CN109099901A - Full-automatic road roller localization method based on multisource data fusion - Google Patents
Full-automatic road roller localization method based on multisource data fusion Download PDFInfo
- Publication number
- CN109099901A CN109099901A CN201810670215.2A CN201810670215A CN109099901A CN 109099901 A CN109099901 A CN 109099901A CN 201810670215 A CN201810670215 A CN 201810670215A CN 109099901 A CN109099901 A CN 109099901A
- Authority
- CN
- China
- Prior art keywords
- road roller
- point
- point cloud
- laser radar
- data
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- Multimedia (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses the full-automatic road roller localization methods based on multisource data fusion, on construction area boundary, boundary marker is set, Multi-path synchronous camera obtains the image comprising boundary marker, and construction area point cloud data is obtained using laser radar, image is matched with laser point cloud data, it identifies construction area mark, calculates construction fence region;Using laser point cloud data and position and posture information of the IMU output information fusion calculation road roller in construction area, realize unmanned road roller in the positioning of construction area.Pass through the localization method of machine vision, by vision data and IMU data fusion, it can adapt to the special construction environment of road roller, it effectively avoids the problem that relying on the signal deletion that GPS positioning generates merely, and positioning and path planning requirement of the unmanned road roller in construction environment can be met in terms of positioning accuracy and speed, it is efficiently applied to road roller construction site, the problem of effective resolving roller positioning accuracy difference.
Description
Technical field
The present invention relates to a kind of full-automatic road roller localization method based on multisource data fusion.
Background technique
Engineering machinery is often in the disaster fields such as earthquake, flood, tsunami and high temperature, high and cold, High aititude, the high evil radiated
It constructs under bad operating condition, not only construction efficiency is low, but also operator usually will also emit life danger.Optimized using unmanned technology
Existing engineering machinery can be with lifting construction quality, reduction personnel cost, reduction security risk, and provides for the update of realization industry
Important base.The road roller being particularly directed in engineering machinery belongs to the scope of road furniture, be widely used in highway,
The embankment compacting operation of the Larger Engineering Projects such as railway, airfield runway, dam, stadium can roll sand, half viscosity and viscous
Property soil, subgrade stability soil and asphalt concrete pavement layer, have extremely wide construction demand and higher task difficulty.Press road
The problem of standardized construction of machine is also each unit in charge of construction in the current whole world and machinery production factory urgent need to resolve, and automatically press road
Machine provides be effectively ensured to solve this problem.
Realize that full-automatic road roller is independently constructed on condition that accurately to obtain self-position.Engineering machinery is main fixed at present
Position method is that positioning is merged with odometer based on GPS positioning or GPS.Due to relying on GPS device positioning, itself exists and miss
Difference, while in the case where signal blocks, for example special operation conditions environment, the GPS signal such as tunnel, bustling urban district can weaken rapidly
It even loses, to be unable to complete the automated construction for meeting homework precision.
There is the deficiencies of positioning accuracy is low, at high cost mostly in current unmanned road roller.Patent publication No. is
CN106127177A discloses a kind of unmanned road roller, is positioned using GPS to road roller, not yet in view of existing when road roller
The problem of GPS signal can not receive when construction operation under particular surroundings causes not making in tunnel, the inferior particular surroundings of bridge
The problem of industry.
Machine vision is to obtain picture using visual sensor and carry out various measurements and judgement using image processing system,
It is an important branch of Computer Subject, combines the technology of optics, machinery, electronics, computer software and hardware etc., be related to
To multiple fields such as computer, image procossing, pattern-recognition, artificial intelligence, signal processing, optical, mechanical and electronic integration.Vision guided navigation
It is that respective handling is carried out to obtain a kind of technology of carrier pose parameter to the image that visual sensor obtains.Vision is led at present
Boat technology is mainly used in the racing contest of mobile robot, industry AGV, the independent navigation of intelligent vehicle and science and techniques of defence technology
Study this four aspects.Patent publication No. CN104835173A proposes a kind of localization method based on machine vision, but the party
Method realizes AGV positioning by vehicle-mounted vision system, and this method relies solely on camera and realizes that vision positioning, stability are inadequate.
Summary of the invention
The purpose of the present invention is overcoming the shortcomings of the prior art, provide a kind of based on the full-automatic of multisource data fusion
Road roller localization method.
