CN114332635B

CN114332635B - Automatic obstacle identification method and system for intelligent transfer robot

Info

Publication number: CN114332635B
Application number: CN202210234548.7A
Authority: CN
Inventors: 张波; 张超; 周晓坤
Original assignee: Innotitan Intelligent Equipment Technology Tianjin Co Ltd
Current assignee: Innotitan Intelligent Equipment Technology Tianjin Co Ltd
Priority date: 2022-03-11
Filing date: 2022-03-11
Publication date: 2022-05-31
Anticipated expiration: 2042-03-11
Also published as: CN114332635A

Abstract

The invention relates to an automatic obstacle identification method and system for an intelligent transfer robot, and belongs to the technical field of obstacle identification. According to the method for automatically identifying the obstacles of the intelligent transfer robot, after a storage space model is constructed, the obstacles in the storage space are classified and stored based on the storage space model, and an obstacle set is obtained; and then, after constructing the barrier set and the storage space model index structure, acquiring the position of the intelligent transfer robot in the storage space in real time, and searching the index structure according to the position to complete the identification of the barrier, thereby solving the problems of complex calculation, incapability of meeting the industrial requirement on precision and the like in the prior art.

Description

Automatic obstacle identification method and system for intelligent transfer robot

Technical Field

The invention relates to the technical field of obstacle identification, in particular to an automatic obstacle identification method and system for an intelligent transfer robot in industrial warehousing.

Background

The rapid development of science and technology and the continuous progress of society have always stimulated the overall reform of the working modes of modern industries such as e-commerce and logistics. Under the stimulation of the development of the science and technology, due to the influence of a series of factors such as working efficiency, intensity, environment, capital and the like, the traditional manual operation mode is gradually changed by modern industry, and the modern industry is further shifted to the mode of replacing manual work by advanced industrial robots to realize the overall automation of the industrial development. With the large-scale application of industrial robots, the robot technology faces more and more problems in practical industrial application, different industrial robots need to be changed correspondingly according to the characteristics of the field working environment, and the dynamic change of the field working environment also brings challenges to the development of the robot application technology. Obstacle identification technology in complex working environment has been a key and difficult point in industrial robot application technology. The traditional technology mainly achieves the purpose of obstacle identification based on means such as deep learning and visual sensor detection, but has the defects of complex calculation, incapability of meeting industrial requirements in accuracy and the like on the whole.

Disclosure of Invention

In order to solve the problems in the prior art, the invention provides an intelligent transfer robot obstacle automatic identification method and system for industrial warehousing.

In order to achieve the purpose, the invention provides the following scheme:

an intelligent transfer robot obstacle automatic identification method comprises the following steps:

constructing a storage space model;

classifying and storing the obstacles in the storage space based on the storage space model to obtain an obstacle set; the obstacle sets comprise a first obstacle set, a second obstacle set and a third obstacle set;

constructing an obstacle set and a storage space model index structure; the index structure comprises leaf elements, branch elements and root elements;

and acquiring the position of the intelligent transfer robot in the storage space in real time, and searching the index structure according to the position to finish obstacle identification.

Preferably, the constructing a warehouse space model specifically includes:

modeling the warehouse space as a rectangle A = (x0, y0, x1, y1) in two-dimensional Euclidean space, wherein (x0, y0) is the first vertex coordinate of the rectangle, and (x1, y1) is the second vertex coordinate of the rectangle; the first vertex is a diagonal vertex of the second vertex.

Preferably, the classifying and storing the obstacles in the storage space based on the storage space model to obtain an obstacle set specifically includes:

modeling each article in the storage space into a minimum bounding rectangle to form a barrier set;

judging the type of each obstacle in the obstacle set;

classifying and storing all obstacles in the obstacle set according to the types of the obstacles to obtain a first obstacle set, a second obstacle set and a third obstacle set.

Preferably, the constructing of the obstacle set and the storage space model index structure specifically includes:

constructing an external circle of the storage space model by taking a central point of the storage space model as a circle center, and determining the radius of the external circle;

equally dividing a direction space of [0 DEG and 360 DEG ] in the circumscribed circle into u direction intervals according to a preset direction interval by taking the positive direction of an X axis as a reference direction of 0 DEG and taking the center of the circumscribed circle as a base point to obtain a first divided circumscribed circle; numbering the intervals in each direction in sequence from 0 degree to 360 degrees in an increasing manner to obtain a direction interval set;

dividing the radius in the first division circumscribed circle into v interval segments according to a preset segment pitch interval to obtain a second division circumscribed circle; the second circumscribed circle comprises u x v storage space sub-pieces; the storage space sub-pieces in each direction interval form an interval sub-piece set, and the interval sub-piece set forms a space sub-piece set;

constructing an interval shard set corresponding to each direction interval in the space shard set into a leaf element, and setting leaf element attributes to obtain a leaf element set; the leaf element attributes include: a hierarchy attribute, an overlay space attribute, a child attribute, and a parent attribute;

sequentially dividing the u-th leaf element of the 1 st layer in the leaf element set into n groups from the reference direction of 0 degrees, constructing a first branch element for each group, and setting the attribute of the first branch element to obtain a first branch element set;

judging whether the number of the branch elements of the jth layer in the first branch element set is 2 or not to obtain a first judgment result;

if the first judgment result shows that the number of the branch elements of the j-th layer in the first branch element set is not 2, sequentially dividing x first branch elements of the j + 1-th layer in the first branch element set into w groups from the reference direction of 0 degrees, constructing a second branch element for each group, and setting the attribute of the second branch element to obtain a second branch element set;

if the first judgment result is that the number of the branch elements at the jth layer in the first branch element set is 2, constructing the circumscribed circle as a root element, and setting a root element attribute; the leaf element attribute, the first branch element attribute, the second branch element attribute, and the root element attribute all include: a hierarchy attribute, an overlay space attribute, a child attribute, and a parent attribute.

