WO2022183871A1

WO2022183871A1 - Construction method for dynamic envelope line for electric locomotive for mine railway

Info

Publication number: WO2022183871A1
Application number: PCT/CN2022/073337
Authority: WO
Inventors: 郑昌陆; 郑益飞; 张华�
Original assignee: 上海申传电气股份有限公司
Priority date: 2021-03-04
Filing date: 2022-02-10
Publication date: 2022-09-09
Also published as: CN113031602A; CN113031602B

Abstract

A construction method for a dynamic envelope line for an electric locomotive for a mine railway, comprising the following steps: a detection system on construction equipment senses environmental information by means of dynamic scanning (S1); an autonomous driving controller performs imaging with respect to the scanned environmental information (S2); the autonomous driving controller performs track recognition by means of dynamic imaging, delineates a safe area for operating forward, and feeds general information back to a whole vehicle controller (S3); and the whole vehicle controller performs autonomous driving by combining a vehicle traveling target and a forward safety envelope line area (S4). This solves the existing problem of the instability of a dynamic imaging technique for an electric locomotive operating track in shaft of a coal mine, the reliance on a track model established by employing a regular scheme, a slow processing speed, the incapability to functionally divide an operational envelope line; and at the same time, the susceptibility of an electric locomotive to bumping due to operational instability caused by fallen gravels and slags on the track in the coal mine.

Description

A Construction Method of Dynamic Envelope of Mine Rail Electric Locomotive

technical field

The invention relates to the field of constructing the dynamic envelope of a rail electric locomotive, in particular to a method for constructing the dynamic envelope of a mining rail electric locomotive.

Background technique

Mine rail locomotives are the main mode of auxiliary transportation. At present, coal mine rail locomotives completely rely on drivers to control the vehicles, and human factors such as driver fatigue can easily lead to safety problems such as non-subjective speeding, running red lights, and rear-end collisions. According to statistics, auxiliary transportation accidents are one of the three major accidents in mines. The number of roadway transportation accidents accounts for 42.14% of the total transportation accidents, and the number of casualties is 41.12%. The electric locomotive accident ranks first in the roadway transportation accidents. Therefore, it is urgent to develop unmanned electric locomotives to reduce the number of people and increase efficiency in coal mines, and to improve the safety of equipment operation, and the unmanned electric locomotives need to construct a dynamic envelope to automatically detect the coal mine track conditions.

However, there are still some defects in the construction process of the dynamic envelope of the existing mining rail locomotives. The dynamic imaging technology of the running track of the underground locomotive in coal mines is unstable. The envelope can not be divided into functions; at the same time, the electric locomotive on the coal mine track is easily unstable and bumpy due to the falling of gravel and slag. On the one hand, it is easy to cause the collected data and imaging information to be inaccurate; There are certain safety hazards due to the derailment of the electric locomotive.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for constructing the dynamic envelope of a mining rail electric locomotive, which can solve the instability of the dynamic imaging technology of the existing coal mine underground electric locomotive running track. It is slow, and the operation envelope cannot be divided into functions; at the same time, the electric locomotive on the coal mine track is prone to unstable operation due to falling gravel and slag. On the other hand, it is easy to cause the derailment of the electric locomotive and there are certain safety hazards.

The object of the present invention can be realized through the following technical solutions:

A construction method for a dynamic envelope of a mining rail electric locomotive, the construction method comprising the following steps:

S1. The detection system on the construction device perceives environmental information through dynamic scanning;

S2. The automatic driving controller images the scanned environmental information;

S3. The automatic driving controller performs track identification through dynamic imaging, demarcates the safe area for running ahead, and feeds the comprehensive information back to the vehicle controller;

S4. The vehicle controller performs automatic driving in combination with the vehicle driving target and the front safety envelope area; wherein,

The construction device further includes a body, a cab and a moving mechanism, the cab is fixed on the top of the body, a power supply is installed on the body, the detection system is fixed on the cab, and the moving mechanism is fixed on the body. At the bottom, the detection system includes a lidar sensor, a millimeter-wave radar, and a processing module, and the processing module is electrically connected to the lidar sensor and the millimeter-wave radar; a sand spreader is installed on the body, and a sand spreader is installed on the outside of the cab buffer.

Preferably, when the ground vehicle in S1 performs environmental perception through the lidar sensor, a stable roadway point cloud model is obtained by superimposing continuous multi-frame point cloud data, and on this basis, the track position is detected, and finally the track is carried out. For the detection of nearby obstacles, according to the detection results, the safe area running ahead is demarcated and the corresponding response mechanism is executed, as follows:

Step 1: Establish a stable roadway model

In the vehicle-mounted automatic driving controller, the model is based on the ROS system, and the real-time data of the point cloud of the laser sensor is collected through the PCL point cloud library function. Sequence; the scanning frequency of the laser sensor is 10Hz, and 100ms can image a point cloud image. The superposition of consecutive multi-frame point cloud images is to superimpose multiple point cloud images according to the order of the point cloud data to form a denser image. Point cloud image; the laser sensor has a built-in IMU sensor. The IMU provides a relative positioning information. Its function is to measure the route relative to the starting point object. The main parameters provided are x, y, z, roll, pitch, yaw , according to these parameters, a three-dimensional space composed of point clouds can be simulated;

Through the communication methods such as can bus, the laser sensor obtains the real-time driving speed of the electric locomotive. According to the driving speed of the electric locomotive at a certain moment, the relative positional relationship between the previous frame and the next frame of the point cloud image can be determined, and the point cloud image of the latter frame can be adjusted and adjusted. It is superimposed on the point cloud image of the previous frame, and by analogy, multiple frames of continuous point cloud images can be superimposed to form a relatively stable and dense point cloud image;

After superimposing the point cloud data, the model restores the point cloud image of the roadway more realistically. When frequent bumps and jitters occur during the driving of the electric locomotive, a virtual three-dimensional space is established, and the plane of the three-dimensional space z=0 is determined as the actual ground; The least squares method is used to fit the point cloud data of a certain vertical plane of a single frame of roadway point cloud image, and the slope of the regression line relative to the virtual plane and the intercept in the vertical direction are obtained, and adjusted according to the slope and intercept. All the point cloud data of the single frame are made to coincide with the virtual plane, and then the adjusted frame data is superimposed, and finally a stable and clear roadway model is obtained;

First set the regression line equation:

z=kx+b

The slope is k, and the y-axis intercept is b; since the ground line cannot be perpendicular to the x-axis, the intercept line equation can be used as the regression line equation; the point set of a vertical plane of a single frame of point cloud image for:

{(x,z)|(x 1,z 1),(x 2,z 2),...(x n,z n)}

Find the sum of squared errors from all points to the line:

It can be seen from the extreme value theorem that the first derivative of the error equation is equal to 0 to obtain the extreme value, so it is derived with respect to k and b respectively, and the values of k and b are solved to make the error function take the minimum value; it can be obtained:

After finishing, you can get:

in

After finding the values of k and b, bring them into the regression equation to get the fitted straight line equation; according to the slope k, adjust the angle of this frame of point cloud image to make it horizontal to the virtual plane; according to the intercept b, translate the point The cloud image is at the same level as the virtual plane; in this way, each frame of point cloud image obtained is at the same level to achieve the maximum degree of overlap, and a pair of consecutive multi-frame superimposed, stable and clear roadway point cloud images are obtained;

Step 2: Track Feature Extraction Model

According to the actual situation, there are two notable features of the underground track, one is that the height of the track is significantly higher than the ground on both sides of the track, and the height difference is 10cm; the other is that the reflection intensity of the track is significantly lower than the ground;

According to the above two features, a model is established, two two-dimensional arrays are established corresponding to the actual ground, and the plane of the three-dimensional space z=0 is gridded. The value of one two-dimensional array gridCell is the point cloud z value, which corresponds to the actual position. Height; the value of another two-dimensional array gridintensity is the curvature value of the point cloud, corresponding to the reflection intensity of the actual position; the grid resolution is set to 8cm;

The amount of superimposed point cloud data is huge, and limiting the detection range can greatly improve the running speed; since the track width is 0.6m and the track is generally within a certain range in front of the track, the width of the detection range is set to 0.8m around the center of the point cloud image, corresponding to Since the grid is 20 grids in the center of a certain row of the image; analyzing the values of two arrays of 20 grids in the center of a certain row, the grid position corresponding to the gridCell value of 0.1 is considered as the track position; the gridintensity value of the array is obviously smaller than the corresponding grid position. The grid position corresponding to the adjacent array value is considered as the track position; the analysis results of the two arrays are integrated to obtain the track position of a row of the grid; and so on, when the electric locomotive is running, each row in the grid is calculated cyclically track position, and finally identify a track in the point cloud image;

Step 3: Obstacle Recognition

Through the track feature extraction model, after the track is correctly identified, obstacles are detected within a certain range near the track; a certain range is delineated near the grid corresponding to the track, and it is detected whether the gridCell array corresponding to the range has a z value, if The z value is greater than the track and less than the height of the vehicle, it is determined as an obstacle; then the distance between the obstacle and the electric locomotive is estimated according to the x value of the obstacle, and the electric locomotive makes a corresponding braking response by communicating with the motor drive system .

Preferably, the moving mechanism includes a base frame, four wheels, and two side frames, the side frames are fixed on the outer walls of both sides of the base frame, and the wheels are installed at the bottom of the base frame between two adjacent wheels. Connected by a rotating shaft, a transmission gear is installed in the middle of the two rotating shafts, a drive gear is meshed with one side of the two transmission gears, and a transmission shaft is penetrated in the middle of the drive gear. Both ends of the drive shaft are connected to a blower fixed at the bottom of the chassis, and the drive shaft is connected to the inside of the blower and connected with the fan blades located inside the blower, and a fan fixed to the bottom frame is provided between the two blowers. a driving motor at the bottom, the output shaft of the driving motor is connected with a driving gear, and the driving gear is meshed with a meshing gear sleeved on the transmission shaft;

Preferably, the wheel includes a rim and a hub body, the rim is located on the outer walls of both sides of the hub body, the outer wall of the hub is provided with a number of evenly distributed air holes, and the rim is close to the side wall of the hub body A number of evenly distributed side holes are arranged on the side wall, a middle cavity is arranged inside the hub body, and a cavity inside the hub body is arranged on one side of the middle cavity, and the cavity is communicated with the middle cavity The middle cavity is communicated with the air hole, and a pneumatic telescopic tube is installed inside each side hole, the pneumatic telescopic tube is communicated with the cavity, and the blower pipe is connected to the middle cavity.

Preferably, a plurality of shock-absorbing springs are fixed on the top of each side frame, the top of the shock-absorbing springs are connected to the bottom of the vehicle body, and two sides of each side frame are provided with a shock-absorbing spring fixed on the side wall of the chassis. A side support, each side support side wall is connected with a connecting rod, the top end of the connecting rod is fixed on the bottom frame side wall, and a plurality of side springs are installed inside each side support.

Preferably, one side of each of the wheels is provided with a brake disc sleeved on the rotating shaft, one side of each of the brake discs is provided with a pneumatic brake for use with it, and the pneumatic brake pipeline is connected at the bottom Cylinder at the bottom of the rack.

Beneficial effects of the present invention: due to the setting of the blower at the bottom of the chassis, and the drive shaft is connected to the inside of the blower and connected with the fan blades located inside the blower, the blower can be driven by the drive shaft to operate, so that the blower can move when the moving mechanism is running. Automatic operation, no need for a separate power source.

Since the rim of the wheel is provided with a side hole with a pneumatic telescopic tube inside, and the hub body of the wheel is provided with an air hole, the blower can send gas into the middle cavity during operation. Since the middle cavity is connected with the cavity, Therefore, the gas entering the middle cavity enters the cavity at the same time, and the gas entering the middle cavity is discharged from the air hole, so that the moving mechanism can use the gas discharged from the wheels to clean the track when it is in operation, so that it can be dropped on the track. The gravel slag on the track is cleaned to avoid the inaccurate image information detected by the detection system due to bumps. At the same time, the gas entering the cavity enters the inside of the pneumatic expansion tube and makes the pneumatic expansion tube eject from the inside of the side hole. Due to the side hole It is located on the side wall of the rim, and the track has a T-shaped structure, so that the pneumatic telescopic tube ejected from the inside of the side hole can be extended to the position below the top of the track, forming a protective structure that can prevent the wheel from derailing and ensuring the detection result of the detection system. At the same time, the present invention adjusts all point cloud data of the single frame according to the slope and intercept to make it coincide with the virtual plane, and then superimposes the adjusted data of this frame to finally obtain a stable and clear roadway Model, the dynamic imaging technology of the running track of the electric locomotive in the coal mine is more stable and efficient, and the operation envelope can be quickly divided into functions.

Description of drawings

In order to facilitate the understanding of those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

Fig. 1 is the module diagram of the present invention;

Fig. 2 is the front view of the overall structure of the construction device of the present invention;

3 is a top view of the overall structure of the construction device of the present invention;

Fig. 4 is the left side view of the overall structure of the construction device of the present invention;

Fig. 5 is the structural schematic diagram of the moving mechanism of the present invention;

6 is a side view of the moving mechanism of the present invention;

Fig. 7 is the top view of the moving mechanism of the present invention;

Figure 8 is a side view of the wheel of the present invention;

9 is a schematic diagram of the internal structure of the wheel of the present invention;

FIG. 10 is a region diagram of constructing a safe driving envelope according to the present invention.

In the picture: 1. body; 1a, sand spreader; 2, cab; 2a, buffer; 3, power supply; 4, detection system; 5, moving mechanism; 6, chassis; 7, wheel; 8, shaft; 9. Transmission gear; 10. Side spring; 11. Side support; 12. Connecting rod; 13. Side frame; 14. Shock spring; 15. Drive motor; 16. Brake disc; 17. Blower; 18. Driving gear ; 19, meshing gear; 20, drive gear; 21, transmission shaft; 22, pneumatic brake; 23, air hole; 24, side hole; 25, pneumatic telescopic tube; 26, cavity; 27, middle cavity.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Please refer to Figure 1-10, a method for constructing the dynamic envelope of a mining rail locomotive, the construction method includes the following steps:

S1, the detection system 4 on the construction device perceives environmental information through dynamic scanning;

When the invention constructs the safe driving envelope, according to the characteristic that the maximum safe braking distance of the mining rail electric locomotive is 40 meters, the operating envelope area of 60 meters in front of the electric locomotive is demarcated;

(1) In the range of 40-60 meters ahead, it is an early warning area. When there are pedestrians or obstacles in the area, the driverless control will issue an early warning message, the locomotive will make various preparations for deceleration, and the driving speed will issue a deceleration command. The speed shall not be higher than 3m/s;

(2) The area within 20-40 meters ahead is the deceleration area. When there are pedestrians or obstacles in the area, the unmanned controller will issue a deceleration command, and the speed of the locomotive shall not be higher than 1m/s;

(3) In the range of 10-20 meters in front, it is the braking area. When there are pedestrians or obstacles in this area, the unmanned controller issues a braking command, and the locomotive needs to decelerate to 0;

(4) The area within 0 to 10 meters ahead is an emergency parking area. When there are pedestrians or obstacles in the area, the driverless controller will issue an emergency braking command, and the locomotive must stop immediately;

The construction device further includes a body 1, a cab 2 and a moving mechanism 5. The cab 2 is fixed on the top of the body 1, a power supply 3 is installed on the body 1, and the detection system 4 is fixed on the cab 2. On the upper side, the moving mechanism 5 is fixed on the bottom of the vehicle body 1, and the detection system 4 includes a lidar sensor, a millimeter-wave radar and a processing module, and the processing module is electrically connected to the lidar sensor and the millimeter-wave radar; There is a sand spreader 1a, and a shock absorber 2a is attached to the outside of the cab 2.

Specifically, when the ground vehicle in S1 performs environmental perception through the lidar sensor, a stable roadway point cloud model is obtained by superimposing multiple frames of point cloud data, and on this basis, the track position is detected, and finally the track is carried out. For the detection of nearby obstacles, according to the detection results, the safe area running ahead is demarcated and the corresponding response mechanism is executed, as follows:

Step 1: Establish a stable roadway model

First set the regression line equation:

z=kx+b

{(x,z)|(x 1,z 1),(x 2,z 2),...(x n,z n)}

Find the sum of squared errors from all points to the line:

After finishing, you can get:

in

Step 2: Track Feature Extraction Model

Step 3: Obstacle Recognition

The moving mechanism 5 includes a base frame 6, four wheels 7, and two side frames 13. The side frames 13 are fixed on the outer walls of both sides of the base frame 6, and the wheels 7 are installed at the bottom of the base frame 6, between two adjacent wheels 7. Connected by a rotating shaft 8, a transmission gear 9 is installed in the middle of the two rotating shafts 8, and a drive gear 20 is meshed with one side of the two transmission gears 9. Both ends of the shaft 21 are connected to a blower 17 fixed at the bottom of the chassis 6, and the transmission shaft 21 is connected to the inside of the blower 17 and is connected with the fan blades located inside the blower 17, and a fixed at the inside of the blower 17 is provided between the two blowers 17. The drive motor 15 at the bottom of the chassis 6, the output shaft of the drive motor 15 is connected to a driving gear 18, and the driving gear 18 meshes with a meshing gear 19 sleeved on the transmission shaft 21;

The wheel 7 includes a rim and a hub body, the rim is located on the outer walls on both sides of the hub body, the outer wall of the hub is provided with a number of evenly distributed air holes 23, and the side wall of the rim close to the hub body is provided with a number of evenly distributed There is a side hole 24 inside the hub body, a middle cavity 27 is arranged inside the hub body, and a cavity 26 located inside the hub body is arranged on one side of the middle cavity 27, the cavity 26 is communicated with the middle cavity 27, and the middle cavity 27 is communicated with the air hole 23. A pneumatic telescopic tube 25 is installed inside each side hole 24 , the pneumatic telescopic tube 25 is communicated with the cavity 26 , and the blower 17 is connected to the middle cavity 27 by the pipeline.

A number of shock-absorbing springs 14 are fixed on the top of each side frame 13 , the top of the shock-absorbing springs 14 is connected to the bottom of the vehicle body 1 , and a side brace 11 fixed on the side wall of the bottom frame 6 is provided on both sides of each side frame 13 A connecting rod 12 is connected to the side wall of each side support 11 , the top end of the connecting rod 12 is fixed on the side wall of the chassis 6 , and several side springs 10 are installed inside each side support 11 .

One side of each wheel 7 is provided with a brake disc 16 sleeved on the rotating shaft 8 , and one side of each brake disc 16 is installed with a pneumatic brake 22 for use with it. of the cylinder.

The operation steps of the construction device of the present invention are as follows:

Step 1: After starting the whole device, the two driving motors 15 inside the moving mechanism 5 both operate, and the driving motor 15 is used to drive the driving gear 18 to rotate. In the process of rotation, the shaft 21 uses the drive gear 20 to drive the transmission gear 9 to rotate, thereby driving the wheel 7 to rotate, so as to realize the position movement of the moving mechanism 5, and the moving mechanism 5 carries the entire device during the position movement process. In the moving position, the transmission shaft 21 is connected to a blower 17 fixed at the bottom of the bottom frame 6 during the rotation of the transmission shaft 21, and the transmission shaft 21 is connected to the inside of the blower 17 and is connected with the fan located inside the blower 17. The blades are connected so that the blower 17 is driven by the transmission shaft 21 to operate. The blower 17 sends gas into the middle cavity 27 during operation, and the gas entering the middle cavity 27 enters the cavity 26 at the same time, and the gas entering the middle cavity 27. It is discharged from the air hole 23, so that when the moving mechanism 5 is in operation, the gas discharged from the wheel 7 is used to clean the track, and the crushed stone slag falling on the track is cleaned, and the gas entering the cavity 26 enters the pneumatic Inside the telescopic tube 25, the pneumatic telescopic tube 25 is pushed out from the inside of the side hole 24. With the operation of the moving mechanism 5, the vehicle body 1 moves forward. At this time, the detection system 4 is activated to detect the internal environment of the mine;

Step 2: Use the lidar sensor and the millimeter-wave radar to perceive the environmental information through dynamic scanning; the automatic driving controller inside the processing module images the environmental information scanned by the lidar and the millimeter-wave radar. Point cloud segmentation is performed on the passable area and the impassable area, so as to plan the driving path and detect obstacles. The underground track of the coal mine is different from the surface road, and the electric locomotive runs on a fixed track. There is no problem of path detection. Nearby obstacle detection is carried out to delineate the safe area for running ahead; the automatic driving controller performs track identification through dynamic imaging, and demarcates the safe area running ahead, that is, the safety envelope area, and feeds back the comprehensive information to the driver. The whole vehicle controller in room 2 is completed to complete the construction. After the construction is completed, the whole vehicle controller combines the vehicle driving target and the front safety envelope area to perform automatic driving.

When the present invention is in use, after the entire device is started, the two drive motors 15 inside the moving mechanism 5 operate, and the drive motor 15 is used to drive the driving gear 18 to rotate, and the driving gear 18 uses the meshing gear 19 to drive the transmission shaft during the rotation. 21 rotates, and the transmission shaft 21 uses the drive gear 20 to drive the transmission gear 9 to rotate during the rotation, thereby driving the wheel 7 to rotate, so as to realize the position movement of the moving mechanism 5, and the moving mechanism 5 carries the entire device during the position movement process. The position is moved on the mining track, and the transmission shaft 21 is connected to a blower 17 fixed at the bottom of the bottom frame 6 during the rotation of the transmission shaft 21 because both ends of the transmission shaft 21, and the transmission shaft 21 is connected to the inside of the blower 17 and is located in the blower. The fan blades inside 17 are connected, so that the blower 17 is driven by the transmission shaft 21 to operate. Since the blower 17 is connected to the middle cavity 27 by the pipeline, the blower 17 can send gas into the middle cavity 27 during operation. The cavities 26 are connected, so that the gas entering the interior of the intermediate cavity 27 enters the interior of the cavity 26 at the same time, and the gas entering the interior of the intermediate cavity 27 is discharged from the air hole 23, so that the moving mechanism 5 can use the gas discharged from the wheel 7 when it is in operation. To clean the track, the gravel and slag falling on the track can be cleaned to avoid the inaccurate image information detected by the detection system 4 due to bumps. At the same time, the gas entering the cavity 26 enters the pneumatic telescopic tube 25. And the pneumatic telescopic tube 25 is pushed out from the inside of the side hole 24. Since the side hole 24 is located on the side wall of the rim and the track has a T-shaped structure, the pneumatic telescopic tube 25 pushed out from the inside of the side hole 24 can extend to the top of the track. The lower position forms a protective structure that can prevent the wheels from derailing. With the operation of the moving mechanism 5, the body 1 moves forward. At this time, the detection system 4 is activated to detect the internal environment of the mine, using lidar sensors and millimeter-wave radars. Perceives environmental information through dynamic scanning; the automatic driving controller inside the processing module images the environmental information scanned by lidar and millimeter-wave radar; the automatic driving controller identifies the trajectory through the dynamic imaging and demarcates the safe area for running ahead That is, the safety envelope area, and the integrated information is fed back to the vehicle controller inside the cab 2 to complete the construction. After the construction is completed, the vehicle controller combines the vehicle driving target and the front safety envelope area to perform automatic driving.

The above-disclosed preferred embodiments of the present invention are provided only to help illustrate the present invention. The preferred embodiments do not exhaust all the details, nor do they limit the invention to only the described embodiments. Obviously, many modifications and variations are possible in light of the content of this specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is to be limited only by the claims and their full scope and equivalents.

Claims

A method for constructing a dynamic envelope of a mining rail locomotive, characterized in that the method for constructing comprises the following steps:

S1. The detection system on the construction device (4) perceives environmental information through dynamic scanning;

S2. The automatic driving controller images the scanned environmental information;

S3. The automatic driving controller performs track identification through dynamic imaging, demarcates the safe area for running ahead, and feeds the comprehensive information back to the vehicle controller;

S4. The vehicle controller performs automatic driving in combination with the vehicle driving target and the front safety envelope area; wherein,

The construction device further comprises a vehicle body (1), a driver's cab (2) and a moving mechanism (5), the driver's cab (2) is fixed on the top of the vehicle body (1), and a power supply is mounted on the vehicle body (1). (3), the detection system (4) is fixed on the cab (2), the moving mechanism (5) is fixed on the bottom of the vehicle body (1), and the detection system (4) includes a lidar sensor, a millimeter-wave radar and a processing module, the processing module is electrically connected to the lidar sensor and the millimeter wave radar; a sand spreader (1a) is installed on the body (1), and a buffer (2a) is installed on the outside of the cab (2) .
The method for constructing the dynamic envelope of a mining rail locomotive according to claim 1, characterized in that, when the ground vehicle in S1 performs environmental perception through a lidar sensor, continuous multi-frame point cloud data is superimposed. In this way, a stable roadway point cloud model is obtained, and on this basis, the track position is detected, and finally obstacles near the track are detected.
Step 1: Establish a stable roadway model In the vehicle-mounted automatic driving controller, the model is based on the ROS system, and the real-time point cloud data of the laser sensor is collected through the PCL point cloud library function. The point cloud data has its own timestamp attribute, which the system determines through this attribute. The sequence of consecutive multi-frame point cloud data; the scanning frequency of the laser sensor is 10Hz, and a point cloud image can be imaged in 100ms. The superposition of consecutive multi-frame point cloud images is to combine multiple point cloud images according to the sequence of point cloud data. It is superimposed to form a denser point cloud image; the laser sensor has a built-in IMU sensor, and the IMU provides a relative positioning information. Its function is to measure the route relative to the starting point object. The main parameters provided are x, y , z, roll, pitch, yaw, according to these parameters, a three-dimensional space composed of point clouds can be simulated;
Through the communication methods such as can bus, the laser sensor obtains the real-time driving speed of the electric locomotive. According to the driving speed of the electric locomotive at a certain moment, the relative positional relationship between the previous frame and the next frame of the point cloud image can be determined, and the point cloud image of the latter frame can be adjusted and adjusted. It is superimposed on the point cloud image of the previous frame, and by analogy, multiple frames of continuous point cloud images can be superimposed to form a relatively stable and dense point cloud image;
After superimposing the point cloud data, the model restores the point cloud image of the roadway more realistically. When frequent bumps and jitters occur during the driving of the electric locomotive, a virtual three-dimensional space is established, and the plane of the three-dimensional space z=0 is determined as the actual ground; The least squares method is used to fit the point cloud data of a certain vertical plane of a single frame of roadway point cloud image, and the slope of the regression line relative to the virtual plane and the intercept in the vertical direction are obtained, and adjusted according to the slope and intercept. All the point cloud data of the single frame are made to coincide with the virtual plane, and then the adjusted frame data is superimposed, and finally a stable and clear roadway model is obtained;
First set the regression line equation:
z=kx+b
The slope is k, and the y-axis intercept is b; since the ground line cannot be perpendicular to the x-axis, the intercept line equation can be used as the regression line equation; the point set of a vertical plane of a single frame of point cloud image for:
{(x,z)|(x1,z1),(x2,z2),...(xn,zn)}
Find the sum of squared errors from all points to the line:

It can be seen from the extreme value theorem that the extreme value is obtained when the first derivative of the error equation is equal to 0, so it is derived with respect to k and b respectively, and the values of k and b are solved to make the error function take the minimum value; it can be obtained:

After finishing, you can get:

in
After finding the values of k and b, bring them into the regression equation to get the fitted straight line equation; according to the slope k, adjust the angle of this frame of point cloud image to make it horizontal to the virtual plane; according to the intercept b, translate the point The cloud image is at the same level as the virtual plane; in this way, each frame of point cloud image obtained is at the same level to achieve the maximum degree of overlap, and a pair of consecutive multi-frame superimposed, stable and clear roadway point cloud images are obtained;
Step 2: Track feature extraction model According to the actual situation, there are two salient features of the underground track, one is that the height of the track is significantly higher than the ground on both sides of the track, and the height difference is 10cm; the other is that the reflection intensity of the track is significantly lower than the ground;
The model is established according to the above two features, and two two-dimensional arrays are established corresponding to the actual ground, so that the plane of the three-dimensional space z=0 is gridded, and the value of one two-dimensional array gridCell is the point cloud z value, which corresponds to the actual position Height; the value of another two-dimensional array gridintensity is the curvature value of the point cloud, corresponding to the reflection intensity of the actual position; the grid resolution is set to 8cm;
The amount of superimposed point cloud data is huge, and limiting the detection range can greatly improve the running speed; since the track width is 0.6m and the track is generally within a certain range in front of the track, the width of the detection range is set to 0.8m around the center of the point cloud image, corresponding to Since the grid is 20 grids in the center of a certain row of the image; analyze the values of the two arrays of 20 grids in the center of a certain row, the grid position corresponding to the gridCell value of 0.1 in the array is considered as the track position; the gridintensity value of the array is obviously smaller than the corresponding grid position. The grid position corresponding to the adjacent array value is considered as the track position; the analysis results of the two arrays are integrated to obtain the track position of a row of the grid; and so on, when the electric locomotive is running, each row in the grid is calculated cyclically track position, and finally identify a track in the point cloud image;
Step 3: Obstacle Recognition Through the track feature extraction model, after the track is correctly identified, obstacles are detected within a certain range near the track; a certain range is delineated near the grid corresponding to the track, and the gridCell array corresponding to the range is detected. There is a z value, if the z value is greater than the track and less than the height of the vehicle, it is determined as an obstacle; then the distance between the obstacle and the electric locomotive is estimated according to the x value of the obstacle, and the electric locomotive is made by communicating with the motor drive system. corresponding braking response.
A method for constructing a dynamic envelope of a mining rail locomotive according to claim 1, wherein the moving mechanism (5) comprises a chassis (6), four wheels (7), two side A frame (13), the side frames (13) are fixed on the outer walls on both sides of the bottom frame (6), and the wheels (7) are installed at the bottom of the bottom frame (6), between two adjacent wheels (7) Connected by a rotating shaft (8), a transmission gear (9) is installed in the middle of the two rotating shafts (8), and a drive gear (20) is meshed with one side of the two transmission gears (9). A transmission shaft (21) is threaded through the middle of the drive gear (20), and both ends of the transmission shaft (21) are connected to a blower (17) fixed at the bottom of the chassis (6), and the transmission The shaft (21) is inserted into the inside of the blower (17) and connected with the fan blades located inside the blower (17), and a drive motor ( 15), the output shaft of the driving motor (15) is connected with a driving gear (18), and the driving gear (18) is engaged with a meshing gear (19) sleeved on the transmission shaft (21).
A method for constructing a dynamic envelope of a mining rail locomotive according to claim 3, wherein the wheel (7) comprises a rim and a hub body, and the rim is located on the outer walls of both sides of the hub body , the outer wall of the hub is provided with a number of evenly distributed air holes (23), the side wall of the rim close to the hub body is provided with a number of evenly distributed side holes (24), the inside of the hub body An intermediate cavity (27) is provided, one side of the intermediate cavity (27) is provided with a cavity (26) located inside the hub body, the cavity (26) is communicated with the intermediate cavity (27), and the The middle cavity (27) is communicated with the air hole (23), a pneumatic telescopic tube (25) is installed inside each of the side holes (24), and the pneumatic telescopic tube (25) is communicated with the cavity (26) , the blower (17) is connected to the intermediate cavity (27) with a pipeline.
A method for constructing a dynamic envelope of a mining rail locomotive according to claim 4, characterized in that, a plurality of damping springs (14) are fixed on the top of each side frame (13), and the The top end of the shock spring (14) is connected to the bottom of the vehicle body (1), each side frame (13) is provided with a side support (11) fixed on the side wall of the bottom frame (6) on both sides, and each A connecting rod (12) is connected to the side wall of the side support (11), the top end of the connecting rod (12) is fixed on the side wall of the bottom frame (6), and each side support (11) is internally Several side springs (10) are installed.
A method for constructing a dynamic envelope of a mining rail locomotive according to claim 5, characterized in that, one side of each of the wheels (7) is provided with a brake sleeved on the rotating shaft (8). Discs (16), one side of each said brake disc (16) is installed with a pneumatic brake (22) matched therewith, and the pneumatic brake (22) is connected with the air cylinder at the bottom of the chassis (6).