The purpose of the present invention is achieved through the following technical solutions:
Full-automatic road roller localization method based on multisource data fusion, it is characterised in that: be arranged on construction area boundary
Boundary marker, as the index point in laser point cloud data and fusing image data positioning auxiliary, Multi-path synchronous camera obtains packet
Image containing boundary marker, and construction area point cloud data is obtained using laser radar, image and laser point cloud data are carried out
Matching identifies construction area mark, calculates construction fence region;Utilize laser point cloud data and IMU output information fusion calculation
Position and posture information of the road roller in construction area, realize unmanned road roller in the positioning of construction area.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein at the construction field (site)
Construction area bordering is set, industrial personal computer and inertial navigation unit are installed on road roller, the surrounding of road roller, which is installed, to be used
Three-dimensional laser radar is set on front side of the camera for obtaining road roller construction area image information, road roller.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein road roller
Camera is distributed in four side of front, rear, left and right, and camera collocation wide-angle lens covers 360 ° of ranges around road roller, utilizes gridiron pattern
Camera is demarcated, obtains camera internal reference, and distortion correction is carried out to camera;
The mesh of three-dimensional laser radar scanning 30 degree of vertical direction and 360 degree of horizontal directions within the scope of 100 meters of radius
Mark, obtains the label to road roller motion profile by way of continuity point cloud Data Matching;
The construction area bordering is cone, using image and the matched method of point cloud data, is obtained to friendship
The effective identification and range measurement of logical cone, realize surveying for construction area boundary;
The inertial navigation unit obtains road roller speed, the information of position, using extended Kalman filter to laser
Radar information is merged with inertial navigation information, realizes synchronization composition and the positioning of unmanned road roller.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein the cone
Pyrometric cone is acquired by three-dimensional laser radar and camera respectively and is regarded at two as construction area bordering for pyrometric cone
Three-dimensional laser point cloud and image under point are closed using the calibration between the laser point cloud data and image data of foundation and obtain triangle
The point cloud for boring marker is semantic, as vehicle location reference frame.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein swashed using three-dimensional
Optical radar obtains the three dimensional point cloud of scene, three-dimensional S LAM composition is realized using LOAM method, in LOAM method, by mentioning
Take calculating coordinate change after Feature Points Matching;Cloud and IMU data are pre-processed first, for extracting characteristic point: primary
The point of scanning is classified by curvature value, and formula is as follows:
Wherein, { L } is the three-dimensional system of coordinate of laser radar, the geometric center originating from laser radar, x-axis direction left side, y
Axis points up, and z-axis is directing forwardly;pkIndicate the point cloud perceived during scanning k, { LkMidpoint i, i ∈ pkCoordinate representation beS is one group of continuity point that laser scanner returns in same scanning process;
Point in scanning sorts and carries out the selection of characteristic point according to C value, wherein choosing the maximum of points of C as edge
Point, the minimum point of C is as planar point;Single pass, which is divided into 4 independent subregions, makes characteristic point be evenly distributed on ring
In border, each subregion at most provides 2 marginal points and 4 planar points, then carries out the registration between adjacent two frames point cloud data,
The association of t moment and t+1 moment point cloud data is completed, and estimates the relative motion relation of radar;The process of point cloud registering are as follows:
For characteristic curve, find a point i point j nearest in t moment point cloud using KD tree, and time near point l is looked for around j, then
(j, l) is known as correspondence of the point i in t moment point cloud;It is similar with characteristic curve for characteristic face, closest approach j is first looked for, is looked for around j
L looks for m around j, and (j, l, m) is known as correspondence of the point i in t moment point cloud;
After finding registration point, the constraint relationship between different moments point cloud is obtained, calculates to correspond to from characteristic point to it and close
The distance of system, since marginal point, for point i, if (j, l) is corresponding edge line, the distance of point to line is calculated are as follows:
WhereinIt is the coordinate of the point i in { L },Be i in last moment corresponding points j,
The coordinate of l;If (j, l, m) is corresponding plane, then the distance of point to face calculates are as follows:
WhereinIt is the coordinate of { L } midpoint m;
To the parameter to be estimated in above formulaIt asks local derviation to obtain Jaccobian matrix, carries out estimation using L-M algorithm
It solves:
Wherein,Rigid motion comprising laser radar on 6DOF,
tx,tyAnd tzRespectively along the x of coordinate system { L }, y and z-axis, θx,θyAnd θzTo rotate angle, it then follows the right-hand rule;Every a line of f
A corresponding characteristic point, d include corresponding distance, and J isFor the Jaccobian matrix of f,Then, pass through
D is leveled off to and zero is obtained by nonlinear iteration:
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein utilize vehicle-mounted three
Dimension laser radar obtains initial data and obtains quasi- road sign set by data filtering and data clustering processing;Then iteration is used
Corresponding relationship in these road signs of closest approach algorithm search and map between road sign calculates two as the corresponding relationship to obtained by
Positional shift T and angle offset r between point set, and utilize the pose of this offset calculating laser radar, it is assumed that it is current to swash
Pose of the optical radar under global coordinate system is expressed as state variable (xL,yL,φL), first two are laser radar in world coordinates
Position under system, Section 3 φLIndicate the direction of advance of laser radar;
Wherein, (xk,yk,φk) it is system state variables, (xL,k,yL,k,φL,k) it is systematic observation variable, T is that position is inclined
It moves, r is angle offset, and v is the observation noise that error matrix is R.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein road roller operation
In the process, due to the limitation of laser radar scanning range, the road roller visual field may without apparent road sign feature, at this time without
Method carries out the pose estimation of road roller;Therefore, the continuity that tracking keeps the estimation of its pose is carried out to road roller pose, using expansion
The data of opening up Kalman filtering fusion odometer and laser radar, count from mileage, calculate the position and side of road roller
Upward increment, as input quantity u=(Δ S, Δ φ)T, it is as follows to establish Vehicular system state equation:
Wherein, w is the process white noise that error matrix is Q.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein the camera is
Big target surface industrial camera.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein described three-dimensional sharp
Optical radar is set on front side of road roller above steel wheel at rack.
Further, the above-mentioned full-automatic road roller localization method based on multisource data fusion, wherein described three-dimensional sharp
Optical radar is installed with tilt angle.
The present invention has significant advantages and beneficial effects compared with prior art, embodies in the following areas:
Vision data and IMU data fusion can adapt to road roller spy by the localization method of machine vision by the present invention
Different construction environment effectively avoids the problem that the signal deletion for relying on GPS positioning generation merely, and in positioning accuracy and speed
Aspect can meet positioning and path planning requirement of the unmanned road roller in construction environment, can be effectively applied to road roller and apply
Work scene;The present invention efficiently solves the problems, such as that road roller positioning accuracy is poor, and can reduce the workload of construction personnel, improves
Labor productivity.
Detailed description of the invention
Fig. 1: structural schematic diagram of the invention.
The meaning of each appended drawing reference see the table below in figure:
Specific embodiment
For a clearer understanding of the technical characteristics, objects and effects of the present invention, specific implementation is now described in detail
Scheme.
The present invention is based on the full-automatic road roller localization methods of multisource data fusion, mark on construction area boundary setting boundary
Will, Multi-path synchronous camera obtains the image comprising boundary marker, and obtains construction area point cloud data using laser radar, will scheme
As being matched with laser point cloud data, identifies construction area mark, calculate construction fence region;Using laser point cloud data with
Position and posture information of the IMU output information fusion calculation road roller in construction area, realize unmanned road roller in construction area
The positioning in domain.
