Method for correcting working face model of inspection robot based on cloud computing

Abstract

The invention discloses a method for correcting a working face model by an inspection robot based on cloud computing. The method comprises the steps of 1, constructing the working face model, and obtaining latest working face data; 2, the inspection robot obtains a displacement increment vector of the inspection robot, and the position of the inspection robot at each moment is obtained; step 3, dividing the coordinates of the open-off cut to obtain n coordinate vectors at equal intervals; 4, the average error of the inspection robot passing through the n hydraulic support positions and the open-off cut positions is calculated; 5, judging whether correction is needed or not according to the average error; if yes, returning to the step 3 after correction; otherwise, executing the step 6; 6, the distance between the two crossheads and the stopping line is obtained; and 7, updating the working face model. The dynamic updating mechanism is formed through mutual feedback of the dynamic data of the inspection robot and the static data of the working face, and the working face dynamically updates the model along with relevant data acquired by the inspection robot.

Description

Technical Field

[0001] The invention belongs to the technical field of mining engineering and relates to a method for correcting a working face model by an inspection robot based on cloud computing. Background Art

[0002] The complexity of coal mine geology brings many safety hazards to the safe and efficient mining of coal mines. For this reason, the National Energy Administration and the State Administration of Work Safety have put forward three-level requirements for transparent geological construction, including the construction of a primary transparent geological model of static detection + three-dimensional geological geometry model, an intermediate transparent geological model of dynamic detection + three-dimensional geological attribute model, and an advanced transparent geological model of monitoring and early warning + three-dimensional geological mining model. Taking the construction of the mining face model as an example, the construction of the basic working face model is mainly based on the static and dynamic data of the working face obtained by drilling, geophysical exploration, measurement and other methods. The advancement of the two chute of the working face, the fluctuation state of the cut eye, etc. are mainly measured manually. How to take effective methods to automatically obtain measurement data to dynamically update the constructed mining face model to achieve safe and efficient coal mine measurement work, reduce costs and increase efficiency, has become an urgent problem to be solved. Summary of the invention

[0003] The purpose of the present invention is to provide a method for calibrating a working surface model of an inspection robot based on cloud computing, so as to solve the problem of automatic updating of the working surface model existing in the prior art.

[0004] The present invention provides a method for calibrating a working surface model by an inspection robot based on cloud computing, the method comprising: constructing a working surface model, and determining the "0" point position of the inspection robot and the working surface model in a spatial coordinate system. Constructing a working surface model refers to constructing a working surface model through data such as geophysical exploration, drilling, measurement, and realism, wherein n x equal-spaced grids are obtained by dividing the number of hydraulic supports (n) mian ,y mian 、z mian The coordinate vector P mian ; According to the distances from the two drifts to the stop line, the two drift models of the working face are fed back to update the working face model; at the same time, according to the P within the error range, mian The data dynamically corrects the cutting data to show the undulating shape of the working surface.

[0005] The specific steps include:

[0006] Step 1: Construct a working face model without feedback from the inspection robot based on drilling, geophysical exploration, measurement, and realism data, and determine the "0" point position of the inspection robot and the working face model in the spatial coordinate system; rotate and translate the constructed working face model to the "0" point position to obtain the latest working face data x mian ,y mian 、zmian .

[0007] Step 2: An infrared transmitter is installed on the body of the inspection robot, and each hydraulic support is installed with an infrared receiver. The inspection robot obtains the displacement increment vector of the inspection robot by using its own strapdown inertial navigation technology and integrating it with the infrared receiver of the hydraulic support, and obtains the position of the inspection robot at each moment through the displacement increment vector at each moment, thereby realizing the positioning of the inspection robot.

[0008] Specifically, the displacement increment vector calculation formula of the inspection robot is as follows:

[0009] S xun =C b *S b (1)

[0010] Where: S xun Represents the displacement increment vector of the inspection robot; C b represents the cosine matrix; S b Represents the displacement increment, obtained using the hydraulic support infrared positioning sensor.

[0011]

[0012] Where: θ, γ, The pitch angle, heading angle, and roll angle of the encoder position are obtained after correction in the unified coordinate system and are obtained using the strapdown inertial navigation technology of the inspection robot.

[0013] The position of the inspection robot at time t is:

[0014] P xun (t) = P xun (t-1)+S xun (t)(3)

[0015] Where: P xun (t), P xun (t-1) represents the position of the inspection robot at time t and time t-1 respectively; S xun (t) represents the displacement increment vector at time t.

[0016] Through the ring network or 5G+wifi xun Transmit to the cloud platform database.

[0017] Step 3: Divide the cut-eye coordinates in the working face model constructed in step 1 according to the number of hydraulic supports (n) to obtain n coordinate vectors P with equal spacing. mian (i.e. the coordinates of each hydraulic support projected onto the working surface model), P mian(t) is the coordinate vector on the working surface model corresponding to the hydraulic support position of the inspection robot at time t.

[0018] Step 4: Calculate the average error E between the inspection robot's position after passing through n hydraulic supports and the cut-eye position of the working surface model ave , the calculation formula is as follows:

[0019]

[0020] Step 5: According to the average error calculated in step 4, the actual production allowable error is set as the error threshold. If the average error is greater than the error threshold, it is considered that correction is required. At this time, the coordinate vector P on the working surface model corresponding to the hydraulic support position where the inspection robot is located at time t is set to mian (t) Take the position P at time t xun (t) Replace and return to step 3; otherwise, if the average error is not greater than the error threshold (i.e., the error is within the error range), it is considered that no correction is required and step 6 is executed.

[0021] Step 6: Get the distance d between the two chute and the stop line 距 , the calculation formula is as follows:

[0022]

[0023] Where: x' xun ,y' xun 、z' xun represents the coordinates of the inspection robot at both ends of the cut eye; x ting ,y ting 、z ting Represents the coordinates of the two end points of the stop mining line.

[0024] Step 7: The distances d from the two drifts to the stop line obtained in step 6 距 , and then fed back to the two chute models of the working face to update the working face model; at the same time, according to the P within the error range finally obtained in step 5 mian The data dynamically corrects the cutting data to show the undulating shape of the working surface.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention forms a dynamic update mechanism by feeding back the dynamic data of the inspection robot and the static data of the working face, so as to realize the dynamic update model of the working face as the inspection robot collects relevant data. A set of digital-analog interactive model update methods is formed, which improves the accuracy of the working face, reduces the workload of manual data collection and update, and improves the safety efficiency of underground coal mine work to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the spatial relationship among the inspection robot, hydraulic support and working surface;

[0028] Figure 2 This is a calibration diagram for the “0” point position. DETAILED DESCRIPTION

[0029] The present invention is further explained below with reference to the accompanying drawings and embodiments.

[0030] Step 1: If Figure 2 The figure shows the correction diagram of the "0" point position of the 81310 working face of a mine after rotation and translation to unify the walking direction of the inspection robot. The direction of the "0" point is determined to be the direction from the rubber transport chute to the auxiliary transport chute. The working face model without feedback from the inspection robot is constructed based on the drilling, geophysical, measurement, and realism data of the 81310 working face, and the "0" point position of the inspection robot and the working face model in the spatial coordinate system is determined; the constructed working face model is rotated and translated to the "0" point position to obtain the latest working face data x mian ,y mian 、z mian .

[0031] Step 2: The inspection robot is equipped with an infrared transmitter on its body, and each hydraulic support is equipped with an infrared receiver. The inspection robot obtains the real-time spatial coordinate x by using the strapdown inertial navigation technology and the infrared receiver of the hydraulic support. xun ,y xun 、z xun , transmitted to the hyper-converged cloud platform database for storage via 5G+wifi.

[0032] Step 3: As shown in Table 1, the working face parameters of 81310 are cut into 122 equally spaced blocks according to the hydraulic supports. Corresponding to the data obtained after the working face is formed by drilling, geophysical exploration, measurement, and artificial realism, the acquired data is divided into 122 blocks at equal intervals to obtain the spatial coordinates of the top and bottom plates of the working face model corresponding to each hydraulic support at the cutting point to form the initial model data.

[0033] Table 1 Data of 81310 working face

[0034]

[0035] Step 4: By comparing the distance the inspection robot travels from one end of the work surface to the other end, use cloud computing technology to calculate the number of times the inspection robot passes through the n hydraulic support positions P xun (t) and the cutting position P of the working surface model mian (t) The average error between the two ave , as shown in Table 2, the data comparison between the inspection robot and the working face cutting eye.

[0036] Table 2 Coordinate data

[0037]

[0038] The data obtained in Table 2 are used to calculate the average error E through formula (4): ave , the average error is 0.47.

[0039] Step 5: According to actual production needs, the model error must be within the range of ±0.5m. The error calculated in step 5 is 0.47, which meets the accuracy requirement. Go to step 6.

[0040] Step 6: Calculate the distance d from the two drifts to the stop line 距 .

[0041] Step 7: The distance d between the two chute and the stop mining line obtained in step 6 距 , and fed back to the two chute models of the working face; according to step 5, obtain P within the error range mian The data dynamic correction updates the geological model cut-eye data, thus showing the undulating shape of the working face. Table 3 shows the comparison of the distances between the two trenches and the stop-mining line before and after correction.

[0042] Table 3 Distance between two drifts and the stop mining line

[0043]

[0044] It can be seen from the results of the embodiments that the present invention uses the inspection robot to calibrate the working face model, which can realize the automatic updating of the working face model, improve the working face accuracy, reduce the workload of manual data collection and update, and improve the safety efficiency of underground coal mine work to a certain extent.

Claims

1. A method for calibrating a working surface model of an inspection robot based on cloud computing, characterized in that: The specific steps include: Step 1: Construct a working face model without feedback from the inspection robot based on drilling, geophysical exploration, measurement, and realism data, and determine the "0" point position of the inspection robot and the working face model in the spatial coordinate system; rotate and translate the constructed working face model to the "0" point position to obtain the latest working face data x mian ,y mian 、z mian ; Step 2: An infrared transmitter is installed on the body of the inspection robot, and an infrared receiver is installed on each hydraulic support; The inspection robot obtains the displacement increment vector of the inspection robot, and obtains the position of the inspection robot at each moment through the displacement increment vector at each moment, that is, realizes the positioning of the inspection robot; Step 3: Divide the cut-eye coordinates in the working face model constructed in step 1 according to the number of hydraulic supports to obtain n coordinate vectors P with equal spacing. mian , P mian (t) is the coordinate vector on the working surface model corresponding to the hydraulic support position where the inspection robot is located at time t; Step 4: Calculate the average error E between the inspection robot's position after passing through n hydraulic supports and the cut-eye position of the working surface model ave ; Step 5: According to the average error calculated in step 4, the actual production allowable error is set as the error threshold. If the average error is greater than the error threshold, it is considered that correction is required. At this time, the coordinate vector P on the working surface model corresponding to the hydraulic support position where the inspection robot is located at time t is set to mian (t) takes the position P at time t xun (t) Replace and return to step 3; otherwise, the error is considered to be within the error range and no correction is required, and step 6 is executed; Step 6: Get the distance d between the two chute and the stop line 距 ; Step 7: The distances d from the two drifts to the stop line obtained in step 6 距 , and then fed back to the two chute models of the working face to update the working face model; at the same time, according to the P within the error range finally obtained in step 5 mian The data dynamically corrects the cutting data to show the undulating shape of the working surface.

2. The method for calibrating a working surface model of an inspection robot based on cloud computing as claimed in claim 1, characterized in that: In step 2, the calculation formula of the displacement increment vector of the inspection robot is as follows: S xun =C b *S b Where: S xun Represents the displacement increment vector of the inspection robot; C b represents the cosine matrix; S b represents the displacement increment; Where: θ, γ, The pitch angle, heading angle, and roll angle of the encoder position obtained after correction in the unified coordinate system; The position of the inspection robot at time t is: P xun (t)=P xun (t-1)+S xun (t) Where: P xun (t), P xun (t-1) represents the position of the inspection robot at time t and time t-1 respectively; S xun (t) represents the displacement increment vector at time t.

3. The method for calibrating a working surface model of an inspection robot based on cloud computing as claimed in claim 1, characterized in that: In step 4, the average error E ave The calculation formula is as follows:

4. The method for calibrating a working surface model of an inspection robot based on cloud computing as claimed in claim 1, characterized in that: In step 6, the distance d between the two drifts and the stop line 距 The calculation formula is as follows: Where: x' xun ,y' xun 、z' xun represents the coordinates of the inspection robot at both ends of the cut eye; x ting ,y ting 、z ting Represents the coordinates of the two end points of the stop mining line.

Images