

[0088] Among them, |xx″|=Δx2, |yy″|=Δy2, |θ-θ″|=Δθ2.

[0089] When the sample position data (x″, y″, θ″) meets the second preset condition (Δx2<5m or Δy2<5m or Δθ2<120°), the sample position data (x″, y″, θ″) constructed at this time is negative sample data, and the positioning result corresponding to the negative sample data is inaccurate positioning.

[0090] According to this method, the relative position information of multiple negative sample data is constructed to form a negative sample data set [P2].

[0091] Finally, the data sets [P0] and [P1] are used as the positive sample position information sets, and [P2] is used as the negative sample position information set.

[0092] It should be noted that the number of positive sample position information in the positive sample position information set and the number of negative sample position information in the negative sample position information set may be the same, different, or proportional, but the present application is not limited thereto.

[0093] In the above implementation process, multiple location information samples are constructed through historical location information when historical positioning is accurate, which can ensure that positive sample data and negative sample data are evenly distributed, and a large amount of sample data can be obtained without manual labeling.

[0094] Each sample position in the sample sets [P0], [P1] and [P2] generates a corresponding position matrix sample, and the multiple relative position matrix samples finally formed are the multiple position information samples of the machine.

[0095] Step 101: According to a plurality of position information samples of a machine and an acquired initial evaluation model, a sample evaluation result of each position information sample is obtained respectively.

[0096] Step 102: According to the sample evaluation results of each position information sample and the positioning result sample, the model parameters of the initial evaluation model are adjusted to obtain a trained positioning evaluation model.

[0097] Specifically, the positioning evaluation model is constructed based on a neural network and is used to evaluate the positioning accuracy of the robot based on the relative position information between each reference laser point and the robot.

[0098] In this way, the robot can be positioned and evaluated using the trained positioning evaluation model. Figure 2 A positioning evaluation method provided in an embodiment of the present application is described.

[0099] See also Figure 2 As shown, Figure 2 This is a flowchart of a method for positioning evaluation provided in an embodiment of the present application. The specific implementation process of the method is as follows:

[0100] Step 200: Obtain the relative position information between the robot and each reference laser point.

[0101] Specifically, a plurality of reference laser points are obtained through the laser currently emitted by the robot, and the relative position information between the current position of the robot and each reference laser point is obtained.

[0102] The relative position information includes: laser distance, mapping distance and cross distance. Each reference laser point is obtained by the laser currently emitted by the robot.

[0103] In one embodiment, the current position of the robot is the current projection position of the robot on a specified two-dimensional plane. The reference laser point is the projection point of the laser point in the specified two-dimensional plane.

[0104] Optionally, the specified 2D plane can be the ground.

[0105] To obtain the relative position information, the following steps may be used:

[0106] S2001: Determine the distance between the robot and each reference laser point as the corresponding laser distance.

[0107] S2002: Determine the shortest distance between each reference laser point and the edge line of the movable path of the robot as the corresponding mapping distance.

[0108] S2003: Determine the distance between each reference laser point and the corresponding laser intersection point as the intersection distance.

[0109] The laser intersection point corresponding to each reference laser point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot.

[0110] It should be noted that in the embodiment of the present application, only the sequential execution of S2001, S2002 and S2003 is used as an example for explanation. In actual application, S2001, S2002 and S2003 can also be executed in other orders or in parallel, which will not be repeated here.

[0111] In one embodiment, a laser sensor is provided in the robot. The laser sensor can emit one or more laser beams in all directions. When each laser beam hits an obstacle in the surrounding environment, a corresponding laser point is generated.

[0112] Optionally, the laser sensor can use a rotating emission method to emit lasers in multiple directions to obtain multiple laser points, or can use a method of emitting lasers at multiple positions of the robot to emit lasers in multiple directions to obtain multiple laser points.

[0113] In one embodiment, before executing step 200, the path that the robot can move is obtained in advance, the path is projected on a specified two-dimensional plane (such as the ground), and the curve or straight line composed of the edge of the path projection is determined as the movable edge line of the robot.

[0114] It should be noted that the movable edge line of the robot may be obtained by the robot through a laser sensor or by other means, which is not limited here.

[0115] In the embodiment of the present application, the laser sensor may be a 2D laser radar sensor or other sensors, which are not limited here.

[0116] See also Figure 3 As shown, a relative position diagram provided by an embodiment of the present application is shown. Point a is the projection of the robot on the ground, i.e., the current position of the robot. Point i is the projection of a laser point on the ground generated by the robot emitting a laser to the surrounding through a laser sensor, i.e., the reference laser point i. Point b is the foot of the perpendicular of point i to the nearest edge line of the robot's movable path. Point c represents the intersection of the line connecting point a and point i and the edge line of the robot's movable path, i.e., the laser intersection point.

[0117] Among them, the distance from point i to point a is the laser distance r corresponding to the robot and the reference laser point i i The distance from point i to point b is the mapping distance d between the robot and the reference laser point i. i The distance from point i to point c is the intersection distance e between the robot and the reference laser point i. i .

[0118] It should be noted that the mapping distance d i It is the shortest distance between the reference laser point and the edge line of the robot's movable path.

[0119] Step 201: Obtain a positioning evaluation result of the robot according to the current relative position information of the robot and a trained positioning evaluation model.

[0120] Specifically, when executing step 201, the following steps may be adopted:

[0121] S2011: Obtaining the current relative position matrix of the robot according to the current relative position information of the robot.

[0122] Specifically, the current relative position information z of the robot i The current relative position matrix D of the robot is composed.

[0123] Among them, the relative position information z i =(r i , d i , e i ) represents the relative position information between the robot and any reference laser point i. i Indicates the laser distance between the robot and the reference laser point i. i represents the mapping distance between the robot and the reference laser point i. i Represents the intersection distance between the robot and the reference laser point i.

[0124] in, The width, height, and number of channels of D are N / 2, 2, and 3 respectively, and N is the total number of reference laser points.

[0125] In the above implementation process, by adjusting the current relative position information of the robot to a current relative position matrix that can be input into the positioning evaluation model, it is convenient to further output the current positioning evaluation result of the robot through the positioning evaluation model.

[0126] S2012: Input the current relative position matrix into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot.

[0127] Specifically, the robot’s current relative position matrix D is input into the trained positioning evaluation model to obtain the robot’s current positioning evaluation result P t .

[0128] It should be noted that the positioning evaluation result P t It can be an evaluation score or a probability value of accurate positioning, but the application is not limited thereto.

[0129] When the positioning evaluation result P t When the evaluation score or the probability value of accurate positioning is greater than or equal to the preset value, it means that the robot is currently positioned accurately; when the positioning evaluation result P t When the evaluation score or the probability value of accurate positioning is less than the preset value, it means that the current positioning of the robot is inaccurate.

[0130] In practical applications, the preset value can be set according to the actual application scenario, such as 0.8, which is not limited here.

[0131] Furthermore, in order to improve the accuracy of the positioning quality assessment, the above current positioning assessment result may be adjusted according to the historical positioning assessment results.

[0132] Specifically, when adjusting the current positioning evaluation result, the following methods can be used:

[0133] At least one historical positioning evaluation result of the robot is obtained, and the current positioning evaluation result and the at least one historical positioning evaluation result are weightedly summed to obtain a new current positioning evaluation result.

[0134] Optionally, when a new current positioning evaluation result P is determined, the following formula may be used:

[0135] P=a1×P t-m +a2×P t-m+1 +…+a m-1 ×P t-2 +a m ×P t-1 +a m+1 ×P t .

[0136] Among them, {Pt-m , P t-m+1 ,…,P t-2 , P t-1} is the current positioning evaluation result P t The previous m historical positioning evaluation results, where m is an integer greater than or equal to 1. t is the current positioning evaluation result, {a1, a2, …, a m-1 , a m , a m+1} is a weight parameter, which can be set based on experience or actual application scenarios and is not limited here.

[0137] In this way, when the robot determines that the current positioning is inaccurate based on the current positioning evaluation results, it can perform operations such as repositioning or stopping the current task, thereby avoiding task errors.

[0138] The specific structure of the above positioning evaluation model is described below. Figure 4 As shown, Figure 4 A schematic diagram of a positioning evaluation model structure provided in an embodiment of the present application, in Figure 4 In the positioning assessment model structure 400 shown, the positioning assessment model includes a first convolution layer 410, a first maximum pooling layer 420, a second convolution layer 430, a second maximum pooling layer 440, a third convolution layer 450, a third maximum pooling layer 460, a fully connected layer 470 and an activation function 480.

[0139] In the embodiment of the present application, only the activation function 480 is taken as the Sigmoid function for illustration. In actual applications, the activation function 480 may also be other activation functions such as LogSigmoid and Tanh function, but the present application is not limited thereto.

[0140] exist Figure 4 , the current relative position matrix D of the robot is sequentially input into the first convolution layer 410, the first maximum pooling layer 420, the second convolution layer 430, the second maximum pooling layer 440, the third convolution layer 450, the third maximum pooling layer 460, the fully connected layer 470 and the activation function 480 in the positioning evaluation model to obtain the current positioning evaluation result P of the robot. t .

[0141] In order to ensure that the sample data for training the initial evaluation model is sufficient and the number of positive sample data and negative sample data is uniform, it takes a lot of work to construct the sample data by manual annotation, and the specific implementation process is difficult to achieve. Therefore, this application uses multiple historical positions corresponding to accurate positioning to construct multiple positive sample location information and negative sample location information. The specific construction process is as follows:

[0142] Based on the same inventive concept, a positioning evaluation device is also provided in the embodiment of the present application. Since the principle of solving the problem by the above-mentioned device and equipment is similar to that of a positioning evaluation method, the implementation of the above-mentioned device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0143] See also Figure 5 As shown, it is a schematic diagram of the structure of a positioning evaluation device provided in an embodiment of the present application. Figure 5 The positioning assessment device 500 is shown with Figure 1 methods, including those that can be implemented Figure 1 Each functional module of the method.

[0144] In one embodiment, Figure 5 The positioning assessment apparatus 500 shown includes:

[0145] An acquisition module 510 is used to acquire relative position information between the robot and corresponding reference laser points, wherein each reference laser point is obtained by the laser currently emitted by the robot;

[0146] The processing module 520 is used to obtain the positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model, wherein the positioning evaluation model is constructed based on a neural network and is used to evaluate the positioning accuracy of the robot based on the relative position information of each reference laser point and the robot.

[0147] In another embodiment, the relative position information includes: laser distance, mapping distance, and intersection distance;

[0148] Obtain the relative position information between the robot and the corresponding reference laser points, including:

[0149] The distance between the robot and each reference laser point is determined as the corresponding laser distance;

[0150] The shortest distance between each reference laser point and the edge line of the movable path of the robot is determined as the corresponding mapping distance;

[0151] The distance between each reference laser point and the corresponding laser intersection point is respectively determined as the intersection distance, wherein the laser intersection point corresponding to each reference laser point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot.

[0152] In another embodiment, the processing module is specifically configured to:

[0153] According to the current relative position information of the robot, the current relative position matrix of the robot is obtained;

[0154] The current relative position matrix is ​​input into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot.

[0155] In another embodiment, the processing module is used to input the current relative position matrix into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot, specifically for:

[0156] Obtain at least one historical positioning evaluation result of the robot;

[0157] A weighted sum is performed on the current positioning evaluation result and at least one historical positioning evaluation result to obtain a new current positioning evaluation result.

[0158] In another embodiment, the positioning assessment device further includes:

[0159] A model training module 530 is used to obtain multiple position information samples of the robot and an actual positioning result of each position information sample in the multiple position information samples of the robot;

[0160] According to the multiple position information samples of the machine and the obtained initial evaluation model, a sample evaluation result of each position information sample is obtained respectively;

[0161] According to the sample evaluation results and actual positioning results of each location information sample, the model parameters of the initial evaluation model are adjusted to obtain a trained positioning evaluation model.

[0162] In another embodiment, the model training module is used to obtain multiple position information samples and corresponding positioning result samples of the robot, specifically for:

[0163] Obtain historical relative position information and corresponding actual positioning results of multiple historical positions corresponding to when the robot's historical positioning is accurate;

[0164] For each historical relative position information, perform the following steps:

[0165] According to the historical relative position information of a historical position, a plurality of positive sample position information and a positioning result representing accurate positioning are obtained, wherein each positive sample position information is located within a first preset range of the historical position;

[0166] According to historical relative position information of a historical position, a plurality of negative sample position information and a positioning result representing inaccurate positioning are obtained, wherein each negative sample position information is located within a second preset range of the historical position;

[0167] Taking each historical relative position information, each positive sample position information and each negative sample position information as position information samples;

[0168] Each actual positioning result, each positioning result corresponding to the position information of the positive sample, and each positioning result corresponding to the position information of the negative sample are taken as positioning result samples.

[0169] It should be noted that Figure 5 The positioning assessment device 500 shown can achieve Figure 1 The method embodiment involves various processes of the positioning assessment method. The operations and / or functions of the various modules in the positioning assessment device 500 are respectively to achieve Figure 1 For details, please refer to the description in the above method embodiment. To avoid repetition, detailed description is appropriately omitted here.

[0170] Please refer to Figure 6 , Figure 6 A schematic diagram of another electronic device structure provided in an embodiment of the present application, Figure 6 The task allocation device 600 shown may include: at least one processor 610, such as a CPU, at least one communication interface 620, at least one memory 630 and at least one communication bus 640. The communication bus 640 is used to realize direct connection and communication between these components. The communication interface 620 of the device in the embodiment of the present application is used to communicate signaling or data with other node devices. The memory 630 may be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The memory 630 may optionally be at least one storage device located away from the aforementioned processor. The memory 630 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 610, the task allocation device executes the above-mentioned Figure 1 The method process is shown.

[0171] The present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a server, Figure 1 The method process shown.

[0172] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are only illustrative. For example, the division of the system devices is only a logical function division. There may be other division methods in actual implementation. For example, multiple devices or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0173] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for positioning assessment, characterized in that: The method is applied to a robot, and the method comprises: Acquire the relative position information between the robot and the corresponding reference laser points, wherein each reference laser point is obtained by the laser currently emitted by the robot; the relative position information includes: laser distance, mapping distance and intersection distance; the laser distance is the distance between the robot and the reference laser point; the mapping distance is the shortest distance between the reference laser point and the edge line of the movable path of the robot; the intersection distance is the distance between the reference laser point and the corresponding laser intersection point; the laser intersection point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot; According to the current relative position information of the robot and the trained positioning evaluation model, the positioning evaluation result of the robot is obtained, wherein the positioning evaluation model is constructed based on a neural network, and is used to evaluate the positioning accuracy of the robot based on the relative position information of each reference laser point and the robot.

2. The method according to claim 1, characterized in that: The obtaining of the positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model includes: Obtaining a current relative position matrix of the robot according to the current relative position information of the robot; The current relative position matrix is ​​input into the trained positioning evaluation model to obtain the current positioning evaluation result of the robot.

3. The method according to claim 2, characterized in that After obtaining the current positioning evaluation result of the robot, the method further includes: Obtaining at least one historical positioning evaluation result of the robot; The current positioning evaluation result and the at least one historical positioning evaluation result are weightedly summed to obtain a new current positioning evaluation result.

4. The method according to claim 1, characterized in that: Before obtaining the positioning evaluation result of the robot according to the current relative position information of the robot and the trained positioning evaluation model, the method further includes: Acquire multiple position information samples of the robot and corresponding positioning result samples; Obtaining a sample evaluation result of each position information sample according to the plurality of position information samples of the robot and the obtained initial evaluation model; According to the sample evaluation results of each position information sample and the positioning result sample, the model parameters of the initial evaluation model are adjusted to obtain a trained positioning evaluation model.

5. The method according to claim 4, characterized in that The obtaining of a plurality of position information samples and corresponding positioning result samples of the robot includes: Obtaining historical relative position information and corresponding actual positioning results of multiple historical positions corresponding to when the historical positioning of the robot is accurate; For each historical relative position information, perform the following steps: According to the historical relative position information of a historical position, a plurality of positive sample position information and a positioning result representing accurate positioning are obtained, wherein each positive sample position information is located within a first preset range of the historical position; According to historical relative position information of a historical position, a plurality of negative sample position information and a positioning result representing inaccurate positioning are obtained, wherein each negative sample position information is located within a second preset range of the historical position; Taking each historical relative position information, each positive sample position information and each negative sample position information as position information samples; Each actual positioning result, each positioning result corresponding to the position information of the positive sample, and each positioning result corresponding to the position information of the negative sample are taken as positioning result samples.

6. A positioning assessment device, characterized in that: The device comprises: An acquisition module is used to acquire relative position information between the robot and corresponding reference laser points, wherein each reference laser point is obtained by the laser currently emitted by the robot; the relative position information includes: laser distance, mapping distance and intersection distance; the laser distance is the distance between the robot and the reference laser point; the mapping distance is the shortest distance between the reference laser point and the edge line of the movable path of the robot; the intersection distance is the distance between the reference laser point and the corresponding laser intersection point; the laser intersection point is the intersection point of the corresponding laser and the edge surface of the movable path of the robot; A processing module is used to obtain a positioning evaluation result of the robot according to the current relative position information of the robot and a trained positioning evaluation model, wherein the positioning evaluation model is constructed based on a neural network and is used to evaluate the positioning accuracy of the robot based on the relative position information of each reference laser point and the robot.

7. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the processor is connected to the memory via the bus, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, they are used to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a server, implements the method as claimed in any one of claims 1 to 5.

Images