CN111947657A - Mobile robot navigation method suitable for dense bent frame environment - Google Patents
Mobile robot navigation method suitable for dense bent frame environment Download PDFInfo
- Publication number
- CN111947657A CN111947657A CN202010532832.3A CN202010532832A CN111947657A CN 111947657 A CN111947657 A CN 111947657A CN 202010532832 A CN202010532832 A CN 202010532832A CN 111947657 A CN111947657 A CN 111947657A
- Authority
- CN
- China
- Prior art keywords
- path
- robot
- shelving
- dense
- environment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 9
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 3
- 230000008521 reorganization Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 7
- 230000002452 interceptive effect Effects 0.000 description 6
- 238000007689 inspection Methods 0.000 description 5
- 230000003542 behavioural effect Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 238000009751 slip forming Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
Images
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
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Manipulator (AREA)
Abstract
The invention relates to a mobile robot navigation method suitable for a dense bent environment. Belongs to the field of mobile robots, and specifically comprises the following steps: (1) acquiring a dense bent frame environment map; (2) positioning the dense bent frame based on a rectangular detection method; (3) constructing a behavior tree selection path mode; (4) and generating a planning path between path reference points by using an A-algorithm, tracking the path by using a prediction control method based on a robot kinematics model to realize navigation, and providing reference by using linear characteristics extracted from local laser observation information under the condition of inaccurate positioning in a long and narrow bent channel. The method realizes the optimal division of the working path of the robot in the dense shelving environment, eliminates errors caused by manually setting the landmark points, can complete the navigation of various mobile robots in the narrow shelving environment compared with the prior art, and is suitable for different cruising speeds.
Description
Technical Field
The invention belongs to the field of mobile robots, and particularly relates to a mobile robot navigation method suitable for a dense bent environment.
Background
When the mobile robot is applied to intensive storage environments such as libraries, supermarkets, warehouses and the like with densely and orderly arranged bookshelves/goods shelves, various tasks such as checking, routing inspection and the like need to be executed, path planning and control on the robot are needed to be realized when the tasks are completed, navigation of the mobile robot under different tasks is realized, and the robot is accurately controlled to move in a long and narrow shelving channel.
At present, most robots working in an intensive shelving environment are based on a fixed path tracking method, for example, in a patent document, "book on shelf automatic inspection robot based on image recognition technology" (application number CN201010617824.5), a camera/photosensitive device is used for acquiring mark information of a robot running line pasted on the ground for line inspection, so that the robot running track tracking method has the advantage of stable operation, but the environment needs to be modified, the robot running track needs auxiliary measures to return to the original track again after being separated from the track line, and the robot running track is dead, which is often not a time-optimal or path-shortest scheme. In the patent document, "a library robot positioning control method" (application number CN201810402240.2), positioning and navigation are realized by establishing a fuzzy inference rule using RFID tags as landmark points, in the method, a large number of RFID tags need to be arranged at the edges of shelves, the correspondence between the tags and the positions needs to be established manually and accurately, and the shelves steel material interferes with the electromagnetic environment, so that the error is large. In recent years, positioning navigation of a robot is realized based on a SLAM method, for example, in patent document "a laser automatic navigation library inventory robot" (application number CN201910987548.2), after a map is obtained, multiple navigation waypoints need to be manually set, time is consumed, inference combination is not performed on relationships between the waypoints, the risk of navigation failure at a certain target point exists, and optimization and improvement are not performed according to environmental characteristics. How to realize the division and the tracking control of the required working path is an urgent problem to be solved in the navigation of the mobile robot in the dense shelving environment.
Disclosure of Invention
Aiming at the problems, the invention provides a mobile robot navigation method suitable for a dense bent frame environment.
The technical scheme of the invention is as follows: a mobile robot navigation method suitable for a dense bent environment comprises the following operation steps:
step (1.1), obtaining a dense bent frame environment map;
step (1.2), positioning the dense bent frame based on a rectangular detection method, and generating a reference path point;
step (1.3), constructing a behavior tree reorganization path reference point sequence and generating a global planning path;
and (1.4) controlling the robot to track the planned path based on the model prediction control method.
Further, in the step (1.2), the dense bent refers to: displaying a regular arrangement of rectangles in an environment map, determining the positions of four corners of the rectangles by a rectangle detection method, generating a reference path point by combining the minimum radius and the safe distance of a robot kinematic model, and using the reference path point
And (3) indicating the positions of an upper left path reference point, an upper right path reference point, a lower right path reference point and a lower left path reference point based on the frame number i, wherein max is the maximum number of the detected rectangles.
Further, the step (1.3) is specifically that a behavior tree is constructed, different robot kinematics models are selected, sequence combinations of reference path points are determined, switching of different path modes is achieved, and the following six modes are preset, (a) all shelving patrolling paths in sequence, (b) shelving patrolling paths on two sides of a designated shelving, (c) a single side shelving patrolling path is designated by the designated shelving, (d) shelving patrolling paths between two shelving, (e) patrolling paths are performed around the peripheral area of a dense shelving, and (f) a plurality of shelving patrolling paths are continuously formed from the designated shelving.
Further, the robot kinematics model constructed comprises a two-wheel differential mobile robot model, a three-wheel omnidirectional wheel kinematics model and a four-wheel Mecanum wheel omnidirectional kinematics model.
Furthermore, the sequence combination of the reference path points is determined by the position and the path mode of the robot together, and the position P of the robot is determined firstlyrobot(x, y) relative to the position of the dense bent, the direction of the bent, andthe order of the sequence of reference path points is selected and, secondly, the shelving reference path points are involved in the determination of the path pattern.
Further, the step (1.4) is specifically that a model predictive control method is used for controlling the robot to track the global planned path, and the reference path position and angle are obtained by sampling from the planned path; when the position of the robot is in a dense shelving area, extracting a straight line characteristic to represent a shelving edge by using local laser observation information, switching a reference angle into an angle deviation between the direction of the straight line characteristic and the motion direction of the robot, artificially setting a position x error to be 0 under a robot coordinate system, and setting a position y error to be a difference value between an expected edgewise motion distance and the distance of the straight line characteristic. The prediction model in the model prediction control method is a robot kinematics model.
The invention has the beneficial effects that: the invention realizes the optimal division of the working path of the mobile robot in the intensive shelving environment, eliminates the error caused by manually setting the landmark points, provides various working paths meeting tasks such as checking, patrol and the like, can complete the navigation of various mobile robots in the narrow shelving environment compared with the prior art, and is suitable for different cruising speeds.
Drawings
FIG. 1 is a flow chart of the architecture of the present invention;
FIG. 2 is a schematic diagram of the construction of a behavior tree for work path partitioning in the present invention;
FIG. 3 is a schematic block diagram of a control system based on a model prediction method according to the present invention;
FIG. 4 is a schematic diagram of a global planned path generated in six path modes according to the present invention;
FIG. 5 is a comparison of the smoothed planned path using a Bezier curve in the present invention;
fig. 6 is a schematic diagram of a preferred mounting location for a twin laser according to the present invention.