As shown in Figure 1, industrial personal computer 2 is set up in the cockpit of road roller 1, fixed inertial navigation unit 5 at industrial personal computer 2,
Surrounding installs the camera 3 for obtaining road roller construction area image information, the front side of road roller 1 at the top of the cockpit of road roller
Three-dimensional laser radar 4 is set;Construction area bordering is set at the construction field (site).Industrial personal computer provides rich for each sensor unit
Rich expansion interface, and each sensing data is handled, camera, three-dimensional laser radar, inertial navigation unit and industry control
It is attached by being issued based on robot operating system (ROS) nodal information with received communication mode between machine, i.e., in ROS
In program process, node is received by publication topic information for other nodes, completes mutual communication.The process flow of data
It is as follows:
Wherein, four side of front, rear, left and right of road roller 1 is distributed with camera 3, and camera 3 uses big target surface industrial camera, phase
The collocation wide-angle lens of machine 3, is covered 360 ° of ranges around road roller, is demarcated using gridiron pattern to camera, obtain camera internal reference,
And distortion correction is carried out to camera.
Three-dimensional laser radar 4 is set to above the front side steel wheel of road roller 1 at rack, is installed with tilt angle, and three-dimensional swashs
Optical radar 4 scans the target of 30 degree of vertical direction and 360 degree of horizontal directions within the scope of 100 meters of radius, passes through continuity point cloud number
The label to road roller motion profile is obtained according to matched mode;It realizes that road roller is accurately positioned, solves GPS signal missing etc. and ask
Topic.
Construction area bordering is cone, using image and the matched method of point cloud data, is obtained to cone
Effective identification and range measurement, realize that construction area boundary surveys;Realize unmanned road roller in the accurate fixed of construction area
Position.
Inertial navigation unit 5 obtains road roller speed, the information of position, using extended Kalman filter to laser radar
Information is merged with inertial navigation information, realizes synchronization composition and the positioning of unmanned road roller.Realize that unmanned road roller is being applied
The accurate positioning in work area domain.
When it is implemented, including following aspect:
A) boundary marker analyte detection
It is specific to use the convolutional neural networks based on region using the object detection method based on deep learning, that is, it combines
The object detection method of region nomination and convolutional neural networks.It is required in terms of comprehensively considering arithmetic speed and accuracy rate two,
Training Faster R-CNN model under caffe frame, Faster-RCNN introduces RPN network, so that extracted region, classification, recurrence
Convolution feature is shared, guarantees to promote arithmetic speed while computational accuracy.
B) camera is merged with laser data
Cone is pyrometric cone, as construction area bordering, is acquired respectively by three-dimensional laser radar and camera
Three-dimensional laser point cloud and image of the pyrometric cone under two viewpoints, using between the laser point cloud data and image data of foundation
The point cloud semanteme for obtaining pyrometric cone marker is closed in calibration, as vehicle location reference frame.
C) the LOAM localization method based on boundary marker object point cloud
The three dimensional point cloud that scene is obtained using three-dimensional laser radar realizes three-dimensional S LAM composition using LOAM method,
In LOAM method, pass through calculating coordinate change after extraction Feature Points Matching;Cloud and IMU data are pre-processed first,
For extracting characteristic point: the point of single pass is classified by curvature value, and formula is as follows:
Wherein, { L } is the three-dimensional system of coordinate of laser radar, the geometric center originating from laser radar, x-axis direction left side, y
Axis points up, and z-axis is directing forwardly;pkIndicate the point cloud perceived during scanning k, { LkMidpoint i, i ∈ pkCoordinate representation beIt is one group of continuity point that laser scanner returns in same scanning process.
Point in scanning sorts and carries out the selection of characteristic point according to C value, wherein choosing the maximum of points of C as edge
Point, for the minimum point of C as planar point, single pass, which is divided into 4 independent subregions, makes characteristic point be evenly distributed on ring
In border, each subregion at most provides 2 marginal points and 4 planar points, then carries out the registration between adjacent two frames point cloud data,
The association of t moment and t+1 moment point cloud data is completed, and estimates the relative motion relation of radar;The process of point cloud registering are as follows:
For characteristic curve, find a point i point j nearest in t moment point cloud using KD tree, and time near point l is looked for around j, then
(j, l) is known as correspondence of the point i in t moment point cloud;It is similar with characteristic curve for characteristic face, closest approach j is first looked for, is looked for around j
L looks for m around j, and (j, l, m) is known as correspondence of the point i in t moment point cloud;
After finding registration point, the constraint relationship between different moments point cloud is obtained, calculates to correspond to from characteristic point to it and close
The distance of system, since marginal point, for point i, if (j, l) is corresponding edge line, the distance put to line be may be calculated:
WhereinIt is the coordinate of the point i in { L },Be i in last moment corresponding points j,
The coordinate of l.If (j, l, m) is corresponding plane, then the distance of point to face calculates are as follows:
WhereinIt is the coordinate of { L } midpoint m.
To the parameter to be estimated in above formulaIt asks local derviation to obtain Jaccobian matrix, carries out estimation using L-M algorithm
It solves:
Wherein,Rigid motion comprising laser radar on 6DOF,
tx,tyAnd tzRespectively along the x of coordinate system { L }, y and z-axis, θx,θyAnd θzTo rotate angle, it then follows the right-hand rule.Every a line of f
A corresponding characteristic point, d include corresponding distance, and J isFor the Jaccobian matrix of f,Then, pass through
D is leveled off to and zero is obtained by nonlinear iteration:
D) road roller athletic posture is continuously estimated
Initial data, which is obtained, using vehicle-mounted three-dimensional laser radar obtains quasi- road by data filtering and data clustering processing
Mark set;Then using the corresponding relationship between road sign in these road signs of iteration closest approach algorithm search and map, by institute
It obtains corresponding relationship and calculates positional shift T and angle offset r between two point sets, and calculate laser thunder using this offset
The pose reached, it is assumed that pose of the present laser radar under global coordinate system is expressed as state variable (xL,yL,φL), first two
For position of the laser radar under global coordinate system, Section 3 φLIndicate the direction of advance of laser radar;
Wherein, wherein (xk,yk,φk) it is system state variables, (xL,k,yL,k,φL,k) it is systematic observation variable, T is position
Offset is set, r is angle offset, and v is the observation noise that error matrix is R.
In road roller operational process, due to the limitation of laser radar scanning range, the road roller visual field may be not bright
Aobvious road sign feature can not carry out the pose estimation of road roller at this time;Therefore, tracking is carried out to road roller pose and keeps its pose
The continuity of estimation is counted from mileage using the data of Extended Kalman filter fusion odometer and laser radar, is calculated
Increment in the position and direction of road roller out, as input quantity u=(Δ S, Δ φ)T, establish Vehicular system state equation such as
Under:
Wherein, w is the process white noise that error matrix is Q.
In conclusion marker is arranged on construction area periphery in the present invention, marker is cone, as laser point cloud number
According to the index point in fusing image data positioning auxiliary;Using the marker of the method identification construction area based on machine learning
And mobile target, construction site image data is acquired, marker and mobile target, the training under caffe frame are marked
Faster R-CNN model identifies construction area marker and mobile target;By the image data of camera acquisition, laser point cloud number
Accordingly and IMU inertial data merges, and solves position and posture information of the road roller in construction area.
By the localization method of machine vision, by vision data and IMU data fusion, it can adapt to that road roller is special to be applied
Work environment effectively avoids the problem that the signal deletion for relying on GPS positioning generation merely, and in terms of positioning accuracy and speed all
It can satisfy positioning and path planning requirement of the unmanned road roller in construction environment, it is existing to can be effectively applied to road roller construction
?.The present invention efficiently solves the problems, such as that road roller positioning accuracy is poor, and can reduce the workload of construction personnel, improves labour
Productivity.
It should be understood that the foregoing is merely the preferred embodiment of the present invention, the power that is not intended to limit the invention
Sharp range;The description above simultaneously, should can be illustrated and implement for the special personage of correlative technology field, thus it is other without departing from
The equivalent change or modification completed under disclosed spirit, should be included in claim.
Claims (10)
1. the full-automatic road roller localization method based on multisource data fusion, it is characterised in that: side is arranged on construction area boundary
Boundary mark will, as the index point in laser point cloud data and fusing image data positioning auxiliary, the acquisition of Multi-path synchronous camera includes
The image of boundary marker, and construction area point cloud data is obtained using laser radar, by image and laser point cloud data progress
Match, identify construction area mark, calculates construction fence region;Utilize laser point cloud data and IMU output information fusion calculation pressure
Position and posture information of the road machine in construction area, realize unmanned road roller in the positioning of construction area.
2. the full-automatic road roller localization method according to claim 1 based on multisource data fusion, it is characterised in that:
Construction area bordering is arranged in construction site, installation industrial personal computer and inertial navigation unit on road roller, and the four of road roller
Camera for obtaining road roller construction area image information is installed in week, three-dimensional laser radar is set on front side of road roller.
3. the full-automatic road roller localization method according to claim 2 based on multisource data fusion, it is characterised in that: pressure
Camera is distributed in four side of front, rear, left and right of road machine, and camera collocation wide-angle lens covers 360 ° of ranges around road roller, utilizes
Gridiron pattern demarcates camera, obtains camera internal reference, and carry out distortion correction to camera;
The target of three-dimensional laser radar scanning 30 degree of vertical direction and 360 degree of horizontal directions within the scope of 100 meters of radius, leads to
The mode for crossing continuity point cloud Data Matching obtains label to road roller motion profile;
The construction area bordering is cone, using image and the matched method of point cloud data, is obtained to cone
Effective identification and range measurement, realize that construction area boundary surveys;
The inertial navigation unit obtains road roller speed, the information of position, using extended Kalman filter to laser radar
Information is merged with inertial navigation information, realizes synchronization composition and the positioning of unmanned road roller.
4. the full-automatic road roller localization method according to claim 3 based on multisource data fusion, it is characterised in that: institute
It states cone and pyrometric cone is acquired by three-dimensional laser radar and camera respectively as construction area bordering for pyrometric cone
Three-dimensional laser point cloud and image under two viewpoints are closed using the calibration between the laser point cloud data and image data of foundation
The point cloud for obtaining pyrometric cone marker is semantic, as vehicle location reference frame.
