Underground coal mine mobile equipment low-computing-power navigation method based on roadway morphological characteristics

Abstract

The invention relates to a low-computing-power navigation method for underground coal mine mobile equipment based on roadway morphological characteristics, belongs to the technical field of underground coal mine navigation, and aims to solve the problems that an existing laser radar scheme is high in computing power cost and an array ranging scheme is interfered by ground sundries. According to the technical scheme, 2N distance sensors are vertically arranged on the left side and the right side of a vehicle body to collect roadway top morphology information; establishing a vehicle body and roadway coordinate system, and calculating the offset and the offset angle of the vehicle body through a coordinate transformation matrix; based on the current offset and the historical offset, the vehicle body speed and the angular speed are dynamically updated through preset parameters; and the arc-shaped roadway is adapted and heightened through the multi-parameter optimization frame. The beneficial effects of the invention are that the system is resistant to ground debris interference; the computing power demand is low; the straightway efficiency and the curve safety are improved through dynamic speed regulation; and the universality is high.

Description

Technical Field

[0001] The present invention belongs to the field of coal mine underground navigation technology, and relates to a low-computing-power navigation method for mobile equipment in coal mines based on tunnel morphology characteristics. Background Art

[0002] The tunnel environment in coal mines is complex, and mobile equipment must precisely control lateral displacement to avoid collisions when moving within narrow tunnels. Existing technologies mainly rely on two types of solutions:

[0003] (1) LiDAR positioning solution

[0004] Navigation is achieved through high-precision point cloud data, but there are significant drawbacks:

[0005] The high cost of equipment increases the burden on enterprises;

[0006] Data processing relies on high-power computing equipment, but the explosion-proof requirements in underground coal mines limit hardware performance, and meeting the computing power requires huge investment.

[0007] (2) Array ranging sensor solution

[0008] A typical example is the channel robot navigation method proposed in patent CN115113629B, which calculates the yaw angle by combining lateral distance sensors arranged on both sides of the vehicle body. Although this method requires low computing power, it has fundamental limitations:

[0009] The sensor relies on ground reflection information. The ground in coal mines is often piled with debris (such as coal blocks and equipment debris), which causes distortion of the ranging signal.

[0010] After being disturbed, the lateral distance data cannot accurately reflect the offset relationship between the vehicle body and the centerline of the lane, which can easily lead to control failure.

[0011] None of the above solutions balance cost and interference resistance. LiDAR solutions are limited by computing power and cost, hindering widespread adoption. Array ranging solutions struggle to operate reliably due to the unique underground environment. Therefore, a low-computing navigation method based on novel sensing methods is urgently needed that can both avoid interference from ground debris and eliminate the need for complex computing equipment. Summary of the Invention

[0012] In light of this, the present invention aims to provide a low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel topography, addressing the high computing power requirements and insufficient computing power of the equipment required for underground LiDAR positioning and mapping. Existing channel robots use distance sensor arrays to determine vehicle information through lateral distance measurements. However, the accumulation of debris in coal mines significantly interferes with this method, making it difficult to apply to navigation within the complex tunnel environments of underground coal mines.

[0013] In order to achieve the above object, the present invention provides the following technical solutions:

[0014] A low-computing-power navigation method for mobile equipment in underground coal mines based on roadway topography features comprises the following steps:

[0015] S1: 2N vertical distance sensors 2 are evenly arranged from front to back on the left and right sides of the vehicle body 1 of the mobile equipment, where N ≥ 3;

[0016] S2: Set the vehicle coordinate system O C -x C y C z C and lane coordinate system O H -x H y H z H , where the origin of the vehicle coordinate system is located at the center of the vehicle and the plane where the sensor value is 0, y C The axis coincides with the front direction of the vehicle body, z C The axis is perpendicular to the ground; the origin of the roadway coordinate system is O H For too O C Draw a line perpendicular to the tunnel axis and parallel to the ground, and the midpoint of the line segment connecting the intersection of the left and right walls of the tunnel. H The axis coincides with the central axis of the roadway, z H The axis is perpendicular to the ground;

[0017] S3: Record the homogeneous coordinates of the roadway contour points measured by the distance sensor in the vehicle coordinate system C P;

[0018] S4: Obtain the homogeneous coordinates of the point in the roadway coordinate system through coordinate transformation H P, where the transformation matrix is:

[0019]

[0020] Among them, δ represents the lateral offset of the vehicle body relative to the center line of the roadway, and θ represents the offset angle of the vehicle body relative to the center line of the roadway;

[0021] S5: Based on the tunnel cross-section equation f(x H ,y H ,z H )=0, use H P calculates the offset δ and the offset angle θ;

[0022] S6: During vehicle navigation, the offset δ based on the current sampling time i and the offset angle θ i , the offset δ of the previous sampling moment i-1 and the offset angle θ i-1, update the vehicle velocity v i and angular velocity ω i :

[0023]

[0024] Among them, p1~p8 are preset information adjustment parameters;

[0025] S7: Determine a stop condition, where the stop condition includes: the distance measured by any distance sensor is less than a minimum threshold, or the vehicle receives a stop command.

[0026] Furthermore, in S1, the arrangement of the distance sensors 2 satisfies the following requirements: N sensors are arranged on the left and right sides, and the distance between the sensors is uniform; wherein, L a Indicates the distance between the front and rear sensors of the vehicle body, L b Indicates the distance between the left and right sensors of the vehicle body; C The expression of P is:

[0027]

[0028] Among them, L L1 ,L L2 ,L L3 is the distance measured by the left sensor, L R1 ,L R2 ,L R3 The distance measured by the right sensor;

[0029] In the S4, the H The expression of P is:

[0030]

[0031] Furthermore, in S5, the tunnel cross-section equation includes:

[0032] For curved roadways, the equation is:

[0033]

[0034] Among them, R H represents the roadway radius, ε H Indicates the roadway height offset, h C Indicates the vehicle height;

[0035] For elevated curved roadways, the equation is:

[0036]

[0037] Furthermore, in S5, calculating the offset δ and the offset angle θ includes:

[0038] Use the least squares method or parameter estimation method to solve implicit relationships;

[0039] or

[0040] Based on the small-angle rotation assumption, the explicit expression is solved using sensor data and average filtering is performed.

[0041] Furthermore, in said S6, the vehicle body speed v is updated i and angular velocity ω i Before, also includes:

[0042] Determine the initial state of the vehicle: all sensor signals are greater than the minimum acceptance threshold, initial velocity v1 and initial angular velocity ω1;

[0043] The sensor data at the i-th sampling moment is recorded at sampling time T, where the k-th sensor data on the left is The k-th sensor data on the right is k=1,2,...,N.

[0044] Furthermore, in S6, the information adjustment parameters p1 to p8 are determined by a multi-parameter optimization method, and the optimization steps include:

[0045] S21: Initialize the parameter combination and set the optimization boundary;

[0046] S22: Construct a variety of roadway training environments, including straight roads with equal width, curved left turns, and curved right turns;

[0047] S23: For each tunnel segment, calculate the evaluation score S of the parameter combination i :

[0048] If the vehicle body has not completed the lane travel, S i =-S r , where S r The percentage of the termination distance to the roadway length is mapped to a fraction from 0 to 50;

[0049] If the vehicle completes its journey, then the distance sequence between the center of the vehicle and the centerline of the lane is E = [e1, e2, ..., e M ], calculate the RMS And map to get S M and S E , then S i =-50-S M -S E ;

[0050] S24: Calculate the total evaluation score S=∑ i S i ;

[0051] S25: If S converges, the optimal parameters are determined; otherwise, the parameter combination is updated using a genetic algorithm, a particle swarm optimization algorithm, or a tuna algorithm, and S23 is repeated.

[0052] Furthermore, the p1 to p8 are configured to dynamically control the movement of the vehicle, specifically including:

[0053] The p1 to p4 are used to generate the speed adjustment, where p1 and p2 respond to the lateral offset δ i and the offset angle θ i , p3 and p4 response offset change rate δ i -δ i-1 and the rate of change of the offset angle θ i -θ i-1 , in order to adjust the vehicle speed according to the characteristics of the roadway shape;

[0054] The p5 to p8 are used to generate the angular velocity adjustment, where p5 and p6 respond to the lateral offset δ i and the offset angle θ i , the response offset change rate δ of p7 and p8 i -δ i-1 and the rate of change of the offset angle θ i -θ i-1 , in order to control the steering direction of the vehicle according to the characteristics of the roadway topography.

[0055] Furthermore, the S1 also includes arranging lateral distance sensors around the vehicle body for safety warning.

[0056] Furthermore, in S1, the vertical distance sensor calculates the offset by receiving the tunnel top topography information to avoid interference of tunnel ground debris on the distance measurement.

[0057] The beneficial effects of the present invention are:

[0058] (1) Vertical distance sensors are used to collect tunnel top topography information, completely avoiding interference from ground debris on the ranging signal. Existing lateral ranging solutions rely on ground reflection data, and accumulated coal blocks and equipment debris in coal mines can cause measurement distortion. However, this invention is based on continuous spatial modeling of the top contour, which makes the offset solution unaffected by the surface environment and ensures navigation reliability.

[0059] (2) It adopts a multi-parameter optimization preset and linear feedback control strategy, eliminating the need for real-time mapping or complex point cloud processing. While the LiDAR solution requires high-computing equipment to support point cloud matching, the present invention transfers the computational workload to the parameter optimization stage above ground, requiring only lightweight matrix operations (such as coordinate transformation and velocity update formulas) to be performed underground, adapting to the low computing power limitations of coal mine explosion-proof equipment.

[0060] (3) Dynamically adjust vehicle speed and steering based on lane topography. Traditional solutions move at a constant speed, while the present invention generates speed adjustments through offset and its rate of change:

[0061] Automatically increase the speed in straight road sections to shorten the machine moving time;

[0062] Actively reduce speed and correct angular velocity in curves or deviation risk areas to avoid collisions.

[0063] Realize intelligent efficiency control of "acceleration in good road conditions and steady driving in bad road conditions".

[0064] (4) Adapting to diverse scenarios through a multi-parameter optimization framework:

[0065] Compatible with various roadways: supports arc, heightened arc and other cross-section equation modeling;

[0066] Universal for multiple vehicles: The parameter optimization process automatically adapts to mobile equipment of different sizes;

[0067] Robust control guarantee: Combination of least squares method and average filtering to resist single-point sensor noise interference.

[0068] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0070] Figure 1 A three-dimensional diagram illustrating the sensor layout;

[0071] Figure 2 Schematic diagram of the information size of each sensor;

[0072] Figure 3 This is a schematic diagram of the information of each sensor in the "arc-shaped" lane;

[0073] Figure 4 This is a schematic diagram of the information of each sensor in the "elevated arc" type laneway;

[0074] Figure 5 This is a schematic diagram of the vehicle coordinate system offset;

[0075] Figure 6 This is a schematic diagram of a lane with a small right turn angle;

[0076] Figure 7This is a flow chart of vehicle navigation based on the longitudinal distance sensor layout;

[0077] Figure 8 Determine the logic diagram for the parameters of vehicle body speed update;

[0078] Figure 9 This is a flow chart of the calculation method for the speed update parameter evaluation system indicators.

[0079] Reference numerals: 1 - vehicle body, 2 - distance sensor, 3 - lane. DETAILED DESCRIPTION

[0080] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0081] Among them, the accompanying drawings are only for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting the present invention. In order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the dimensions of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the accompanying drawings.

[0082] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", "front", "back", etc. indicating directions or positional relationships, they are based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0083] See also Figures 1 to 9 On the left and right sides of the mobile equipment, 2N (N≥3) distance sensors 2 are evenly arranged from front to back, and the measurement direction of the distance sensors is the vertical direction. (In addition, lateral distance sensors can be arranged on all sides to measure the distance from the vehicle body 1 for safety warning. It is better to have lateral distance measurement on all sides, which seems more reasonable, but it has no impact on the core idea of ​​this solution).

[0084] The vehicle body is traveling in lane 3. Through the signals of the sensors arranged in the vertical direction, the offset and offset angle can be obtained by using the coordinate system transformation idea. Figures 1 to 5 Taking the arrangement of 6 sensors as an example, the steps for solving from signal to offset are as follows:

[0085] S1: Set coordinate system: Set vehicle coordinate system O C -x C y C z C like Figure 1 As shown, the origin is located at the center of the vehicle body and the sensor value is 0, y C The axis coincides with the front direction of the vehicle body, z C The axis direction is perpendicular to the ground. Set the roadway coordinate system O H -x H y H z H , over O C Draw a straight line perpendicular to the axis of the tunnel and parallel to the ground, and the midpoint of the line segment connecting the intersection points of the left and right walls of the tunnel is the origin O. H ,y H The axis coincides with the central axis of the roadway, y H The axis is perpendicular to the ground. Figure 5 As shown, from O H -x H y H z H to O C -x C y C z C The transformation along x H The axis is translated by δ, and then around the translated z H Axis rotation f(x H ,y H , z H )=0, then the corresponding coordinate system O H -x H y H z H To coordinate system O C -x C y C z C The transformation matrix is:

[0086]

[0087] S2: Record the 6 points of the roadway profile measured by the 6 sensors in the vehicle coordinate system O C -x C y C z C The homogeneous coordinates in , that is,

[0088]

[0089] Through coordinate transformation, we get 6 points in the lane coordinate system O H -x H y H z H The homogeneous coordinates in

[0090] H P=T H→C · C P

[0091]

[0092] S3: Knowing the shape of the tunnel section, and using the infinitesimal idea, assuming that the L before and after the coordinate system a The length is a straight segment, then we can get H -x H y H z H Down, The equation of the segment is:

[0093] f(x H ,y H , z H )=0

[0094] by Figure 3 Taking the “arc” roadway shown as an example, its equation is:

[0095] z H ∈[-h C , R H -ε H -h C ]

[0096] |x H |≤R H , z H ∈[0, R H ]

[0097] by Figure 4 Taking the “elevated arc” tunnel shown as an example, in reality only the complete semicircle is the effective measurement area, so its equation is:

[0098] z H ∈[ε H -h C , R H +ε H -h C ]

[0099] S4: Substitute the coordinates of the six points in the roadway coordinate system obtained in S2 into the equation in S3 to obtain the implicit relationship between the offset and the deflection angle. Using the least squares method or other parameter estimation methods, the estimated values ​​of the offset and the deflection angle can be obtained; or it is reasonable to assume that the vehicle body offset is small and the angle of rotation is small, that is, the three sensors on the left are in the coordinate system O H x H y H z H The horizontal coordinate is always less than 0. The three sensors on the right are in the coordinate system O H -x H y H z H If the horizontal axis of is always greater than 0, two of the six sensors can be used to solve the displayed expressions for the offset and offset angle, and the remaining four sensors can be used to solve them together. The obtained offsets and offset angles are then averaged and filtered. Finally, the offset δ and offset angle θ can be calculated using parameter estimation or the small angle hypothesis method.

[0100] In the specific implementation, when the robot receives a forward or backward command, the drilling robot can use the signals collected by the arranged sensors to plan the path and control the feedback to realize autonomous forward or backward movement along the center of the tunnel until it receives a stop command or reaches an inaccessible area.

[0101] Taking forward autonomous assisted navigation walking as an example, the specific steps include:

[0102] S11: Determine the initial state of the robot: the position where the signals of all sensors are greater than the minimum acceptance threshold, the initial velocity v1, and the initial angular velocity ω1.

[0103] S12: Let the sampling time be T, and record the data of 2N distance measuring sensors at the i-th sampling time. Among them, the data measured by the distance sensor in the k-th row on the left is The data measured by the distance sensor in the kth row on the right is k=1,2,...,N。When i=1,the above method is used to calculate the lateral offset δ1 and the offset angle θ1 of the vehicle body relative to the center line of the lane, and the vehicle moves forward with the initial velocity v1; when i≥2,the lateral offset δ at the i-th sampling moment is calculated. i , offset angle θ i , and the offset δ at the i-1th sampling moment i-1 , offset angle θ i-1 The data is combined to judge the degree of deviation and the speed of deviation, and the speed and angular velocity at the i-th sampling moment are updated accordingly.

[0104] One of the feasible methods for updating speed and angular velocity is as follows:

[0105]

[0106] Among them, p1, p2, ..., p8 represent information adjustment parameters; p1, p2, p3, p4 represent speed adjustment coefficients according to the offset value, and it is expected that the speed size is adjusted according to the offset situation to achieve increased efficiency; p5, p6, p7, p8 represent angular velocity adjustment coefficients according to the offset value, and the positive and negative and size of the angular velocity are determined according to the offset situation; p1, p2, p5, p6 represent proportional factors of the offset value; p3, p4, p7, p8 represent proportional factors of the differential of the offset value, and represent adjustments to the rate of change.

[0107] S13: Determine whether the stop condition has been met. If so, the process ends. If not, the process continues with S12. The stop condition can be one of the following: ① Any distance sensor is less than the minimum threshold (indicating an impending / imminent collision); ② The vehicle receives a stop command.

[0108] In addition, the invention should also include a method for solving parameters p1, p2, ..., p8, and the specific optimization steps are as follows:

[0109] S21: Initialize the parameter combination and determine the optimization boundary of the parameters.

[0110] S22: Calculate an evaluation score obtained by the parameter combination, where the evaluation score takes into account the degree of walking completion, the degree of deviation from the center during successful completion, and the speed of successful completion.

[0111] One possible evaluation system is as follows:

[0112] Construct a variety of lane sections, including straight roads with equal width, left turns with different radii, right turns with different radii, etc. For a specific parameter combination, under the i-th specific lane section,

[0113] S31: If the current lane cannot be completed smoothly, the basic score is recorded as 0, and the percentage of the ending distance to the lane length is mapped to a range of 0 to 50 points, recorded as S r , then the fraction of lanes recorded in the i-th lane is:

[0114] S i =-S r

[0115] S32: If the process is completed successfully, record the sequence of the distance between the vehicle center and the lane centerline at the completion time.

[0116] E=[e1,e2,…,e M-1 , e M ] T

[0117] Among them, e i It represents the distance between the center of the vehicle and the center line of the lane at the moment, M represents the number of moments from the beginning to the end, and the root mean square of the sequence E is recorded.

[0118]

[0119] Then construct a mapping, map M to the range of 0 to 50, and record it as S M , E RMS Construct a geometric mapping to the range of 0 to 50, recorded as S E , then the current evaluation score is recorded as

[0120] S i =-50-S M -S E

[0121] After training all the typical tunnel sections, the evaluation score of the current parameter combination can be calculated as

[0122]

[0123] S33: Determine whether the evaluation score converges. If so, determine that the current parameters are the optimal parameters and end. If not, update the parameter combination according to a multi-parameter optimization method such as a genetic algorithm, a particle swarm optimization algorithm, or a tuna algorithm, and continue to execute S32.

[0124] The above-mentioned vehicle body sensor arrangement scheme, vehicle body navigation framework, and weight coefficient determination method together form the low-computing power navigation method based on lane morphology features of the present invention.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology features, characterized by: The following steps are involved: S1: 2N vertical distance sensors (2) are evenly arranged on the left and right sides of the vehicle body (1) of the mobile equipment from front to back, where N ≥ 3; S2: Set the vehicle coordinate system O C -x C y C z C and lane coordinate system O H -x H y H z H , where the origin of the vehicle coordinate system is located at the center of the vehicle and the plane where the sensor value is 0, y C The axis coincides with the front direction of the vehicle body, z C The axis is perpendicular to the ground; the origin of the roadway coordinate system is O H For too O C Draw a line perpendicular to the tunnel axis and parallel to the ground, and the midpoint of the line segment connecting the intersection of the left and right walls of the tunnel. H The axis coincides with the central axis of the roadway, z H The axis is perpendicular to the ground; S3: Record the homogeneous coordinates of the roadway contour points measured by the distance sensor in the vehicle coordinate system C P; S4: Obtain the homogeneous coordinates of the point in the roadway coordinate system through coordinate transformation H P, where the transformation matrix is: Among them, δ represents the lateral offset of the vehicle body relative to the center line of the roadway, and θ represents the offset angle of the vehicle body relative to the center line of the roadway; S5: Based on the tunnel cross-section equation f(x H ,y H ,z H )=0, use H P calculates the offset δ and the offset angle θ; S6: During vehicle navigation, the offset δ based on the current sampling time i and the offset angle θ i , the offset δ of the previous sampling moment i-1 and the offset angle θ i-1 , update the vehicle velocity v i and angular velocity ω i : Among them, p1~p8 are preset information adjustment parameters; S7: Determine a stop condition, where the stop condition includes: the distance measured by any distance sensor is less than a minimum threshold, or the vehicle receives a stop command.

2. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S1, the arrangement of the distance sensors (2) satisfies the following requirements: N sensors are arranged on the left and right sides, and the distance between the sensors is uniform; wherein, L a Indicates the distance between the front and rear sensors of the vehicle body, L b Indicates the distance between the left and right sensors of the vehicle body; C The expression of P is: Among them, L L1 ,L L2 ,L L3 is the distance measured by the left sensor, L R1 ,L R2 ,L R3 The distance measured by the right sensor; In the S4, the H The expression of P is:

3. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S5, the tunnel cross-section equation includes: For curved roadways, the equation is: Among them, R H represents the roadway radius, ε H Indicates the roadway height offset, h C Indicates the vehicle height; For elevated curved roadways, the equation is:

4. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S5, calculating the offset δ and the offset angle θ includes: Use the least squares method or parameter estimation method to solve implicit relationships; or Based on the small-angle rotation assumption, the explicit expression is solved using sensor data and average filtering is performed.

5. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S6, the vehicle body speed v is updated. i and angular velocity ω i Before, also includes: Determine the initial state of the vehicle: all sensor signals are greater than the minimum acceptance threshold, initial velocity v1 and initial angular velocity ω1; The sensor data at the i-th sampling moment is recorded at sampling time T, where the k-th sensor data on the left is The k-th sensor data on the right is k=1,2,...,N.

6. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S6, the information adjustment parameters p1 to p8 are determined by a multi-parameter optimization method, and the optimization steps include: S21: Initialize the parameter combination and set the optimization boundary; S22: Construct a variety of roadway training environments, including straight roads with equal width, curved left turns, and curved right turns; S23: For each tunnel segment, calculate the evaluation score S of the parameter combination i : If the vehicle body has not completed the lane travel, S i =-S r , where S r The percentage of the termination distance to the roadway length is mapped to a fraction from 0 to 50; If the vehicle completes its journey, then the distance sequence between the center of the vehicle and the centerline of the lane is E = [e1, e2, ..., e M ], calculate the RMS And map to get S M and S E , then S i =-50-S M -S E ; S24: Calculate the total evaluation score S=∑ i S i ; S25: If S converges, the optimal parameters are determined; otherwise, the parameter combination is updated using a genetic algorithm, a particle swarm optimization algorithm, or a tuna algorithm, and S23 is repeated.

7. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: The p1 to p8 are configured to dynamically control the movement of the vehicle, specifically including: The p1 to p4 are used to generate the speed adjustment, where p1 and p2 respond to the lateral offset δ i and the offset angle θ i , p3 and p4 response offset change rate δ i -δ i-1 and the rate of change of the offset angle θ i -θ i-1 , in order to adjust the vehicle speed according to the characteristics of the roadway shape; The p5 to p8 are used to generate the angular velocity adjustment, where p5 and p6 respond to the lateral offset δ i and the offset angle θ i , the response offset change rate δ of p7 and p8 i -δ i-1 and the rate of change of the offset angle θ i -θ i-1 , in order to control the steering direction of the vehicle according to the characteristics of the roadway topography.

8. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: Said S1 also includes arranging lateral distance sensors around the vehicle body for safety warning.

9. The low-computing-power navigation method for mobile equipment in underground coal mines based on tunnel morphology characteristics according to claim 1 is characterized by: In S1, the vertical distance sensor calculates the offset by receiving the tunnel top topography information to avoid interference of tunnel ground debris on the distance measurement.

Images