CN109460032A - A kind of localization method and autonomous charging of robots method based on laser-correlation - Google Patents
A kind of localization method and autonomous charging of robots method based on laser-correlation Download PDFInfo
- Publication number
- CN109460032A CN109460032A CN201811449494.6A CN201811449494A CN109460032A CN 109460032 A CN109460032 A CN 109460032A CN 201811449494 A CN201811449494 A CN 201811449494A CN 109460032 A CN109460032 A CN 109460032A
- Authority
- CN
- China
- Prior art keywords
- laser
- robot
- pose
- grid
- candidate
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0225—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving docking at a fixed facility, e.g. base station or loading bay
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0234—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using optical markers or beacons
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
- G05D1/027—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means comprising intertial navigation means, e.g. azimuth detector
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
- G05D1/0272—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means comprising means for registering the travel distance, e.g. revolutions of wheels
Abstract
The invention discloses a kind of localization method based on laser-correlation and autonomous charging of robots methods, belong to intelligent robot technology field.Localization method of the invention enters positioning when charging house is charged for robot, comprising the following steps: judges the step of whether robot has entered charging house；The step of coarse positioning is carried out to robot；The step of fine positioning is carried out to robot: the laser aiming signal that the laser emitter where search target issues；When the laser pickoff of the laser pickoff dot matrix middle finger number of delimiting the organizational structure in robot receives laser aiming signal, robot is mobile towards the direction of laser aiming signal, guarantees that laser aiming signal is correctly received in laser pickoff dot matrix in moving process.Using method of the invention, autonomous charging of robots suitable for trackless navigator fix realizes automation, the intelligence of entire charging process；No matter whether charging house wall is smooth, very high navigation and positioning accuracy may have.
Description
Technical field
The invention belongs to intelligent robot technology fields, and in particular to a kind of localization method and machine based on laser-correlation
Device people's recharging method.
Background technique
Currently, the recharging of Intelligent Mobile Robot mostly uses contact recharging technology.Work as crusing robot
When needing to charge, specified charging zone can be driven towards automatically, the power supply contact of robot charging connection contact and recharging base is automatically quasi-
Really docks and implement to charge.Crusing robot is detached from recharging base automatically and drives towards workspace or the area that awaits orders after charging complete.
How to realize that the location navigation of crusing robot is the key that robot realizes recharging.Currently used scheme
It is the navigation locating method based on magnetic orbital guidance.The premise that this method is implemented is to be laid with track, not only brings a large amount of base
Infrastructure construction work, and there are the inflexible defects of robot ambulation route.In consideration of it, laser navigation location technology is got over
Come in more navigator fix operation process for being applied to crusing robot.Laser navigation location technology gets rid of track limitation,
While saving cost, robot can neatly adjust walking route.
Localization method of the robot in charging house scene mostly uses the aggregate information for constructing charging house, utilizes charging house
Edge feature or the position of key point robot is positioned.For example, Chinese patent CN2016112609096 discloses one
Kind crusing robot localization method and automatic recharging method, first by secondary set partitioning algorithm rapidly and efficiently to continuous
Laser data collection is split, and obtains higher three edge features of robustness, then charging house model is reconstructed, and is utilized
The corresponding key point of reconstruct charging house model positions robot.While guaranteeing positioning accuracy, do not directly rely on
In charging house corner inflection point, charging house building technology is relied on lower.
The above method has the following problems:
(1) it if charging house or so wall out-of-flatness, will lead to fitting a straight line and deviation occur, influences charging house corner and turns
The accuracy of point position, keeps the navigation and positioning accuracy of robot undesirable, to robot charging connection contact and charging occur
The power supply contact of pedestal docks the phenomenon that failure.It is right when carrying out the navigator fix and recharging of robot using the above method
The planarization of charging house wall is more demanding, and robot is poor to the adaptability of environment.
(2) using the environmental information of dimension sensor acquisition charging house, detection range is small, and the Limited information of data volume is made
It is difficult to meet the needs of actual scene at robot navigation's positioning accuracy is not ideal enough.
Summary of the invention
Object of the present invention is to: in view of the deficiencies of the prior art, provide a kind of localization method and base based on laser-correlation
In the autonomous charging of robots method of the localization method.Robot autonomous suitable for trackless navigator fix of this method is filled
Electricity realizes automation, the intelligence of entire charging process；No matter whether charging house wall is smooth, very high lead may have
Boat positioning accuracy.
On the one hand, the localization method provided by the invention based on laser-correlation, enters charging house for robot and is filled
Positioning when electric, comprising the following steps:
Judge the step of whether robot has entered charging house:
Metope, left side metope and right side metope establish three-dimensional laser grid immediately ahead of robot in charging house；Work as basis
When the three-dimensional laser sensor of robot scans three-dimensional laser data acquisition obtained by the metope to standard feature grid, it is believed that machine
Device people has entered charging house；
The step of coarse positioning is carried out to robot:
Three of three-dimensional laser sensor during robot is mobile to object pose by each frame from robot
After dimension laser data is processed into laser point cloud data, the laser point cloud data in 3 d grid space is projected into two-dimensional laser grid
In lattice plane, according to pose calculating robot velocity of displacement in the two continuous frames laser data on the two-dimensional surface and rotation speed
Degree, and pass through former frame pose, the prediction pose initial value of velocity of displacement and rotation speed calculating next frame；
It is counted according to newest received mileage and school is carried out to prediction pose initial value with newest received inertial navigation data
Just obtaining the estimated value of pose；
Positioning scanning window size is calculated, each scanning angle on each map grid is determined according to positioning scanning window size
Pose, as all possible candidate pose；After often obtaining laser data of the frame from the two-dimensional laser sensor of robot,
Calculate coordinate of each laser reflection point in map coordinates system under each scanning angle；And map datum is combined to calculate each scanning angle
Under coordinate of the corresponding map grid of each laser reflection point in map coordinates system, the discrete scanning number as each scanning angle
According to；
The confidence level for calculating each possible candidate pose, selects the highest pose of confidence level score value as the optimal position of robot
Appearance estimation, the i.e. current pose of robot；
The step of fine positioning is carried out to robot:
It is specified to judge whether robot arrived distance objective pose according to the current pose of the robot and object pose
In range；
The laser aiming signal that laser emitter where searching for target issues；
When the laser pickoff of the laser pickoff dot matrix middle finger number of delimiting the organizational structure in robot receives laser aiming signal,
Robot is mobile towards the direction of laser aiming signal, guarantees that laser pickoff dot matrix is correctly received laser and draws in moving process
Lead signal.
Furthermore, described that three-dimensional laser data obtained by the metope are scanned according to the three-dimensional laser sensor of robot
Obtain standard feature grid method include:
The metope is scanned using three-dimensional laser sensor；
Laser point cloud data in 3 d grid space is projected on two-dimensional laser raster plane；
Horizontal direction and the orthogonal continuous grey grid of vertical direction are extracted on two-dimensional laser raster plane；
It verifies the horizontal direction extracted and whether the orthogonal continuous grey grid of vertical direction is in same water
On horizontal line, meet condition is standard feature grid；When being unsatisfactory for condition, returns and scan institute using three-dimensional laser sensor
The step of stating metope carries out above-mentioned processing to next frame laser data.
Furthermore, the laser point cloud data in 3 d grid space is projected and is wrapped on two-dimensional laser raster plane
It includes,
Laser point cloud data in 3 d grid space is projected in the two-dimensional grid plane among laser where harness,
Obtaining robot coordinate in global coordinate system X0Y is p (x, y, z), and wherein x, the value of y, z are as follows:
Wherein, point S is the distance of a certain laser point-to-point 0 in 3 d grid space, and α is the laser point relative to level
The angle in direction, θ are angle of the laser point relative to vertical direction.
Furthermore, the laser point cloud data by 3 d grid space projects on two-dimensional laser raster plane
Before, further includes: after often obtaining a frame laser data, first check whether the frame per second of laser data meets the requirement of threshold value, such as
Fruit is less than the requirement of threshold value then sufficient threshold value with thumb down, and at this time without using the laser data and report and alarm, waiting is received down
One frame laser data.
Furthermore, the laser point cloud data by 3 d grid space projects on two-dimensional laser raster plane
Before, further includes: after often obtaining a frame laser data, if the frame per second of the laser data meets the requirement of threshold value, then first to this
Each laser reflection point in laser data is filtered, and is removed in each laser reflection point at a distance of closer point and farther away point, is remained
Remaining each laser reflection point is used further to subsequent calculating.
Furthermore, the confidence level for calculating each possible candidate pose method particularly includes:
According to the confidence level of each map grid in the discrete scan data of the corresponding scanning angle of each candidate's pose, meter
The confidence level candidate_probability of each candidate pose is calculated, formula is as follows:
Wherein, m is the sum of the map grid in the discrete scan data of the corresponding scanning angle of each candidate pose, if
The map reference of n-th of map grid is (xn, yn), then the map grid confidence level is
The corresponding confidence level of each candidate pose is calculated with the pose difference of the estimated value of pose according to each candidate pose
Weight candidate_weight, formula are as follows:
Wherein, x_offset is the displacement between each candidate pose and the estimated value of pose along x-axis, and y_offset is each
Along the displacement of y-axis, trans between candidate pose and the estimated value of poseweightIt is displacement weight, candiate.rotation is to wait
Angle, rot are rotated between bit selecting appearance and the estimated value of poseweightIt is rotation angle weight；
By the confidence level candidate_probability and confidence weight candidate_ of each candidate pose
Confidence level score value of the product of weight as current pose, formula are as follows:
Score=candidate_probability*candidate_weight
The highest pose of confidence level score value score is selected to estimate as optimal pose.
Furthermore, guarantee that the side of laser aiming signal is correctly received in laser pickoff dot matrix in the moving process
Method includes:
In robot into object pose moving process, judge whether the laser pickoff of specified number all receives laser
Guide signal: if so, robot is moved towards object pose；If it is not, then robot adjusts pose, so that the laser of specified number
Receiver receives laser aiming signal.
Furthermore, the robot adjusts pose, so that the laser pickoff of specified number receives laser aiming
The step of signal includes:
When the laser pickoff for being located at the specified laser pickoff side numbered receives laser aiming signal, robot
It is mobile to side, so that the laser pickoff of specified number all receives laser aiming signal；
When the laser pickoff for being located at the specified laser pickoff other side numbered receives laser aiming signal, machine
People is mobile to the other side, so that the laser pickoff of specified number all receives laser aiming signal；
When receiving the number interval of two laser pickoffs of laser aiming signal greater than specified numerical value, robot tune
Oneself whole deflection angle, so that it is divided into specified numerical value between receiving the numbers of two laser pickoffs of laser aiming signal,
Laser aiming signal is searched for later, and whether the laser pickoff for reentering the specified number of judgement all receives laser aiming signal
The step of.
Furthermore, the height of the laser emitter and laser pickoff apart from ground is identical, so that laser pick-off
The laser aiming signal of its horizontal direction that can receive laser emitter sending.
Furthermore, the laser emitter of the specified number is 2, is distributed horizontally to target two sides.
Furthermore, the laser pickoff dot matrix includes the laser pickoff of several horizontal homogeneous distribution.
On the other hand, the present invention also provides a kind of method for carrying out autonomous charging of robots based on three-dimensional laser grid, benefits
It is started to charge with the above-mentioned localization method based on laser-correlation so that robot is moved to the predetermined charge position of charging pile.
Beneficial effects of the present invention are as follows: the localization method and autonomous charging of robots of the invention based on laser-correlation
Method judges machine using laser raster geometrical characteristic using the environmental characteristic data of three-dimensional laser sensor acquisition charging house
Whether people enters charging house, after determining that robot enters charging house, passes through the two continuous frames laser from three-dimensional laser sensor
Data, calculating robot's velocity of displacement and rotation speed, and it is next by former frame pose, velocity of displacement and rotation speed calculating
The prediction pose initial value of frame is counted with newest received inertial navigation data according to newest received mileage at the beginning of prediction pose
Value is corrected to obtain the estimated value of pose；The position of each scanning angle on each map grid is determined according to positioning scanning window size
Appearance, as all possible candidate pose；The confidence level for calculating each possible candidate pose, selects the highest position of confidence level score value
Appearance estimates and envisions the object pose of setting according to the optimal pose of robot, robot is not as the optimal pose estimation of robot
The position of disconnected adjustment and deflection angle, when reaching immediately ahead of distance objective position in specified range, the positioning strategy of robot
It is adjusted to fine positioning strategy, fine positioning is carried out to robot using laser-correlation switch.Robot believes according to the data of fine positioning
Breath constantly adjusts oneself pose, and until reaching designated position, the power supply contact of charging connection contact and recharging base is to tapping into
Row charging.After tested, the positioning accuracy of robot greatly improves, the power supply contact of robot charging connection contact and recharging base
Docking success rate reaches 99% or more.
Detailed description of the invention
Fig. 1 is the location algorithm flow chart of the embodiment of the present invention.
Fig. 2 be the embodiment of the present invention judge whether enter charging house flow chart.
Fig. 3 is the feature laser raster schematic diagram of the embodiment of the present invention.
Fig. 4 is the grid schematic diagram of the embodiment of the present invention being mapped to after plane.
Fig. 5 is the three-dimensional laser point cloud data mapping schematic diagram of the embodiment of the present invention.
Fig. 6 is the robot coarse positioning flow chart of the embodiment of the present invention.
Fig. 7 is the robot fine positioning flow chart of the embodiment of the present invention.
Fig. 8 is the schematic diagram of the laser emitter of the embodiment of the present invention.
Fig. 9 is the schematic diagram of the laser pickoff of the embodiment of the present invention.
Figure 10 is that the laser pickoff dot matrix of the embodiment of the present invention is properly received guidance signal schematic representation.
Label in figure: 1- charging door, the left metope of 2- charging house, the right metope of 3- charging house, 4- charging house left comer are fallen, 5-
Charging house right corner is fallen, 6- robot, 7- laser emitter 1,8- laser emitter 2,9- laser pickoff.
Specific embodiment
Below with reference to embodiment and referring to attached drawing, present invention is further described in detail.
Embodiment 1:
One embodiment of the present of invention, describe it is a kind of based on three-dimensional laser grid positioning Intelligent Mobile Robot from
Main charging method.
Technical term related to the present embodiment is defined as follows:
Grid: close adjacent grid array uniform in size.
Laser raster: there is the grid of laser irradiation.
Grey grid: meet the laser raster of confidence level requirement.
Raster plane: grid are established in some identical height of metope, left side metope and right side metope immediately ahead of the charging house
Lattice, a raster plane of formation.
Continuous grey grid: closely coupled grey grid.
Feature grid: the grid of at L-shaped grey raster series is organized.
Standard feature grid: the feature grid on the right angle of feature raster series, and water is on raster plane
Square to same straight line on.
Premise: field personnel the position by the dimension data of charging house, charging pile in charging house (robot
Object pose) typing robot industrial personal computer in.Laser pickoff is installed in robot, Laser emission is installed on charging pile
Device.Robot interior has electric power detection module, when total electricity is lower than certain value (such as 10%), judges that robot needs fill
Electricity.When robot requires supplementation with electric power, charging house is driven towards by location and navigation technology automatically, robot host computer sends instruction
To the electrical cabinet of charging house, electrical cabinet opens charging door after receiving instruction, and robot, which is fallen back, to be entered in charging house.Into
After charging house, using, based on the localization method of laser-correlation, obtaining coordinate information and the deflection of robot described in following steps
Angle information.
The overall flow of localization method based on laser-correlation is as shown in Figure 1, specific implementation step is as follows:
1, judge whether charging house is entered to robot.As shown in Figure 2, the specific steps are as follows:
1-1) utilize metope, left side metope and right side metope immediately ahead of robot in three-dimensional laser sensor scanning charging house
Information.Metope, left side metope and right side metope establish the grid that side length is a, a ∈ respectively immediately ahead of charging house robot
[1cm, 2cm].As shown in figure 3, each grid indicates that the lattice point is idle or occupies state with s=0 or s=1.For each
It is a it is observed that grid, the grid identifier that laser point cloud is fallen in be grey, grey grid expression detect the presence of barrier,
The grid is the state being occupied；The grid identifier of no laser point cloud is white, and barrier (example is not detected in white grid expression
The door opened such as charging house), which is idle state.
The laser point cloud data in 3 d grid space 1-2) is projected into two dimension in the plane among laser where harness
It is in X ' O ' Y ' plane in robot coordinate system in laser raster plane.As shown in figure 4, grey parts are that laser point cloud reflects
Grid region after being mapped to two-dimensional laser raster plane.As shown in figure 5, indicating a certain laser point p in three-dimensional space with point S
To the distance for arriving origin 0, α, θ are the horizontal direction angle value and vertical direction angle value of the laser.It is mapped to plane z0It is afterwards p
(x ', y ', z '), wherein x ', y ', z ' the following formula of value shown in.
To three-dimensional laser point cloud carry out mapping processing it is available following the utility model has the advantages that
Using three-dimensional laser point cloud data, the rich of laser data under sparse environment can be increased, thus reduce it is fixed
The error of position；
For there is the object of shield portions, it is still able to detect that the reflectance data of the object after mapping, can increase
The robustness and stability of location algorithm；
By the way that three-dimensional laser data are mapped to two-dimensional surface, the operational efficiency of location algorithm can be improved.
Horizontal direction and the orthogonal continuous grey colour grid of vertical direction 1-3) are extracted on two-dimensional laser raster plane
Lattice, rectangular shaped grid.The feature grid met the requirements is as shown in the upper left corner in Fig. 4 and upper right corner grey grid.
1-4) laser raster feature verifies: whether two feature grids for verifying extraction are in same horizontal line, meet
Condition is standard feature grid, i.e., expression robot comes into charging house.When being unsatisfactory for condition, next frame is waited to swash
Light data and return step 1-1).
2, coarse positioning is carried out to robot using three-dimensional laser data.
Entire coarse positioning process is divided into estimation initial value, positioning scanning window, calculates confidence level three phases, in conjunction with Fig. 6 pairs
The detailed step of robot coarse positioning is described.
2-1) estimate initial value
2-1-1) pose of robot 6DOF can be expressed as p (x, y, z, α, beta, gamma) in three dimensions, wherein
X, y, z are respectively robot x-axis, y-axis, coordinate in z-axis in global coordinate system XOY, Eulerian angles α, beta, gamma respectively correspond around
The rotation angle of three coordinate axial directions.Due to the ground smooth planar in charging house, the movement of robot is in a plane.
Therefore, robot z-axis coordinate, Eulerian angles α, β three degree of freedom can be limited, z=z is made0, α=0, β=0, wherein z0For three-dimensional
The mounting height of laser sensor.The pitch angle of robot and tilt angle are 0 at this time, are equivalent to robot and are merely able in z
=z0Height on move, only have 3 freedom degrees.Robot pose is represented by p (x after limiting freedom degreet, yt, γt)。
During robot is mobile to object pose, three-dimensional laser sensor of the frame from robot is often obtained
After laser data, laser data is handled to obtain laser point cloud data, which is every frame laser data institute
The general designation for the laser data point information for including reflects coordinate of each laser reflection point in laser sensor coordinate system.For
The characteristics of sparse environment, laser point cloud data can be filtered, that is, remove each laser reflection point in laser point cloud data
In corresponding noise (at a distance of closer point and farther away point), residue is used as available point, and the confidence level of location estimation can be improved.
For laser data, the detection of Yao Jinhang frame per second, i.e., after often obtaining a frame laser data, check laser data frame per second whether
Less than threshold value, respective handling then is re-started after report and alarm and waiting receive next frame laser data if it is less than threshold value.
For example, if the frame per second of definition is 25Hz, and the time for obtaining adjacent two frames laser data is greater than 40ms, then laser data is not inconsistent
It closes and requires, the failures such as overheating occurs in possible laser sensor need report and alarm and wait next frame laser data.
Planar Mapping 2-1-2) is carried out for satisfactory laser point cloud data, by the laser point in 3 d grid space
It is x ' o ' y ' in robot coordinate system in plane where cloud data projection to harness among laser in two-dimensional grid plane
Plane.
Robot 2-1-3) is respectively indicated in the pose at t-1, t, t+1 moment with P (t-1), P (t), P (t+1), passes through P
(t) displacement of x-axis, y-axis direction calculates separately the speed velocity_x of Robot x-axis, the speed along y-axis between P (t-1)
Velocity_y is spent, the difference of angle is rotated around z-axis come the rotation of calculating robot by robot between P (t) and P (t-1)
Speed velocity γ.P (t+1) is calculated by P (t), velocity_x, velocity_y and velocity_ γ, obtains machine
People is in the prediction pose P (t+1) of time point t+1, referred to as prediction pose initial value pose_prediction.
2-1-4) after obtaining mileage and counting (including being displaced and rotate angle), counted according to two frame mileages between
Difference calculates the corrected value that mileage counts, for optimizing P (t+1).
Current velocity of displacement 2-1-5) is corrected according to the linear acceleration Information acceleration that inertial navigation unit obtains
Velocity_x and velocity_y corrects current rotation speed velocity_ γ according to angular acceleration, will be at the beginning of prediction pose
Value pose_prediction is updated to the estimated value pose_estimated of pose.
2-2) position scanning window
Positioning scanning window 2-2-1) is arranged according to the estimated value pose_estimated of robot pose.Anchor window is swept
Using displacement sweep parameter linear_search_window and angle scanning parameter when retouching, wherein angle scanning parameter includes sweeping
Retouch angular dimension angular_search_window and scanning angle step-length angular_step.Displacement sweep parameter is for limiting
The displacement range for positioning scanning window is the linear_search_ centered on the estimated value pose_estimated of pose
The square of window size.Angle scanning parameter is used to limit the angular range of positioning scanning window as with the estimated value of pose
Pose_estimated is center angle, respectively deviates the angle of angular_search_window size up and down.Positioning scanning window
Mouth size determines the pose of each scanning angle on each map grid in positioning scanning window, as all possible candidate pose
possible_candidates。
2-2-2) (swept by the estimated value pose_estimated of pose, angle according to the positioning scanning window of robot
Retouch parameter and displacement sweep parameter determine) and laser point cloud data calculate the coordinate of the laser point cloud data of each scanning angle.Root
It is calculated according to the estimated value pose_estimated of pose and the laser point cloud data coordinate of each scanning angle each under each scanning angle
The corresponding map grid positions of laser reflection point (calculate coordinate of each map grid in map coordinates system), as respectively sweeping
Retouch the discrete scan data of angle.For the discrete scan data of some scanning angle, if wherein thering is repetition to fall in samely
Multiple laser reflection points of figure grid positions only take the corresponding map grid of one of laser reflection point in map coordinates system
Coordinate, the confidence calculations for subsequent step.
2-3) calculate confidence level.
According to confidence level (the map grid of the corresponding each map grid of each candidate's pose possible_candidate
Confidence value it is related to map structuring process, in position fixing process, for the value having determined), calculate each candidate pose
Confidence level candidate_probability, formula are as follows:
Wherein, m is the sum of the map grid in the discrete scan data of some corresponding scanning angle of candidate's pose, by
In z=z0The value of z coordinate is not considered during calculating pose confidence level, if wherein the map reference of n-th of grid is (xn,
yn), which isValue range is 0.1~0.9.
According to the pose of each candidate pose possible_candidate and the estimated value pose_estimated of pose
Difference calculates the corresponding confidence weight candidate_weight of each candidate pose, and formula is as follows:
Wherein, x_offset is the displacement between each candidate pose and the estimated value pose_estimated of pose along x-axis,
Y_offset is the displacement between each candidate pose and the estimated value pose_estimated of pose along y-axis, transweightIt is position
Weight is moved, candiate.rotation is to rotate angle between candidate pose and the estimated value pose_estimated of pose,
rotweightIt is rotation angle weight；
By the confidence level candidate_probability and confidence weight candidate_ of each candidate pose
Confidence level score value of the product of weight as current pose, formula is as follows,
Score=candidate_probability*candidate_weight (3)
The highest pose of confidence level score value score is selected to update the estimated value pose_estimated of pose, as optimal
Pose estimation, that is, be used as P (t+1), coarse positioning terminates.
3, fine positioning is carried out to robot.As shown in fig. 7, detailed process is as follows:
3-1) according to preset object pose, the position and deflection angle that robot constantly adjusts, be confirmed whether to
Up in distance objective pose specified range, such as 0.5m.
3-2) as shown in figure 8, installing laser emitter on charging pile, 2 level point of laser emitter 1 and laser emitter
Cloth is in charging electrode two sides.As shown in Figure 9 and Figure 10, the receiver of laser opposite-radiation photoelectric sensor is installed behind in robot
Dot matrix, number from left to right are 1~10, and laser pickoff array horizontal homogeneous is distributed in above charging card slot, and transmitter
Height with receiver apart from ground is equal.
Robot walks according to path as defined in navigation algorithm, searches for laser aiming signal.As shown in Figure 10, work as robot
On laser pickoff dot matrix in number be 4 and 7 laser pickoff when receiving laser aiming signal, robot is towards swashing
Light guides the direction of signal to retreat, and keeps laser pickoff dot matrix correctly to receive laser aiming signal, machine in fallback procedures
Pose is adjusted in accordance with the following steps in people's fallback procedures.
3-2-1) mobile robot judges No. 4 and when whether No. 7 laser pickoffs all receives laser aiming signal, if
It is that then explanation finds accurate position, robot retreats at this time, moves towards charging pile direction, otherwise robot is according to very unwise move
Pose is slightly constantly adjusted, is finally reached laser pickoff 4, laser pickoff 7 receives laser aiming signal.
When number is (Isosorbide-5-Nitrae) or the receiver of (2,5) or (3,6) receives laser aiming signal, illustrate robot
Tend to charging pile right.Robot moves to left, so that No. 4 and No. 7 laser pickoffs all receive laser aiming signal.
When number is (5,8) or the receiver of (6,9) or (7,10) receives laser aiming signal, illustrate robot
Tend to charging pile left.Robot moves to right, so that No. 4 and No. 7 laser pickoffs all receive laser aiming signal.
When receiving the number interval of two receivers of laser aiming signal greater than 2, illustrate robot and two
There are deflection angles between the line of laser emitter.The deflection angle of robot or so adjustment oneself, so that receiving laser aiming
2 are divided between the number of two receivers of signal, adjusts the pose of oneself according to step 3-2-1) later.
4, robot is moved towards charging pile to be stopped after colliding charging electrode, and robot stretches out charging arm, insertion
Into charging pile.When robot detects that charging arm connect good with charging pile, start to charge；Otherwise, robot host computer is sent out
It send instruction to robot, withdraws charging arm, again according to the pose of abovementioned steps adjustment robot.
Recharging method based on laser-correlation of the invention utilizes the environment of three-dimensional laser sensor acquisition charging house
Characteristic judges whether robot enters charging house using laser raster geometrical characteristic, after determining that robot enters charging house,
The pose of robot predict and the data of odometer and inertial navigation unit is combined to be corrected prediction pose, is obtained
Robot pose volume estimated value sets positioning scanning window according to robot pose estimated value, calculates and wait in positioning scanning window
The confidence level score value of bit selecting appearance chooses coarse positioning optimal pose of the highest candidate pose of score value as robot.Work as robot
Close to after preset object pose, fine positioning is carried out to robot using laser-correlation switch.Robot is according to fine positioning
Data information constantly adjust oneself pose, until reaching designated position, the power supply touching of charging connection contact and recharging base
Point docking is charged.After tested, the positioning accuracy of robot greatly improves, robot charging connection contact and recharging base
Power supply contact docking success rate reaches 99% or more.
Although the present invention has been described by way of example and in terms of the preferred embodiments, embodiment is not for the purpose of limiting the invention.Not
It is detached from the spirit and scope of the present invention, any equivalent change or retouch done also belongs to the protection scope of the present invention.Cause
This protection scope of the present invention should be based on the content defined in the claims of this application.
Claims (12)
1. a kind of localization method based on laser-correlation enters positioning when charging house is charged, feature for robot
It is, comprising the following steps:
Judge the step of whether robot has entered charging house:
Metope, left side metope and right side metope establish three-dimensional laser grid immediately ahead of robot in charging house；When according to machine
When the three-dimensional laser sensor of people scans three-dimensional laser data acquisition obtained by the metope to standard feature grid, it is believed that robot
Charging house is entered；
The step of coarse positioning is carried out to robot:
The three-dimensional of three-dimensional laser sensor during robot is mobile to object pose by each frame from robot swashs
After light data is processed into laser point cloud data, the laser point cloud data in 3 d grid space is projected into two-dimensional laser grid and is put down
On face, according to pose calculating robot velocity of displacement and rotation speed in the two continuous frames laser data on the two-dimensional surface,
And the prediction pose initial value of next frame is calculated by former frame pose, velocity of displacement and rotation speed；
It is counted according to newest received mileage and prediction pose initial value is corrected with newest received inertial navigation data
To the estimated value of pose；
Positioning scanning window size is calculated, the position of each scanning angle on each map grid is determined according to positioning scanning window size
Appearance, as all possible candidate pose；After often obtaining laser data of the frame from the two-dimensional laser sensor of robot, meter
Calculate coordinate of each laser reflection point in map coordinates system under each scanning angle；And map datum is combined to calculate under each scanning angle
Coordinate of the corresponding map grid of each laser reflection point in map coordinates system, the discrete scan data as each scanning angle；
The confidence level for calculating each possible candidate pose, selects the highest pose of confidence level score value to estimate as the optimal pose of robot
Meter, the i.e. current pose of robot；
The step of fine positioning is carried out to robot:
Judge whether robot arrived distance objective pose specified range according to the current pose of the robot and object pose
It is interior；
The laser aiming signal that laser emitter where searching for target issues；
When the laser pickoff of the laser pickoff dot matrix middle finger number of delimiting the organizational structure in robot receives laser aiming signal, machine
People is mobile towards the direction of laser aiming signal, guarantees that laser aiming letter is correctly received in laser pickoff dot matrix in moving process
Number.
2. the localization method according to claim 1 based on laser-correlation, which is characterized in that described according to the three of robot
Tieing up the method that laser sensor scans three-dimensional laser data acquisition standard feature grid obtained by the metope includes:
The metope is scanned using three-dimensional laser sensor；
Laser point cloud data in 3 d grid space is projected on two-dimensional laser raster plane；
Horizontal direction and the orthogonal continuous grey grid of vertical direction are extracted on two-dimensional laser raster plane；
It verifies the horizontal direction extracted and whether the orthogonal continuous grey grid of vertical direction is in same horizontal line
On, meet condition is standard feature grid；When being unsatisfactory for condition, returns and scan the wall using three-dimensional laser sensor
The step of face, carries out above-mentioned processing to next frame laser data.
3. the localization method according to claim 2 based on laser-correlation, which is characterized in that will be in 3 d grid space
Laser point cloud data projects on two-dimensional laser raster plane,
Laser point cloud data in 3 d grid space is projected in the two-dimensional grid plane among laser where harness, is obtained
Robot coordinate in global coordinate system XOY is p (x, y, z), and wherein x, the value of y, z are as follows:
Wherein, point S be 3 d grid space in a certain laser point-to-point 0 distance, α be the laser point relative to horizontal direction
Angle, θ be angle of the laser point relative to vertical direction.
4. the localization method according to claim 3 based on laser-correlation, which is characterized in that described by 3 d grid space
In laser point cloud data project on two-dimensional laser raster plane before, further includes: after often obtaining a frame laser data, first
Check whether the frame per second of laser data meets the requirement of threshold value, if it is less than the requirement of threshold value then sufficient threshold value with thumb down, at this time
Without using the laser data and report and alarm, waiting receives next frame laser data.
5. the localization method according to claim 3 based on laser-correlation, which is characterized in that described by 3 d grid space
In laser point cloud data project on two-dimensional laser raster plane before, further includes: after often obtaining a frame laser data, such as
The frame per second of the laser data meets the requirement of threshold value, then is first filtered, removes to each laser reflection point in the laser data
At a distance of closer point and farther away point in each laser reflection point, remaining each laser reflection point is used further to subsequent calculating.
6. the localization method according to claim 1 based on laser-correlation, which is characterized in that described to calculate each possible time
The confidence level of bit selecting appearance method particularly includes:
According to the confidence level of each map grid in the discrete scan data of the corresponding scanning angle of each candidate's pose, calculate every
The confidence level candidate_probability of a candidate's pose, formula are as follows:
Wherein, m is the sum of the map grid in the discrete scan data of the corresponding scanning angle of each candidate pose, if n-th
The map reference of a map grid is (xn, yn), then the map grid confidence level is
The corresponding confidence weight of each candidate pose is calculated with the pose difference of the estimated value of pose according to each candidate pose
Candidate_weight, formula are as follows:
Wherein, x_offset is the displacement between each candidate pose and the estimated value of pose along x-axis, and y_offset is each candidate
Along the displacement of y-axis, trans between pose and the estimated value of poseweightIt is displacement weight, candiate.rotation is candidate bit
Angle, rot are rotated between appearance and the estimated value of poseweightIt is rotation angle weight；
By the confidence level candidate_probability and confidence weight candidate_weight of each candidate pose
Confidence level score value of the product as current pose, formula are as follows:
Score=candidate_probability*candidate_weight
The highest pose of confidence level score value score is selected to estimate as optimal pose.
7. the localization method according to claim 1 based on laser-correlation, which is characterized in that guarantee in the moving process
The method that laser aiming signal is correctly received in laser pickoff dot matrix includes:
In robot into object pose moving process, judge whether the laser pickoff of specified number all receives laser aiming
Signal: if so, robot is moved towards object pose；If it is not, then robot adjusts pose, so that the laser pick-off of specified number
Device receives laser aiming signal.
8. the localization method according to claim 7 based on laser-correlation, which is characterized in that the robot adjusts position
Appearance, so that the laser pickoff of specified number includes: the step of receiving laser aiming signal
When being located at the laser pickoff of laser pickoff side of specified number and receiving laser aiming signal, robot is to one
Side is mobile, so that the laser pickoff of specified number all receives laser aiming signal；
When being located at the laser pickoff of the laser pickoff other side of specified number and receiving laser aiming signal, robot to
The other side is mobile, so that the laser pickoff of specified number all receives laser aiming signal；
When receiving the number interval of two laser pickoffs of laser aiming signal greater than specified numerical value, robot adjustment is certainly
Oneself deflection angle, so that being divided into specified numerical value between receiving the numbers of two laser pickoffs of laser aiming signal, later
Laser aiming signal is searched for, whether the laser pickoff for reentering the specified number of judgement all receives the step of laser aiming signal
Suddenly.
9. the localization method according to claim 1 based on laser-correlation, which is characterized in that the laser emitter and swash
Height of the optical receiver apart from ground is identical, enable laser pick-off its receive laser emitter sending horizontal direction swash
Light guides signal.
10. the localization method according to claim 7 based on laser-correlation, which is characterized in that the specified number swashs
Optical transmitting set is 2, is distributed horizontally to target two sides.
11. the localization method according to claim 1 based on laser-correlation, which is characterized in that the laser pickoff point
Laser pickoff of the battle array comprising the distribution of several horizontal homogeneous.
12. a kind of method of the autonomous charging of robots based on laser-correlation, which is characterized in that utilize claim 1~11 times
It is started to charge based on the localization method of laser-correlation so that robot is moved to the predetermined charge position of charging pile described in one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811449494.6A CN109460032A (en) | 2018-11-29 | 2018-11-29 | A kind of localization method and autonomous charging of robots method based on laser-correlation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811449494.6A CN109460032A (en) | 2018-11-29 | 2018-11-29 | A kind of localization method and autonomous charging of robots method based on laser-correlation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109460032A true CN109460032A (en) | 2019-03-12 |
Family
ID=65612029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811449494.6A Pending CN109460032A (en) | 2018-11-29 | 2018-11-29 | A kind of localization method and autonomous charging of robots method based on laser-correlation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109460032A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109974712A (en) * | 2019-04-22 | 2019-07-05 | 广东亿嘉和科技有限公司 | It is a kind of that drawing method is built based on the Intelligent Mobile Robot for scheming optimization |
| CN109991980A (en) * | 2019-04-01 | 2019-07-09 | 珠海市一微半导体有限公司 | The forming method of the signal quantization distribution map of cradle |
| CN110031822A (en) * | 2019-04-22 | 2019-07-19 | 上海禾赛光电科技有限公司 | It can be used for noise recognition methods and the laser radar system of laser radar |
| CN110133681A (en) * | 2019-05-07 | 2019-08-16 | 深圳越登智能技术有限公司 | Recharging guidance system and its recharge bootstrap technique based on laser radar |
| CN110196044A (en) * | 2019-05-28 | 2019-09-03 | 广东亿嘉和科技有限公司 | It is a kind of based on GPS closed loop detection Intelligent Mobile Robot build drawing method |
| CN110824491A (en) * | 2019-10-24 | 2020-02-21 | 北京迈格威科技有限公司 | Charging pile positioning method and device, computer equipment and storage medium |
| CN110989596A (en) * | 2019-12-04 | 2020-04-10 | 上海高仙自动化科技发展有限公司 | Pile alignment control method and device, intelligent robot and storage medium |
| CN111026130A (en) * | 2019-12-25 | 2020-04-17 | 劢微机器人科技(深圳)有限公司 | AGV positioning deviation correction control method and device and readable storage medium |
| CN111781930A (en) * | 2020-07-10 | 2020-10-16 | 广州今甲智能科技有限公司 | Method for accurately positioning charging pile by intelligent robot and intelligent robot |
| CN112214011A (en) * | 2019-07-11 | 2021-01-12 | 珠海市一微半导体有限公司 | System and method for positioning charging seat of self-moving robot |
| CN112327842A (en) * | 2020-10-29 | 2021-02-05 | 深圳市普渡科技有限公司 | Method and system for positioning charging pile by robot |
| WO2022267681A1 (en) * | 2021-06-22 | 2022-12-29 | 速感科技（北京）有限公司 | Automatic recharging method and system for autonomous mobile device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045263A (en) * | 2015-07-06 | 2015-11-11 | 杭州南江机器人股份有限公司 | Kinect-based robot self-positioning method |
| CN106020188A (en) * | 2016-05-17 | 2016-10-12 | 杭州申昊科技股份有限公司 | Substation patrol robot autonomous charging method based on laser navigation |
| CN106356944A (en) * | 2016-10-14 | 2017-01-25 | 四川超影科技有限公司 | Automatic charging laser aligning system of patrol check robot and aligning method |
| CN106786938A (en) * | 2016-12-30 | 2017-05-31 | 亿嘉和科技股份有限公司 | A kind of crusing robot localization method and automatic recharging method |
| CN107340522A (en) * | 2017-07-10 | 2017-11-10 | 浙江国自机器人技术有限公司 | A kind of method, apparatus and system of laser radar positioning |
| CN108253958A (en) * | 2018-01-18 | 2018-07-06 | 亿嘉和科技股份有限公司 | A kind of robot real-time location method under sparse environment |
| US20180306587A1 (en) * | 2017-04-21 | 2018-10-25 | X Development Llc | Methods and Systems for Map Generation and Alignment |
| CN108873001A (en) * | 2018-09-17 | 2018-11-23 | 江苏金智科技股份有限公司 | A kind of accurate method for judging robot localization precision |
-
2018
- 2018-11-29 CN CN201811449494.6A patent/CN109460032A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105045263A (en) * | 2015-07-06 | 2015-11-11 | 杭州南江机器人股份有限公司 | Kinect-based robot self-positioning method |
| CN106020188A (en) * | 2016-05-17 | 2016-10-12 | 杭州申昊科技股份有限公司 | Substation patrol robot autonomous charging method based on laser navigation |
| CN106356944A (en) * | 2016-10-14 | 2017-01-25 | 四川超影科技有限公司 | Automatic charging laser aligning system of patrol check robot and aligning method |
| CN106786938A (en) * | 2016-12-30 | 2017-05-31 | 亿嘉和科技股份有限公司 | A kind of crusing robot localization method and automatic recharging method |
| US20180306587A1 (en) * | 2017-04-21 | 2018-10-25 | X Development Llc | Methods and Systems for Map Generation and Alignment |
| CN107340522A (en) * | 2017-07-10 | 2017-11-10 | 浙江国自机器人技术有限公司 | A kind of method, apparatus and system of laser radar positioning |
| CN108253958A (en) * | 2018-01-18 | 2018-07-06 | 亿嘉和科技股份有限公司 | A kind of robot real-time location method under sparse environment |
| CN108873001A (en) * | 2018-09-17 | 2018-11-23 | 江苏金智科技股份有限公司 | A kind of accurate method for judging robot localization precision |
Non-Patent Citations (1)
| Title |
|---|
| 张国良 等: "《移动机器人的SLAM与VSLAM方法》", 30 October 2018, 西安交通大学出版社 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109991980A (en) * | 2019-04-01 | 2019-07-09 | 珠海市一微半导体有限公司 | The forming method of the signal quantization distribution map of cradle |
| CN109991980B (en) * | 2019-04-01 | 2022-07-08 | 珠海一微半导体股份有限公司 | Method for forming signal quantization distribution diagram of charging seat |
| WO2020215369A1 (en) * | 2019-04-22 | 2020-10-29 | 上海禾赛光电科技有限公司 | Noise point recognition method applicable to lidar and lidar system |
| CN110031822A (en) * | 2019-04-22 | 2019-07-19 | 上海禾赛光电科技有限公司 | It can be used for noise recognition methods and the laser radar system of laser radar |
| CN109974712A (en) * | 2019-04-22 | 2019-07-05 | 广东亿嘉和科技有限公司 | It is a kind of that drawing method is built based on the Intelligent Mobile Robot for scheming optimization |
| CN110133681A (en) * | 2019-05-07 | 2019-08-16 | 深圳越登智能技术有限公司 | Recharging guidance system and its recharge bootstrap technique based on laser radar |
| CN110196044A (en) * | 2019-05-28 | 2019-09-03 | 广东亿嘉和科技有限公司 | It is a kind of based on GPS closed loop detection Intelligent Mobile Robot build drawing method |
| CN112214011B (en) * | 2019-07-11 | 2022-05-10 | 珠海一微半导体股份有限公司 | System and method for positioning charging seat of self-moving robot |
| CN112214011A (en) * | 2019-07-11 | 2021-01-12 | 珠海市一微半导体有限公司 | System and method for positioning charging seat of self-moving robot |
| CN110824491A (en) * | 2019-10-24 | 2020-02-21 | 北京迈格威科技有限公司 | Charging pile positioning method and device, computer equipment and storage medium |
| CN110824491B (en) * | 2019-10-24 | 2022-07-29 | 北京迈格威科技有限公司 | Charging pile positioning method and device, computer equipment and storage medium |
| CN110989596A (en) * | 2019-12-04 | 2020-04-10 | 上海高仙自动化科技发展有限公司 | Pile alignment control method and device, intelligent robot and storage medium |
| CN110989596B (en) * | 2019-12-04 | 2023-06-06 | 上海高仙自动化科技发展有限公司 | Pile alignment control method and device, intelligent robot and storage medium |
| CN111026130A (en) * | 2019-12-25 | 2020-04-17 | 劢微机器人科技(深圳)有限公司 | AGV positioning deviation correction control method and device and readable storage medium |
| CN111026130B (en) * | 2019-12-25 | 2023-08-15 | 劢微机器人科技(深圳)有限公司 | AGV positioning deviation correction control method, AGV positioning deviation correction control device and readable storage medium |
| CN111781930A (en) * | 2020-07-10 | 2020-10-16 | 广州今甲智能科技有限公司 | Method for accurately positioning charging pile by intelligent robot and intelligent robot |
| CN112327842A (en) * | 2020-10-29 | 2021-02-05 | 深圳市普渡科技有限公司 | Method and system for positioning charging pile by robot |
| CN112327842B (en) * | 2020-10-29 | 2022-06-03 | 深圳市普渡科技有限公司 | Method and system for positioning charging pile by robot |
| WO2022267681A1 (en) * | 2021-06-22 | 2022-12-29 | 速感科技（北京）有限公司 | Automatic recharging method and system for autonomous mobile device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460032A (en) | A kind of localization method and autonomous charging of robots method based on laser-correlation | |
| CN109828280A (en) | A kind of localization method and autonomous charging of robots method based on three-dimensional laser grid | |
| CN109901590B (en) | Recharging control method of desktop robot | |
| JP6915209B2 (en) | How to create a map of a mobile robot and how to plan a route based on the map | |
| CN107463173B (en) | Storage AGV navigation method and device, computer equipment and storage medium | |
| CN109730590B (en) | Cleaning robot and method for automatically returning and charging same | |
| CN105115497B (en) | A kind of reliable indoor mobile robot precision navigation positioning system and method | |
| CN102866706B (en) | Cleaning robot adopting smart phone navigation and navigation cleaning method thereof | |
| CN111693050B (en) | Indoor medium and large robot navigation method based on building information model | |
| CN109443351A (en) | A kind of robot three-dimensional laser positioning method under sparse environment | |
| CN105700522B (en) | A kind of robot charging method and its charging system | |
| Fruh et al. | Fast 3D model generation in urban environments | |
| CN108919825A (en) | The unmanned plane indoor locating system and method for having barrier avoiding function | |
| CN112945137B (en) | Storage ore heap scanning method based on single-line laser radar and range finder equipment | |
| Xu et al. | SLAM of Robot based on the Fusion of Vision and LIDAR | |
| CN110477825A (en) | Clean robot, recharging method, system and readable storage medium storing program for executing | |
| CN102116625A (en) | GIS (geographic information system)-GPS (global position system) navigation method of inspection robot | |
| CN109282808A (en) | Unmanned plane and Multi-sensor Fusion localization method for the detection of bridge Cut-fill | |
| CN108801269A (en) | A kind of interior cloud Algorithms of Robots Navigation System and method | |
| WO2020003304A1 (en) | A computerized system for guiding a mobile robot to a docking station and a method of using same | |
| CN112819943A (en) | Active vision SLAM system based on panoramic camera | |
| CN112484746A (en) | Monocular vision-assisted laser radar odometer method based on ground plane | |
| CN114047750A (en) | Express delivery warehousing method based on mobile robot | |
| CN111862200B (en) | Unmanned aerial vehicle positioning method in coal shed | |
| CN114034299A (en) | Navigation system based on active laser SLAM |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190312 |
|
| RJ01 | Rejection of invention patent application after publication |