Detailed Description
In order to more clearly illustrate the technical solution of the present invention, the present invention will be further described below; obviously, the following description is only a part of the embodiments, and it is obvious for a person skilled in the art to apply the technical solutions of the present invention to other similar situations without creative efforts; in order to more clearly illustrate the technical solution of the present invention, the following detailed description is made with reference to the accompanying drawings:
as shown in the attached figure 1, the mobile robot navigation method suitable for the dense shelving environment is realized by the following steps,
(1) acquiring a dense bent frame environment map;
(2) positioning the dense bent frame based on a rectangular detection method to generate a reference path point; the method specifically comprises the following steps: displaying a regular arrangement of rectangles in an environment map, determining the positions of four corners of the rectangles by a rectangle detection method, generating a reference path point by combining the minimum radius and the safe distance of a robot kinematic model, and using the reference path point
Indicating the positions of an upper left path reference point, an upper right path reference point, a lower right path reference point and a lower left path reference point based on the number i of the row frames, wherein max is the maximum number of the detected rectangles;
(3) acquiring a user interaction instruction, dividing a path mode, constructing a behavior tree recombination path reference point sequence, and generating a global planning path; constructing a behavior tree, selecting different robot kinematics models, determining sequence combinations of reference path points, realizing switching of different path modes, and presetting six modes, wherein the six modes comprise (a) all shelving sequential patrol paths, (b) appointing shelving paths on two sides of a shelving, (c) appointing a shelving appointing one-side patrol path, (d) patrol paths between two shelving, (e) patrol paths around the peripheral area of a dense shelving, and (f) a plurality of shelving patrol paths from the appointed shelving;
wherein the default direction of the frame patrol of the six path modes is Pi(x1,y1)、Pi(x4,y4)、Pi(x3,y3)、Pi(x2,y2) At the position P of the robotrobot(x, y) relative toThe position of the dense bent frame meets the following conditions:
if the condition (1) is met, the direction of the frame is opposite, and the frame is replaced by Pi(x4,y4)、Pi(x1,y1)、Pi(x2,y2)、Pi(x3,y3)。
When the position P of the robotrobot(x, y) the position relative to the dense bent satisfies the following condition:
and (5) recording as a condition (2), reordering the reference path point sequence in a reverse order.
Each path mode starts from a sequencer node, a path reference point sequence is obtained according to path mode parameters, a left sub-tree is a condition judgment node, a reference path point sequence loads a mark, a right sub-tree is a selector node, the selection condition is whether a kinematics model is complete constraint or not, if so, the left sub-tree is executed, an A algorithm is sequentially used from the starting position of the robot to the reference path point to generate a global planning path, otherwise, the right sub-tree is executed, the A algorithm is sequentially used from the starting position of the robot to the reference path point sequence, a Bezier curve smoothing path is used before and after a path control point to generate a global planning path;
(a) all bent frames sequentially patrol the frame path, which is a default path mode under the state without specified parameters, and the reference path point sequence group is a starting point and P1(x1,y1)、P1(x4,y4)、P1(x3,y3)、P1(x2,y2)、……、Pi(x1,y1)、Pi(x4,y4)、Pi(x3,y3)、Pi(x2,y2)、……、Pmax(x1,y1)、Pmax(x4,y4)、Pmax(x3,y3)、Pmax(x2,y2)；
(b) In the routing path on two sides of the appointed bent, a user needs to appoint a parameter as an appointed bent serial number i, i is more than or equal to 1 and less than or equal to max, a reference path point sequence is set as a starting point and Pi(x1,y1)、Pi(x4,y4)、Pi(x3,y3)、Pi(x2,y2)；
(c) Appointing a bent frame to appoint a unilateral frame patrol path, wherein the user needs to appoint parameters including an appointed bent frame serial number i and a bent frame left and right side, i is more than or equal to 1 and less than or equal to max, a left _ side flag bit is 1 to represent the left side, and 0 to represent the right side, the reference path point serial number is a starting point, Pi(x1,y1)、Pi(x4,y4) Or starting point, Pi(x2,y2)、Pi(x3,y3)；
(d) A user needs to designate a parameter as a designated shelving serial number i, i is more than or equal to 1 and less than or equal to max-1, a reference path point sequence is set as a starting point and Pi(x2,y2)、Pi(x3,y3)、Pi+1(x4,y4)、Pi+1(x1,y1)；
(e) And carrying out patrol route around the peripheral area of the dense shelving, wherein the user can specify parameters as serial numbers i and j of the specified shelving, i is more than or equal to 1 and j is less than or equal to max, the default parameter is that i is 1, j is max, the reference route point sequence is set as a starting point and P is set as the reference route point sequencei(x1,y1)、Pi(x4,y4)、Pj(x3,y3)、Pj(x2,y2)、Pi(x1,y1)；