Preferably, the acquiring a position of the intelligent transfer robot in the warehousing space in real time and searching the index structure according to the position to complete obstacle identification specifically includes:

acquiring the positions of all the carrying robots in the storage space according to a preset time interval to obtain a robot position set, and acquiring the positions of the obstacles in the second type of obstacle set to obtain an obstacle position set;

determining the distance between the position of the transfer robot at the first moment and the position of the transfer robot at the second moment to obtain a first distance;

judging whether the first distance is greater than a distance threshold value, if so, storing the position of the transfer robot at the second moment into a new robot position set;

determining the distance between the position of the obstacle at the first moment and the position of the obstacle at the second moment to obtain a second distance;

judging whether the second distance is greater than a distance threshold value, if so, storing the position of the obstacle at the second moment into a new obstacle position set;

updating the index structure from the root element according to the position of the first-time transfer robot, the position of the second-time transfer robot, the position of the first-time obstacle and the position of the second-time obstacle;

constructing an intermediate set, and initializing the intermediate set by using a root element of an index structure;

acquiring a terminal point of a road section where the position of the transfer robot is located at a second moment, acquiring a passing road section of the transfer robot, and determining a direction range of the passing road section relative to the center of the circumscribed circle;

acquiring an index structure contained in the direction range, and judging the type of the index structure;

when the index structure is a branch element, acquiring child attributes of the branch element to obtain a child attribute set;

acquiring children in the child attribute set and coverage space attributes corresponding to the children;

judging whether the direction range is intersected with a space range in the coverage space attribute corresponding to the child;

if the direction range is intersected with the space range in the coverage space attribute, storing the children in the child attribute set into the intermediate set to obtain a result set;

if the direction range is not intersected with the space range in the coverage space attribute, judging whether the child in the child attribute set is the last child in the child attribute set;

if the child in the child attribute set is not the last child in the child attribute set, returning to execute the step of acquiring the child in the child attribute set and the coverage space attribute corresponding to the child;

if the child in the child attribute set is the last child in the child attribute set, judging whether the intermediate set is empty;

if the intermediate set is not empty, returning to execute the steps of obtaining the index structure contained in the direction range and judging the type of the index structure;

if the intermediate set is empty, updating the passing path of the transfer robot according to the result set;

when the index structure is a leaf element, acquiring the coverage space attribute of the leaf element;

judging whether the direction range intersects with a space range in the coverage space attribute of the leaf element, if so, storing the leaf element into an intermediate set to obtain a result set, and if not, judging whether the intermediate set is empty;

and if the intermediate set is empty, updating the passing path of the transfer robot according to the result set.

Preferably, the updating the index structure from the root element according to the position of the first-time transfer robot, the position of the second-time transfer robot, the position of the first-time obstacle, and the position of the second-time obstacle specifically includes:

traversing the index structure from a root element by taking the position of the transfer robot at the first moment as a target to obtain an interval fragment containing a third type of obstacle set, and recording the interval fragment as a first interval fragment;

removing the position of the transfer robot at the first moment from the set of obstacles of the third type of the first interval slice;

traversing the index structure from the root element by taking the position of the transfer robot at the second moment as a target to obtain interval fragments intersected with the position of the transfer robot at the second moment, and marking the interval fragments as second interval fragments;

storing the position of the transfer robot at the second moment into a third type of obstacle set in the second interval slicing;

traversing the index structure from the root element by taking the position of the obstacle at the first moment as a target to obtain an interval fragment containing a second type of obstacle set, and marking as a third interval fragment;

removing the position of the obstacle at the first time from the second type obstacle set of the third interval slices;

traversing the index structure from the root element by taking the position of the obstacle at the second moment as a target to obtain interval fragments intersected with the position of the obstacle at the second moment, and recording the interval fragments as fourth interval fragments;

and storing the position of the obstacle at the second moment into the second obstacle set of the fourth interval fragment.

Preferably, the updating the passage of the transfer robot according to the result set specifically includes:

sequentially acquiring a branch element in the result set, and acquiring child attributes of the branch element to obtain a second child attribute set;

sequentially acquiring an interval fragment in the second child attribute set, and judging whether the current passing road section of the intelligent transfer robot passes through the interval fragment;

when the current passing road section of the intelligent transfer robot passes through the interval fragment, acquiring the fixed barrier attribute of the interval fragment to obtain a first type of barrier set, and judging whether the first type of barrier set is empty or not;

when the first type of barrier set is empty, acquiring the semi-fixed barrier attribute of the interval fragment to obtain a second type of barrier set, and judging whether the second type of barrier set is empty;

when the second type of barrier set is empty, obtaining the activity barrier attribute of the interval fragment to obtain a third type of barrier set, and judging whether the third type of barrier set is empty;

when the third type of barrier set is empty, judging whether the interval fragment is the last interval fragment in the second child attribute set;

when the interval fragment is the last interval fragment in the second child attribute set, returning to execute the steps of sequentially acquiring one branch element in the result set and acquiring the child attribute of the branch element to obtain a second child attribute set;

when the interval fragment is not the last interval fragment in the second child attribute set, returning to execute the steps of sequentially acquiring one interval fragment in the second child attribute set and judging whether the current passing road section of the intelligent transfer robot passes the interval fragment;

when the first type of obstacle set is not empty, acquiring a first obstacle in the first type of obstacle set, and adjusting the passing path of the intelligent transfer robot according to the position of the first obstacle;

when the second type of obstacle set is not empty, outputting a signal that a second obstacle exists;

and when the third type of obstacle set is not empty, generating a blocking signal of a third obstacle in the third type of obstacle set, and adjusting the passing path of the intelligent transfer robot or generating a standby signal of the intelligent transfer robot according to the position of the third obstacle.

According to the specific embodiment provided by the invention, the invention discloses the following technical effects:

according to the automatic obstacle identification method for the intelligent transfer robot, after a storage space model is constructed, obstacles in a storage space are classified and stored based on the storage space model, and an obstacle set is obtained; and then, after constructing the barrier set and the storage space model index structure, acquiring the position of the intelligent transfer robot in the storage space in real time, and searching the index structure according to the position to complete the identification of the barrier, thereby solving the problems of complex calculation, incapability of meeting the industrial requirement on precision and the like in the prior art.

In addition, the invention also provides an automatic obstacle identification system of the intelligent transfer robot, which comprises a processor and a memory;

the processor is connected with the memory; the memory is stored with a computer software program for implementing the automatic obstacle identification method for the intelligent transfer robot; the processor is used for calling and executing the computer software program.

The technical effect achieved by the intelligent conveying robot obstacle automatic identification system provided by the invention is the same as that achieved by the intelligent conveying robot obstacle automatic identification method, so that the details are not repeated herein.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a flowchart of an obstacle automatic identification method for an intelligent transfer robot according to the present invention;

fig. 2 is a display diagram of a storage space model and various obstacle models according to an embodiment of the present invention;

FIG. 3 is a schematic diagram illustrating the direction range division of the circumscribed circle based on Δ α according to the embodiment of the invention;

FIG. 4 is a schematic diagram of the warehouse space slicing construction using the patient as the standard;

FIG. 5 is a schematic diagram of a storage space and an index structure thereof including obstacles according to an embodiment of the present invention;

fig. 6 is a schematic structural view of an intelligent transfer robot obstacle automatic identification system provided in the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention aims to provide an automatic obstacle identification method and system for an intelligent transfer robot, which can solve the problems that in the prior art, calculation is complex, precision cannot meet industrial requirements and the like.

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

As shown in fig. 1, the method for automatically identifying an obstacle of an intelligent transfer robot according to the present invention includes:

and S1, constructing a storage space model. Specifically, in the two-dimensional euclidean space, the warehouse space is modeled as a rectangle a = (x0, y0, x1, y1), where (x0, y0) is the first vertex coordinate of the rectangle, and (x1, y1) is the second vertex coordinate of the rectangle. The first vertex is a diagonal vertex of the second vertex.

And S2, classifying and storing the obstacles in the storage space based on the storage space model to obtain an obstacle set. The obstacle sets comprise a first type obstacle set, a second type obstacle set and a third type obstacle set. Specifically, the method comprises the following steps:

and S20, modeling each article in the storage space into a minimum enclosing rectangle to form an obstacle set.

And S21, judging the type of each obstacle in the obstacle set.

And S22, classifying and storing all obstacles in the obstacle set according to the obstacle types to obtain a first-class obstacle set, a second-class obstacle set and a third-class obstacle set. The first type of obstacle set is a set of fixed type obstacles, the second type of obstacle set is a set of semi-fixed type obstacles, and the third type of obstacle set is an active type obstacle set.

And S3, constructing an obstacle set and a storage space model index structure. The index structure includes leaf elements, branch elements, and root elements. Specifically, the method comprises the following steps:

and S30, constructing an external circle of the warehouse space model by taking the central point of the warehouse space model as the center of a circle, and determining the radius of the external circle.

And S31, taking the positive direction of the X axis as a 0-degree reference direction, taking the center of the circumscribed circle as a base point, and equally dividing the direction space of [0 degrees and 360 degrees ] in the circumscribed circle into u direction intervals according to a preset direction interval to obtain a first divided circumscribed circle. And numbering is sequentially carried out on each direction interval from 0 degree to 360 degrees in an increasing mode to obtain a direction interval set.

And S32, dividing the radius in the first division circumscribed circle into v interval segments according to a preset segment spacing interval to obtain a second division circumscribed circle. The second circumscribed circle comprises uxv storage space segments. The storage space sub-pieces in each direction interval form an interval sub-piece set, and the interval sub-piece sets form a space sub-piece set.

S33, constructing the interval shard set corresponding to each direction interval in the space shard set into a leaf element, and setting the attribute of the leaf element to obtain the leaf element set. The leaf element attributes include: a hierarchy attribute, an overlay space attribute, a child attribute, and a parent attribute.

And S34, sequentially dividing the u-th leaf element of the layer 1 in the leaf element set into n groups from the reference direction of 0 degree, constructing a first branch element for each group, and setting the attribute of the first branch element to obtain a first branch element set.

S35, judging whether the number of the branch elements of the jth layer in the first branch element set is 2 or not, and obtaining a first judgment result.

S36, if the first judgment result shows that the number of the branch elements of the j-th layer in the first branch element set is not 2, sequentially dividing x first branch elements of the j + 1-th layer in the first branch element set into w groups from the reference direction of 0 degrees, constructing a second branch element for each group, and setting the attribute of the second branch element to obtain a second branch element set.

S37, if the first judgment result is that the number of the branch elements of the jth layer in the first branch element set is 2, constructing the circumscribed circle as a root element, and setting the attribute of the root element. The leaf element attribute, the first branch element attribute, the second branch element attribute and the root element attribute all comprise: a hierarchy attribute, an overlay space attribute, a child attribute, and a parent attribute.

And S4, acquiring the position of the intelligent transfer robot in the warehousing space in real time, and searching an index structure according to the position to finish obstacle identification. Specifically, the method comprises the following steps:

and S40, acquiring the positions of all the transfer robots in the warehousing space according to the preset time interval to obtain a robot position set, and acquiring the positions of the obstacles in the second type of obstacle set to obtain an obstacle position set.

And S41, determining the distance between the position of the transfer robot at the first moment and the position of the transfer robot at the second moment to obtain a first distance.

And S42, judging whether the first distance is larger than the distance threshold value, and if so, storing the position of the transfer robot at the second moment into a new robot position set.

And S43, determining the distance between the position of the obstacle at the first moment and the position of the obstacle at the second moment to obtain a second distance.

And S44, judging whether the second distance is greater than the distance threshold value, and if so, storing the position of the obstacle at the second moment into a new obstacle position set.

And S45, updating the index structure from the root element according to the position of the first-time transfer robot, the position of the second-time transfer robot, the position of the first-time obstacle and the position of the second-time obstacle. The method specifically comprises the following steps:

and S450, traversing the index structure from the root element by taking the position of the transfer robot at the first moment as a target to obtain an interval fragment containing the third type of obstacle set, and recording the interval fragment as the first interval fragment.

And S451, removing the position of the transfer robot at the first moment from the third obstacle set of the first interval slice.

And S452, traversing the index structure from the root element by taking the position of the transfer robot at the second moment as a target, and obtaining interval slices intersecting the position of the transfer robot at the second moment, wherein the interval slices are marked as second interval slices.

And S453, storing the position of the transfer robot at the second moment into the third obstacle set in the second interval slice.

And S454, traversing the index structure from the root element by taking the position of the obstacle at the first moment as a target, obtaining an interval fragment containing the second type of obstacle set from the root element traversal index structure, and marking as a third interval fragment.

And S455, removing the position of the obstacle at the first moment from the second type obstacle set of the third interval slice.

And S456, traversing the index structure from the root element by taking the position of the obstacle at the second moment as a target, obtaining a spacing section intersected with the position of the obstacle at the second moment, and recording as a fourth spacing section.

And S457, storing the position of the obstacle at the second moment into a second obstacle set of a fourth interval slice.

S46, constructing an intermediate set, and initializing the intermediate set by using the root element of the index structure.

And S47, acquiring the end point of the road section where the position of the transfer robot is located at the second moment, acquiring the passing road section of the transfer robot, and determining the direction range of the passing road section relative to the center of the circumscribed circle.

And S48, acquiring the index structure contained in the direction range, and judging the type of the index structure.

S49, when the index structure is a branch element, acquiring the child attributes of the branch element to obtain a child attribute set.

S410, acquiring the child in the child attribute set and the coverage space attribute corresponding to the child.

S411, judging whether the direction range is intersected with the space range in the coverage space attribute corresponding to the child.

S412, if the direction range is intersected with the space range in the coverage space attribute, storing the children in the child attribute set into the middle set to obtain a result set.

S413, if the direction range does not intersect the space range in the coverage space attribute, determining whether the child in the child attribute set is the last child in the child attribute set.

And S414, if the child in the child attribute set is not the last child in the child attribute set, returning to execute the step S410.

S415, if the child in the child attribute set is the last child in the child attribute set, judging whether the middle set is empty.

And S416, if the intermediate set is not empty, returning to the step S48.

And S417, if the intermediate set is empty, updating the passing path of the transfer robot according to the result set.

And S418, when the index structure is the leaf element, acquiring the coverage space attribute of the leaf element.

S419, judging whether the direction range intersects with the space range in the coverage space attribute of the leaf elements, if so, storing the leaf elements into an intermediate set to obtain a result set, and if not, judging whether the intermediate set is empty.

And S420, if the intermediate set is not empty, returning to the step S48.

And S421, if the intermediate set is empty, updating the passing path of the transfer robot according to the result set.

The specific process of updating the passage of the transfer robot based on the result set in step S417 and step S421 includes:

and sequentially acquiring one branch element in the result set, and acquiring the child attributes of the branch element to obtain a second child attribute set.

And sequentially acquiring an interval fragment in the second child attribute set, and judging whether the current passing road section of the intelligent transfer robot passes through the interval fragment.

When the current passing road section of the intelligent transfer robot passes through the interval fragment, acquiring the fixed barrier attribute of the interval fragment to obtain a first type of barrier set, and judging whether the first type of barrier set is empty or not.

And when the first type of obstacle set is empty, acquiring the semi-fixed obstacle attribute of the interval fragment to obtain a second type of obstacle set, and judging whether the second type of obstacle set is empty or not.

And when the second type of obstacle set is empty, obtaining the activity obstacle attribute of the interval fragment to obtain a third type of obstacle set, and judging whether the third type of obstacle set is empty or not.

And when the third type of barrier set is empty, judging whether the interval fragment is the last interval fragment in the second child attribute set.

And when the interval fragment is the last interval fragment in the second child attribute set, returning to execute the steps of sequentially acquiring one branch element in the result set and acquiring the child attributes of the branch element to obtain a second child attribute set.

And when the interval fragment is not the last interval fragment in the second child attribute set, returning to execute the step of sequentially acquiring one interval fragment in the second child attribute set and judging whether the current passing road section of the intelligent transfer robot passes through the interval fragment.

And when the first type of obstacle set is not empty, acquiring a first obstacle in the first type of obstacle set, and adjusting the passing path of the intelligent transfer robot according to the position of the first obstacle.

And when the second type of obstacle set is not empty, outputting a signal that a second obstacle exists.

In addition, the present invention also provides an intelligent transfer robot obstacle automatic identification system, as shown in fig. 6, which includes a processor 120 and a memory 121.

The processor 120 is coupled to the memory 121. The memory 121 stores a computer software program for implementing the above-described obstacle automatic recognition method for the intelligent transfer robot. The processor 120 is used to retrieve and execute computer software programs.

In the following, a practical implementation process of the above-mentioned method for automatically identifying an obstacle of an intelligent transfer robot is described with an example of a hardware implementation, and in a practical application process, the method is not limited to the parameter range selection provided in this embodiment.

The implementation process of the intelligent transfer robot obstacle automatic identification method comprises the following steps:

s1, constructing a barrier model, identifying categories and storing the categories of the whole storage space and various articles contained in the storage space in a classified mode, wherein the specific operation steps are as follows:

s1a, modeling the entire warehouse space as a rectangle a = (x0, y0, x1, y1) in two-dimensional euclidean space, where (x0, y0) and (x1, y1) are coordinates of the lower left vertex and the upper right vertex of the rectangle, respectively, but not limited thereto.

S1b, each item in the storage space is further modeled as a small bounding rectangle, each bounding rectangle is represented by a quadruple formed by coordinates of top left and bottom right points of the bounding rectangle, so that an obstacle set O = { O1, O2,. once, on ' }isformed, wherein n ' is the assumed number of obstacles, and any element oa (1 ≦ a ≦ n ') in the set O represents an obstacle which is represented by the small bounding rectangle of the actual obstacle.

S1c, further performing category identification and classified storage for each obstacle in the obstacle set O, and sequentially removing one obstacle oa from O.

S1d, determining oa as a fixed type obstacle (i.e. the first obstacle), and if the fixed type obstacle is mainly composed of shelves, storing oa in the obstacle set O1, i.e. O1 ← O1 { [ oa }.

S1e, determining oa as a semi-fixed type obstacle (i.e. the second obstacle), where the semi-fixed type obstacle mainly consists of articles that do not move frequently, and storing oa in an obstacle set O2, i.e. O2 ← O2 { oa }.

S1f, determining oa as an activity type obstacle (i.e. a third obstacle), where the activity type obstacle is mainly composed of temporarily placed articles, active robots or workers, and the like, and then storing oa in an obstacle set O3, that is, O3 ← O3 ≈ oa }.

S1g, judging whether O is empty, if O is empty, continuing to execute the following steps, otherwise, jumping to S1 c.

And S1h, returning to the three obstacle sets O1, O2 and O3 and ending.

In the embodiment of the present invention, as shown in fig. 2, the storage space is modeled as a rectangle in a two-dimensional plane space, and various articles contained in the storage space are further modeled and classified and identified, including fixed obstacles, i.e., shelves, and the like, semi-fixed obstacles, i.e., articles whose positions do not change for a long time, and movable obstacles, i.e., obstacles whose positions move, e.g., robots, and the like. Further, the circumscribed circle space is divided into a series of fragments according to a certain rule by constructing an outside circle for the storage space in consideration of the direction and distance factors, and a certain hierarchy construction is carried out on each fragment to organize the whole storage space and the classification obstacles contained in the storage space.

Further, fig. 3-5 illustrate a direction and distance aware spatial index structure constructed for the stocker space and the classification obstacles contained therein. As shown in fig. 3, the whole circumcircle is firstly divided into 16 direction ranges { d1, d 2., d16} with Δ α as the standard, and as shown in fig. 4, the circumcircle divided in the above directions is further divided into 6 concentric rings with Δ d as the standard, and finally 16 × 6 segments are obtained. As shown in fig. 5, the segments are organized layer by layer according to their positions and directions to construct the organization structures of the leaf elements, the branch elements and the final root elements.

S2, constructing an index structure for the whole storage space and various barriers contained in the storage space, and specifically performing the following operation steps:

and S2a, constructing a circumscribed circle = (c, r) of the warehouse space (rectangle, namely the shape described by A) A by taking a center point c = ((x0+ x1)/2, (y1+ y2)/2) of the warehouse space A = (x0, y0, x1, y1) as a center, and calculating the radius of the circumscribed circle as r.

S2b, a direction interval construction unit, equally dividing the [0 DEG and 360 DEG direction space around the circle center c into u = 360/[ alpha ] direction intervals by taking the Alpha as a standard and the positive X axis direction as a 0 DEG reference direction, and sequentially increasing the number of the direction intervals from 0 DEG to 360 DEG to obtain a series of direction intervals D = { D1, D2, ·, du } of 1, 2, 3, ·.

S2c, the warehousing space segment constructing unit divides the radius r into v = r/D interval segments by taking D as a standard, so as to divide the circumscribed circle into v rings, the rings all use c as a circle center, the difference between the inner radius and the outer radius of the rings is Δ D, the operation further divides the direction into v segments at intervals, and the segments S = { S1, S2, Su } and Si = { Si1, Si2,. sj } are sequentially added to the direction interval di ∈ D in an increasing mode and are numbered i1, i2, i3,. sj, iv, and any element sij in the set Si is one segment and is equal to or less than 1 j or equal to v, so that u × v warehousing space segments are obtained, and the segment attribute setting unit executes the following.

The S2d and the tile attribute setting unit set any tile sij in the set S = { S1, S2,. Su }, as a setting range attribute sij, range of any tile sij in the set S = { S1, Si2,. Su }, the space coverage of which is one tuple, the fixed obstacle attribute sij.o1 is a set formed by all fixed obstacle elements intersecting sij.range in O1, the semi-fixed obstacle attribute sij.o2 is a set formed by all semi-fixed obstacle elements intersecting sij.range in O2, and the movable obstacle attribute sij.o3 is a set formed by all movable obstacle elements intersecting sij.range in O3.

S2e, a level 1 leaf element constructing unit, i.e. a leaf element attribute setting module that constructs the sliced set Si corresponding to each directional interval di in the set S = { S1, S2., Su } into one leaf element leaf i (1 ≦ i ≦ u) and performs the following steps thereon, thereby obtaining u leaf elements of leaf = { leaf1, leaf 2., leaf } and each leaf element is continuous in direction.

S2f, a leaf element attribute setting module, a level attribute leaf.lay =1, a coverage space attribute leaf.cover = (c, r, gamma +. alpha) is a fan coverage range formed by leaf with [ gamma, gamma +. delta) as an angle range (gamma is a starting angle, gamma +. delta) as an ending angle) and c as a center and r as a radius, a child attribute is a set formed by v slices of leaf.child = Si = { Si1, Si 2.,. siv }, and a parent attribute leaf.parent is a branch element of which the 2 nd layer can completely cover the leaf.cover.

S2g, a level 2 branch element construction unit, i.e., a branch element attribute setting module that sequentially divides u leaf elements of the level 1 in the set leaf = { leaf1, leaf2,. and leaf } into n groups of L1= { leaf1,. and leaf }, L2= { leaf +1,. and leaf2M },. and Ln = { leaf (n-1) M +1,. and leaf } (M = u/n, M ≦ M, M being a positive integer of the setting), further constructs one branch element brani ″ for each group Li', and performs the following steps, starting from a reference direction of 0 °.

S2h, a branch element attribute setting module, a hierarchical attribute brani ' lay =2, a coverage space attribute brani ' cover = (c, r, a, β) is a sector coverage area formed by brani ' with [ a, β) as an angle range, c as a center, and r as a radius, a child attribute brani ' child = Li is a leaf element not greater than M at level 1 in Li fully covered by brani ' cover, and a parent attribute brani ' parent is a branch element at level 3 capable of fully covering brani ' cover.

S2i, a jth-level branch element constructing unit, i.e., a branch element attribute setting module that sequentially divides x branch elements of a jth-1 level in a set BranchS = { bran1, bran2,. and branx } into w groups B1= { bran1,. and brany }, B2= { brany +1,. and bran2y },. and Bw = { bran (n-1) y +1,. and branx } (w = x/y, y ≦ M, M being a positive integer of the setting), further constructs one branch element brani ″ for each group Bi ″' and performs the following steps, starting from a reference direction of 0 °.

The method comprises the following steps of S2j, a branch element attribute setting module, a hierarchical attribute brani ' lay = j, a coverage space attribute brani ' cover = (c, r, alpha, beta) is a sector coverage area formed by brani taking [ alpha, beta) as an angle range, c as a center and r as a radius, a child attribute brani ' child = Bi is not more than M branch elements at the j-1 th layer of Bi ' completely covered by brani ' cover, and a parent attribute brani ' parent is a branch element or a root element at the j +1 th layer capable of completely covering brani ' cover.

S2k, judging whether the number of the branch elements of the j-th layer is 2, if so, continuing to execute the following steps, otherwise, j +1 and jumping to the step S2 i.

S2l, a highest h-layer root element construction unit, namely a root element attribute setting module which constructs the circumscribed circle into a root element root and executes the following steps for the root element root.

S2m, a root element attribute setting module, a hierarchical attribute root.lay = h, a coverage space attribute root.cover = (c, r) which is a space region covered by a circumscribed circle, a child attribute root.child = { bra 1, bra 2} which is 2 branch elements in the h-1 th layer covered by root.cover, and the following end unit is continuously executed.

And S2n, ending the index structure construction by an ending unit and returning to a root element root.

S3, refreshing the positions of the transfer robots in the storage space, searching the index structure according to the positions of the transfer robots, identifying obstacles and performing subsequent processing, wherein the specific operation steps are as follows:

and S3a, a position refreshing unit, namely refreshing and acquiring the positions of all the transfer robots in the storage space every t1 to obtain a position set Q = { Q1, Q2,. once, qk }, wherein any element qz (z is more than or equal to 1 and less than or equal to k, and k is the total number of the robots) is the position of one transfer robot, and simultaneously refreshing and acquiring the position O2= { O1, O2,. once, ox } (the position 1 of one semi-fixed obstacle in the storage space is more than or equal to j 'and less than or equal to x) of the semi-fixed obstacle (any oj' is the position 1 of one semi-fixed obstacle in the storage space), and further executing the following position processing unit.

S3b, robot position processing unit, for any robot, calculating the distance distq between the positions qz' qz before and after refreshing, judging whether the distq is larger than the distance threshold xi, if yes, storing qz into a new set Qnew.

S3c, the position processing unit of the semi-fixed barrier calculates the distance disto between the position oa' and oa before and after refreshing any semi-fixed barrier, judges whether the disto is larger than the distance threshold xi, and stores the oa into a new set O2new if the disto is larger than the distance threshold xi.

S3d, an index structure updating unit, which updates the index structure from top to bottom from the root element root according to the positions qz ' and qz before and after the robot refresh and the positions oa ' and oa ' before and after the semi-fixed obstacle refresh, where a distance between qz ' and qz is greater than a distance threshold ξ and a distance between oa ' and oa is also greater than a distance threshold ξ, and specifically includes:

s3d1, targeting qz ', finding a slice from the root element root that the active barrier attribute set contains qi ' from top to bottom and removing qi ' from the active barrier attribute set of this slice.

And S3d2, with qz as a target, finding a slice which intersects qz from top to bottom from the root element root, and storing qi in the active obstacle attribute set of the slice.

S3d3, targeting oa ', finds a slice of the semi-fixed barrier property set containing oa ' from top to bottom starting from the root element root and removes oa ' from this sliced semi-fixed barrier property set.

And S3d4, aiming at oa, finding a fragment intersected with oa from the root element root from top to bottom, and storing the oa into the semi-fixed barrier attribute set of the fragment.

And S3d5, judging whether the positions before and after the refreshing of the movable barrier and the semi-fixed barrier which can cause the index updating are processed completely, if so, ending and returning to a new root element root, otherwise, repeating the steps S3d1-S3d 5.

S3e, a parallel processing unit, that is, allocating one CPU core to each robot position in the set Q = { Q1, Q2,. and qk } and executing the following obstacle identifying unit in parallel.

S3f, constructing an empty set G = { } by a robot qi obstacle identification unit, initializing the set by using a root element root of an index structure, namely G = { root }, further acquiring an end point e of a road section where the robot qz is located, obtaining the road section [ qz, e ] which the robot is going to pass through, and calculating a direction range [ phi 1, phi 2] of the road section [ qz, e ] relative to a circle center c, wherein phi 1 and phi 2 are respectively a starting direction and an ending direction.

And S3G, removing one index element node from the set G at a time and carrying out type judgment on the node.

And S3h, judging that the node is a branch element, and acquiring a child attribute node of the node.

S3i, sequentially obtaining a child chi in the set node.child, and obtaining a coverage space attribute chi.cover = (c, r, a, β), where c and r are a circle center and a radius of the circumscribed circle, respectively, and a and β are a start direction and an end direction, respectively.

S3j, judging whether the two direction ranges [ phi 1, phi 2] and [ alpha, beta ] are intersected, if so, storing chi into a set G, otherwise, continuing to execute the following steps.

And S3k, judging whether chi is the last child in the node set, if so, jumping to S3n, and otherwise, jumping to S3 i.

S3l, judging that the node is a leaf element, and acquiring the coverage space attribute node.

S3m, judging whether the two direction ranges [ phi 1, phi 2] and [ gamma, gamma +. alpha ] are intersected, if so, storing the node into the result set retSet, otherwise, continuing to execute the following steps.

S3n, judging whether the set G is empty, if so, continuing to execute the following steps, otherwise, jumping to the step S3G.

S3o, executing an obstacle recognition unit for qi according to the set retSet and executing different operations aiming at different obstacle types, wherein the method specifically comprises the following steps:

and S3o1, sequentially obtaining a branch element leaf in the set retSet, further obtaining the child attribute leaf, and continuing to execute the following steps, otherwise, jumping to the step S3o 8.

And S3o2, sequentially acquiring a segment si in the set leaf and judging whether the road section [ qi, e ] passes through the segment, wherein qz is the current robot position, e is the end point of the road section where the robot qz is located, executing the following steps if the segment is passed, otherwise, jumping to the step S3o 7.

S3o3, the road section [ qi, e ] obtains the fixed obstacle attribute si.O1 of si and executes S3o4 through slicing si, obtains the semi-fixed obstacle attribute si.O2 of si and executes S3o5, obtains the activity obstacle attribute si.O3 of si and executes S3o 6.

And S3o4, judging whether the fixed type obstacle set si.O1 is empty, if not, acquiring the fixed obstacles in the si.O1, adjusting the moving path of the robot qz according to the positions of the obstacles, and if so, continuing to execute the following steps.

And S3o5, judging whether the semi-fixed type barrier set si.O2 is empty, if not, sending a barrier signal to a worker for the worker to perform subsequent processing, and if so, continuing to execute the following steps.

And S3o6, judging whether the activity type barrier set si.O3 is empty, if not, sending a blocking signal to the barrier, entering a waiting state or changing a moving path, and if so, continuing to execute the following steps.

And S3o7, judging whether si is the last fragment in the set leaf, if so, continuing to execute the following steps, otherwise, jumping to the step S3o 2.

And S3o8, judging whether the current branch element leaf is the last element in the set retSet, if so, ending, otherwise, jumping to the step S3o 1.

Compared with the prior art, the invention has the beneficial effects that:

1. the intelligent transfer robot barrier automatic identification technology applied to industrial warehousing breaks through the limitation of a training model in barrier identification processing means in the traditional technology based on machine learning and the like, well solves the difficulties in organization and management of the industrial warehousing environment and the barriers contained in the industrial warehousing environment due to the characteristics of complexity, dynamic change, diversified types and the like, and brings great challenges to the development of the robot technology;

2. the invention designs an organization management technology aiming at complex dynamic storage space and complex multi-element barriers, and realizes the rapid identification and reaction of a transfer robot to the barriers in the industrial storage environment based on the technical means and the designed dynamic barrier identification and rapid reaction technology, thereby further improving the working efficiency of the transfer robot in the industrial storage;

3. the index structure designed by the invention fully considers the characteristics of complexity, dynamics and diversification of the storage space and the contained barriers, and divides the storage space and constructs the organization structure by utilizing the circumscribed circle, the direction and the distance constraint in a slicing mode, thereby further providing support for the identification technology of the subsequent barriers and further improving the organization management efficiency of the storage space and various barriers;

the design indexing and searching scheme of the invention has strong expansibility and convenient maintenance, and has wide application value and practical significance.

The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.

The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims

1. An intelligent transfer robot obstacle automatic identification method is characterized by comprising the following steps:

constructing a storage space model;

constructing an index structure of a barrier set and an index structure of a storage space model; the index structure of the barrier set and the index structure of the storage space model both comprise leaf elements, branch elements and root elements;

acquiring the position of the intelligent transfer robot in the storage space in real time, and searching the index structure according to the position to finish obstacle identification;

the method for constructing the index structure of the obstacle set and the index structure of the storage space model specifically comprises the following steps:

equally dividing the direction space of [0 degrees and 360 degrees ] in the circumscribed circle into u direction intervals according to a preset direction interval by taking the positive direction of the X axis as a reference direction of 0 degree and the center of the circumscribed circle as a base point to obtain a first divided circumscribed circle; numbering the intervals in each direction in sequence from 0 degree to 360 degrees in an increasing manner to obtain a direction interval set;

dividing the radius in the first division circumscribed circle into v interval segments according to a preset segment pitch interval to obtain a second division circumscribed circle; the second circumscribed circle comprises u × v storage space sub-pieces; the storage space sub-pieces in each direction interval form an interval sub-piece set, and the interval sub-piece set forms a space sub-piece set;

2. The method according to claim 1, wherein the constructing a warehouse space model specifically comprises:

modeling the warehouse space as a rectangle A = (x0, y0, x1, y1) in two-dimensional Euclidean space, wherein (x0, y0) is a first vertex coordinate of the rectangle, and (x1, y1) is a second vertex coordinate of the rectangle; the first vertex is a diagonal vertex of the second vertex.

3. The automatic obstacle recognition method for the intelligent transfer robot according to claim 2, wherein the method for classifying and storing the obstacles in the storage space based on the storage space model to obtain an obstacle set specifically comprises:

judging the type of each obstacle in the obstacle set;

classifying and storing all obstacles in the obstacle set according to the types of the obstacles to obtain a first-class obstacle set, a second-class obstacle set and a third-class obstacle set.

4. The method according to claim 1, wherein the acquiring a position of the intelligent transfer robot in the warehousing space in real time and searching the index structure according to the position to complete obstacle recognition specifically comprises:

judging whether the direction range intersects with a space range in the coverage space attribute of the leaf elements, if so, storing the leaf elements into an intermediate set to obtain a result set, and if not, judging whether the intermediate set is empty;

5. The intelligent transfer robot obstacle automatic identification method according to claim 4, wherein the updating of the index structure from the root element according to the position of the first-time transfer robot, the position of the second-time transfer robot, and the position of the first-time obstacle and the position of the second-time obstacle specifically comprises:

6. The intelligent transfer robot obstacle automatic recognition method according to claim 5, wherein the updating of the passage of the transfer robot based on the result set specifically includes:

when the interval shard is the last interval shard in the second child attribute set, returning to execute the steps of sequentially acquiring one branch element in the result set and acquiring the child attribute of the branch element to obtain a second child attribute set;

7. An automatic obstacle identification system of an intelligent transfer robot is characterized by comprising a processor and a memory;

the processor is connected with the memory; the memory stores a computer software program for implementing the intelligent transfer robot obstacle automatic identification method according to any one of claims 1 to 6; the processor is used for calling and executing the computer software program.