5. the full-automatic road roller localization method according to claim 3 based on multisource data fusion, it is characterised in that: benefit
The three dimensional point cloud that scene is obtained with three-dimensional laser radar realizes three-dimensional S LAM composition using LOAM method, in LOAM method
In, pass through calculating coordinate change after extraction Feature Points Matching;Cloud and IMU data are pre-processed first, for extracting spy
Levy point: the point of single pass is classified by curvature value, and formula is as follows:
Wherein, the three-dimensional system of coordinate of { L } for laser radar, the geometric center originating from laser radar, x-axis are directed toward left side, and y-axis refers to
Upwards, z-axis is directing forwardly;pkIndicate the point cloud perceived during scanning k, { LkMidpoint i, i ∈ pkCoordinate representation beS is one group of continuity point that laser scanner returns in same scanning process;
Point in scanning sorts and carries out the selection of characteristic point according to C value, wherein choose the maximum of points of C as marginal point, C's
Minimum point is as planar point;Single pass, which is divided into 4 independent subregions, is uniformly distributed characteristic point in the environment, often
Sub-regions at most provide 2 marginal points and 4 planar points, then carry out the registration between adjacent two frames point cloud data, i.e. completion t
The association at moment and t+1 moment point cloud data, and estimate the relative motion relation of radar;The process of point cloud registering are as follows: for spy
Line is levied, a point i point j nearest in t moment point cloud is found using KD tree, and look for time near point l around j, then (j, l) is claimed
For correspondence of the point i in t moment point cloud;It is similar with characteristic curve for characteristic face, closest approach j is first looked for, l is looked for around j, at j weeks
It encloses and looks for m, (j, l, m) is known as correspondence of the point i in t moment point cloud;
After finding registration point, the constraint relationship between different moments point cloud is obtained, is calculated from characteristic point to its corresponding relationship
Distance, since marginal point, for point i, if (j, l) is corresponding edge line, the distance of point to line is calculated are as follows:
WhereinIt is the coordinate of the point i in { L },It is corresponding points j, l of i in last moment
Coordinate;If (j, l, m) is corresponding plane, then the distance of point to face calculates are as follows:
WhereinIt is the coordinate of { L } midpoint m;
To the parameter to be estimated in above formulaIt asks local derviation to obtain Jaccobian matrix, carries out estimation solution using L-M algorithm:
Wherein,Rigid motion comprising laser radar on 6DOF,tx,ty
And tzRespectively along the x of coordinate system { L }, y and z-axis, θx,θyAnd θzTo rotate angle, it then follows the right-hand rule;Every a line of f is corresponding
One characteristic point, d include corresponding distance, and J isFor the Jaccobian matrix of f,Then, by non-thread
D is leveled off to and zero is obtained by property iteration:
6. the full-automatic road roller localization method according to claim 3 based on multisource data fusion, it is characterised in that: benefit
Initial data, which is obtained, with vehicle-mounted three-dimensional laser radar obtains quasi- road sign set by data filtering and data clustering processing;Then
Using the corresponding relationship between road sign in these road signs of iteration closest approach algorithm search and map, pass through the corresponding relationship meter to obtained by
The positional shift T and angle offset r between two point sets are calculated, and calculates the pose of laser radar using this offset, it is false
If pose of the present laser radar under global coordinate system is expressed as state variable (xL,yL,φL), first two exist for laser radar
Position under global coordinate system, Section 3 φLIndicate the direction of advance of laser radar;
Wherein, (xk,yk,φk) it is system state variables, (xL,k,yL,k,φL,k) it is systematic observation variable, T is positional shift, r
For angle offset, v is the observation noise that error matrix is R.
7. the full-automatic road roller localization method according to claim 6 based on multisource data fusion, it is characterised in that: pressure
In the machine operational process of road, due to the limitation of laser radar scanning range, the road roller visual field may be special without apparent road sign
Sign can not carry out the pose estimation of road roller at this time;Therefore, tracking is carried out to road roller pose and keeps the continuous of its pose estimation
Property, it using the data of Extended Kalman filter fusion odometer and laser radar, counts from mileage, calculates road roller
Increment in position and direction, as input quantity u=(Δ S, Δ φ)T, it is as follows to establish Vehicular system state equation:
Wherein, w is the process white noise that error matrix is Q.
8. the full-automatic road roller localization method according to claim 2 based on multisource data fusion, it is characterised in that: institute
Stating camera is big target surface industrial camera.
9. the full-automatic road roller localization method according to claim 2 based on multisource data fusion, it is characterised in that: institute
Three-dimensional laser radar is stated to be set on front side of road roller above steel wheel at rack.
10. the full-automatic road roller localization method according to claim 2 based on multisource data fusion, it is characterised in that:
The three-dimensional laser radar is installed with tilt angle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810670215.2A CN109099901B (en) | 2018-06-26 | 2018-06-26 | Full-automatic road roller positioning method based on multi-source data fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810670215.2A CN109099901B (en) | 2018-06-26 | 2018-06-26 | Full-automatic road roller positioning method based on multi-source data fusion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109099901A true CN109099901A (en) | 2018-12-28 |
CN109099901B CN109099901B (en) | 2021-09-24 |
Family
ID=64845012
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810670215.2A Active CN109099901B (en) | 2018-06-26 | 2018-06-26 | Full-automatic road roller positioning method based on multi-source data fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109099901B (en) |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110082739A (en) * | 2019-03-20 | 2019-08-02 | 深圳市速腾聚创科技有限公司 | Method of data synchronization and equipment |
CN110147095A (en) * | 2019-03-15 | 2019-08-20 | 广东工业大学 | Robot method for relocating based on mark information and Fusion |
CN110244322A (en) * | 2019-06-28 | 2019-09-17 | 东南大学 | Pavement construction robot environment sensory perceptual system and method based on Multiple Source Sensor |
CN110389590A (en) * | 2019-08-19 | 2019-10-29 | 杭州电子科技大学 | A kind of AGV positioning system and method merging 2D environmental map and sparse artificial landmark |
CN110568447A (en) * | 2019-07-29 | 2019-12-13 | 广东星舆科技有限公司 | Visual positioning method, device and computer readable medium |
CN110749327A (en) * | 2019-08-08 | 2020-02-04 | 南京航空航天大学 | Vehicle navigation method in cooperation environment |
CN110782497A (en) * | 2019-09-06 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Method and device for calibrating external parameters of camera |
CN110849362A (en) * | 2019-11-28 | 2020-02-28 | 湖南率为控制科技有限公司 | Laser radar and vision combined navigation algorithm based on vehicle-mounted inertia |
CN111364549A (en) * | 2020-02-28 | 2020-07-03 | 江苏徐工工程机械研究院有限公司 | Synchronous drawing and automatic operation method and system based on laser radar |
WO2020154970A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Deep learning–based feature extraction for lidar localization of autonomous driving vehicles |
CN111551976A (en) * | 2020-05-20 | 2020-08-18 | 四川万网鑫成信息科技有限公司 | Method for automatically completing abnormal positioning by combining various data |
CN111595333A (en) * | 2020-04-26 | 2020-08-28 | 武汉理工大学 | Modularized unmanned vehicle positioning method and system based on visual inertial laser data fusion |
CN111709355A (en) * | 2020-06-12 | 2020-09-25 | 北京百度网讯科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
CN111754798A (en) * | 2020-07-02 | 2020-10-09 | 上海电科智能***股份有限公司 | Method for realizing detection of vehicle and surrounding obstacles by fusing roadside laser radar and video |
CN111929699A (en) * | 2020-07-21 | 2020-11-13 | 北京建筑大学 | Laser radar inertial navigation odometer considering dynamic obstacles and mapping method and system |
CN111947647A (en) * | 2020-08-26 | 2020-11-17 | 四川阿泰因机器人智能装备有限公司 | Robot accurate positioning method integrating vision and laser radar |
CN111998832A (en) * | 2020-08-12 | 2020-11-27 | 河北雷神科技有限公司 | Laser point cloud-based inspection method for accurately positioning target object by using unmanned aerial vehicle |
CN112254729A (en) * | 2020-10-09 | 2021-01-22 | 北京理工大学 | Mobile robot positioning method based on multi-sensor fusion |
CN112334790A (en) * | 2019-08-21 | 2021-02-05 | 深圳市大疆创新科技有限公司 | Positioning system and positioning method for movable object, and storage medium |
CN112540382A (en) * | 2019-09-07 | 2021-03-23 | 山东大学 | Laser navigation AGV auxiliary positioning method based on visual identification detection |
WO2021056190A1 (en) * | 2019-09-24 | 2021-04-01 | Beijing Didi Infinity Technology And Development Co., Ltd. | Semantic-assisted multi-resolution point cloud registration |
CN112904395A (en) * | 2019-12-03 | 2021-06-04 | 青岛慧拓智能机器有限公司 | Mining vehicle positioning system and method |
CN113091736A (en) * | 2021-04-02 | 2021-07-09 | 京东数科海益信息科技有限公司 | Robot positioning method, device, robot and storage medium |
CN113359782A (en) * | 2021-05-28 | 2021-09-07 | 福建工程学院 | Unmanned aerial vehicle autonomous addressing landing method integrating LIDAR point cloud and image data |
CN113777600A (en) * | 2021-09-09 | 2021-12-10 | 北京航空航天大学杭州创新研究院 | Multi-millimeter-wave radar cooperative positioning tracking method |
CN114429432A (en) * | 2022-04-07 | 2022-05-03 | 科大天工智能装备技术(天津)有限公司 | Multi-source information layered fusion method and device and storage medium |
CN114663992A (en) * | 2022-03-18 | 2022-06-24 | 福建工程学院 | Multi-source data fusion expressway portal positioning method |
WO2022142827A1 (en) * | 2020-12-30 | 2022-07-07 | 华为技术有限公司 | Road occupancy information determination method and apparatus |
CN115235478A (en) * | 2022-09-23 | 2022-10-25 | 武汉理工大学 | Intelligent automobile positioning method and system based on visual label and laser SLAM |
WO2022256976A1 (en) * | 2021-06-07 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Method and system for constructing dense point cloud truth value data and electronic device |
US11550058B2 (en) | 2020-04-10 | 2023-01-10 | Caterpillar Paving Products Inc. | Perception system three lidar coverage |
CN116299367A (en) * | 2023-05-18 | 2023-06-23 | 中国测绘科学研究院 | Multi-laser space calibration method |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101017080A (en) * | 2002-09-30 | 2007-08-15 | 石川岛播磨重工业株式会社 | Device of measuring object |
CN101469998A (en) * | 2007-12-27 | 2009-07-01 | 爱信艾达株式会社 | Feature information collecting apparatus and feature information collecting program, and own vehicle position recognition apparatus and navigation apparatus |
CN103389103A (en) * | 2013-07-03 | 2013-11-13 | 北京理工大学 | Geographical environmental characteristic map construction and navigation method based on data mining |
CN105210128A (en) * | 2013-04-10 | 2015-12-30 | 谷歌公司 | Mapping active and inactive construction zones for autonomous driving |
CN105512646A (en) * | 2016-01-19 | 2016-04-20 | 腾讯科技(深圳)有限公司 | Data processing method, data processing device and terminal |
CN105719284A (en) * | 2016-01-18 | 2016-06-29 | 腾讯科技(深圳)有限公司 | Data processing method, device and terminal |
CN106127153A (en) * | 2016-06-24 | 2016-11-16 | 南京林业大学 | The traffic sign recognition methods of Vehicle-borne Laser Scanning cloud data |
CN106774335A (en) * | 2017-01-03 | 2017-05-31 | 南京航空航天大学 | Guiding device based on multi-vision visual and inertial navigation, terrestrial reference layout and guidance method |
CN106932780A (en) * | 2017-03-14 | 2017-07-07 | 北京京东尚科信息技术有限公司 | Object positioning method, device and system |
CN107111885A (en) * | 2014-12-30 | 2017-08-29 | 诺基亚技术有限公司 | For the method for the position for determining portable set |
CN107463918A (en) * | 2017-08-17 | 2017-12-12 | 武汉大学 | Lane line extracting method based on laser point cloud and image data fusion |
CN107688184A (en) * | 2017-07-24 | 2018-02-13 | 宗晖(上海)机器人有限公司 | A kind of localization method and system |
CN108196535A (en) * | 2017-12-12 | 2018-06-22 | 清华大学苏州汽车研究院(吴江) | Automated driving system based on enhancing study and Multi-sensor Fusion |
CN108628206A (en) * | 2017-03-17 | 2018-10-09 | 通用汽车环球科技运作有限责任公司 | Road construction detecting system and method |
-
2018
- 2018-06-26 CN CN201810670215.2A patent/CN109099901B/en active Active
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101017080A (en) * | 2002-09-30 | 2007-08-15 | 石川岛播磨重工业株式会社 | Device of measuring object |
CN101469998A (en) * | 2007-12-27 | 2009-07-01 | 爱信艾达株式会社 | Feature information collecting apparatus and feature information collecting program, and own vehicle position recognition apparatus and navigation apparatus |
CN105210128A (en) * | 2013-04-10 | 2015-12-30 | 谷歌公司 | Mapping active and inactive construction zones for autonomous driving |
CN107976200A (en) * | 2013-04-10 | 2018-05-01 | 伟摩有限责任公司 | The method and system of vehicle of the operation with autonomous driving pattern |
CN103389103A (en) * | 2013-07-03 | 2013-11-13 | 北京理工大学 | Geographical environmental characteristic map construction and navigation method based on data mining |
CN107111885A (en) * | 2014-12-30 | 2017-08-29 | 诺基亚技术有限公司 | For the method for the position for determining portable set |
CN105719284A (en) * | 2016-01-18 | 2016-06-29 | 腾讯科技(深圳)有限公司 | Data processing method, device and terminal |
CN105512646A (en) * | 2016-01-19 | 2016-04-20 | 腾讯科技(深圳)有限公司 | Data processing method, data processing device and terminal |
CN106127153A (en) * | 2016-06-24 | 2016-11-16 | 南京林业大学 | The traffic sign recognition methods of Vehicle-borne Laser Scanning cloud data |
CN106774335A (en) * | 2017-01-03 | 2017-05-31 | 南京航空航天大学 | Guiding device based on multi-vision visual and inertial navigation, terrestrial reference layout and guidance method |
CN106932780A (en) * | 2017-03-14 | 2017-07-07 | 北京京东尚科信息技术有限公司 | Object positioning method, device and system |
CN108628206A (en) * | 2017-03-17 | 2018-10-09 | 通用汽车环球科技运作有限责任公司 | Road construction detecting system and method |
CN107688184A (en) * | 2017-07-24 | 2018-02-13 | 宗晖(上海)机器人有限公司 | A kind of localization method and system |
CN107463918A (en) * | 2017-08-17 | 2017-12-12 | 武汉大学 | Lane line extracting method based on laser point cloud and image data fusion |
CN108196535A (en) * | 2017-12-12 | 2018-06-22 | 清华大学苏州汽车研究院(吴江) | Automated driving system based on enhancing study and Multi-sensor Fusion |
Cited By (45)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020154970A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Deep learning–based feature extraction for lidar localization of autonomous driving vehicles |
CN110147095A (en) * | 2019-03-15 | 2019-08-20 | 广东工业大学 | Robot method for relocating based on mark information and Fusion |
CN110082739A (en) * | 2019-03-20 | 2019-08-02 | 深圳市速腾聚创科技有限公司 | Method of data synchronization and equipment |
CN110082739B (en) * | 2019-03-20 | 2022-04-12 | 深圳市速腾聚创科技有限公司 | Data synchronization method and device |
CN110244322A (en) * | 2019-06-28 | 2019-09-17 | 东南大学 | Pavement construction robot environment sensory perceptual system and method based on Multiple Source Sensor |
CN110568447A (en) * | 2019-07-29 | 2019-12-13 | 广东星舆科技有限公司 | Visual positioning method, device and computer readable medium |
CN110749327A (en) * | 2019-08-08 | 2020-02-04 | 南京航空航天大学 | Vehicle navigation method in cooperation environment |
CN110389590A (en) * | 2019-08-19 | 2019-10-29 | 杭州电子科技大学 | A kind of AGV positioning system and method merging 2D environmental map and sparse artificial landmark |
CN110389590B (en) * | 2019-08-19 | 2022-07-05 | 杭州电子科技大学 | AGV positioning system and method integrating 2D environment map and sparse artificial landmark |
CN112334790A (en) * | 2019-08-21 | 2021-02-05 | 深圳市大疆创新科技有限公司 | Positioning system and positioning method for movable object, and storage medium |
CN110782497A (en) * | 2019-09-06 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Method and device for calibrating external parameters of camera |
CN110782497B (en) * | 2019-09-06 | 2022-04-29 | 腾讯科技(深圳)有限公司 | Method and device for calibrating external parameters of camera |
CN112540382B (en) * | 2019-09-07 | 2024-02-13 | 山东大学 | Laser navigation AGV auxiliary positioning method based on visual identification detection |
CN112540382A (en) * | 2019-09-07 | 2021-03-23 | 山东大学 | Laser navigation AGV auxiliary positioning method based on visual identification detection |
WO2021056190A1 (en) * | 2019-09-24 | 2021-04-01 | Beijing Didi Infinity Technology And Development Co., Ltd. | Semantic-assisted multi-resolution point cloud registration |
CN110849362A (en) * | 2019-11-28 | 2020-02-28 | 湖南率为控制科技有限公司 | Laser radar and vision combined navigation algorithm based on vehicle-mounted inertia |
CN110849362B (en) * | 2019-11-28 | 2022-01-04 | 湖南率为控制科技有限公司 | Laser radar and vision combined navigation algorithm based on vehicle-mounted inertia |
CN112904395A (en) * | 2019-12-03 | 2021-06-04 | 青岛慧拓智能机器有限公司 | Mining vehicle positioning system and method |
CN111364549B (en) * | 2020-02-28 | 2021-11-09 | 江苏徐工工程机械研究院有限公司 | Synchronous drawing and automatic operation method and system based on laser radar |
CN111364549A (en) * | 2020-02-28 | 2020-07-03 | 江苏徐工工程机械研究院有限公司 | Synchronous drawing and automatic operation method and system based on laser radar |
US11550058B2 (en) | 2020-04-10 | 2023-01-10 | Caterpillar Paving Products Inc. | Perception system three lidar coverage |
CN111595333A (en) * | 2020-04-26 | 2020-08-28 | 武汉理工大学 | Modularized unmanned vehicle positioning method and system based on visual inertial laser data fusion |
CN111551976A (en) * | 2020-05-20 | 2020-08-18 | 四川万网鑫成信息科技有限公司 | Method for automatically completing abnormal positioning by combining various data |
CN111709355B (en) * | 2020-06-12 | 2023-08-29 | 阿波罗智联(北京)科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
CN111709355A (en) * | 2020-06-12 | 2020-09-25 | 北京百度网讯科技有限公司 | Method and device for identifying target area, electronic equipment and road side equipment |
CN111754798A (en) * | 2020-07-02 | 2020-10-09 | 上海电科智能***股份有限公司 | Method for realizing detection of vehicle and surrounding obstacles by fusing roadside laser radar and video |
CN111929699A (en) * | 2020-07-21 | 2020-11-13 | 北京建筑大学 | Laser radar inertial navigation odometer considering dynamic obstacles and mapping method and system |
CN111998832A (en) * | 2020-08-12 | 2020-11-27 | 河北雷神科技有限公司 | Laser point cloud-based inspection method for accurately positioning target object by using unmanned aerial vehicle |
CN111947647B (en) * | 2020-08-26 | 2024-05-17 | 四川阿泰因机器人智能装备有限公司 | Precise positioning method for vision and laser radar integrated robot |
CN111947647A (en) * | 2020-08-26 | 2020-11-17 | 四川阿泰因机器人智能装备有限公司 | Robot accurate positioning method integrating vision and laser radar |
CN112254729A (en) * | 2020-10-09 | 2021-01-22 | 北京理工大学 | Mobile robot positioning method based on multi-sensor fusion |
WO2022142827A1 (en) * | 2020-12-30 | 2022-07-07 | 华为技术有限公司 | Road occupancy information determination method and apparatus |
CN113091736A (en) * | 2021-04-02 | 2021-07-09 | 京东数科海益信息科技有限公司 | Robot positioning method, device, robot and storage medium |
CN113359782B (en) * | 2021-05-28 | 2022-07-29 | 福建工程学院 | Unmanned aerial vehicle autonomous addressing landing method integrating LIDAR point cloud and image data |
CN113359782A (en) * | 2021-05-28 | 2021-09-07 | 福建工程学院 | Unmanned aerial vehicle autonomous addressing landing method integrating LIDAR point cloud and image data |
WO2022256976A1 (en) * | 2021-06-07 | 2022-12-15 | 深圳市大疆创新科技有限公司 | Method and system for constructing dense point cloud truth value data and electronic device |
CN113777600B (en) * | 2021-09-09 | 2023-09-15 | 北京航空航天大学杭州创新研究院 | Multi-millimeter wave radar co-location tracking method |
CN113777600A (en) * | 2021-09-09 | 2021-12-10 | 北京航空航天大学杭州创新研究院 | Multi-millimeter-wave radar cooperative positioning tracking method |
CN114663992A (en) * | 2022-03-18 | 2022-06-24 | 福建工程学院 | Multi-source data fusion expressway portal positioning method |
CN114429432B (en) * | 2022-04-07 | 2022-06-21 | 科大天工智能装备技术(天津)有限公司 | Multi-source information layered fusion method and device and storage medium |
CN114429432A (en) * | 2022-04-07 | 2022-05-03 | 科大天工智能装备技术(天津)有限公司 | Multi-source information layered fusion method and device and storage medium |
CN115235478A (en) * | 2022-09-23 | 2022-10-25 | 武汉理工大学 | Intelligent automobile positioning method and system based on visual label and laser SLAM |
CN115235478B (en) * | 2022-09-23 | 2023-04-07 | 武汉理工大学 | Intelligent automobile positioning method and system based on visual label and laser SLAM |
CN116299367B (en) * | 2023-05-18 | 2024-01-26 | 中国测绘科学研究院 | Multi-laser space calibration method |
CN116299367A (en) * | 2023-05-18 | 2023-06-23 | 中国测绘科学研究院 | Multi-laser space calibration method |
Also Published As
Publication number | Publication date |
---|---|
CN109099901B (en) | 2021-09-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109099901A (en) | Full-automatic road roller localization method based on multisource data fusion | |
CN110084272B (en) | Cluster map creation method and repositioning method based on cluster map and position descriptor matching | |
US10897575B2 (en) | Lidar to camera calibration for generating high definition maps | |
CN108571971B (en) | AGV visual positioning system and method | |
CN106651953B (en) | A kind of vehicle position and orientation estimation method based on traffic sign | |
CN111652179A (en) | Semantic high-precision map construction and positioning method based on dotted line feature fusion laser | |
CN103411609B (en) | A kind of aircraft return route planing method based on online composition | |
CN106296814B (en) | Highway maintenance detection and virtual interactive interface method and system | |
CN105667518A (en) | Lane detection method and device | |
CN104729485A (en) | Visual positioning method based on vehicle-mounted panorama image and streetscape matching | |
CN110308729A (en) | The AGV combined navigation locating method of view-based access control model and IMU or odometer | |
CN103837143B (en) | Super-mapping machine | |
CN103885455B (en) | Tracking measurement robot | |
CN103868504B (en) | Autonomous surveying and mapping machine | |
CN110263607A (en) | A kind of for unpiloted road grade global context drawing generating method | |
CN106705962A (en) | Method and system for acquiring navigation data | |
CN111721279A (en) | Tail end path navigation method suitable for power transmission inspection work | |
CN112749584B (en) | Vehicle positioning method based on image detection and vehicle-mounted terminal | |
CN116883604A (en) | Three-dimensional modeling technical method based on space, air and ground images | |
Hara et al. | Vehicle localization based on the detection of line segments from multi-camera images | |
Alshaiba et al. | Automatic manhole extraction from MMS data to update basemaps | |
CN112446915A (en) | Picture-establishing method and device based on image group | |
CN110415299B (en) | Vehicle position estimation method based on set guideboard under motion constraint | |
CN112530270B (en) | Mapping method and device based on region allocation | |
KR20210098534A (en) | Methods and systems for creating environmental models for positioning |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20190404 Address after: Room 525, Building 333 Xingpu Road, Suzhou Industrial Park, Jiangsu Province Applicant after: Zhongke Weiyi (Suzhou) Intelligent Technology Co., Ltd. Address before: 215021 Linquan Street 399, Suzhou Industrial Park, Jiangsu Province Applicant before: Suzhou Road Agent Intelligent Technology Co., Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |