CN109959377A - A kind of robot navigation's positioning system and method - Google Patents
A kind of robot navigation's positioning system and method Download PDFInfo
- Publication number
- CN109959377A CN109959377A CN201711420969.4A CN201711420969A CN109959377A CN 109959377 A CN109959377 A CN 109959377A CN 201711420969 A CN201711420969 A CN 201711420969A CN 109959377 A CN109959377 A CN 109959377A
- Authority
- CN
- China
- Prior art keywords
- robot
- map
- path
- path planning
- positioning
- 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.)
- Withdrawn
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
- G01C21/20—Instruments for performing navigational calculations
-
- 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
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
-
- 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
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- 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
- G05D1/0274—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means using mapping information stored in a memory device
Abstract
The present invention discloses a kind of robot navigation's positioning system and method, map structuring, positioning and path planning are carried out for a robot, its method includes the following steps: S100, positioning step: robot detects ambient condition information by multiple sensors, the then SLAM algorithm based on adaptive particle filter, and match different odometers and complete real-time map building and positioning；And S200, path planning step: use the path planning algorithm based on bipolar mixture state A*, comprising: S210 after progress path planning obtains path length and expanding node number on the map of rasterizing, passes through the map that parsing extension obtains higher rasterizing；And S220 gains enlightenment formula weight using the path length of acquisition and expanding node number as the input of fuzzy reasoning by fuzzy reasoning, the input of the search as second stage carries out path planning on the map of higher rasterizing.The present invention can not only be adapted to varying environment, and being capable of active path planning.
Description
Technical field
The present invention relates to a kind of robot navigation's positioning system and methods, specifically, being to be related to one kind to be adapted to difference
Environment, active path planning robot navigation's positioning system and method.
Background technique
One perfect mobile service robot system of idealization is usually made of 4 parts, respectively mobile mechanism,
Functional entity, sensory perceptual system and control system.
Mobile mechanism provides locomotive function for robot, and there are ratcheting mechanism, crawler type mechanism, joint in common mobile mechanism
Formula mechanism and mixed organization etc..Functional entity realizes various service functions, such as nursing, carrying etc. for robot, common
Form is mechanical arm etc..Sensory perceptual system is made of various sensors, acquires itself and external environmental information for service robot.Clothes
Sensors have camera, laser radar, ultrasonic sensor, contact and proximity sensor, inertia measurement list in business robot
Member, odometer etc..Control system is equivalent to " brain " of service robot.Autonomous service robot has image recognition, environment
Clothes independently can be advanced and be completed to the multinomial technology such as perception, path planning, detection of obstacles according to the functional requirement of setting
Business task.
Navigation is the core and key technology of mobile robot, and reflection mobile robot realizes that intelligentized key refers to
Mark.Mobile Robotics Navigation is exactly the cartographic information that robot can independently according to storage inside it, or is obtained according to sensor
The external environment signal obtained cooks up movement routine, and can be moved to along the path in the case where no manual intervention
Predetermined target point.Since service robot working environment often has the feature that structure is complicated, more than dynamic object, therefore, how
It is correct in such environment, to be safely completed the functions such as positioning and map structuring, active path planning immediately be also at present urgently
One of solve the problems, such as.Current location technology can only mostly obtain good locating effect in some environments, cannot
Enough efficiently solve the orientation problem of the robot under all environment.
The path planning of mobile robot refers to that robot will avoid the barrier in environment during mobile to target point
Hinder object, an optimal or suboptimum road from starting point to target point is gone out with a certain index planning such as distance, time, energy consumption
Diameter.The technology is still in the incomplete problem of path planning in the case where barrier obstruction, environment dynamic change at present.
Summary of the invention
The object of the present invention is to provide a kind of robot navigation's positioning system and methods, can not only be adapted to different rings
Border, and being capable of active path planning.
To achieve the goals above, robot navigation's localization method of the invention, for a robot carry out map structuring,
Positioning and path planning comprising following steps:
S100, positioning step: robot detects ambient condition information by multiple sensors, is then based on adaptive particle
The SLAM algorithm of filtering, and match different odometers and complete real-time map building and positioning；
Path planning step: S200 uses the path planning algorithm based on bipolar mixture state A*, comprising: S210, in grid
After progress path planning obtains path length and expanding node number on the map formatted, passes through parsing extension and obtain higher rasterizing
Map；S220 is opened using the path length of acquisition and expanding node number as the input of fuzzy reasoning by fuzzy reasoning
Hairdo weight, the input of the search as second stage carry out path planning on the map of higher rasterizing.
In one embodiment of above-mentioned robot navigation's localization method, the step S100 includes the following steps: S110,
Map is established using the data of sensor feedback；S120 carries out dead reckoning using odometer, and particle filter obtains boat position and pushes away
Information and the initial estimation as robot pose are calculated, utilizes the current frame data of sensor and local map progress later
Match, eliminates the accumulated error of odometer；
Current frame information, while benefit is then added when present frame matching result reaches a matching threshold in S130 in map
With current frame information eliminate map in original error message, when matching result persistently reaches the matching threshold, then continue into
Row map rejuvenation；When present frame matching result does not reach the matching threshold, map is remained unchanged, the position of update robot
Appearance information.
In one embodiment of above-mentioned robot navigation's localization method, in the step S100, adaptive particle filter
SLAM algorithm carry out dead reckoning using wheeled odometer, visual odometry or inertial navigation device.
In one embodiment of above-mentioned robot navigation's localization method, the dead reckoning process table of wheeled odometer is used
It is shown as:
Wherein,For tkThe estimated value of moment robot location,For tkThe corner of moment robot, B are left and right wheels
Between distance,Respectively tk-1To tkThe distance of revolver and right wheel passed by period.
In one embodiment of above-mentioned robot navigation's localization method, the step of dead reckoning is carried out using visual odometry
Suddenly are as follows: acquire new frame image first, extract the ORB characteristic point in image, and calculate corresponding BRIEF description of characteristic point；
Then the characteristic point of present image is matched with the characteristic point of previous frame image；Finally carry out the solution of 3D-2D pose, i.e., it is logical
It crosses and minimizes the rotational translation matrix that re-projection error obtains before and after frames, so that accumulative obtain robot current pose.
In one embodiment of above-mentioned robot navigation's localization method, dead reckoning is carried out using inertial navigation device
Step are as follows: the acceleration and angular speed that robot is obtained by gyroscope and accelerometer carries out integral to it to calculate machine
The current pose of device people.
In one embodiment of above-mentioned robot navigation's localization method, the step S130 further includes following steps: more
A sensor is set as at least one ultrasonic sensor, at least one laser radar and at least one RGB-D camera, when one
The deviation of the data of ultrasonic sensor detects in region data and laser radar and RGB-D phase machine testing exceeds a deviation threshold
It when value, then repeats to check those data, if those data are unstable, then it is assumed that there are reflective/transparent/translucents in the region
Object, and be labeled in map.
In one embodiment of above-mentioned robot navigation's localization method, the step S220 further includes parsing spread step
221: ignoring the barrier in environment, calculate present node to the path of destination node, if the path is collisionless, general
The node in the path is directly appended in node listing.
In one embodiment of above-mentioned robot navigation's localization method, further include after the step parsing spread step
Following processing Optimization Steps S222: the smooth planning path of conjugate gradient method is used；Interpolation is carried out to increase the close of point range in path
Degree；Reusing conjugate gradient method makes path far from barrier.
Robot navigation's positioning system of the invention, is connected in a robot, carries out map structure for the robot
It builds, position and path planning comprising:
Sensor unit is connected in the robot；Positioning unit, including odometer module, connect the sensor
Unit, the robot detects ambient condition information by the sensor unit, then based on adaptive particle filter
SLAM algorithm, and match the odometer module and complete real-time map building and positioning；And path planning unit, described in connection
Sensor unit and positioning unit, after progress path planning obtains path length and expanding node number on the map of rasterizing,
The map of higher rasterizing is obtained by parsing extension；And using the path length of acquisition and expanding node number as fuzzy reasoning
Input, gain enlightenment formula weight by fuzzy reasoning, the input of the search as second stage, in the map of higher rasterizing
Upper carry out path planning.
In one embodiment of above-mentioned robot navigation's positioning system, the odometer module include wheeled odometer,
Visual odometry and inertial navigation device.
Beneficial functional of the invention is:
1, positioning aspect, proposes a kind of SLAM frame based on adaptive particle filter, which can match difference
Odometer technology complete real-time map building and positioning function under a variety of environment such as outer indoors.Its basic thought is by dividing
Present frame laser data structure feature is analysed, present frame is carried out and is matched with local map, the accumulated error of odometer is eliminated, to reach
To the function being accurately positioned with composition.This method has also merged laser radar, RGB-D camera, ultrasonic sensor and odometer etc.
Multiple sensors information solves ambient light variation, dynamic barrier, transparent/translucent object that conventional method can not be coped with
The problem of the various aspects such as body.In addition, map edit host computer has been write in this project design, it can with the environment of human-edited's building
Figure realizes a series of functions such as map optimization, virtual wall setting, the ease for use for improving robot is reused by map.
2, in terms of path planning, a kind of path planning algorithm based on bipolar mixture state A* is proposed.The algorithm is to biography
The heuristic A * searching algorithm of system is improved, and is assessed first environment complexity, and calculates the heuristic power of algorithm
Weight, then in conjunction with heuristic weight, robot current state, dbjective state and robot kinematics' model environment high score
Higher-dimension path planning is carried out in resolution map, directly exports the robot target pose and target velocity of following a period of time.
The algorithm considers the kinematics model of robot, and institute's planning path meets kinematical constraint, therefore can guarantee in the process of moving
The ride comfort of robot operation.Simultaneously as higher-dimension program results include target speed information, there is no need to redesign complexity
Motion controller can make robot to path carry out high precision tracking.
Below in conjunction with the drawings and specific embodiments, the present invention will be described in detail, but not as a limitation of the invention.
Detailed description of the invention
Fig. 1 is the structural schematic diagram for the embodiment that robot navigation's positioning system of the invention is applied to a robot；
Fig. 2 a and Fig. 2 b are respectively the decomposition explosion figure of the different angle of structure shown in Fig. 1；
Fig. 3 is the composition figure of robot navigation's positioning system of the invention；
Fig. 4 is robot navigation's localization method block diagram of the invention；
Fig. 5 is the composition figure of the positioning unit of robot navigation's positioning system of the invention；
Fig. 6 is the two-wheeled differential steering illustraton of model of the odometer module of robot navigation's positioning system of the invention；
Fig. 7 is the vision differential steering illustraton of model of the odometer module of robot navigation's positioning system of the invention；
Fig. 8 is the exemplary diagram of the update timing of map of the present invention；
Fig. 9 is that map rejuvenation of the present invention and history of errors eliminate schematic diagram；
Figure 10 is map constructed by the present invention；
Figure 11 is VGG-Net schematic network structure；
Figure 12 is the algorithm architecture diagram based on bipolar mixture state A* algorithm of robot navigation's localization method of the invention；
Figure 13 is the path search algorithm of traditional A* based on grating map；
Figure 14 is the path planning example of traditional A* based on grating map；
Figure 15 is the method for the present invention based on bipolar mixture state A* algorithm path planning example；
Figure 16 is the single cycle flow chart based on bipolar mixture state A* algorithm of the method for the present invention；
Figure 17 is using the path comparison diagram before and after interpolation；
Figure 18 is that robot uses the route programming result figure based on bipolar mixture state A* under actual conditions.
Wherein, appended drawing reference
10 robots
100 fixed plates
200 walking mechanisms
300 drive control mechanisms
400 collision detection mechanisms
401 ultrasonic sensors
402 RGB-D cameras
500 laser radar mechanisms
501 laser radars
600 battery means
700 extension mechanisms
20 robot navigation's positioning systems
23 sensor units
24 sensor units
241 odometer modules
242 particle filter modules
243 map rejuvenation modules
25 path planning units
S100, S110-S130, S200, S210-S 220, S221-S222 step
Specific embodiment
Technical solution of the present invention is described in detail in the following with reference to the drawings and specific embodiments, to be further understood that
The purpose of the present invention, scheme and effect, but it is not intended as the limitation of scope of the appended claims of the present invention.
As shown in Fig. 1 and Fig. 2 a, Fig. 2 b, robot 10 includes fixed plate 100, walking mechanism 200, drive control mechanism
300, collision detection mechanism 400, laser radar mechanism 500, battery means 600 and extension mechanism 700.It need to illustrate yes, figure
It is shown as a robot moving platform, can be attached by extension mechanism 700 and a variety of robot functional components, example is formed
Such as educational robot, domestic robot, welcome guide robot, meal delivery robot.
Robot navigation's positioning system 20 of the invention is connected in robot 10, carries out map structure for robot 10
It builds, position and path planning, as shown in figure 3, robot navigation's positioning system 20 includes sensor unit 23, sensor unit
24 and path planning unit 25.Sensor unit 23 is connected in robot 10.Sensor unit 23 is, for example, laser radar
The laser radar 501 of mechanism 500, multiple ultrasonic sensors 401 of collision detection mechanism 400, RGB-D camera 402 etc..
Sensor unit 24 connects sensor unit 23, and sensor unit 24 includes odometer module 241, and robot 10 is logical
It crosses sensor unit 23 and detects ambient condition information, then the SLAM algorithm based on adaptive particle filter, and match odometer
Module 241 completes real-time map building and positioning.Path planning unit 25 connects sensor unit 23 and sensor unit 24, road
Diameter planning unit 25 carries out passing through parsing and expanding after path planning obtains path length and expanding node number on the map of rasterizing
Exhibition obtains the map of higher rasterizing；And using the path length of acquisition and expanding node number as the input of fuzzy reasoning, pass through
Fuzzy reasoning gains enlightenment formula weight, the input of the search as second stage, in the enterprising walking along the street diameter of the map of higher rasterizing
Planning.
As shown in figure 4, robot navigation's localization method of the invention, for a robot carry out map structuring, positioning with
And path planning comprising following steps:
S100, positioning step: robot detects ambient condition information by multiple sensors, is then based on adaptive particle
The SLAM algorithm of filtering, and match different odometers and complete real-time map building and positioning；
Path planning step: S200 uses the path planning algorithm based on bipolar mixture state A*, comprising:
S210 passes through parsing after progress path planning obtains path length and expanding node number on the map of rasterizing
Extension obtains the map of higher rasterizing；
S220 is obtained using the path length of acquisition and expanding node number as the input of fuzzy reasoning by fuzzy reasoning
Heuristic weight, the input of the search as second stage carry out path planning on the map of higher rasterizing.
The step S100 includes the following steps:
S110 establishes map using the data of sensor feedback；
S120 carries out dead reckoning using odometer, and particle filter obtains dead reckoning information and as robot position
The initial estimation of appearance is matched using the current frame data of sensor with local map later, and the accumulation for eliminating odometer misses
Difference；
Current frame information, while benefit is then added when present frame matching result reaches a matching threshold in S130 in map
With current frame information eliminate map in original error message, when matching result persistently reaches the matching threshold, then continue into
Row map rejuvenation；When present frame matching result does not reach the matching threshold, map is remained unchanged, the position of update robot
Appearance information.
Laser radar 501 that SLAM algorithm based on Multi-sensor Fusion of the invention mainly uses robot 10 to carry,
The combination S LAM algorithm of RGB-D camera 402 and odometer positions robot 10, while using ultrasonic sensor 401
Reflective/transparent/translucent object that vision-based detection is had any problem is detected, is supplemented based on optical detection method.
SLAM algorithm based on Multi-sensor Fusion proposes the CRSPF-SLAM based on laser radar 501 as main fixed
Position means.The algorithm is based on adaptive particle filter, can match different odometer technologies indoors under a variety of environment such as outer
Complete real-time map building and positioning function.Its basic thought is the laser data structure feature by analyzing present frame, will be worked as
Previous frame is matched with local map, eliminates the accumulated error of odometer, to reach the function of accurate positioning and composition.
It will be described above content below.
As shown in figure 5, positioning unit 21 of the invention mainly includes three modules, respectively odometer module 241, grain
Subfilter module 242 and map rejuvenation module 243, it is specific as follows.
1), odometer module 241
SLAM algorithm based on Multi-sensor Fusion has used three kinds of different odometers respectively --- wheeled odometer, view
Feel that odometer and inertial navigation device (IMU) carry out dead reckoning.Three kinds of odometers can satisfy a variety of different platforms or
Application environment, they are mainly used to provide initial 10 posture information of robot for particle filter.
Wherein, wheeled odometer module obtains the real-time attitude of rough robot 10 by dead reckoning, and two-wheeled is differential
Steering model is as shown in Figure 6.
The differential model dead reckoning procedural representation of two-wheeled are as follows:
Wherein,For tkThe estimated value of moment robot location,For tkThe corner of moment robot 10, B are left and right
Distance between wheel,Respectively tk-1To tkThe distance of revolver and right wheel passed by period.
Wheeled odometer is suitable for Land leveling and has the environment of certain frictional force, if occurring to beat in the movement of robot 10
Precision can reduce when sliding or ground out-of-flatness.
The main algorithm process of visual odometry are as follows: new frame image is acquired first, extracts the ORB characteristic point in image,
And calculate corresponding BRIEF description of characteristic point.Then the characteristic point of present image and the characteristic point of previous frame image are carried out
Matching, since there may be error hidings, purifies matched characteristic point using RANSAC algorithm.Finally carry out 3D-
2D pose solves, i.e., the rotational translation matrix of before and after frames is obtained by minimizing re-projection error, so that accumulative obtain robot
10 current poses.
As shown in fig. 6, P=[XYZ]TFor the known three-dimensional point in space,WithTo be obtained by characteristic matching
Three-dimensional point P in -1 frame of kth and kth frame corresponding points,The P being calculated by pinhole camera model for P point is in the second frame
The practical subpoint of image.Ideally, the match point in the first secondary figureCorresponding three-dimensional point P projects to the second width figure
As obtainingIt should be with characteristic matching pointIt is stringent to be overlapped, actually due to the presence of error,WithIt can't weigh
It closes, but there are certain distance e, add up for the error distance of characteristic matching point all in two images, and minimum
Changing this error, to can be obtained by optimization aim as follows
By optimizing the objective function, pose transformation relation between available adjacent two field pictures, and then can calculate
The mileage and course information of robot 10 out.
Visual odometry suitable for the environment of feature rich, if environmental colors are single or light too it is bright it is too dark all can shadow
Ring the precision of odometer.
Inertial navigation device can obtain the acceleration and angular speed of robot 10 by gyroscope and accelerometer, to its into
The capable current pose for integrating to calculate robot 10, but micro electronmechanical (MEMS) gyroscope and accelerometer that the present invention uses
There are biggish drift error, it is only capable of providing the accurate displacement in the short period and posture information.
2), particle filter module 242
Particle filter (Particle Filter) obtains the dead reckoning information of above-mentioned three kinds of odometers, as machine
The initial estimation of 10 pose of people is matched with cartographic information using the data of laser radar 501 later, obtains more accurate pose,
The accumulated error of odometer is eliminated again.In the system of this project, which has been substantially carried out present laser structural analysis, is based on
IPAN algorithm finds the key structure point in present laser frame, increases the impact factor of system point, is then based on Monte Carlo
Local iteration's convergence process finds optimum attitude of the robot 10 under local space.Wherein, each particle posture passes through following formula
It obtains:
Sm(t)=[pm(1)...pm(i)...pm(Ns(t))]
Sb(t)=[pb(1)...pb(i)...pb(Ns(t))]
pm(i)=[xm(i) ym(i)]T
pb(i)=[xb(i) yb(i)]T
Wherein Sm(t) for as particle to be matched, SbIt (t) is Current observation data, pmAnd pbEach laser point respectively in particle
Each laser point coordinates in coordinate and observation data.By to continuous coupling frame analyze, search out the optimal key frame of present period and
Its matching result M (t).Meanwhile the Optimum Matching result is used as value of feedback, acts on next frame particle matching process, particle
Distribution (XM,YM,AM) and the number of iterations (loopnum) pass through history match value and parameter preset method of determination are as follows:
On the other hand, which will also be returned to odometer module 241 and map rejuvenation module 243, be respectively used to accumulate
Tired error concealment and map information update.
3), map rejuvenation module 243
Map rejuvenation module (Map Update) 223 mainly includes that three classes execute movement: when key frame matching result is enough
When excellent, then present frame scanning information is not only added in global map, while map can be eliminated using present frame scanning information
In original error message；When matching value persistently keeps excellent, then map rejuvenation is persistently carried out；When key frame matching result not
When enough excellent, global map will be remained unchanged, the posture information of update robot 10.
The exemplary diagram of the update timing of map is as shown in Figure 8.
1. figure middle line is the value of each moment matching result M (t)；It indicates to carry out map rejuvenation when figure middle line is 1 2., is 0
When indicate without update.
It is as shown in Figure 9 that map rejuvenation and history of errors eliminate schematic diagram.Figure grey area range is current key frame
Scanning range, the map original information in the region will be eliminated as history of errors.
Three modules work independently under the different clock cycle respectively, by Real time data share and feedback to realize
Positioning and map structuring function, building result schematic diagram are as shown in Figure 10 in real time for robot 10.
In addition, 10 navigation positioning system of robot, which uses, surrounds robot 10 for the object of reflective/transparent/translucent
Multiple ultrasonic sensors 401 on mobile mechanism's platform detect it.Since the sensitizing range of ultrasonic sensor 401 is one
A sector, therefore when the data of distance and laser radar 501 and RGB-D camera 402 that ultrasonic wave returns have relatively large deviation, it is
System just checks these data, if the unstable i.e. measurement variance of data is larger at this time, then it is assumed that exist in the sector anti-
Light/transparent/translucent object, system will be labeled in map at this time.
The observation station of same a collection of target is obtained to ensure that each sensor observation data obtained are synchronizations, need by
Observation data synchronize.Synchronizing main for data includes two aspects: spatial registration and time synchronization.Spatially, it observes
On the basis of the coordinate of data is all the local coordinate system locating for each sensor, for the ease of data fusion, need each biography
The position of sensor and angle information are unified into global coordinate system.The static data that the present invention is obtained by each sensor, according to
The relative pose that each sensor can be sought by the registration of static data carries out off-line calibration to the relative position of each sensor.
Use local zone time as this frame data when control system obtains each sensor information for the time synchronization for guaranteeing data
Timestamp realizes that clock is synchronous.
SLAM algorithm based on Multi-sensor Fusion have preferable robustness, however this method building 2D point cloud for
The ability of scene Recognition is weaker.If the inadequate robust of scene Recognition algorithm, robot 10 would have to require when starting every time
Robot 10 is still in a certain fixed position of map, otherwise possibly can not be properly positioned.And the scene Recognition algorithm of robust can
So that robot 10 obtains more accurate pose, start robot 10 at an arbitrary position.
10 navigation positioning system of robot uses the scene recognition method based on vision to the initial position of robot 10
Estimated with initial attitude.While carrying out map structuring, the every operation a distance of robot 10 is just by imaging sensor
The image of acquisition and current position and posture are recorded, and image data base is saved in.It is opened every time in robot 10 in this way
After machine, a series of history images for saving and the letter such as pose of robot 10 when shooting the image when load is constructed map
Breath searches scene similar with present image, finds similar scene by the image in comparison present image and image data base
Perspective n point location problem (PnP) is solved afterwards, obtains the pose transformation relation of history image in present image and image data base,
Due to pose of the history image in global map it is known that the therefore current position in global map of available robot 10
Estimation.Later, algorithm using present laser radar points cloud by Monte Carlo localization algorithm to the location estimation of scene Recognition into
It advances a successive step, obtains the locating effect of degree of precision, realize the function of starting whenever and wherever possible.
SLAM algorithm based on Multi-sensor Fusion mainly passes through bag of words and identifies to scene image, bag of words
It is broadly divided into off-line training in application process and uses two parts online.
Off-line training part is to pass through training in the way of offline by clustering algorithm to a large amount of image datas.Number is extracted first
According to the FAST characteristic point for concentrating all pictures, the image block around characteristic point in 64 × 64 pixel coverages is extracted later, by going
The convolutional neural networks VGG-Net for falling full articulamentum obtains characteristic pattern, and characteristic pattern is carried out spatial pyramid pond, obtains the spy
Corresponding description of sign point, clusters description of characteristic points all in image, the classification number n clustered is visual word
The word number of allusion quotation.VGG-Net schematic network structure is as shown in figure 11.
It is online to carry out winding detection using the vision word that off-line training obtains using part, it is freshly harvested for each
Image acquires the characteristic point in image, and description of each characteristic point is calculated with method identical with training process, is obtained most
The vision word of neighbour, the corresponding vision word of all characteristic points, is indicated with n dimensional feature vector in statistical picture, if two
The feature vector of width image is closely located, it may be considered that two images are more similar.The advantage of vision bag of words is to lead to
The dimension of data can be reduced by crossing cluster, increase storage efficiency, and can increase system to the robust of brightness or visual angle change
Property.
The path planning of mobile robot 10 refers to that robot 10 will avoid in environment during mobile to target point
Barrier, an optimal or suboptimum from starting point to target point is gone out with a certain index planning such as distance, time, energy consumption
Path.The path planning of mobile robot 10 is divided into global path planning and two kinds of local paths planning.Global path planning is
Refer in the completely known situation of environmental information, finds the optimal path from starting point to target point.Global path planning is to meter
The requirement of real-time of calculation is not high, but but very big to the dependence of environmental map.But the working environment one of mobile robot 10
As be all non-structured environment, and there are dynamic changes, therefore rely solely on global path planning often and can not obtain very well
Planning effect.Local paths planning does not need whole Environmental Map Informations, according to the information of sensor feedback in part
Route searching planning is carried out in environmental map, this requires robots 10 to have preferable information processing capability, to the reality of calculating
The requirement of when property is relatively high, while to have preferable robustness.
For mobile-robot system, a reasonable solution of comparison be first in global map according to it is long when not
The cartographic information of change carries out a Global motion planning, is then held in robot local map along the global path cooked up
Continuous local paths planning solves the influence of environmental change and dynamic object to robot with this.
In Global motion planning, in order to rapidly search for out arriving at the path of task point, using based on fuzzy reasoning
Bipolar mixture state A* algorithm, algorithm architecture diagram are as shown in figure 12.
So-called two-phase refers to that algorithm is divided into two stages, and the search of first stage is on low resolution map using tradition
A* algorithm carries out path planning and obtains path length and expanding node number.Using the two amounts as the input of fuzzy reasoning, pass through
Fuzzy reasoning gains enlightenment formula weight, and the input of the search as second stage carries out path planning on High Resolution Ground Map.
Traditional A* path search algorithm based on grating map is introduced first, as shown in figure 13.
A* algorithm uses the thought of heuristic search, and so-called heuristic search is exactly to be evaluated to each in state space
The position estimated is assessed, and best position is therefrom obtained, then is scanned for and assessed until reaching target position from the position.
For well-established lower resolution grid map, the OPEN table and CLOSE table of path node are established, starting point is placed in CLOSE
Table then by point merging OPEN table a certain range of around CLOSE table, and successively calculates its corresponding path scoring.Path
Cost function is f (n)=g (n)+h (n), wherein g (n) represents the actual range (G cost) from starting point to any vertex n, h
(n) estimated distance (H cost) of any vertex n of expression to representative points.H (n) can obtain difference using different distance functions
Program results.By the way that the node with minimum cost is constantly placed in CLOSE table, while its neighborhood interior nodes cost is calculated,
And node connection relationship is adjusted, the final path for obtaining the minimum cost from terminal to starting point.
Bipolar mixture state A* algorithm based on fuzzy reasoning of the invention is the innovatory algorithm of traditional A* algorithm.Traditional
A* planning algorithm is by map rasterizing, then using grid midpoint as search node, causes planning path discontinuous, such as Figure 14 institute
Show.
Bipolar mixture state A* based on fuzzy reasoning of the invention has then contacted each grid and a continuous state
Come, these continuous states utilize robot kinematics' model, are calculated by forward simulation come therefore, of the invention based on mould
The path of the bipolar mixture state A* planning of paste reasoning is smooth and meets the kinematical constraint of robot, as shown in figure 15.
The single cycle flow chart for the bipolar mixture state A* algorithm based on fuzzy reasoning that this system method uses is as schemed
Shown in 16.
Identical as traditional A*, algorithm first associates the continuous state of current robot with starting grid, then leads to
It crosses forward simulation and calculates the sub- state of current continuous state, and calculate the grid that this little state is fallen into, if the grid is never
It appears in OPEN list, then being added in OPEN list；If the grid is already present in OPEN list, count
The current G cost of the grid is calculated, if it is less than original G cost, updates the cost and father node of the grid, and again right
OPEN list is ranked up.Since forward simulation is difficult accurately to arrive at dbjective state, and also to further increase mixing shape
The real-time and path smoothness of state A* search, have also carried out parsing spread step 221, which neglects during point spread
Barrier slightly in environment calculates present node to the path Reed-Shepp of destination node, if the path be it is collisionless,
Then the direct point in the path is added in OPEN list.
Post-processing in Optimization Steps 222 includes three steps, and the first step is using the smooth planning path of conjugate gradient method, second step
It is the density for carrying out interpolation to increase point range in path, third step is to reuse conjugate gradient method to make path far from barrier.
The optimization of the first step uses the smooth planning path of conjugate gradient method, objective function to be optimized are as follows:
Wherein κmaxFor minimum turning radius, σκIt is cost coefficient, ωκAnd ωsIt is weighting coefficient, N is the waypoint in path
Quantity, Δ xiWith Δ φiIt is position vector and angle vector respectively, calculation method is as follows:
Δxi=xi-xi-1
Wherein x and y is the cross of waypoint, ordinate respectively, and in order to improve the real-time of admixture A* search, search is used
Grating map resolution ratio it is usually lower, therefore the waypoint spacing in path is larger, be unfavorable for underlying programs and carry out tracing control,
The optimization of second step, i.e. interpolation operation are carried out thus.Path comparison before and after interpolation is as shown in figure 17.
Two figures are a kind of smoothing methods in chassis path, and wherein left figure is smooth later path, right figure be smoothly with
Preceding, every each paths of figure.Path is that many equally spaced points link up to be formed, between the path point and point in right figure
Interval it is very big, be actually to be spliced by many short and small straight lines although to look at be a smooth-path to naked eyes, machine
If people will very acutely according to this route direction change, can interleaving in original point and point by difference operation
Enter many intermediate points, so that the interval put and put in new route becomes smaller, such path is just more smooth, and result is exactly left figure.
In order to enable planning path on the basis of collisionless further away from barrier, algorithm carries out third step optimization,
Conjugate gradient method path optimizing is reused, objective function to be optimized is
Wherein σoIt is cost coefficient, ωoIt is weighting coefficient, oiIt is obstacle article coordinate, dmaxBe robot and barrier most
Big distance.
In sector planning, this system use based on the calculation of the admixture A* path planning of target point range ambiguity reasoning
Method.The input of the local paths planning includes current state (coordinate, posture, the speed of established grating map, robot
Deng), the dbjective state of robot, output includes planning path, i.e. the status switch of robot and the corresponding control of each state
Amount processed is used for tracing control.
The local paths planning method is using last stage Global motion planning as a result, to learn robot current state and mesh
The distance of mark state, the rear heuristic weight of reasoning is prepared for it；Then admixture is carried out on local high resolution map
A* search, which has comprehensively considered robot coordinate, attitude angle and speed, and combines the kinematics of robot
Model can cook up a safe and efficient, collisionless smooth-path.In admixture A* search, Rational choice is inspired
The speed of service of searching algorithm can be improved in formula weight, by the expanding node number and path length of global stage tradition A* search output
Degree carries out Fuzzy processing, then carries out fuzzy reasoning, last de-fuzzy can online choose heuristic weight.
In addition, bipolar mixture state A* search is when carrying out collision detection, it is contemplated that the volume size of robot, therefore advise
The path drawn not will lead to robot strikes obstacles, and search dbjective state is arranged in global path and can be returned to after avoidance
Correct path.As previously mentioned, bipolar mixture state A* search considers the kinematics model of robot, institute's planning path meets fortune
It is dynamic to learn constraint, therefore can guarantee the ride comfort of robot operation in the process of moving.
Figure 18 illustrates route programming result of the robot based on admixture A* under actual conditions.Circle R is represented in figure
Robot current location, short-term direction represent robot course；Rectangular and short-term T represents robot target pose；Solid black lines L1
For global path planning as a result, black curve L2 is local paths planning result.
Wherein, scheme in (a), since moving obstacle being not present in environment, the local path L2 of robot can fast convergence
To global path L1.And scheme in (b), since there are dynamic barrier D (black laser points on the L1 of robot global path
Cloud), therefore in local path L2 planning process, robot can bypass barrier D, then transport further along global path L1
Row.
Due to using the A* planing method of bipolar mixture state that can export a series of optimum states of robot, including position
Therefore appearance and speed are pushed away by the way that the kinematics model of robot is counter, can be obtained the optimal control parameter of robot.So
And there are many X factors, the coarse smooth degree on ground etc. for robot workplace, this lead to robot must
Closed loop controller is crossed to realize the accurate tracking of driving path and state.
Engineering in practice, the closed loop controller being most widely used be ratio, integral, derivative controller, abbreviation PID control
Device processed.PID controller appearance has nearly 70 years history so far, since structure is simple, stability is good, reliable operation, adjustment side
Just become one of the major technique of Industry Control.PID controller is a kind of linear controller, it is defeated according to given value and reality
Value constitutes control deviation out.The proportional, integral term and differential term of deviation are constituted into control amount by linear combination.
The A* method of bipolar mixture state uses fuzzy controller, needs to pass through experiment before the online use of controller
Or experience obtains the fuzzy reasoning relationship between these three parameters of ratio, integral, differential and error, error rate, this is pushed away
Knowledge base is made in reason relationship.Online in use, error and error rate are blurred by controller first, it is inputted knowledge base
In, fuzzy reasoning output is obtained as a result, three ratio, integral, differential parameters will be obtained after result defuzzification, realizes parameter
On-line tuning to meet the requirement of different errors and error rate, and improves the dynamic and steady-state performance of controlled process.It is fuzzy
The input of controller is course, location error and its change rate of robot, and output quantity is the setting value of two wheel speed of robot.
For the stability for guaranteeing robot operation, path trace module is also required to carry out the disposal of gentle filter to control parameter, to avoid
Danger caused by robot acceleration acute variation as caused by emergency stop, anxious starting.
When needing to dispose robot, first using host computer remote-controlled robot scan affiliated area, constructing environment map,
Then editorial optimization is carried out using host computer to map, virtual wall is such as set to limit robot working space, finally according to industry
Business demand sets several task points.
Machine man-hour, client are obtained by human-computer interaction and are serviced, and top layer operation system can be according to customer demand guide
Send instructions under boat system, the programmed decision-making module of navigation system is according to task point and self poisoning, in the map built
Carry out path planning.Meanwhile positioning immediately calculates current pose using the information of various kinds of sensors acquisition with map structuring module,
And pose and ambient condition information are fed back together to programmed decision-making module.Programmed decision-making module synthesis task point, robot work as
Preceding state and ambient condition information carry out path planning and generate motion control amount, be sent to robot motion's system.In addition,
Programmed decision-making module can monitor execution status of task in real time, and state be fed back to the financial business application of top layer, and pass through people
Machine interactive function informs user.
Certainly, the present invention can also have other various embodiments, without deviating from the spirit and substance of the present invention, ripe
It knows those skilled in the art and makes various corresponding changes and modifications, but these corresponding changes and change in accordance with the present invention
Shape all should fall within the scope of protection of the appended claims of the present invention.
Claims (11)
1. a kind of robot navigation's localization method carries out map structuring, positioning and path planning, feature for a robot
It is, includes the following steps:
S100, positioning step: robot detects ambient condition information by multiple sensors, is then based on adaptive particle filter
SLAM algorithm, and match different odometer and complete real-time map building and positioning；And
Path planning step: S200 uses the path planning algorithm based on bipolar mixture state A*, comprising:
S210 after progress path planning obtains path length and expanding node number on the map of rasterizing, passes through parsing and extends
Obtain the map of higher rasterizing；And
S220 gains enlightenment using the path length of acquisition and expanding node number as the input of fuzzy reasoning by fuzzy reasoning
Formula weight, the input of the search as second stage carry out path planning on the map of higher rasterizing.
2. robot navigation's localization method according to claim 1, which is characterized in that the step S100 includes following step
It is rapid:
S110 establishes map using the data of sensor feedback；
S120 carries out dead reckoning using odometer, and particle filter obtains dead reckoning information and as robot pose
Initial estimation is matched using the current frame data of sensor with local map later, eliminates the accumulated error of odometer；
When present frame matching result reaches a matching threshold current frame information is then added, while utilizing and working as in S130 in map
Preceding frame information eliminates original error message in map, when matching result persistently reaches the matching threshold, then lasting to carry out ground
Figure updates；When present frame matching result does not reach the matching threshold, map is remained unchanged, the pose letter of update robot
Breath.
3. robot navigation's localization method according to claim 1 or 2, which is characterized in that adaptive in the step S100
The SLAM algorithm of particle filter is answered to carry out dead reckoning using wheeled odometer, visual odometry or inertial navigation device.
4. robot navigation's localization method according to claim 3, which is characterized in that pushed away using the boat position of wheeled odometer
Calculate procedural representation are as follows:
Wherein,For tkThe estimated value of moment robot location,For tkThe corner of moment robot, B is between left and right wheels
Distance,Respectively tk-1To tkThe distance of revolver and right wheel passed by period.
5. robot navigation's localization method according to claim 3, which is characterized in that carry out boat position using visual odometry
The step of reckoning are as follows: acquire new frame image first, extract the ORB characteristic point in image, and calculate the corresponding BRIEF of characteristic point
Description；Then the characteristic point of present image is matched with the characteristic point of previous frame image；3D-2D pose is finally carried out to ask
Solution obtains the rotational translation matrix of before and after frames by minimizing re-projection error, so that accumulative obtain robot current pose.
6. robot navigation's localization method according to claim 3, which is characterized in that navigated using inertial navigation device
The step of position calculates are as follows: the acceleration and angular speed that robot is obtained by gyroscope and accelerometer, it is integrated from
And calculate the current pose of robot.
7. robot navigation's localization method according to claim 2, which is characterized in that the step S130 further includes as follows
Step: multiple sensors are set as at least one ultrasonic sensor, at least one laser radar and at least one RGB-D phase
Machine, when the deviation of the data of ultrasonic sensor detects in a region data and laser radar and RGB-D phase machine testing is super
It out when a deviation threshold, then repeats to check those data, if those data are unstable, then it is assumed that there are reflective/saturating in the region
Bright/translucent object, and be labeled in map.
8. robot navigation's localization method according to claim 1, which is characterized in that the step S220 further includes parsing
Spread step 221: ignoring the barrier in environment, calculates present node to the path of destination node, if the path is that nothing is touched
It hits, then the node in the path is directly appended in node listing.
9. robot navigation's localization method according to claim 8, which is characterized in that step parsing spread step it
After further include following processing Optimization Steps S222: use the smooth planning path of conjugate gradient method；Interpolation is carried out to increase in path
The density of point range；Reusing conjugate gradient method makes path far from barrier.
10. a kind of robot navigation's positioning system, is connected in a robot, map structuring is carried out for the robot, is determined
Position and path planning characterized by comprising
Sensor unit is connected in the robot；
Positioning unit, including odometer module, connect the sensor unit, and the robot is examined by the sensor unit
Ambient condition information is surveyed, then the SLAM algorithm based on adaptive particle filter, and matches the odometer module and complete in real time
Map structuring and positioning；And
Path planning unit connects the sensor unit and positioning unit, and path planning is carried out on the map of rasterizing and is obtained
To after path length and expanding node number, the map of higher rasterizing is obtained by parsing extension；And by the path length of acquisition
Degree and input of the expanding node number as fuzzy reasoning gain enlightenment formula weight by fuzzy reasoning, as searching for second stage
The input of rope carries out path planning on the map of higher rasterizing.
11. robot navigation's positioning system according to claim 10, which is characterized in that the odometer module includes wheel
Formula odometer, visual odometry and inertial navigation device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711420969.4A CN109959377A (en) | 2017-12-25 | 2017-12-25 | A kind of robot navigation's positioning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711420969.4A CN109959377A (en) | 2017-12-25 | 2017-12-25 | A kind of robot navigation's positioning system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109959377A true CN109959377A (en) | 2019-07-02 |
Family
ID=67021025
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711420969.4A Withdrawn CN109959377A (en) | 2017-12-25 | 2017-12-25 | A kind of robot navigation's positioning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109959377A (en) |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110260867A (en) * | 2019-07-29 | 2019-09-20 | 浙江大华技术股份有限公司 | Method, equipment and the device that pose is determining in a kind of robot navigation, corrects |
| CN110531766A (en) * | 2019-08-27 | 2019-12-03 | 熵智科技(深圳)有限公司 | Based on the known continuous laser SLAM composition localization method for occupying grating map |
| 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 |
| CN110632919A (en) * | 2019-08-28 | 2019-12-31 | 广东工业大学 | Autonomous positioning navigation method based on crawler-type rescue robot |
| CN110648353A (en) * | 2019-08-30 | 2020-01-03 | 北京影谱科技股份有限公司 | Monocular sensor-based robot indoor positioning method and device |
| CN110763225A (en) * | 2019-11-13 | 2020-02-07 | 内蒙古工业大学 | Trolley path navigation method and system and transport vehicle system |
| CN110823225A (en) * | 2019-10-29 | 2020-02-21 | 北京影谱科技股份有限公司 | Positioning method and device under indoor dynamic situation |
| CN111012251A (en) * | 2019-12-17 | 2020-04-17 | 哈工大机器人（合肥）国际创新研究院 | Planning method and device for full-coverage path of cleaning robot |
| CN111060108A (en) * | 2019-12-31 | 2020-04-24 | 江苏徐工工程机械研究院有限公司 | Path planning method and device and engineering vehicle |
| CN111142514A (en) * | 2019-12-11 | 2020-05-12 | 深圳市优必选科技股份有限公司 | Robot and obstacle avoidance method and device thereof |
| CN111176288A (en) * | 2020-01-07 | 2020-05-19 | 深圳南方德尔汽车电子有限公司 | Reedsshepp-based global path planning method and device, computer equipment and storage medium |
| CN111309015A (en) * | 2020-02-25 | 2020-06-19 | 华南理工大学 | Transformer substation inspection robot positioning navigation system integrating multiple sensors |
| CN111308490A (en) * | 2020-02-05 | 2020-06-19 | 浙江工业大学 | Balance car indoor positioning and navigation system based on single-line laser radar |
| CN111432341A (en) * | 2020-03-11 | 2020-07-17 | 大连理工大学 | Environment self-adaptive positioning method |
| CN111427047A (en) * | 2020-03-30 | 2020-07-17 | 哈尔滨工程大学 | Autonomous mobile robot S L AM method in large scene |
| CN111487986A (en) * | 2020-05-15 | 2020-08-04 | 中国海洋大学 | Underwater robot cooperative target searching method based on global information transfer mechanism |
| CN111812669A (en) * | 2020-07-16 | 2020-10-23 | 南京航空航天大学 | Winding inspection device, positioning method thereof and storage medium |
| CN111812668A (en) * | 2020-07-16 | 2020-10-23 | 南京航空航天大学 | Winding inspection device, positioning method thereof and storage medium |
| CN111830977A (en) * | 2020-07-02 | 2020-10-27 | 中国兵器科学研究院 | Autonomous navigation software framework and navigation method for mobile robot |
| CN111964682A (en) * | 2020-08-10 | 2020-11-20 | 北京轩宇空间科技有限公司 | Fast path planning method and device adapting to unknown dynamic space and storage medium |
| CN112130445A (en) * | 2020-11-24 | 2020-12-25 | 四川写正智能科技有限公司 | Intelligent watch and method for carrying out safety early warning based on driving route of child |
| CN112146660A (en) * | 2020-09-25 | 2020-12-29 | 电子科技大学 | Indoor map positioning method based on dynamic word vector |
| CN112336267A (en) * | 2019-08-07 | 2021-02-09 | 杭州萤石软件有限公司 | Cleaning robot and control method thereof |
| CN112476433A (en) * | 2020-11-23 | 2021-03-12 | 深圳怪虫机器人有限公司 | Mobile robot positioning method based on array boundary identification |
| CN112509056A (en) * | 2020-11-30 | 2021-03-16 | 中国人民解放军32181部队 | Dynamic battlefield environment real-time path planning system and method |
| CN112652001A (en) * | 2020-11-13 | 2021-04-13 | 山东交通学院 | Underwater robot multi-sensor fusion positioning system based on extended Kalman filtering |
| CN113050140A (en) * | 2019-12-27 | 2021-06-29 | 中移智行网络科技有限公司 | Positioning method, positioning device, storage medium and server |
| CN113093761A (en) * | 2021-04-08 | 2021-07-09 | 浙江中烟工业有限责任公司 | Warehouse robot indoor map navigation system based on laser radar |
| CN113109821A (en) * | 2021-04-28 | 2021-07-13 | 武汉理工大学 | Mapping method, device and system based on ultrasonic radar and laser radar |
| CN113189977A (en) * | 2021-03-10 | 2021-07-30 | 新兴际华集团有限公司 | Intelligent navigation path planning system and method for robot |
| CN113232025A (en) * | 2021-06-07 | 2021-08-10 | 上海大学 | Mechanical arm obstacle avoidance method based on proximity perception |
| CN113359714A (en) * | 2021-05-25 | 2021-09-07 | 国网江苏省电力有限公司电力科学研究院 | Routing inspection robot dynamic path planning method and device based on particle filter algorithm |
| CN113608170A (en) * | 2021-07-07 | 2021-11-05 | 云鲸智能(深圳)有限公司 | Radar calibration method, radar, robot, medium, and computer program product |
| CN113835428A (en) * | 2021-08-27 | 2021-12-24 | 华东交通大学 | Robot path planning method for restaurant |
| CN114035562A (en) * | 2021-07-20 | 2022-02-11 | 新兴际华集团有限公司 | Multi-information fusion acquisition robot for explosive environment |
| CN114217622A (en) * | 2021-12-16 | 2022-03-22 | 南京理工大学 | Robot autonomous navigation method based on BIM |
| CN114489036A (en) * | 2021-07-25 | 2022-05-13 | 西北农林科技大学 | Indoor robot navigation control method based on SLAM |
| CN115290098A (en) * | 2022-09-30 | 2022-11-04 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
| WO2022247045A1 (en) * | 2021-05-28 | 2022-12-01 | 浙江大学 | Laser radar information-based mobile robot location re-identification method |
| CN115884853A (en) * | 2020-09-23 | 2023-03-31 | 应用材料公司 | Robot joint space diagram path planning and movement execution |
| CN116541574A (en) * | 2023-07-07 | 2023-08-04 | 湖北珞珈实验室 | Intelligent extraction method, device, storage medium and equipment for map sensitive information |
| CN117405178A (en) * | 2023-12-15 | 2024-01-16 | 成都电科星拓科技有限公司 | Mobile monitoring platform and method for automatically detecting indoor environment parameters |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110125323A1 (en) * | 2009-11-06 | 2011-05-26 | Evolution Robotics, Inc. | Localization by learning of wave-signal distributions |
| CN103644903A (en) * | 2013-09-17 | 2014-03-19 | 北京工业大学 | Simultaneous localization and mapping method based on distributed edge unscented particle filter |
| CN104236551A (en) * | 2014-09-28 | 2014-12-24 | 北京信息科技大学 | Laser range finder-based map establishing method of snake-like robot |
| CN106446815A (en) * | 2016-09-14 | 2017-02-22 | 浙江大学 | Simultaneous positioning and map building method |
| CN107092264A (en) * | 2017-06-21 | 2017-08-25 | 北京理工大学 | Towards the service robot autonomous navigation and automatic recharging method of bank's hall environment |
-
2017
- 2017-12-25 CN CN201711420969.4A patent/CN109959377A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110125323A1 (en) * | 2009-11-06 | 2011-05-26 | Evolution Robotics, Inc. | Localization by learning of wave-signal distributions |
| CN103644903A (en) * | 2013-09-17 | 2014-03-19 | 北京工业大学 | Simultaneous localization and mapping method based on distributed edge unscented particle filter |
| CN104236551A (en) * | 2014-09-28 | 2014-12-24 | 北京信息科技大学 | Laser range finder-based map establishing method of snake-like robot |
| CN106446815A (en) * | 2016-09-14 | 2017-02-22 | 浙江大学 | Simultaneous positioning and map building method |
| CN107092264A (en) * | 2017-06-21 | 2017-08-25 | 北京理工大学 | Towards the service robot autonomous navigation and automatic recharging method of bank's hall environment |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110260867A (en) * | 2019-07-29 | 2019-09-20 | 浙江大华技术股份有限公司 | Method, equipment and the device that pose is determining in a kind of robot navigation, corrects |
| CN112336267A (en) * | 2019-08-07 | 2021-02-09 | 杭州萤石软件有限公司 | Cleaning robot and control method thereof |
| CN110531766A (en) * | 2019-08-27 | 2019-12-03 | 熵智科技(深圳)有限公司 | Based on the known continuous laser SLAM composition localization method for occupying grating map |
| CN110531766B (en) * | 2019-08-27 | 2022-06-28 | 熵智科技(深圳)有限公司 | Continuous laser SLAM (Simultaneous laser mapping) composition positioning method based on known occupied grid map |
| CN110632919A (en) * | 2019-08-28 | 2019-12-31 | 广东工业大学 | Autonomous positioning navigation method based on crawler-type rescue robot |
| CN110648353A (en) * | 2019-08-30 | 2020-01-03 | 北京影谱科技股份有限公司 | Monocular sensor-based robot indoor positioning method and device |
| CN110823225A (en) * | 2019-10-29 | 2020-02-21 | 北京影谱科技股份有限公司 | Positioning method and device under indoor dynamic situation |
| CN110763225A (en) * | 2019-11-13 | 2020-02-07 | 内蒙古工业大学 | Trolley path navigation method and system and transport vehicle system |
| CN111142514A (en) * | 2019-12-11 | 2020-05-12 | 深圳市优必选科技股份有限公司 | Robot and obstacle avoidance method and device thereof |
| CN111142514B (en) * | 2019-12-11 | 2024-02-13 | 深圳市优必选科技股份有限公司 | Robot and obstacle avoidance method and device thereof |
| CN111012251A (en) * | 2019-12-17 | 2020-04-17 | 哈工大机器人（合肥）国际创新研究院 | Planning method and device for full-coverage path of cleaning robot |
| CN113050140A (en) * | 2019-12-27 | 2021-06-29 | 中移智行网络科技有限公司 | Positioning method, positioning device, storage medium and server |
| CN111060108B (en) * | 2019-12-31 | 2021-10-12 | 江苏徐工工程机械研究院有限公司 | Path planning method and device and engineering vehicle |
| CN111060108A (en) * | 2019-12-31 | 2020-04-24 | 江苏徐工工程机械研究院有限公司 | Path planning method and device and engineering vehicle |
| CN111176288A (en) * | 2020-01-07 | 2020-05-19 | 深圳南方德尔汽车电子有限公司 | Reedsshepp-based global path planning method and device, computer equipment and storage medium |
| CN111308490B (en) * | 2020-02-05 | 2021-11-19 | 浙江工业大学 | Balance car indoor positioning and navigation system based on single-line laser radar |
| CN111308490A (en) * | 2020-02-05 | 2020-06-19 | 浙江工业大学 | Balance car indoor positioning and navigation system based on single-line laser radar |
| CN111309015A (en) * | 2020-02-25 | 2020-06-19 | 华南理工大学 | Transformer substation inspection robot positioning navigation system integrating multiple sensors |
| CN111432341A (en) * | 2020-03-11 | 2020-07-17 | 大连理工大学 | Environment self-adaptive positioning method |
| CN111427047A (en) * | 2020-03-30 | 2020-07-17 | 哈尔滨工程大学 | Autonomous mobile robot S L AM method in large scene |
| CN111427047B (en) * | 2020-03-30 | 2023-05-05 | 哈尔滨工程大学 | SLAM method for autonomous mobile robot in large scene |
| CN111487986A (en) * | 2020-05-15 | 2020-08-04 | 中国海洋大学 | Underwater robot cooperative target searching method based on global information transfer mechanism |
| CN111830977A (en) * | 2020-07-02 | 2020-10-27 | 中国兵器科学研究院 | Autonomous navigation software framework and navigation method for mobile robot |
| CN111812668A (en) * | 2020-07-16 | 2020-10-23 | 南京航空航天大学 | Winding inspection device, positioning method thereof and storage medium |
| CN111812669A (en) * | 2020-07-16 | 2020-10-23 | 南京航空航天大学 | Winding inspection device, positioning method thereof and storage medium |
| CN111964682A (en) * | 2020-08-10 | 2020-11-20 | 北京轩宇空间科技有限公司 | Fast path planning method and device adapting to unknown dynamic space and storage medium |
| CN115884853A (en) * | 2020-09-23 | 2023-03-31 | 应用材料公司 | Robot joint space diagram path planning and movement execution |
| CN115884853B (en) * | 2020-09-23 | 2024-01-16 | 应用材料公司 | Robot joint space diagram path planning and mobile execution |
| CN112146660B (en) * | 2020-09-25 | 2022-05-03 | 电子科技大学 | Indoor map positioning method based on dynamic word vector |
| CN112146660A (en) * | 2020-09-25 | 2020-12-29 | 电子科技大学 | Indoor map positioning method based on dynamic word vector |
| CN112652001A (en) * | 2020-11-13 | 2021-04-13 | 山东交通学院 | Underwater robot multi-sensor fusion positioning system based on extended Kalman filtering |
| CN112476433B (en) * | 2020-11-23 | 2023-08-04 | 深圳怪虫机器人有限公司 | Mobile robot positioning method based on identification array boundary |
| CN112476433A (en) * | 2020-11-23 | 2021-03-12 | 深圳怪虫机器人有限公司 | Mobile robot positioning method based on array boundary identification |
| CN112130445A (en) * | 2020-11-24 | 2020-12-25 | 四川写正智能科技有限公司 | Intelligent watch and method for carrying out safety early warning based on driving route of child |
| CN112509056A (en) * | 2020-11-30 | 2021-03-16 | 中国人民解放军32181部队 | Dynamic battlefield environment real-time path planning system and method |
| CN112509056B (en) * | 2020-11-30 | 2022-12-20 | 中国人民解放军32181部队 | Dynamic battlefield environment real-time path planning system and method |
| CN113189977A (en) * | 2021-03-10 | 2021-07-30 | 新兴际华集团有限公司 | Intelligent navigation path planning system and method for robot |
| CN113093761A (en) * | 2021-04-08 | 2021-07-09 | 浙江中烟工业有限责任公司 | Warehouse robot indoor map navigation system based on laser radar |
| CN113109821A (en) * | 2021-04-28 | 2021-07-13 | 武汉理工大学 | Mapping method, device and system based on ultrasonic radar and laser radar |
| CN113359714A (en) * | 2021-05-25 | 2021-09-07 | 国网江苏省电力有限公司电力科学研究院 | Routing inspection robot dynamic path planning method and device based on particle filter algorithm |
| WO2022247045A1 (en) * | 2021-05-28 | 2022-12-01 | 浙江大学 | Laser radar information-based mobile robot location re-identification method |
| CN113232025B (en) * | 2021-06-07 | 2022-04-22 | 上海大学 | Mechanical arm obstacle avoidance method based on proximity perception |
| CN113232025A (en) * | 2021-06-07 | 2021-08-10 | 上海大学 | Mechanical arm obstacle avoidance method based on proximity perception |
| CN113608170A (en) * | 2021-07-07 | 2021-11-05 | 云鲸智能(深圳)有限公司 | Radar calibration method, radar, robot, medium, and computer program product |
| CN113608170B (en) * | 2021-07-07 | 2023-11-14 | 云鲸智能(深圳)有限公司 | Radar calibration method, radar, robot, medium and computer program product |
| CN114035562A (en) * | 2021-07-20 | 2022-02-11 | 新兴际华集团有限公司 | Multi-information fusion acquisition robot for explosive environment |
| CN114489036A (en) * | 2021-07-25 | 2022-05-13 | 西北农林科技大学 | Indoor robot navigation control method based on SLAM |
| CN113835428A (en) * | 2021-08-27 | 2021-12-24 | 华东交通大学 | Robot path planning method for restaurant |
| CN114217622B (en) * | 2021-12-16 | 2023-09-01 | 南京理工大学 | BIM-based robot autonomous navigation method |
| CN114217622A (en) * | 2021-12-16 | 2022-03-22 | 南京理工大学 | Robot autonomous navigation method based on BIM |
| CN115290098B (en) * | 2022-09-30 | 2022-12-23 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
| CN115290098A (en) * | 2022-09-30 | 2022-11-04 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
| CN116541574A (en) * | 2023-07-07 | 2023-08-04 | 湖北珞珈实验室 | Intelligent extraction method, device, storage medium and equipment for map sensitive information |
| CN116541574B (en) * | 2023-07-07 | 2023-10-03 | 湖北珞珈实验室 | Intelligent extraction method, device, storage medium and equipment for map sensitive information |
| CN117405178A (en) * | 2023-12-15 | 2024-01-16 | 成都电科星拓科技有限公司 | Mobile monitoring platform and method for automatically detecting indoor environment parameters |
| CN117405178B (en) * | 2023-12-15 | 2024-03-15 | 成都电科星拓科技有限公司 | Mobile monitoring method for automatically detecting indoor environment parameters |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109959377A (en) | A kind of robot navigation's positioning system and method | |
| Thorpe et al. | Vision and navigation for the Carnegie-Mellon Navlab | |
| Gao et al. | Review of wheeled mobile robots’ navigation problems and application prospects in agriculture | |
| Huang et al. | Visual odometry and mapping for autonomous flight using an RGB-D camera | |
| Weiss | Vision based navigation for micro helicopters | |
| Facchinetti et al. | Self-Positioning Robot Navigation Using Ceiling Images Sequences | |
| CN107092264A (en) | Towards the service robot autonomous navigation and automatic recharging method of bank's hall environment | |
| CN106840148A (en) | Wearable positioning and path guide method based on binocular camera under outdoor work environment | |
| CN106406338A (en) | Omnidirectional mobile robot autonomous navigation apparatus and method based on laser range finder | |
| Sales et al. | Adaptive finite state machine based visual autonomous navigation system | |
| CN106767791A (en) | A kind of inertia/visual combination air navigation aid using the CKF based on particle group optimizing | |
| Holz et al. | Continuous 3D sensing for navigation and SLAM in cluttered and dynamic environments | |
| CN113189613B (en) | Robot positioning method based on particle filtering | |
| CN113116224A (en) | Robot and control method thereof | |
| Burnett et al. | Zeus: A system description of the two‐time winner of the collegiate SAE autodrive competition | |
| CN109655058A (en) | A kind of inertia/Visual intelligent Combinated navigation method | |
| CN115639823A (en) | Terrain sensing and movement control method and system for robot under rugged and undulating terrain | |
| Thorpe et al. | The new generation system for the CMU Navlab | |
| Liu et al. | Dloam: Real-time and robust lidar slam system based on cnn in dynamic urban environments | |
| CN113239072A (en) | Terminal equipment positioning method and related equipment thereof | |
| Bajracharya et al. | Target tracking, approach, and camera handoff for automated instrument placement | |
| Wang et al. | Agv navigation based on apriltags2 auxiliary positioning | |
| Sun et al. | Indoor Li-DAR 3D mapping algorithm with semantic-based registration and optimization | |
| Han et al. | Novel cartographer using an OAK-D smart camera for indoor robots location and navigation | |
| Sethuramasamyraja et al. | Automatic calibration and neural networks for robot guidance |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190702 |