(f) A plurality of continuous shelving rack inspection paths from the designated bookshelf, wherein the user needs to designate parameters of a designated shelving serial number i and the number n, i is more than or equal to 1 and less than or equal to max, n is more than 0 and less than or equal to max-i, and the reference path point sequenceColumn set is the starting point, Pi(x1,y1)、Pi(x4,y4)、Pi(x3,y3)、Pi(x2,y2)、Pi+1(x1,y1)、Pi+1(x4,y4)、Pi+1(x3,y3)、Pi+1(x2,y2)、……、Pi+n(x1,y1)、Pi+n(x4,y4)、Pi+n(x3,y3)、Pi+n(x2,y2)；
The robot kinematics model comprises a two-wheel differential mobile robot model, a three-wheel omnidirectional wheel kinematics model and a four-wheel Mecanum wheel omnidirectional kinematics model; the sequence combination of the reference path points is determined by the position and the path mode of the robot together, and the position P of the robot is determined firstlyrobot(x, y) selecting the frame patrol direction and the sequence of the reference path point sequence relative to the position relation of the dense frames, and determining the reference path point of the frame in combination with the path mode;
(4) and finally, controlling the robot to track the planned path based on a model prediction control method: specifically, a model prediction control method is used for controlling the robot to track a global planned path, a prediction model is a robot kinematics model, and reference path positions and angles are obtained by sampling from the planned path; when the position of the robot is within the dense bent area, i.e. the position of the robot is within the area defined by P1(x1,y1)、P1(x4,y4)、Pmax(x3,y3)、Pmax(x2,y2) When the rectangular area is formed, local laser observation information is used for extracting straight line characteristics to represent the edge of the bent frame, the reference angle is switched into the angle deviation between the direction of the straight line characteristics and the motion direction of the robot, the position x error is set to be 0 artificially under the coordinate system of the robot, and the position y error is set to be the difference value between the expected edge motion distance and the distance of the straight line characteristics. The prediction model in the model prediction control method is a robot kinematics model.
Setting a robot two-wheel differential mobile robot kinematic model, setting a starting point at (8, 92), setting the minimum radius r _ min of the robot kinematic model to be 3, setting the safe distance safe _ distance to be 1, and providing robot position information by using an adaptive Monte Carlo positioning algorithm; firstly, establishing an environment grid map by a mapping algorithm, realizing rectangle detection by using an opencv library, carrying out edge detection on the grid map by using a canny operator, then finding out a contour by using a findContours function, carrying out polygon fitting by using contour points, further judging the maximum cosine of an angle between two adjacent sides under the condition that the contour is a convex quadrangle and the area is within a threshold value range, obtaining a rectangle when the maximum cosine is less than 0.3, taking the coordinates of four corner points of the rectangle under an image coordinate system, converting the coordinates under the image coordinate system into a map coordinate system, extending the length of a robot to be 4 in the same direction to obtain a reference path point sequence,
indicating the positions of an upper left path reference point, an upper right path reference point, a lower right path reference point and a lower left path reference point based on the number i of the row frames, wherein the max of the detected rectangles is 6; determining the position P of the robotrobot(x, y) the position relative to the dense bent does not satisfy the condition (1), and the direction of the bent is Pi(x1,y1)、Pi(x4,y4)、Pi(x3,y3)、Pi(x2,y2) Also, the condition (2) is not satisfied, and the sequence order of the reference path points is not changed.
The behavioral tree diagram shown in fig. 2 completes division of a working path, starting from a root node, a left sub-tree of the behavioral tree diagram is a parallelizer node, a rectangular detection is executed in parallel to obtain an action node of a dense shelving reference path point and obtain an action node of a robot kinematics model, a right sub-tree of the behavioral tree diagram is a selector node for constructing the working path, user interaction parameters are obtained, and a corresponding path mode is entered, wherein the path mode comprises six types, namely, (a) sequentially patrolling paths of all shelving, (b) appointing shelving paths on two sides of shelving, (c) appointing shelving to appoint a single-side shelving path, (d) patrolling paths between shelving, (e) patrolling paths around a peripheral area of the dense shelving, and (f) continuously patrolling paths of a plurality of shelving from the appointed shelving.
After the global planned path is generated, a model prediction control-based method is adopted to control the robot to track the planned path, and as shown in the attached figure 3, a control flow chart of the whole system is described; when in the dense bent area, i.e. the self-positioning position of the robot is in the area indicated by P1(x1,y1)、P1(x4,y4)、Pmax(x3,y3)、Pmax(x2,y2) When the rectangular area is formed, local laser observation information is used for extracting straight line characteristics to represent the edge of the bent frame, the reference angle is switched into the angle deviation between the direction of the straight line characteristics and the motion direction of the robot, the position x error is set to be 0 artificially under the coordinate system of the robot, and the position y error is set to be the difference value between the expected edge motion distance and the distance of the straight line characteristics.
The kinematics model of the robot is taken as a prediction model, the differential mobile robot is taken as an example,
wherein L is the wheelbase between two wheels; the state space of the model predictive control system is defined as [ x y theta upsilope re]Respectively representing x coordinate, y coordinate, angle, linear velocity, position error and angle error of the robot position, wherein the position error is represented by the error of a cubic polynomial fitting planning path curve, the reference angle value is obtained by solving the slope of a connecting line of two points and then solving arctan of the connecting line, and when the robot position is in a dense racking area, namely the robot position is in a P-th-order arrangement area1(x1,y1)、P1(x4,y4)、Pmax(x3,y3)、Pmax(x2,y2) When the rectangular area is formed, local laser observation information is used for extracting linear characteristics to represent the edge of the bent frame, the reference angle is switched into the angle deviation between the direction of the linear characteristics and the motion direction of the robot, and the reference angle is positioned under the coordinate system of the robotThe x error is artificially set to be 0, the position y error is set to be the difference value of the expected edgewise motion distance and the distance of the straight line characteristic, and the input quantity of the system is defined as [ nu w [ ]]And respectively represent linear and angular velocities.
The three-wheel omnidirectional wheel mobile robot has the kinematics model that,
wherein L is the distance from the center of the triangle formed by the three omni wheels to each wheel.
The four Mecanum wheel omnidirectional mobile robot kinematic models which are arranged orthogonally to each other are as follows,
wherein a is the distance from the center of the robot to the Mecanum wheels on the same side, and b is the distance from the center of the robot to the front ends of the Mecanum wheels.
The user selects (a) all shelving sequential patrol paths on the interactive interface, and as shown in fig. 4(a), the planned path generated by the robot in the path mode is shown, and the reference path point sequence group is a starting point and P1(x1,y1)、P1(x4,y4)、P1(x3,y3)、P1(x2,y2)、P2(x1,y1)、P2(x4,y4)、P2(x3,y3)、P2(x2,y2)、P3(x1,y1)、P3(x4,y4)、P3(x3,y3)、P3(x2,y2)、P4(x1,y1)、P4(x4,y4)、P4(x3,y3)、P4(x2,y2)、P5(x1,y1)、P5(x4,y4)、P5(x3,y3)、P5(x2,y2)、P6(x1,y1)、P6(x4,y4)、P6(x3,y3)、P6(x2,y2)。
The user selects a rack patrol path on two sides of the designated rack on the interactive interface, the designated parameter is designated as a designated rack sequence 2, as shown in fig. 4(b), the planned path generated by the robot in the path mode is shown, and the reference path point sequence is set as a starting point and P2(x1,y1)、P2(x4,y4)、P2(x3,y3)、P2(x2,y2)。
The user selects a designated rack designated one-side rack inspection path on the interactive interface, the designated parameter is designated rack serial number 4, the left _ side flag bit is 1, as shown in fig. 4(c), the planned path generated by the robot in the path mode is shown, the reference path point serial number is set as a starting point, and P is set as a reference path point serial number4(x1,y1)、P4(x4,y4)。
The user selects a rack walking path between two racks on the interactive interface, the designated parameter is designated as a designated rack serial number 4, as shown in fig. 4(d), the planned path generated by the robot in the path mode is shown, the reference path point serial number is set as a starting point, P4(x2,y2)、P4(x3,y3)、P5(x4,y4)、P5(x1,y1)。
The user selects a patrol path around the peripheral area of the dense bent frame on the interactive interface, does not specify parameters, uses the default parameters i equal to 1 and j equal to max equal to 6, and as shown in fig. 4(e), the planned path generated by the robot in the path mode is shown, the reference path point sequence is set as a starting point and P is set as a reference path point sequence1(x1,y1)、P1(x4,y4)、P6(x3,y3)、P6(x2,y2)、P1(x1,y1)。
User' sSelecting a plurality of continuous shelving polling paths from the designated bookshelf on the interactive interface, designating a shelving number 3 and a quantity 2, and as shown in fig. 4(f), selecting a planned path generated by the robot in the path mode, wherein the reference path point sequence is a starting point and P3(x1,y1)、P3(x4,y4)、P3(x3,y3)、P3(x2,y2)、P4(x1,y1)、P4(x4,y4)、P4(x3,y3)、P4(x2,y2)、P5(x1,y1)、P5(x4,y4)、P5(x3,y3)、P5(x2,y2)。
When there are two or more dense shelving areas, regarding the areas with close and regular spatial distribution as an area, merging shelving path point information, as shown in fig. 5 (a); processing the dense shelving areas which are regarded as independent dense shelving areas with far spatial distribution according to independent areas, as shown in fig. 5(b), after generating a path in the first upper area, taking the last path point as a starting point of the robot, and planning the path in the second lower area; meanwhile, the planned path of the fully constrained mobile robot is shown in fig. 5 (a).
The above embodiment example uses a laser sensor of 360 degree range; more generally, due to the limitation of appearance packaging, the range of a single laser installed on the robot does not meet 360 degrees, so that the environmental information around the robot can be collected only by the double lasers before and after installation, the double laser data is cut and unified to a coordinate system with the robot as a center, see fig. 6, the distance from the intersection point C1/C2 of the double laser data to the robot center needs to meet the condition that the distance is more than or equal to the minimum radius of a robot kinematic model, and thus, good and stable continuous straight line features can be extracted for feedback guidance of the motion of the robot.
Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of embodiments of the present invention; other variations are possible within the scope of the invention; thus, by way of example, and not limitation, alternative configurations of embodiments of the invention may be considered consistent with the teachings of the present invention; accordingly, the embodiments of the invention are not limited to the embodiments explicitly described and depicted.
Claims (6)
1. A mobile robot navigation method suitable for a dense bent frame environment is characterized in that: the operation steps are as follows:
step (1.1), obtaining a dense bent frame environment map;
step (1.2), positioning the dense bent frame based on a rectangular detection method, and generating a reference path point;
step (1.3), constructing a behavior tree reorganization path reference point sequence and generating a global planning path;
and (1.4) controlling the robot to track the planned path based on the model prediction control method.
2. The method for navigating the mobile robot suitable for the dense shelving environment as claimed in claim 1, wherein: in the step (1.2), the dense bent refers to: displaying a regular arrangement of rectangles in an environment map, determining the positions of four corners of the rectangles by a rectangle detection method, generating a reference path point by combining the minimum radius and the safe distance of a robot kinematic model, and using the reference path point
And (3) indicating the positions of an upper left path reference point, an upper right path reference point, a lower right path reference point and a lower left path reference point based on the frame number i, wherein max is the maximum number of the detected rectangles.
3. The method for navigating the mobile robot suitable for the dense shelving environment as claimed in claim 1, wherein: the step (1.3) is specifically that a behavior tree is constructed, different robot kinematics models are selected, sequence combinations of reference path points are determined, switching of different path modes is achieved, and the following six modes are preset, (a) all shelving sequential patrol paths, (b) shelf patrol paths on two sides of a designated shelving, (c) a shelf patrol path on one side of the designated shelving, (d) shelf patrol paths between two shelving, (e) patrol paths are conducted around the peripheral area of the dense shelving, and (f) a plurality of shelving patrol paths are conducted continuously from the designated bookshelf.
4. The method for navigating the mobile robot suitable for the dense shelving environment as claimed in claim 3, wherein: the robot kinematics model constructed comprises a two-wheel differential mobile robot model, a three-wheel omnidirectional wheel kinematics model and a four-wheel Mecanum wheel omnidirectional kinematics model.
5. The method for navigating the mobile robot suitable for the dense shelving environment as claimed in claim 3, wherein: the sequence combination of the reference path points is determined by the position and the path mode of the robot together, and the position P of the robot is determined firstlyrobot(x, y) relative to the position relationship of the dense shelving, selecting the shelving direction and the sequence of the reference path point sequence, and determining the reference path point of the shelving according to the path mode.
6. The method for navigating the mobile robot suitable for the dense shelving environment as claimed in claim 1, wherein: the step (1.4) is specifically that a model predictive control method is used for controlling the robot to track the global planned path, and the reference path position and angle are obtained by sampling from the planned path; when the position of the robot is in a dense shelving area, extracting a straight line characteristic to represent a shelving edge by using local laser observation information, switching a reference angle into an angle deviation between the direction of the straight line characteristic and the motion direction of the robot, artificially setting a position x error to be 0 under a robot coordinate system, and setting a position y error to be a difference value between an expected edgewise motion distance and the distance of the straight line characteristic. The prediction model in the model prediction control method is a robot kinematics model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010532832.3A CN111947657B (en) | 2020-06-12 | 2020-06-12 | Mobile robot navigation method suitable for compact shelving environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010532832.3A CN111947657B (en) | 2020-06-12 | 2020-06-12 | Mobile robot navigation method suitable for compact shelving environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111947657A true CN111947657A (en) | 2020-11-17 |
| CN111947657B CN111947657B (en) | 2024-04-19 |
Family
ID=73337102
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010532832.3A Active CN111947657B (en) | 2020-06-12 | 2020-06-12 | Mobile robot navigation method suitable for compact shelving environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111947657B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914068A (en) * | 2013-01-07 | 2014-07-09 | 中国人民解放军第二炮兵工程大学 | Service robot autonomous navigation method based on raster maps |
| CN107167141A (en) * | 2017-06-15 | 2017-09-15 | 同济大学 | Robot autonomous navigation system based on double line laser radars |
| CN108960364A (en) * | 2018-06-25 | 2018-12-07 | 天津大学 | A kind of books are made an inventory the checking method of robot |
| US20190220043A1 (en) * | 2019-03-26 | 2019-07-18 | Intel Corporation | Methods and apparatus to facilitate autonomous navigation of robotic devices |
| WO2019154119A1 (en) * | 2018-02-06 | 2019-08-15 | 广州科语机器人有限公司 | Map creation method for mobile robot based on laser ranging sensor |
| CN110744559A (en) * | 2019-10-17 | 2020-02-04 | 上海飒智智能科技有限公司 | Laser automatic navigation library inventory robot |
-
2020
- 2020-06-12 CN CN202010532832.3A patent/CN111947657B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914068A (en) * | 2013-01-07 | 2014-07-09 | 中国人民解放军第二炮兵工程大学 | Service robot autonomous navigation method based on raster maps |
| CN107167141A (en) * | 2017-06-15 | 2017-09-15 | 同济大学 | Robot autonomous navigation system based on double line laser radars |
| WO2019154119A1 (en) * | 2018-02-06 | 2019-08-15 | 广州科语机器人有限公司 | Map creation method for mobile robot based on laser ranging sensor |
| CN108960364A (en) * | 2018-06-25 | 2018-12-07 | 天津大学 | A kind of books are made an inventory the checking method of robot |
| US20190220043A1 (en) * | 2019-03-26 | 2019-07-18 | Intel Corporation | Methods and apparatus to facilitate autonomous navigation of robotic devices |
| CN110744559A (en) * | 2019-10-17 | 2020-02-04 | 上海飒智智能科技有限公司 | Laser automatic navigation library inventory robot |
Non-Patent Citations (4)
| Title |
|---|
| LI RENJUN ET AL.: "AuRoSS: an Autonomous Robotic Shelf Scanning System", 《2015 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)》, pages 6100 - 6105 * |
| LI SUIQUN ET AL.: "Study on the substation inspection robot navigation system based on Fuzzy-PID control", 《PROCEEDINGS OF THE 2015 4TH INTERNATIONAL CONFERENCE ON COMPUTER, MECHATRONICS, CONTROL AND ELECTRONIC ENGINEERING (ICCMCEE 2015)》, vol. 37, pages 1385 - 1389 * |
| 赵健等: "基于典型栅格地图的代价地图改进方法", 《机械与电子》, vol. 36, no. 12, pages 73 - 80 * |
| 赵志光;: "图书馆智能化机器人三项关键技术的研究与探讨", 山东图书馆学刊, no. 05, pages 69 - 72 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111947657B (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qin et al. | Autonomous exploration and mapping system using heterogeneous UAVs and UGVs in GPS-denied environments | |
| CN111103887B (en) | Multi-sensor-based multi-mobile-robot scheduling system design method | |
| JP2677600B2 (en) | Method to determine collision-free path in three-dimensional space | |
| Yahja et al. | An efficient on-line path planner for outdoor mobile robots | |
| Bogdan Rusu et al. | Leaving Flatland: Efficient real‐time three‐dimensional perception and motion planning | |
| Tariq et al. | Construction of cellular automata over hexagonal and triangular tessellations for path planning of multi-robots | |
| Bopardikar et al. | Dynamic vehicle routing for translating demands: Stability analysis and receding-horizon policies | |
| Guney et al. | Dynamic prioritized motion coordination of multi-AGV systems | |
| Yang et al. | A novel path planning algorithm for warehouse robots based on a two-dimensional grid model | |
| CN112633590B (en) | Intelligent warehousing method and system for four-way shuttle | |
| Sharma et al. | Coordination of multi-robot path planning for warehouse application using smart approach for identifying destinations | |
| Guo et al. | Intelligent path planning for automated guided vehicles system based on topological map | |
| Dharmasiri et al. | Novel implementation of multiple automated ground vehicles traffic real time control algorithm for warehouse operations: djikstra approach | |
| Pivtoraiko et al. | Differentially constrained motion replanning using state lattices with graduated fidelity | |
| Jeon et al. | An entry-exit path planner for an autonomous tractor in a paddy field | |
| Klaas et al. | Simulation aided, knowledge based routing for AGVs in a distribution warehouse | |
| Okada et al. | Exploration and observation planning for 3D indoor mapping | |
| CN111947657B (en) | Mobile robot navigation method suitable for compact shelving environment | |
| Nagatani et al. | Sensor-based navigation for car-like mobile robots based on a generalized Voronoi graph | |
| Chen et al. | Simple and efficient algorithm for drone path planning | |
| CN115235483A (en) | Method for constructing grid map, path planning method, processor and device | |
| Meyer et al. | Trajectory-based traveling salesman problem for multirotor UAVs | |
| Dichtl et al. | Robot navigation with polymap, a polygon-based map format | |
| Fleckenstein et al. | Smooth local planning incorporating steering constraints | |
| Gregorić et al. | Autonomous hierarchy creation for computationally feasible near-optimal path planning in large environments |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |