CN113581320A - Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method - Google Patents
Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method Download PDFInfo
- Publication number
- CN113581320A CN113581320A CN202110566142.4A CN202110566142A CN113581320A CN 113581320 A CN113581320 A CN 113581320A CN 202110566142 A CN202110566142 A CN 202110566142A CN 113581320 A CN113581320 A CN 113581320A
- Authority
- CN
- China
- Prior art keywords
- surveying
- unmanned aerial
- current frame
- point
- aerial vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D63/00—Motor vehicles or trailers not otherwise provided for
- B62D63/02—Motor vehicles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R11/00—Arrangements for holding or mounting articles, not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D63/00—Motor vehicles or trailers not otherwise provided for
- B62D63/02—Motor vehicles
- B62D63/04—Component parts or accessories
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/003—Navigation within 3D models or images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/08—Indexing scheme for image data processing or generation, in general involving all processing steps from image acquisition to 3D model generation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Transportation (AREA)
- Geometry (AREA)
- Remote Sensing (AREA)
- Combustion & Propulsion (AREA)
- Computer Graphics (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to an autonomous three-dimensional surveying unmanned vehicle for a mine and a surveying method, wherein the autonomous three-dimensional surveying unmanned vehicle comprises the following steps: in the process that the unmanned vehicle travels in a mine, a laser radar on the unmanned vehicle collects point cloud data of the current environment in real time and sends the corresponding point cloud data to a host frame by frame; screening out high-value data points from the data points by the host, and respectively acquiring a plane characteristic point set and an edge characteristic point set according to the high-value data points; obtaining the attitude of the unmanned vehicle corresponding to the current frame through iterative pose optimization calculation, and simultaneously creating a three-dimensional map of the current frame; and planning the traveling route of the unmanned vehicle in real time according to the three-dimensional map, and controlling the unmanned vehicle to travel through a controller according to the traveling route. The invention can autonomously complete the surveying and mapping task in the mine and has high efficiency and high precision.
Description
Technical Field
The invention relates to the field of mining industry, in particular to an autonomous three-dimensional surveying and mapping unmanned vehicle for a mine and a surveying and mapping method.
Background
The mine measurement work is an important link in the geological exploration and mine exploitation processes, and has an instructive effect on the mine production work.
The mine surveying and mapping operation has the following characteristics:
1. the working environment is severe. Due to the limitation of the mine measurement working environment, most of mine geographic positions are remote, transportation and supply are inconvenient, workers need to carry a large amount of material equipment when entering a mountain, and the accuracy and precision of mine measurement can be influenced by uncontrollable factors such as environment, climate and the like. In addition, mine surveying and mapping precision is related to professional qualities of instruments and equipment and workers.
2. The working strength is high. Due to the fact that the workload is large, a single worker cannot acquire complete data in a short time, accuracy of the acquired data cannot be guaranteed, multiple times of measurement are needed, and the mine measurement is difficult due to the fact that the working intensity is too large to a certain extent.
At present, in the field of mine surveying and mapping, manual surveying and mapping are mainly adopted for operation, for example, electronic theodolite, total station type instruments, GPS receivers and the like are used for surveying and mapping. In recent years, three-dimensional laser radar has also begun to be used in mine surveying as a non-contact measurement technique. For example, forestry and other [1] utilize network RTK technology to control and measure a certain open mine, and adopt three-dimensional laser scanning technology to monitor the mining and stripping face of the mine regularly, through the point cloud data processing, the three-dimensional geological model of the mining and stripping face is constructed, and through carrying out the superposition analysis with the model of each period, the dynamic monitoring of the mine reserves is realized; gispanze et al [2] used three-dimensional laser scanning technology to carry on the dynamic experimental monitoring of resource reserves to a certain strip mine in Dandong, think this technology is superior to the traditional observation method in observing efficiency, and in observing the precision; the three-dimensional laser scanning technology is adopted to dynamically monitor the resource reserves of a rock quarry in Xiyin year and the like [3], the technology is considered to be adopted for observation, so that the problem of reserve management under the condition of complex terrain can be effectively solved, and the method has certain advantages compared with the traditional observation method.
Even so, the above measurement technique still has the following disadvantages to be optimized: 1. the manual mapping mode has high working strength and low efficiency. The high-precision mapping has higher requirements on the performance of instruments and equipment and the professional quality of workers; 2. the result of measurement by the manual mapping mode is generally a plane graph, the information is not completely filed, and the displayed data is not intuitive enough. The mapping precision also has certain limitation; 3. the three-dimensional laser mapping equipment adopted at present still needs manual operation, needs longer man-hour in the survey and drawing operation, and is not high in efficiency.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the autonomous three-dimensional surveying unmanned vehicle and the surveying method for the mine can autonomously complete the surveying task in the mine and have the advantages of high efficiency and high precision.
In order to solve the technical problems, the invention adopts the technical scheme that:
a surveying method of an autonomous three-dimensional surveying unmanned aerial vehicle for a mine comprises the following steps:
in the process that the unmanned aerial vehicle travels in a mine, a laser radar on the unmanned aerial vehicle collects point cloud data of the current environment in real time, and sends the corresponding point cloud data to a host in the unmanned aerial vehicle frame by frame;
after receiving the point cloud data of the current frame, the host screens out high-value data points;
the host computer respectively extracts plane feature points and edge feature points by calculating the local smoothness of the high-value data points, and obtains a plane feature point set and an edge feature point set corresponding to the current frame;
according to the plane feature point set and the edge feature point set corresponding to the current frame, obtaining the attitude of the unmanned aerial vehicle corresponding to the current frame through iterative pose optimization calculation, and meanwhile creating a three-dimensional map of the current frame;
the host machine plans the advancing route of the unmanned aerial vehicle in real time according to the three-dimensional map of the current frame, and controls the unmanned aerial vehicle to advance through the controller according to the advancing route.
Preferably, the screening out high value data points therefrom comprises:
the host computer filters out point cloud data close to the edge of the FOV field of view, point cloud data with overlarge or undersize intensity, point cloud data with an incident angle close to pi or 0 and point cloud data hidden behind an object by calculating the characteristics of each point cloud data of the current frame in the FLU coordinate system, and high-value data points are obtained.
Preferably, the obtaining of the pose of the unmanned aerial vehicle corresponding to the current frame through iterative pose optimization calculation according to the plane feature point set and the edge feature point set corresponding to the current frame, and creating the three-dimensional map of the current frame at the same time includes:
s01: the method comprises the steps that a host computer obtains a plane feature point set and an edge feature point set of a current frame, a plane feature point set and an edge feature point set of an existing map, and the posture of a previous frame of the unmanned aerial vehicle;
s02: taking a feature point p in the edge feature point set of the current framelAccording to the formula pw＝Rkpl+tkCalculating the feature point plPoint p projected into existing mapw(ii) a Wherein, the (R) isk， tk) The pose of the unmanned aerial vehicle when the last data point of the current frame is sampled;
s03: determining a distance p from the point in a set of edge feature points of an existing mapwThe nearest 5 feature points;
s04: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-edge residual error of (1);
s05: determining a distance p from the point in a set of planar feature points of an existing mapwThe nearest 5 feature points;
s06: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-surface residual error of;
s07: returning to execution S02, recalculating feature pointsplPoint-to-edge residual and point-to-surface residual of (1);
s08: according to preset iteration times, calculating to obtain multiple groups of point-to-edge residual errors and point-to-surface residual errors corresponding to the iteration times;
s09: calculating to obtain the attitude of the unmanned aerial vehicle corresponding to the current frame according to the groups of point-to-side residual errors and point-to-surface residual errors, and removing abnormal values in a plane feature point set and an edge feature point set corresponding to the current frame;
s10: projecting the plane feature point set and the edge feature point set corresponding to the current frame without the abnormal value to the existing map to obtain a three-dimensional map corresponding to the current frame;
preferably, the host machine plans a traveling route of the unmanned aerial vehicle in real time according to the three-dimensional map of the current frame, and controls the unmanned aerial vehicle to travel through the controller according to the traveling route, and the method includes:
the host calculates the advancing route of the unmanned aerial vehicle in the three-dimensional map of the current frame according to a rapid search random number algorithm;
the host machine determines the linear velocity and the angular velocity corresponding to the unmanned aerial vehicle according to the traveling route and transmits the linear velocity and the angular velocity to the controller;
the controller controls the unmanned aerial vehicle according to the received linear velocity and the received angular velocity.
Preferably, the laser radar is fixedly mounted on the vehicle body through a fixed corner brace; laser radar's quantity is 4 at least, respectively towards unmanned aerial vehicle's the place ahead, top, left side and right side.
The invention provides another technical scheme as follows:
the mine autonomous three-dimensional surveying and mapping unmanned aerial vehicle applied to the surveying and mapping method comprises a vehicle body, an aluminum section support, a laser radar and a moving mechanism, wherein a micro-host, a controller, a switching power supply, a rear fastener, a remote control receiver, a drive plate, a lithium battery and a front fastener are arranged in the vehicle body, a front baffle plate is arranged at the front part of the vehicle body, a rear baffle plate is arranged at the rear part of the vehicle body, the micro-host and the controller are fixed in the vehicle body, communication lines are arranged between the micro-host and the controller for connection, the switching power supply is installed at the tail part of the vehicle body and provided with an inner hexagonal wrench for switching on or off the power supply in a rotating way, the remote control receiver is fixed at the tail part for receiving remote control signals, the drive plate is close to the switching power supply, the lithium battery is used for supplying power and installed in the vehicle body, the front fastener and the rear fastener are respectively located at the front baffle plate and the rear baffle plate of the vehicle body, and a top cover plate is arranged at the top of the vehicle body, and be connected with back fastener and preceding fastener for the fixed top apron of locking, aluminium section bar support mounting is at the top of top apron, moving mechanism includes the cross country tire, the cross country tire is totally four installs in the bottom of automobile body for drive automobile body motion, laser radar is connected with micro-host computer and controller, laser radar is totally four installs on aluminium section bar support, and moves towards different angles respectively, laser radar surveys and draws a survey and drawing and route planning and navigation, the controller is connected intelligent drive direct current motor operation with direct current motor.
Preferably, the laser radar is a solid laser radar, and a laser radar base is arranged at the bottom of the laser radar and connected with the aluminum section bar support.
Preferably, lidar includes top lidar, left side orientation lidar, right side orientation lidar and bottom orientation lidar, and the direction contained angle is sixty degrees.
Preferably, moving mechanism still includes direct current motor, reduction gear, curb plate, rubber seal, wheel hub flange, reduction gear output shaft and reduction gear outer joint piece, direct current motor is connected with the reduction gear, is fixed in on the curb plate, the gap between curb plate and the reduction gear outer joint piece is sealed by rubber seal, cooperation rotation output power between reduction gear output shaft and the wheel hub flange, cross country tire passes through and links between bolt and the wheel hub flange.
Preferably, a fixed corner brace is arranged between the laser radar and the aluminum section bracket for connection.
Preferably, the communication line is an RS232 serial communication line.
The invention has the beneficial effects that: the underground mine surveying and mapping system is based on an innovative unmanned aerial vehicle and SLAM technology, can realize accurate positioning and path planning in an underground closed space, and flexibly and accurately completes the surveying and mapping task in the mine on the premise of ensuring the safety of personnel, so that the underground mine surveying and mapping system has the capability of autonomously completing the surveying and mapping in the mine with high efficiency and high precision.
Drawings
Fig. 1 is an external structural schematic diagram of a mine autonomous three-dimensional surveying and mapping unmanned aerial vehicle in an embodiment of the invention;
fig. 2 is a schematic diagram of an internal structure of the autonomous three-dimensional surveying and mapping unmanned aerial vehicle for a mine in an embodiment of the present invention;
fig. 3 is a schematic structural view of an external local method of the autonomous three-dimensional surveying and mapping unmanned aerial vehicle for a mine in an embodiment of the present invention;
fig. 4 is a schematic side plate structure view of the autonomous three-dimensional surveying and mapping unmanned aerial vehicle for a mine in an embodiment of the present invention;
fig. 5 is a schematic side plate structure view of the autonomous three-dimensional surveying and mapping unmanned aerial vehicle for a mine in an embodiment of the present invention;
fig. 6 is a schematic flow chart of the autonomous three-dimensional surveying and mapping method for a mine according to an embodiment of the present invention.
Description of reference numerals:
1. a top laser radar; 2. a laser radar base; 3. an aluminum profile bracket; 4. Left side facing lidar; 5. the bottom faces the laser radar; 6. fixing the corner connectors; 7. Off-road tires; 8. a top cover plate; 9. a front baffle; 10. a tire securing flange; 11. The right side faces the laser radar; 12. a micro-mainframe; 13. a controller; 14. a switching power supply; 15. a fastener; 16. a remote control receiver; 17. a drive plate; 18. a lithium battery; 19. a front fastener; 20. A direct current motor; 21. a speed reducer; 22. a side plate; 23. a rubber seal ring; 24. a hub flange; 25. A reduction gearbox output shaft; 26. the outer connecting sheet of the speed reducer.
Detailed Description
In order to explain technical contents, achieved objects, and effects of the present invention in detail, the following description is made with reference to the accompanying drawings in combination with the embodiments.
Example one
Referring to fig. 1 to 5, the embodiment provides an autonomous three-dimensional surveying and mapping unmanned mine vehicle, which includes a vehicle body, an aluminum section bracket 3, a laser radar, and a moving mechanism, wherein a micro-host 12, a controller 13, a switching power supply 14, a rear fastener 15, a remote control receiver 16, a driving board 17, a lithium battery 18, and a front fastener 19 are disposed in the vehicle body, a front baffle 9 is disposed at a front portion of the vehicle body, a rear baffle is disposed at a rear portion of the vehicle body, the micro-host 12 and the controller 13 are fixed in the vehicle body, and are connected by a communication line, the switching power supply 14 is installed at a rear portion of the vehicle body, and is provided with an allen wrench for turning on or off the power supply, the remote control receiver 16 is fixed at the rear portion and receives a remote control signal, the driving board 17 abuts against the switching power supply 14, the lithium battery 18 is used for supplying power and installed in the vehicle body, and the front fastener 19 and the rear fastener 15 are respectively located at the front baffle 9 and the rear baffle of the vehicle body, automobile body top is equipped with top apron 8, and is connected with back fastener 15 and preceding fastener 19 for lock fixed top apron 8, aluminium section bar support 3 installs at the top of top apron 8, moving mechanism includes cross country tire 7, cross country tire 7 is totally four installs the bottom at the automobile body for drive automobile body motion, lidar is connected with micro-host 12 and controller 13, lidar totally installs on aluminium section bar support 3 for four, and moves towards different angles respectively, lidar carries out survey and drawing and route planning and navigation, controller 13 is connected intelligent drive direct current motor 20 operation with direct current motor 20.
In this embodiment, laser radar is solid-state laser radar, the laser radar bottom is equipped with laser radar base 2 and is connected with aluminium section bar support 3, and then improves the stability that laser radar used.
In this embodiment, laser radar includes top laser radar 1, left side orientation laser radar 4, right side orientation laser radar 11 and bottom orientation laser radar 5, and the direction contained angle is sixty degrees, and then improves laser radar's detection scope.
In this embodiment, the moving mechanism further includes a dc motor 20, a speed reducer 21, a side plate 22, a rubber seal ring 23, a hub flange 24, a speed reducer output shaft 25 and a speed reducer external connection piece 26, the dc motor 20 is connected with the speed reducer 21 and fixed on the side plate 22, a gap between the side plate 22 and the speed reducer external connection piece 26 is sealed by the rubber seal ring 23, the speed reducer output shaft 25 and the hub flange 24 are matched to rotate to output power, and the off-road tire 7 is linked with the hub flange 24 through bolts, so that the moving stability effect of the vehicle body is improved.
In this embodiment, be equipped with fixed angle sign indicating number 6 between lidar and the aluminium section support 3 and be connected, and then improve lidar and install the stability on aluminium section support 3.
In this embodiment, the communication line is an RS232 serial communication line, thereby improving the communication effect between the micro-host 12 and the controller 13.
To sum up, when the autonomous three-dimensional surveying and mapping unmanned mine vehicle is used, when the autonomous three-dimensional surveying and mapping unmanned mine vehicle enters an autonomous navigation mode, four laser radars are all fixed on an aluminum section bar support 3 through fixed angle codes 6, in order to obtain an optimal view field, a top laser radar 1 is installed on a laser radar base 2 with an inclination angle of 10 degrees and is fixed on the aluminum section bar support 3 by using a ship-shaped nut lock, the left side of the ship-shaped nut lock faces towards a laser radar 4, the bottom of the ship-shaped nut lock faces towards a laser radar 5, the right side of the ship-shaped nut lock faces towards a laser radar 11, the detection directions of the laser radars respectively face towards sixty degrees on the left side of the vehicle body, the laser radars incline downwards by ten degrees in the middle and face towards the sixty-degree direction on the right side of the vehicle body, a lithium battery 18 supplies power to the laser radar and a micro-host 12 and transmits data to the micro-host 12 for processing, and executes an SLAM algorithm program; the micro-host 12 is communicated with the controller 13 under the control of the switching power supply 14 and outputs signals to the drive plate 17, the drive plate 17 controls the forward and reverse rotation and acceleration and deceleration actions of the direct current motor 20 after receiving the signals, the torque of the direct current motor 20 can be increased by matching with the reducer 21, the load capacity of the vehicle body is improved, the reducer output shaft 25 is matched with the hub flange 24, the off-road tires 7 are fixed with the hub flange 24 through bolts, the forward and reverse rotation of the vehicle body is realized through the forward rotation and reverse rotation of the four direct current motors 20 at the front and the rear, the turning is realized through the differential rotation of the off-road tires 7 at the left side and the right side of the vehicle body, when a manual control mode is used, a receiver arranged in the vehicle body is installed through a remote controller, the forward and reverse movement and the like of the trolley are directly controlled under the manual mode, and the working conditions of autonomous surveying and mapping are met.
In this embodiment, the autonomous mapping process based on the above unmanned aerial vehicle includes:
step 1: the method comprises the steps that point cloud data of a current working environment area are collected in real time through a top laser radar 1, a left side-oriented laser radar 4, a bottom-oriented laser radar 5 and a right side-oriented laser radar 11;
step 2: the point cloud data are transmitted to an upper computer NUC micro host 12 for processing, obstacles in the current working area are determined through an execution algorithm, the advancing route of the unmanned aerial vehicle is planned, and meanwhile, a three-dimensional model of the internal structure of the mine is constructed in real time;
and step 3: the lower computer controller receives a path planning command from the upper NUC micro host 12 through a mavlink protocol, and converts the command into a driving signal of the controller 13, so as to control the operation of the driving direct current motor 20;
and 4, step 4: the direct current motor 20 drives the wheels to rotate, and the wheels drive the trolley to move integrally, so that the whole mine hole surveying and mapping task is completed.
And circularly executing the steps 1-4 until the mapping task is completed.
Example two
Referring to fig. 6, the present embodiment provides a method for autonomous three-dimensional surveying and mapping of a mine based on an unmanned aerial vehicle according to the first embodiment, including:
in the process that the unmanned aerial vehicle travels in a mine, all laser radars on the unmanned aerial vehicle acquire point cloud data of the current environment in real time, and send the corresponding point cloud data to a host frame by frame;
after receiving the point cloud data of the current frame, the host screens out high-value data points;
preferably, the host computer filters out point cloud data close to the edge of the FOV field of view, point cloud data with too high or too low intensity, point cloud data with an incident angle close to pi or 0, and point cloud data hidden behind the object by calculating the features of each point cloud data of the current frame in the FLU coordinate system, thereby acquiring high-value data points.
Then, the host computer respectively extracts plane feature points and edge feature points by calculating the local smoothness of the high-value data points, and obtains a plane feature point set and an edge feature point set corresponding to the current frame;
according to the plane feature point set and the edge feature point set corresponding to the current frame, obtaining the attitude of the unmanned aerial vehicle corresponding to the current frame through iterative pose optimization calculation, and meanwhile creating a three-dimensional map of the current frame;
optionally, the steps specifically include:
s01: the method comprises the steps that a host computer obtains a plane feature point set and an edge feature point set of a current frame, a plane feature point set and an edge feature point set of an existing map, and the posture of a previous frame of the unmanned aerial vehicle;
s02: taking a feature point p in the edge feature point set of the current framelAccording to the formula pw＝Rkpl+tkCalculating the feature point plPoint p projected into existing mapw(ii) a Wherein, the (R) isk， tk) The pose of the unmanned aerial vehicle when the last data point of the current frame is sampled;
s03: determining a distance p from the point in a set of edge feature points of an existing mapwThe nearest 5 feature points;
s04: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-edge residual error of (1);
s05: determining a distance p from the point in a set of planar feature points of an existing mapwThe nearest 5 feature points;
s06: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-surface residual error of;
s07: returning to execution S02, feature point p is recalculatedlPoint-to-edge residual and point-to-surface residual of (1);
s08: according to preset iteration times, calculating to obtain multiple groups of point-to-edge residual errors and point-to-surface residual errors corresponding to the iteration times;
s09: calculating to obtain the attitude of the unmanned aerial vehicle corresponding to the current frame according to the groups of point-to-side residual errors and point-to-surface residual errors, and removing abnormal values in a plane feature point set and an edge feature point set corresponding to the current frame;
s10: and projecting the plane feature point set and the edge feature point set corresponding to the current frame without the abnormal values into the existing map to obtain the three-dimensional map corresponding to the current frame.
And finally, planning the advancing route of the unmanned aerial vehicle in real time by the host according to the three-dimensional map of the current frame, and controlling the unmanned aerial vehicle to advance through the controller according to the advancing route.
Optionally, the step may specifically include:
the host calculates the advancing route of the unmanned aerial vehicle in the three-dimensional map of the current frame according to a rapid search random number algorithm;
the host machine determines the linear velocity and the angular velocity corresponding to the unmanned aerial vehicle according to the traveling route and transmits the linear velocity and the angular velocity to the controller;
the controller controls the unmanned aerial vehicle according to the received linear velocity and the received angular velocity.
The surveying and mapping method based on the autonomous three-dimensional surveying and mapping unmanned aerial vehicle for the mine is different from the problems of long time consumption, high cost, high precision and the like of the existing traditional manual operation mode, and the unmanned aerial vehicle has the functions of surveying and mapping operation, path planning and autonomous navigation, so that high-precision mapping can be autonomously performed in the mine, and a three-dimensional model capable of reflecting the internal structure of the mine is generated; and the system can be well adapted to the bad road conditions in the mine, can stably operate, and can ensure the safety of personnel while improving the survey efficiency.
EXAMPLE III
The embodiment provides a method for completing autonomous borehole mapping based on the unmanned aerial vehicle provided in the first embodiment or the second embodiment, and the method specifically includes the following steps:
step 1: the method comprises the steps of collecting point cloud data under the current working environment in real time through a top laser radar 1, a left side orientation laser radar 4, a bottom orientation laser radar 5 and a right side orientation laser radar 11 on an unmanned aerial vehicle, and transmitting the collected point cloud data of each frame to a NUC micro host 12 to execute a SLAM algorithm in real time.
Specifically, after the NUC micro-host 12 receives a frame of point cloud data, the performing SLAM algorithm process includes:
step 11: screening out high-value data points by calculating the characteristics of laser radar point cloud P [ x, y, z ] in a Front-Left-Up (FLU) coordinate system;
specifically, screening is performed according to the result of calculation of each point cloud data according to the following formula:
radar ranging information:
laser deflection angle Φ (angle between laser ray and X axis):
incident angle θ (local plane angle of laser to the periphery of the measured point):
according to the calculation result of each point cloud data, the following characteristic points are removed to improve the accuracy of positioning and mapping:
1. points near the edge of the FOV field of view, i.e. feature points where e.g. Φ (p) ≧ 17 °, in such regions the scan trajectory has a large curvature, resulting in reduced reliability of feature extraction;
2. points of too great or too small strength, i.e. e.g. I (p). ltoreq.7X 10-3，I(p)≤1×10-1Characteristic points of Livox MID 40. This is achieved byThe reason is that: on one hand, the intensity directly represents the intensity of the received laser signal, and the excessive intensity (signal) usually causes the saturation or distortion of the receiving circuit and reduces the ranging accuracy; on the other hand, too small intensity (signal) generally results in a reduced signal-to-noise ratio and also reduced ranging accuracy.
3. The point where the incident angle is close to pi or 0, i.e. e.g. theta (p) ≦ 5 deg., theta (p) ≧ 175 deg., characteristic point of Livox MID 40. This is because the laser spot caused by the non-zero divergence angle of the laser beam will cause the laser spot to be greatly lengthened.
4. Hidden at a point behind the object.
The points in the point cloud data are preliminarily screened, the most valuable data points are extracted, interference is eliminated, and positioning and mapping accuracy is improved.
Step 12: and (5) extracting features.
And after the high-value data points are screened out, performing feature extraction. Specifically, the local smoothness of the candidate points is calculated to extract the plane feature points and the edge feature points.
Preferably, to mitigate match degradation due to the limited number of features resulting from FoV and point selection, LiDAR reflectivity is used as the 4-dimensional measurement. If the reflectivity of a 3D point is very different from the nearby points, it is considered an edge feature point (an edge in reflectivity due to a material change, as opposed to an edge in geometry due to a shape change). In certain degenerate cases, such as facing walls with closed doors and windows, such feature points facilitate feature extraction.
Step 13: and (5) iterative pose optimization.
Due to the non-repeated scanning mode adopted by the solid-state laser radar, even under a static environment, the scanned track and the feature point are different from the previous frame, and the point cloud features extracted between the two frames cannot be used for matching. Iterative pose optimization is therefore used in operation to estimate the pose of the lidar to achieve a real-time mapping at 20 Hz.
13.1 residual between edges
Using epsilonkSet of edge feature points representing a current frameUsing epsilonmRepresenting a set of edge feature points in an existing map; for epsilonkEach point p inwObtaining epsilonmThe 5 points closest thereto. To increase the search speed, epsilon will be constructed in advancemKD tree of (1). Furthermore, the KD-tree is built by another parallel thread upon receiving the last registration frame/sub-frame. Thus, the KD-tree is immediately available when a new frame is received.
Let pl be ε of the current frame (kth frame)kOne point of (a) due to ekP in (1)lPoints are located in a local LiDAR framework and εmPoints are registered in the global map, e.g. to findmIn (c) plNeeds to be projected into the global map by the following formula (4) transformation.
pw＝RkPl+tkFormula (4)
Where (Rk, tk) is the LiDAR pose at which the last point of the current frame was sampled, and needs to be determined by pose optimization. Here, the LiDAR pose of the last point in the frame is used to represent the pose of the entire frame, and all points in the frame are projected onto a global map using this pose. Note also that the last point in the current frame is essentially the first point in the next frame.
Let p beiRepresents epsilonmP of (a)wThe ith closest point of (a). To ensure piExactly on one line, calculate pwThe m closest points of (a) form a mean μ and covariance matrix ∑. Preferably, m is set to 5 in operation. If the maximum eigenvalue of sigma is three times larger than the second maximum eigenvalue, p can be guaranteedwThe closest point of (1) forms pwThe line of the location. The corresponding point-to-edge residual is then calculated by equation (5) below and then added to the pose optimization.
13.2 plane-to-plane residual
And edge feature pointSimilarly, the planar feature set for the current frame
At a point in the map of the set of planar features
The 5 closest points are found. By calculating their covariance matrix sigma it is ensured that the 5 nearest points are indeed in the same plane. If the minimum eigenvalue of Σ is three times smaller than the second minimum eigenvalue, the distance from the plane point in the current frame to the plane formed by 5 points in the same plane is calculated as shown in the following equation (6), and this residual is added to the optimization.
13.3 Intra motion Compensation
As previously described, as LiDAR motion continues, the 3D points are sampled at different times (i.e., motor blur) for different poses. In order to eliminate the effect of motion blur, the present embodiment proposes the following two methods:
13.3.1 segmentation process
A simple and effective way to remove the effect of motion blur is a segmentation process. By dividing the input frame into three consecutive sub-frames; these three sub-frames are then respectively matched to the same map accumulated so far. During scan matching of each sub-frame, all of its points are projected to the global map using the LiDAR pose at the end of the sub-frame. Thus, the time interval of each sub-frame is 1/3 of the original frame. The benefit of this segmentation process is that by parallelizing the matching of each sub-frame, a multi-core structure can be utilized in modern CPUs.
13.3.2 linear interpolation:
another commonly used motion compensation method is linear interpolation: with (R)k，tk) Indicating the LiDAR pose at the last point in the current frame, (R)k-1，tk-1) In order to be the one frame ahead,
for the relative rotation and translation between the previous frame and the current frame, then, the calculation is made by the following equation (7):
let tk-1Is the sample time of the last point in the previous frame. For any point sampled at time t of the current frame, there is t ∈ [ t ]k-1，tk]Calculating S ═ t (t-t)k-1)/(tk-tk-1) Then the linear interpolation pose at time t is:
where θ is the intensity and ω is
The unit vector of the rotation axis of (a).
Is a symmetric matrix of ω. According to the formula of Rodrigue:
this means that sin (s θ) and cos (s θ) need only be calculated for each point of the current frame, while the rest remains unchanged. This simplifies the calculation. For the
The LiDAR pose at the current time is:
the point at time t can then be projected onto the global map by an interpolation gesture, as follows, equation (10) is calculated:
pw(t)＝Rtpl+ttformula (11)
Step 14: removing outliers and filtering dynamic objects;
to avoid degrading the accuracy of scan matching by moving objects in a real-world environment, the following dynamic object filtering is performed:
in each iteration of the iterative pose optimization, the nearest neighbors of each feature point are re-found and the residual equations (5) and the inter-plane residual equations (6) of the edge-to-edge objective function are added, and the pose optimization is performed first in a small number of iterations (e.g., 2 are used in the experiment). Using the optimization results, two residuals in (5) and (6) are calculated, and the top 20% of the largest residual is removed. After removing the outliers, a full pose optimization is finally performed.
Step 2: and (3) acquiring the pose and map information of the current autonomous surveying and mapping trolley by the algorithm in the step 1, and using the information for planning the path of the autonomous surveying and mapping trolley.
Specifically, the step may include:
step 21: creating a prior map according to the initial pose and the point cloud data of the laser radar;
step 22: computing a global path in the map according to an RRT (fast search random tree algorithm);
and step 3: the global path calculated by the NUC micro host of the upper computer is converted into the corresponding linear velocity, and the angular velocity information is transmitted to the lower computer controller for execution through the Mallink protocol.
And 4, step 4: the lower computer controller outputs high-level and low-level signals to the driving plate, and the driving plate controls the speed and the forward and reverse rotation of the motor.
And 5: the motor is decelerated by the reducer, and power is output to the wheels to drive the trolley to move, so that the surveying and mapping task is completed.
The above steps are repeatedly executed before the program is not suspended.
In conclusion, the autonomous three-dimensional surveying unmanned vehicle and the surveying method for the mine provided by the invention can realize rapid three-dimensional scene reconstruction and surveying and mapping work in a mine environment by carrying the high-precision laser radar for mapping on the unmanned aerial vehicle, and can realize autonomous navigation and route planning by a laser SLAM algorithm; and the autonomous completion of the surveying and mapping task can be realized. Compared with the traditional manual mapping mode, the working intensity of workers is greatly reduced, and the time is saved; shorten the survey and drawing engineering time limit for a project, practice thrift cost in business.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (10)
1. A surveying and mapping method of an autonomous three-dimensional surveying and mapping unmanned aerial vehicle for a mine is characterized by comprising the following steps:
in the process that the unmanned aerial vehicle travels in a mine, a laser radar on the unmanned aerial vehicle collects point cloud data of the current environment in real time, and sends the corresponding point cloud data to a host in the unmanned aerial vehicle frame by frame;
after receiving the point cloud data of the current frame, the host screens out high-value data points;
the host computer respectively extracts plane feature points and edge feature points by calculating the local smoothness of the high-value data points, and obtains a plane feature point set and an edge feature point set corresponding to the current frame;
according to the plane feature point set and the edge feature point set corresponding to the current frame, obtaining the attitude of the unmanned aerial vehicle corresponding to the current frame through iterative pose optimization calculation, and meanwhile creating a three-dimensional map of the current frame;
the host machine plans the advancing route of the unmanned aerial vehicle in real time according to the three-dimensional map of the current frame, and controls the unmanned aerial vehicle to advance through the controller according to the advancing route.
2. The method for surveying and mapping of the mine autonomous three-dimensional surveying and mapping drone of claim 1, wherein the screening out high value data points therefrom comprises:
the host computer filters out point cloud data close to the edge of the FOV field of view, point cloud data with overlarge or undersize intensity, point cloud data with an incident angle close to pi or 0 and point cloud data hidden behind an object by calculating the characteristics of each point cloud data of the current frame in the FLU coordinate system, and high-value data points are obtained.
3. The method for surveying and mapping the mine autonomous three-dimensional surveying and mapping unmanned aerial vehicle according to claim 1, wherein the obtaining of the pose of the unmanned aerial vehicle corresponding to the current frame through iterative pose optimization calculation according to the plane feature point set and the edge feature point set corresponding to the current frame and the creating of the three-dimensional map of the current frame comprise:
s01: the method comprises the steps that a host computer obtains a plane feature point set and an edge feature point set of a current frame, a plane feature point set and an edge feature point set of an existing map, and the posture of a previous frame of the unmanned aerial vehicle;
s02: taking a feature point p in the edge feature point set of the current framelAccording to the formula pw＝Rkpl+tkCalculating the feature point plPoint p projected into existing mapw(ii) a Wherein, the (R) isk，tk) The pose of the unmanned aerial vehicle when the last data point of the current frame is sampled;
s03: determining a distance p from the point in a set of edge feature points of an existing mapwThe nearest 5 feature points;
s04: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-edge residual error of (1);
s05: determining a distance p from the point in a set of planar feature points of an existing mapwThe nearest 5 feature points;
s06: if the 5 characteristic points are on the same straight line, according to the point-to-edge residual error formula
Calculating the point pwPoint-to-surface residual error of;
s07: returning to execution S02, feature point p is recalculatedlPoint-to-edge residual and point-to-surface residual of (1);
s08: according to preset iteration times, calculating to obtain multiple groups of point-to-edge residual errors and point-to-surface residual errors corresponding to the iteration times;
s09: calculating to obtain the attitude of the unmanned aerial vehicle corresponding to the current frame according to the groups of point-to-side residual errors and point-to-surface residual errors, and removing abnormal values in a plane feature point set and an edge feature point set corresponding to the current frame;
s10: and projecting the plane feature point set and the edge feature point set corresponding to the current frame without the abnormal values into the existing map to obtain the three-dimensional map corresponding to the current frame.
4. The method for surveying and mapping the mine autonomous three-dimensional surveying and mapping unmanned aerial vehicle as claimed in claim 1, wherein the host machine plans a travel route of the unmanned aerial vehicle in real time according to the three-dimensional map of the current frame, and controls the unmanned aerial vehicle to travel through the controller according to the travel route, comprising:
the host calculates the advancing route of the unmanned aerial vehicle in the three-dimensional map of the current frame according to a rapid search random number algorithm;
the host machine determines the linear velocity and the angular velocity corresponding to the unmanned aerial vehicle according to the traveling route and transmits the linear velocity and the angular velocity to the controller;
the controller controls the unmanned aerial vehicle according to the received linear velocity and the received angular velocity.
5. The mine autonomous three-dimensional surveying and mapping unmanned aerial vehicle applied to the surveying and mapping method according to any one of claims 1-4 comprises a vehicle body, an aluminum section support (3), a laser radar and a moving mechanism, wherein a micro-host (12), a controller (13), a switching power supply (14), a rear fastener (15), a remote control receiver (16), a driving plate (17), a lithium battery (18) and a front fastener (19) are arranged in the vehicle body, a front baffle (9) is arranged at the front part of the vehicle body, a rear baffle is arranged at the rear part of the vehicle body, the micro-host (12) and the controller (13) are fixed in the vehicle body, a communication line is arranged between the micro-host and the controller for connection, the switching power supply (14) is installed at the tail part of the vehicle body and provided with an inner hexagon wrench for rotationally connecting or disconnecting the power supply, the remote control receiver (16) is fixed at the tail part for receiving remote control signals, the driving plate (17) is close to a switching power supply (14), the lithium battery (18) is used for supplying power and is installed inside a vehicle body, the front fastener (19) and the rear fastener (15) are respectively located on a front baffle plate (9) and a rear baffle plate of the vehicle body, a top cover plate (8) is arranged at the top of the vehicle body and is connected with the rear fastener (15) and the front fastener (19) and used for locking and fixing the top cover plate (8), the aluminum section bar support (3) is installed at the top of the top cover plate (8), the moving mechanism comprises off-road tires (7), the off-road tires (7) are installed at the bottom of the vehicle body in total, and are used for driving the vehicle body to move, the laser radar is connected with a micro host (12) and a controller (13), the laser radar is installed on the aluminum section bar support (3) in total in four parts and respectively faces different angles, and carries out surveying and mapping, path planning and navigation, the controller (13) is connected with the direct current motor (20) to intelligently drive the direct current motor (20) to operate.
6. The mining autonomous three-dimensional surveying unmanned aerial vehicle of claim 5, characterized in that the lidar is a solid-state lidar, and a lidar base (2) is arranged at the bottom of the lidar and connected with an aluminum section bracket (3).
7. The mine autonomous three-dimensional surveying unmanned vehicle according to claim 5, characterized in that the lidar comprises a top lidar (1), a left-side-oriented lidar (4), a right-side-oriented lidar (11) and a bottom-oriented lidar (5), and the included angle of direction is sixty degrees.
8. The mine autonomous three-dimensional surveying unmanned vehicle of claim 5, characterized in that the moving mechanism further comprises a DC motor (20), a speed reducer (21), a side plate (22), a rubber sealing ring (23), a hub flange (24), a speed reducer output shaft (25) and a speed reducer external connection plate (26), the DC motor (20) is connected with the speed reducer (21) and fixed on the side plate (22), a gap between the side plate (22) and the speed reducer external connection plate (26) is sealed by the rubber sealing ring (23), the speed reducer output shaft (25) and the hub flange (24) are matched to rotate to output power, and the off-road tire (7) is connected with the hub flange (24) through bolts.
9. The mine autonomous three-dimensional surveying unmanned vehicle according to claim 5, characterized in that a fixed corner brace (6) connection is provided between the lidar and the aluminium profile support (3).
10. The mine autonomous three-dimensional surveying unmanned vehicle of claim 5, wherein the communication line is an RS232 serial communication line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110566142.4A CN113581320A (en) | 2021-05-24 | 2021-05-24 | Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110566142.4A CN113581320A (en) | 2021-05-24 | 2021-05-24 | Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113581320A true CN113581320A (en) | 2021-11-02 |
Family
ID=78243358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110566142.4A Pending CN113581320A (en) | 2021-05-24 | 2021-05-24 | Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113581320A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114115287A (en) * | 2021-12-06 | 2022-03-01 | 西安航空学院 | Unmanned vehicle-unmanned aerial vehicle air-ground cooperative patrol and guidance system |
| CN114705171A (en) * | 2022-03-28 | 2022-07-05 | 南通市融信信息科技有限公司 | Three-dimensional terrain surveying system and surveying method |
-
2021
- 2021-05-24 CN CN202110566142.4A patent/CN113581320A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114115287A (en) * | 2021-12-06 | 2022-03-01 | 西安航空学院 | Unmanned vehicle-unmanned aerial vehicle air-ground cooperative patrol and guidance system |
| CN114115287B (en) * | 2021-12-06 | 2023-09-22 | 西安航空学院 | Unmanned vehicle-unmanned aerial vehicle air-ground collaborative patrol and guide system |
| CN114705171A (en) * | 2022-03-28 | 2022-07-05 | 南通市融信信息科技有限公司 | Three-dimensional terrain surveying system and surveying method |
| CN114705171B (en) * | 2022-03-28 | 2024-02-02 | 南通市融信信息科技有限公司 | Three-dimensional topographic mapping system and mapping method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110411462B (en) | GNSS/inertial navigation/lane line constraint/milemeter multi-source fusion method | |
| CN113945206A (en) | Positioning method and device based on multi-sensor fusion | |
| CN113581320A (en) | Autonomous three-dimensional surveying and mapping unmanned vehicle for mine and surveying and mapping method | |
| Pfaff et al. | Towards mapping of cities | |
| Jacobson et al. | Semi-supervised slam: Leveraging low-cost sensors on underground autonomous vehicles for position tracking | |
| CN114526745A (en) | Drawing establishing method and system for tightly-coupled laser radar and inertial odometer | |
| CN111338342A (en) | Automatic tracking driving control system and method for wheel type engineering machinery | |
| CN115993825A (en) | Unmanned vehicle cluster control system based on air-ground cooperation | |
| CN116560357A (en) | Tunnel inspection robot system based on SLAM and inspection control method | |
| CN114563795B (en) | Positioning tracking method and system based on laser odometer and label fusion algorithm | |
| CN114383598B (en) | Tunnel construction operation car and automatic driving system thereof | |
| Srinara et al. | Performance analysis of 3D NDT scan matching for autonomous vehicles using INS/GNSS/3D LiDAR-SLAM integration scheme | |
| Wang et al. | Rail vehicle localization and mapping with LiDAR-vision-inertial-GNSS fusion | |
| Filip et al. | Lidar SLAM Comparison in a Featureless Tunnel Environment | |
| Androulakis et al. | Navigation system for a semi-autonomous shuttle car in room and pillar coal mines based on 2D LiDAR scanners | |
| Guo et al. | Occupancy grid based urban localization using weighted point cloud | |
| CN115562076B (en) | Simulation system, method and storage medium for unmanned mine car | |
| CN110058587B (en) | Coal mine fully-mechanized mining face inspection unmanned vehicle based on SLAM technology and autonomous inspection method | |
| Wang et al. | A robust lidar state estimation and map building approach for urban road | |
| CN214823750U (en) | Mine autonomous three-dimensional surveying and mapping unmanned vehicle | |
| CN114296059A (en) | Laser radar point cloud distortion correction method for large-scale rescue equipment | |
| Xiang et al. | Where am I? SLAM for mobile machines on a smart working site | |
| Patel et al. | LiDAR-guided object search and detection in Subterranean Environments | |
| Chen et al. | Lidar inertial odometry and mapping for autonomous vehicle in GPS-denied parking lot | |
| Xu et al. | Challenges of Autonomous Navigation and Perception Technology for Unmanned Special Vehicles in Underground Mine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |