CN112347664B - Modeling method and device for irregular loading space - Google Patents
Modeling method and device for irregular loading space Download PDFInfo
- Publication number
- CN112347664B CN112347664B CN202110027801.7A CN202110027801A CN112347664B CN 112347664 B CN112347664 B CN 112347664B CN 202110027801 A CN202110027801 A CN 202110027801A CN 112347664 B CN112347664 B CN 112347664B
- Authority
- CN
- China
- Prior art keywords
- loading space
- concave
- convex
- information
- irregular
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Automatic Assembly (AREA)
Abstract
The invention provides a modeling method and a device of a non-regular loading space, wherein the modeling method comprises the following steps: scanning the irregular loading space to construct a first loading space model; fitting a concave-convex part of the loading space, and constructing a second loading space model; and correcting the second loading space model according to the concave-convex part information of the irregular loading space to obtain a corrected model of the irregular loading space. The modeling method provided by the invention can be suitable for space modeling of irregular loading space, ensures smooth operation of an automatic loading method and improves the safety of an automatic loading device; concave-convex information is obtained based on the fitting of the scanning data, accurate concave-convex information is obtained, and meanwhile, the system cost is reduced.
Description
Technical Field
The invention relates to the field of automatic loading in the logistics industry, in particular to a modeling method and device for irregular loading space.
Background
The development of economy provides higher requirements for the loading and transporting industry, however, manual loading is long in time, labor intensity is large, turnover is slow, in order to improve loading and transporting efficiency, an automatic loading robot appears, and the automatic loading robot realizes automatic loading and stacking of goods by means of mechanical arms or other mechanical structures. Generally speaking, the product of producing on the production line is deposited according to the form of pile up neatly, or deposits the box according to the form of pile up neatly after packing into cases the product, when the product needs the transportation, uses fork truck to transport the good product of pile up neatly to disassembling the department usually, disassembles the good product of pile up neatly and transmits to carriage position after, puts to the carriage according to predetermineeing the pile up neatly type pile up neatly behind the arrangement of product by automatic loading robot, realizes the automatic loading of product.
In the product loading process, a laser sensor is usually used for scanning the outside of a carriage, a space model of the carriage is constructed, and stacking and walking path planning are carried out, wherein the scanning process mainly scans the carriage at the outside, and stacking and walking are carried out in the carriage, so that even if the thickness difference is considered, the accurate information of the space inside the loading carriage cannot be obtained by adopting the method; the other mode is to adopt a high-precision sensor to scan the interior of the carriage to acquire the information of the interior space of the carriage, and the mode has higher requirement on the precision of the sensor and higher cost. In addition, the conventional car space modeling method can be applied only to the car models having a regular inner surface or a simple inner space structure, such as a general box truck, for the carriage with irregular internal space, such as the container, semi-trailer traction, flat (naked) plate and the carriage with special-shaped vehicle body, the irregular environment is not considered in the existing building process of the loading space model, the stack type design and the walking path design both depend on a loading space model, the loading space model has a larger error with the actual space size or cannot reflect the environment of the actual space, the length of the finished product is larger than that of the loading space, the loading cannot be smoothly carried out, the loading space model does not consider the irregular environment, the loading robot is easy to collide with the inner wall when walking according to the preset path, and the equipment is damaged, and for irregular environment, the loading robot cannot stack the goods to the position corresponding to the designed stack shape. If the irregular loading space actual model is used in the automatic loading control method, the corresponding methods such as navigation, stack type planning and the like need to be redesigned, and the development cost is high.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a modeling method of a non-regular loading space, which comprises the following steps:
step 1, scanning the irregular loading space, and constructing a first loading space model; the irregular loading space is a loading space with at least one concave-convex part on the inner surface of the loading space;
step 2, fitting a concave-convex part of the loading space and constructing a second loading space model;
step 3, determining concave-convex part information of the irregular loading space based on the second loading space model;
and 4, correcting the second loading space model according to the information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space.
Preferably, the first loading space model is a discrete model determined by sampling point positions; the second loading space model is a continuous model.
Preferably, the concave-convex portion information includes a height of the concave-convex portion, and/or a length of the concave-convex portion, and/or position information of the concave-convex portion.
Preferably, the concave-convex portion of the loading space is fitted based on a periodic function.
Preferably, the step 4 specifically includes:
determining at least one concave-convex inner surface where the concave-convex portion exists, based on the position information of the concave-convex portion;
determining a concave-convex portion having the highest convex height among the concave-convex portions existing on the inner surface of each concave-convex portion;
determining maximum concave-convex height information based on concave-convex part information of a concave-convex part with the highest convex height;
constructing at least one correction plane parallel to each of the at least one concave-convex inner surfaces based on the maximum concave-convex height information;
replacing at least one concave-convex inner surface corresponding thereto where the concave-convex portion exists with at least one corrected plane;
and connecting the at least one correction plane and the inner surface without the concave-convex part to form a communication region, thereby obtaining a correction model of the irregular loading space.
Preferably, the step 4 specifically includes:
judging whether a concave-convex part exists on the first inner surface or not according to the second loading space model;
if not, continuing to search the second inner surface until the judgment of all the inner surfaces of the irregular loading space is finished;
if so, obtaining a first concave-convex portion P existing on the first inner surface based on the second loading space model1A second concave-convex part P2…, n-th concave-convex part PnObtaining a first concavo-convex part P1To the n-th concave-convex part PnThe concave-convex portion information of (1);
according to the first concavo-convex part P1To the n-th concave-convex part PnDetermines a distribution of irregularities of the first inner surface;
according to the first concavo-convex part P1To the n-th concave-convex part PnThe length of the second loading space model is used for carrying out truncation processing on the first inner surface in the second loading space model, and two end points of a truncation part are communicated;
based on the first concave-convex part P1To the n-th concave-convex part PnDetermining maximum concave-convex height information according to the height information, and moving the first inner surface which is subjected to truncation processing and communicated with the truncation part by a preset distance towards the convex direction of the concave-convex part corresponding to the maximum concave-convex height information;
returning to the step of judging whether concave-convex parts exist on the first inner surface or not according to the second loading space model, searching and correcting the second inner surface until all inner surfaces of the irregular loading space are judged, and finally obtaining a corrected model of the irregular loading space.
Preferably, in step 1, scanning the irregular loading space by using a plurality of sensors, before scanning the irregular loading space, the method further includes:
when the vehicle stays at the preset position, recording the first time t when the vehicle reaches the preset position1；
Acquiring loading order information corresponding to the vehicle;
extracting departure time information t from the order information2；
Calculating the departure time information t2And the first time t1The difference between them;
determining a scanning mode of the sensor based on the difference, the scanning mode including at least a fast mode and a power saving mode;
if the difference value is smaller than or equal to a first time threshold value, determining that the scanning mode is a fast mode, and starting the sensors with the number exceeding a first preset number to scan the irregular loading space;
and if the difference value is larger than a first time threshold value, determining that the scanning mode is the power-saving mode, and starting the sensors less than a second preset number to scan the irregular loading space.
Preferably, in step 1, the scanning the irregular loading space with a sensor further includes, before the scanning the irregular loading space: the method comprises the steps of obtaining order information corresponding to a loading vehicle, extracting the size information of the loading vehicle from the order information, and determining a data acquisition period T of a sensor according to the size information of the loading vehicle.
Preferably, in the step 1, the irregular loading space is scanned, after the scanning data is obtained, the scanning data is verified, and a first loading space model of the irregular loading space is constructed based on the verified data;
the verification specifically comprises:
judging whether data are missing or not according to the data acquisition period T and the time for acquiring each data in the scanning data;
if data loss exists between data corresponding to two adjacent times, height information in the data corresponding to the two adjacent times is obtained;
judging whether the difference value of the height information in the data corresponding to the two adjacent times is greater than a first height threshold value or not;
if so, the height information of the missing data is the average value of the height information in the data corresponding to the two adjacent times;
and if not, the height information of the missing data is any one of the height information in the data corresponding to the two adjacent times.
The invention aims to solve the problems in the prior art and provides a modeling device of a non-regular loading space, which is characterized by comprising one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing a modeling method for a non-structured car loading space.
The invention has the beneficial effects that: firstly, the modeling method provided by the invention can be suitable for space modeling of irregular loading space, on one hand, a model without concave-convex parts is constructed on the basis of considering the actual internal environment of the irregular loading space by the finally obtained correction model, a regular rectangular space is formed, the method can be matched with the automatic loading control method of the carriage which is suitable for regular internal surfaces or has a simple internal space structure in the prior art for use, the automatic loading control method does not need to be redesigned aiming at the irregular loading space, the development cost is saved, and the loading safety is ensured; on the other hand, the modeling method provided by the invention can be suitable for the transport vehicles with irregular loading spaces by the steps of scanning, constructing the first loading space model, fitting the concave-convex part, constructing the second loading space model and obtaining the correction model, so that the smooth operation of the automatic loading method is ensured, the use of high-precision scanning equipment is avoided, the cost of the system is reduced, and the safety of the automatic loading device is improved. Secondly, the fitting based on the scanning data provided by the invention obtains the concave-convex information, avoids the distortion fitting generated by linear fitting, takes the periodic function as the fitting basis, corrects the periodic function curve through the discrete point piecewise fitting for special conditions, and generates a more accurate fitting curve, thereby obtaining the accurate concave-convex information and taking the balance of the fitting accuracy and the fitting efficiency into account.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the invention, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a modeling method of an irregular loading space according to an embodiment of the invention;
FIG. 2 is a schematic diagram of the calculation of the sum of squares of the fitted curves of discrete points and periodic functions in an embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced otherwise than as specifically described herein, and thus the scope of the present invention is not limited by the specific embodiments disclosed below.
Example one
The first embodiment of the invention provides a modeling method of a non-regular loading space, which comprises the following steps of:
step 1, scanning the irregular loading space, and constructing a first loading space model; the irregular loading space is a loading space with at least one concave-convex part on the inner surface of the loading space. The first loading space model is a discrete model established based on sampling points.
The loading space is in a cubic form, and for each plane forming the loading space, an inner surface and an outer surface forming the loading space exist. The irregular loading space refers to a loading space with concave-convex parts on the inner surface, such as a container, a semi-trailer traction, a flat (naked) plate, a carriage of a special vehicle body and the like. Therefore, the loading space is not a regular cuboid, and in a top view, compared with a standard rectangle, the concave-convex part in the vehicle body, particularly the possible periodic, aperiodic and asymmetric concave-convex part greatly influences the real area range of the loading space which can actually load goods. In order to accurately establish a model of a loading space, firstly, the loading space is scanned by using a sensor, the sensor is preferably a 3D laser sensor or an infrared depth of field sensor, initial 3D data of the loading space is obtained, and a first loading space model of the loading space is established based on the initial 3D data.
And 2, fitting the concave-convex part of the loading space to construct a second loading space model.
The judgment method of the concave-convex part comprises the steps of dividing the second loading model into a plurality of sections, and fitting data collected by the first loading space model based on a periodic function and a linear function respectively through a least square method. And when the difference between the fitting degree based on the periodic function fitting and the fitting degree by adopting the linear function is lower than a preset value, the interval is considered to be a concave-convex part. And fitting the scanned data based on a periodic function, such as a trigonometric function or other self-defined periodic functions, determining the wavelength and amplitude of the scanned data to obtain a second loading space model, wherein the second loading space model is a continuous model. The method has the advantages that the continuous loading space model is obtained by fitting the scanning data based on the discrete regression algorithm, high-precision scanning equipment is avoided, the cost of the system is reduced, the loading space model is accurately obtained, and the accuracy of the model and the cost of the system are both considered.
Step 3, determining concave-convex part information of the irregular loading space based on the second loading space model;
and acquiring information of each concave-convex part on each inner surface based on the second vehicle-mounted space model, wherein the information of the concave-convex part can be specifically the height and/or the length of the concave-convex part and/or the position information of the concave-convex part, the height of the concave-convex part refers to the distance information of the concave-convex part from the plane of the concave-convex part in the direction perpendicular to the plane of the concave-convex part, and the length of the concave-convex part refers to the length information of the concave-convex part in the direction parallel to the plane of the concave-convex part.
And 4, correcting the second loading space model according to the information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space.
And (3) correcting the initial model of the loading space according to the concave-convex part information determined in the step (3) to obtain a corrected model, and taking the corrected model as a new loading space model for the pallet type planning, the loading robot walking path planning and the like. Specifically, an inner surface S for the loading space1Determining the inner surface S based on the information of the concave and convex portions1And obtaining an inner surface S1Information of concave and convex portions on the inner surface S1The information of the concave and convex portions is determined and compared with the inner surface S1Corresponding modified inner surface S1'; judging whether concave-convex parts exist on the inner surfaces of the loading space one by one, correcting the inner surfaces with the concave-convex parts according to the concave-convex information to obtain a corrected model of the loading space, enabling the loading space modeling to consider the actual concave-convex information in the loading space, enabling the final corrected model not to contain the concave-convex parts, and finally taking the corrected model as the actual space model of the loading space.
Based on the information of the concave-convex part, the initial model is corrected, so that the concave-convex part in the carriage is directly shielded off on the model level while the size of the internal space of the loading is accurately reflected by the final loading space model, a regular loading space model is obtained, the method can be directly matched with other automatic loading control algorithms in the prior art for use, algorithm design does not need to be carried out again in each link, and the modeling method provided by the invention can be suitable for various irregular loading spaces, and is wide in application range and strong in compatibility. The loading space model established by the method fully considers the concave-convex information in the irregular loading space, the established loading space model accurately reflects the actual environment in the loading space, the internal environment of the loading space can be considered by the correction model, the loading space model which ensures that the automatic loading robot can successfully stack and safely walk in the loading space is obtained, the damage of the automatic loading device is avoided, the maintenance and replacement cost of the automatic loading device is reduced, and the automatic loading efficiency is improved.
Example two
The embodiment of the present invention provides a modeling method for a non-regular loading space, which is consistent with the embodiment and is not repeated in this embodiment, and the modeling method for a non-regular loading space provided in the embodiment two includes the following steps:
step 1, scanning the irregular loading space, and constructing a first loading space model; the irregular loading space is a loading space with at least one concave-convex part on the inner surface of the loading space; which is the same as step 1 of the embodiment and will not be described again in this embodiment.
Step 2, fitting a concave-convex part of the loading space and constructing a second loading space model;
fitting the concave-convex part of the loading space based on the scanning of the loading space obtained in step 1, specifically, fitting the scanned initial 3D data based on a periodic function, which may be a trigonometric function.
The initial 3D data is fitted through a periodic function, for example, a trigonometric function, so that amplitude and wavelength information can be obtained, the amplitude information and the wavelength information correspond to the height of the concave-convex part and the length of the concave-convex part, on one hand, the amplitude and the wavelength can be respectively regarded as the height and the length of the concave-convex part, on the other hand, noise and error information existing in sensor equipment in the scanning process are considered, and in order to ensure that the corrected space model can enable the automatic loading equipment to smoothly and safely walk and stack, the height and the length of the concave-convex part can be in proportional corresponding relation with the amplitude and the wavelength obtained through fitting. In addition, if asymmetric opposite space, such as space formed by other obstacles such as semi-trailer traction, flat (bare) plates and the like, is too much and a single periodic function cannot be well and efficiently and accurately fitted to the loading space, discrete random point piecewise fitting curves are introduced and connected to realize the fitting of the final loading space.
Firstly, fitting the space in the vehicle by adopting a periodic function, and starting a discrete random point piecewise fitting curve process when the weighted square sum of each discrete point and a periodic function fitting curve is greater than a preset boundary threshold value.
As shown in fig. 2, the first m points (m is more than 5% of the total number of samples) at which the square sum of each discrete point and the periodic function fitting curve of each internal surface is maximum are determined, and the concentrated region where the m points are located is determined by: and sequencing the m points based on the position sequence of the x-axis coordinate to obtain a region formed by the first point to the last point along the x-axis direction from the 1 st to the m-th points as a concentrated region. The square sum of the fitting curve of the discrete points and the periodic function is calculated byA discrete point (x)p,yp) Selecting a point (x) on the fitting curve of the periodic functionp,yp) Closest point (x)p0,yp0) Calculating a discrete point (x)p,yp) Points (x) on a curve fitted to a periodic functionp0,yp0) As the sum of the squares of the curves fitted to the discrete points and the periodic function:
the calculation method of the square sum of the discrete point and the adjacent discrete points is that for any discrete point (x)p,yp) Selecting the other discrete points as the neutralization point (x)p,yp) Closest point (x)p1,yp1) Calculating a discrete point (x)p,yp) With the closest of the other discrete points (x)p1,yp1) The distance between the two points is taken as the square sum of the discrete point and the adjacent discrete point:
segmenting the concentrated region, determining the range of at least one segment fit: calculating the sum of squares L of the 1 st to mth point to periodic function fitting curves1To LmFor the 1 st to m points, the distance K between each point and its nearest neighbor is calculated1To KmI.e. the sum of squares of each point and adjacent discrete points is calculated. The x-axis coordinate of the interval to be piecewise fitted is determined based on a specified constraint threshold H according to equation (1).
Where h is the constraint value, p is an integer, p ϵ [1, m ], q is an integer, q ϵ [ p +1, m ], i denotes the ith point, i = p,2, …, q. And designating p as 1, calculating a constraint value H when q is [2, m ], then designating p as 2, calculating a constraint value H when q is [3, m ], calculating a constraint value H of m-1 points by analogy, screening n ranges meeting the constraint value greater than a designated constraint threshold value H and enabling | q-p | to be maximum, and taking the n ranges as a segmentation result of the concentrated region to obtain a plurality of segmentation fitting ranges.
According to the method for determining the piecewise fitting range, the periodic function curve is corrected based on the sum of squares of the fitting curve of each discrete point and the periodic function, only a part of segments are selected to correct the fitting curve, the calculation amount is reduced, and the calculation amount of accurately fitting the discrete points is also reduced; in addition, the concentrated region is segmented based on the constraint threshold and the constraint value, which is obviously different from the conventional clustering algorithm, so that the scattered point interval with the most serious influence on the modeling effect can be more efficiently and accurately obtained, and a better basis is provided for segment fitting aiming at the small-range discrete points.
And if the screened n ranges are overlapped, taking the range with the maximum constraint value h in the overlapped range as a piecewise fitting range to obtain at least one non-overlapped piecewise fitting range. Through the selection of the overlapping range, only the concentrated distribution interval of the scattered points which have the most serious influence on the modeling effect is refitted, so that the fitting curve of the periodic function is corrected, the calculation amount is reduced, and the fitting precision is improved.
And fitting a plurality of discrete points in each piecewise fitting range to form at least one piecewise fitting curve formula. The piecewise-fit curve formula may be a periodic function or a non-periodic function, wherein, to ensure the fitting efficiency, a quadratic function or a cubic function is the preferred choice of piecewise-fit curve.
And connecting the periodic function and at least one piecewise fitting curve formula to determine the concave-convex part information of the irregular loading space.
According to the method, data obtained after the loading space is scanned are fitted based on the periodic function, so that the length of the concave-convex part and the height information of the concave-convex part are obtained, and aiming at the situation that the asymmetric opposite space, especially the space formed by other obstacles such as semi-trailer traction, flat (naked) plates and the like is too much, and a single periodic function cannot be well and efficiently and accurately fitted into the loading space, when the fitting effect of the periodic function is poor, discrete random point piecewise fitting curves and the periodic function can be further adopted to jointly fit discrete point data obtained by scanning, the discrete random point piecewise fitting curves are fitted, and the fitting curves are connected to realize fitting of the final loading space so as to determine the concave-convex part information of the irregular loading space. The information of the concave-convex parts is determined by fitting the scanned discrete data based on a discrete regression algorithm, the problem that the concave-convex parts exist on the inner surface is neglected by adopting a straight line fitting mode is solved, the information of the concave-convex parts in the loading space is considered during the construction of the space model, and the safety of the automatic loading and stacking method is ensured, so that the modeling method provided by the invention can be suitable for the irregular loading space. In addition, in the process of fitting the discrete points, the point with the maximum square sum of the periodic function curve is selected, so that the fitting data volume is reduced, the situation that the optimal fitting curve cannot be obtained due to the fact that all the discrete points are considered is avoided, the data volume of operation is reduced, and the data fitting speed is improved. According to the fitting method provided by the invention, on the basis of a periodic function curve, the periodic function is corrected in a small range through the piecewise fitting curve, so that a larger calculated amount of refitting is avoided, and the fitted curve can better accord with the distribution condition of the discrete points.
And 3, determining concave-convex part information of the irregular loading space based on the second loading space model.
The information of the concave-convex part determined in the step 3 comprises the height of the concave-convex part, and/or the length of the concave-convex part and/or the position information of the concave-convex part.
And 4, correcting the second loading space model according to the information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space.
The modifying the second loading space model based on the concave-convex information may include: determining at least one concave-convex inner surface where the concave-convex portion exists, based on the position information of the concave-convex portion; determining, for at least one concave-convex inner surface where the concave-convex portions exist, a concave-convex portion having the highest convex height among the concave-convex portions existing for the respective concave-convex inner surfaces, respectively; determining maximum concave-convex height information based on concave-convex part information of a concave-convex part with the highest convex height; constructing at least one correction plane parallel to each concave-convex inner surface based on the maximum concave-convex height information; replacing the concave-convex inner surface corresponding to the concave-convex inner surface with a correction plane; and forming a communication area by each correction plane and the inner surface without the concave-convex part to obtain a correction model of the irregular loading space.
Specifically, for any concave-convex inner surface having a concave-convex portion, the concave-convex portion having the highest protrusion height among the concave-convex portions present on the concave-convex inner surface is determined, the maximum concave-convex height information is determined based on the concave-convex portion information of the concave-convex portion, the correction plane parallel to the concave-convex inner surface is constructed based on the maximum concave-convex height information, the second loading space model of the loading space constructed in step 2 is corrected based on each correction plane corresponding to the concave-convex inner surface having the concave-convex portion, the communication region can be reconstructed with the inner surface having no concave-convex portion by using each correction plane instead of the corresponding inner surface, and the corrected model of the irregular loading space can be obtained.
As an alternative embodiment, the second loading space model is modified based on the concave-convex information, and the following method may be adopted: judging whether a concave-convex part exists on the first inner surface or not according to the second loading space model; if not, continuing to search the second inner surface until the judgment of all the inner surfaces of the irregular loading space is finished; if yes, obtaining the concave-convex part P existing on the first inner surface judged currently according to the second loading space model1、P2、…、PnObtaining a concave-convex portion P1To PnThe concave-convex portion information of (1); according to the concave-convex part P1To PnDetermining the concave-convex distribution condition of the currently judged first inner surface according to the position information; according to the concave-convex part P1To PnThe length of the second loading space model is used for carrying out truncation processing on the currently judged first inner surface in the second loading space model, and two end points of a truncation part are communicated; based on the concave-convex part P1To PnDetermining maximum concave-convex height information according to the height information, and moving a first inner surface which is communicated with a cut-off part after the cut-off treatment to a preset distance in the convex direction of a concave-convex part corresponding to the maximum concave-convex height information; returning to the step of judging whether the concave-convex part exists on the first inner surface according to the second loading space model, and searching and correcting the next inner surface until all inner surfaces of the irregular loading space are judged, and finally obtaining a corrected model of the irregular loading space. For the inner surface S with the concave-convex part2Searching for the presence of irregularities P on the inner surface1、P2、…、PnObtaining a concave-convex portion P1-PnBased on the information of the concave-convex part P1To PnDetermines the inner surface S based on the position information2According to the distribution of the uneven portion P1-PnLength information of (a) to the inner surface S in the second loading space model2Cutting off to obtain the inner surface S2'. Specifically, the distribution of the concave-convex portions and the concave-convex portions P1To PnLength information of (1), the inner surface S2At a position corresponding thereto, a prescribed length, which may be specifically a concave-convex portion P1To PnAnd connecting the two end points of the cut-off to obtain the inner surface S2', realizing the inner surface S2And removing the concave-convex parts. Then based on the concave-convex part P1To PnDetermines the maximum concave-convex height information, and cuts off the processed inner surface S2The information of the maximum concave-convex height is moved to the direction of the concave-convex part corresponding to the information of the maximum concave-convex height by a preset distance which is more than or equal to the information of the maximum concave-convex height. And (3) performing the processing mode on the inner surface with the concave-convex part in each inner surface forming the loading space, so as to realize the second loading space model correction of the loading space constructed in the step (2) and obtain a correction model of the irregular loading space.
By the correction mode, the inner surface with the concave-convex part is corrected based on a simple plane processing mode, information of the concave-convex part is prevented from being used as a constraint condition in a design and planning algorithm, and the design process of the irregular loading space automatic loading method is simplified.
The modeling method for the irregular loading space provided by the embodiment corrects the second loading space model according to the height information, the length information and the position information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space, corrects the second loading space model according to the information of the concave-convex part, so that the corrected model does not comprise the concave-convex part and becomes a regular rectangular space, can be matched with the automatic loading control method for the carriage with simple regular inner surface or inner space structure in the prior art, and does not need to redesign the automatic loading control method for the irregular loading space; through the correction steps, the modeling method provided by the invention can be suitable for transport vehicles with irregular loading spaces, the smooth operation of the automatic loading method is ensured, and the safety of the automatic loading device is improved.
EXAMPLE III
The third embodiment of the present invention provides a modeling method for a non-regular loading space, which is not repeated in this embodiment consistent with the first and second embodiments, and the third embodiment of the present invention provides a modeling method for a non-regular loading space, which includes the following steps:
step 1, scanning the irregular loading space, and constructing a first loading space model; the irregular loading space is a loading space with at least one concave-convex part on the inner surface of the loading space;
in the step 1, in order to establish a model of a loading space, firstly, a plurality of sensors are used for scanning the loading space, data of the plurality of sensors are fused after scanning, initial 3D data obtained after scanning of the loading space are obtained, the sensors are specifically 3D laser sensors or infrared depth sensors, and a first loading space model of the loading space is established based on the initial 3D data obtained after scanning. The sensor for scanning the loading space can be arranged on the automatic loading robot, and once the transport vehicle stops at a preset position, the automatic loading robot automatically enters the interior of the automatic loading robot to complete the scanning of the loading space; in addition, the sensor for scanning the loading space can be arranged on another device, once the transport vehicle stops at a preset position, an operator can carry the device or the device automatically enters the carriage, and the scanning of the loading space is automatically completed.
The scanning of the irregular loading space at least comprises two modes, namely a fast mode and a power-saving mode. In the fast mode, in order to improve the speed of scanning the loading space, the sensors exceeding the first preset number are started to scan the loading space, and in the power-saving mode, the sensors less than the second preset number are started to scan the loading space. Before scanning irregular loading space, when the vehicle stays at the preset position, recording the first time t when the vehicle reaches the preset position1Obtaining the order information of the vehicle loading from the server or the local storage equipment of the system, and extracting the departure time information t from the order information2According to time t2And time t1The difference between them determines the scanning pattern if time t2And time t1The difference value between the first time threshold and the second time threshold is less than or equal to the first time threshold, and the scanning mode is determined to be a fast mode; if time t2And time t1And if the difference value between the two is larger than the first time threshold value, determining that the scanning mode is the power-saving mode, and scanning the irregular loading space according to the scanning mode. The method comprises the steps of determining departure time according to actual order information of loading and transportation, judging the relation between actual time for loading cargoes and a first time threshold value, wherein the first time threshold value can be set or calculated according to an automatic loading method, and the scanning mode of a loading space can be adjusted according to different time arrival times, so that the scanning speed of the loading space is adjusted according to actual conditions.
Further, before scanning the irregular loading space, when the vehicle stays at a preset position, obtaining order information of loading of the vehicle from a server or a local storage device of the system, extracting vehicle size information from the order information, and determining a sensor data acquisition period T according to the size information.
And after the scanning data are obtained, verifying the scanning data, and thus constructing a first loading space model of the irregular loading space based on the verified data.
The scanning data is checked, including data missing judgment, whether the data are missing or not is judged according to the data acquisition period T and the data acquisition time in the scanning data, and if the data are missing or not, the adjacent data acquisition time T is3、t4If the difference between the first time threshold and the second time threshold is greater than a second time threshold, which may be a preset multiple of the data acquisition period, it is determined that there is data loss. If there is a data miss, according to t3、t4The corresponding scan data is calculated as missing data. Further, for height information of missing data, the time t is acquired3、t4Scanning data of the corresponding loading space to obtain height information h vertical to the inner surface3、h4Judging the height information h3、h4If the difference between the first and second height thresholds is greater than the first height threshold, if yes, h is selected3、h4As height information of missing data; if not, h is added3Or h4Height information considered as missing data. In the event of data loss, the invention differs from the conventional mean or previous point information substitution, first based on t3、t4And the height difference value judges whether the data missing position belongs to a concave-convex part or a plane, and different correction methods are selected according to different areas of the missing position on the inner surface of the irregular loading space, so that the particularity of the irregular loading space environment is fully considered, and the accuracy of the correction method is improved.
Step 2, fitting a concave-convex part of the loading space and constructing a second loading space model;
step 3, determining concave-convex part information of the irregular loading space based on the second loading space model;
and 4, correcting the second loading space model according to the information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space.
Which is the same as the steps 2-4 in the second embodiment, and the description is omitted here.
According to the modeling method of the irregular loading space provided by the embodiment, the irregular loading space is scanned, before scanning, the scanning mode of the sensor is determined according to order information and vehicle arrival time information, different scanning speeds can be selected according to different requirements, and therefore the loading speed is influenced integrally; and determining a scanning period according to the order information, after scanning, performing missing judgment on the initial data, and calculating the height information of the missing data according to the height information of the point data adjacent to the missing time, so that the height information of the missing data is obtained by fully considering the irregular actual scene of the inner surface of the loading space, the accuracy of the scanned data is improved, and the accuracy of the first loading space model is further ensured.
Example four
The embodiment provides a modeling device of non-regular loading space, includes: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the methods of embodiments one through three.
The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the embodiments. It will be apparent, however, to one skilled in the art that the embodiments may be practiced without the specific details. Thus, the foregoing descriptions of specific embodiments described herein are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. It will be apparent to those skilled in the art that many modifications and variations are possible in light of the above teaching. Further, as used herein to refer to the position of a component, the terms above and below, or their synonyms, do not necessarily refer to an absolute position relative to an external reference, but rather to a relative position of the component with reference to the drawings.
Moreover, the foregoing drawings and description include many concepts and features that may be combined in various ways to achieve various benefits and advantages. Thus, features, components, elements and/or concepts from various different figures may be combined to produce embodiments or implementations not necessarily shown or described in this specification. Furthermore, not all features, components, elements and/or concepts shown in a particular figure or description are necessarily required to be in any particular embodiment and/or implementation. It is to be understood that such embodiments and/or implementations fall within the scope of the present description.
Claims (9)
1. A modeling method of an irregular loading space is characterized by comprising the following steps:
step 1, scanning the irregular loading space, and constructing a first loading space model; the irregular loading space is a loading space with at least one concave-convex part on the inner surface of the loading space;
step 2, fitting a concave-convex part of the loading space and constructing a second loading space model;
step 3, determining concave-convex part information of the irregular loading space based on the second loading space model;
step 4, correcting the second loading space model according to the information of the concave-convex part of the irregular loading space to obtain a corrected model of the irregular loading space;
in step 1, scanning the irregular loading space by using a plurality of sensors, wherein before scanning the irregular loading space, the method further includes:
when the vehicle stays at the preset position, recording the first time when the vehicle reaches the preset position
；
Acquiring loading order information corresponding to the vehicle;
extracting departure time information from the order information
；
Calculating the departure time information
And the first time
The difference between them;
determining a scanning mode of the sensor based on the difference, the scanning mode including at least a fast mode and a power saving mode;
if the difference value is smaller than or equal to a first time threshold value, determining that the scanning mode is a fast mode, and starting the sensors with the number exceeding a first preset number to scan the irregular loading space;
and if the difference value is larger than a first time threshold value, determining that the scanning mode is the power-saving mode, and starting the sensors less than a second preset number to scan the irregular loading space.
2. The modeling method of an irregular loading space according to claim 1, wherein the first loading space model is a discrete model determined by sampling point positions; the second loading space model is a continuous model.
3. The modeling method of irregular loading space according to claim 1,
the concave-convex portion information includes a height of the concave-convex portion, and/or a length of the concave-convex portion, and/or position information of the concave-convex portion.
4. The modeling method of irregular loading space according to claim 1,
fitting the concave-convex part of the loading space based on a periodic function.
5. The modeling method for the irregular loading space according to claim 3, wherein the step 4 specifically comprises:
determining at least one concave-convex inner surface where the concave-convex portion exists, based on the position information of the concave-convex portion;
determining a concave-convex portion having the highest convex height among the concave-convex portions existing on the inner surface of each concave-convex portion;
determining maximum concave-convex height information based on concave-convex part information of a concave-convex part with the highest convex height;
constructing at least one correction plane parallel to each of the at least one concave-convex inner surfaces based on the maximum concave-convex height information;
replacing at least one concave-convex inner surface corresponding thereto where the concave-convex portion exists with at least one corrected plane;
and connecting the at least one correction plane and the inner surface without the concave-convex part to form a communication region, thereby obtaining a correction model of the irregular loading space.
6. The modeling method for the irregular loading space according to claim 3, wherein the step 4 specifically comprises:
judging whether a concave-convex part exists on the first inner surface or not according to the second loading space model;
if not, continuing to search the second inner surface until the judgment of all the inner surfaces of the irregular loading space is finished;
if so, obtaining a first concave-convex portion existing on the first inner surface based on the second loading space model
A second concave-convex part
…, the first
Concave-convex part
Obtaining a first concavo-convex part
To the first
Concave-convex part
The concave-convex portion information of (1);
according to the first concavo-convex part
To the first
Concave-convex part
Determines a distribution of irregularities of the first inner surface;
according to the first concavo-convex part
To the first
Concave-convex part
The length of the second loading space model is used for carrying out truncation processing on the first inner surface in the second loading space model, and two end points of a truncation part are communicated;
based on the first concave-convex part
To the first
Concave-convex part
Determining maximum concave-convex height information according to the height information, and moving the first inner surface which is subjected to truncation processing and communicated with the truncation part by a preset distance towards the convex direction of the concave-convex part corresponding to the maximum concave-convex height information;
returning to the step of judging whether concave-convex parts exist on the first inner surface or not according to the second loading space model, searching and correcting the second inner surface until all inner surfaces of the irregular loading space are judged, and finally obtaining a corrected model of the irregular loading space.
7. The modeling method of irregular loading space according to claim 1,
in step 1, scanning the irregular loading space by using a sensor, wherein before scanning the irregular loading space, the method further includes: the method comprises the steps of obtaining order information corresponding to a loading vehicle, extracting the size information of the loading vehicle from the order information, and determining a data acquisition period T of a sensor according to the size information of the loading vehicle.
8. The modeling method of irregular loading space according to claim 1,
in the step 1, scanning the irregular loading space, checking the scanning data after the scanning data is obtained, and constructing a first loading space model of the irregular loading space based on the checked data;
the verification specifically comprises:
according to the data acquisition period
Judging whether the data is missing or not according to the time for acquiring each data in the scanning data;
if data loss exists between data corresponding to two adjacent times, height information in the data corresponding to the two adjacent times is obtained;
judging whether the difference value of the height information in the data corresponding to the two adjacent times is greater than a first height threshold value or not;
if so, the height information of the missing data is the average value of the height information in the data corresponding to the two adjacent times;
and if not, the height information of the missing data is any one of the height information in the data corresponding to the two adjacent times.
9. A modeling apparatus for a non-structured vehicle loading space, the modeling apparatus comprising one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the modeling methods of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027801.7A CN112347664B (en) | 2021-01-11 | 2021-01-11 | Modeling method and device for irregular loading space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110027801.7A CN112347664B (en) | 2021-01-11 | 2021-01-11 | Modeling method and device for irregular loading space |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112347664A CN112347664A (en) | 2021-02-09 |
| CN112347664B true CN112347664B (en) | 2021-03-23 |
Family
ID=74428212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110027801.7A Active CN112347664B (en) | 2021-01-11 | 2021-01-11 | Modeling method and device for irregular loading space |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112347664B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102980531A (en) * | 2012-12-07 | 2013-03-20 | 中国铁道科学研究院 | Volume measurement method and device based on three-dimensional laser scanning |
| CN104180771A (en) * | 2014-09-09 | 2014-12-03 | 南京金创有色金属科技发展有限公司 | High-speed and high-precision tank volume measurement method and device based on three-dimensional laser scanning |
| CN108007406A (en) * | 2017-11-22 | 2018-05-08 | 湖南三德科技股份有限公司 | Compartment localization method and system |
| CN109902857A (en) * | 2019-01-22 | 2019-06-18 | 江苏徐工工程机械研究院有限公司 | Automatic planning method and system for loading point of transport vehicle |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN208496970U (en) * | 2018-06-07 | 2019-02-15 | 苏州澳昆智能机器人技术有限公司 | Laser guidance Full automatic loading machine people's system |
| CN109035390B (en) * | 2018-07-06 | 2023-04-25 | 上海盎维信息技术有限公司 | Modeling method and device based on laser radar |
-
2021
- 2021-01-11 CN CN202110027801.7A patent/CN112347664B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102980531A (en) * | 2012-12-07 | 2013-03-20 | 中国铁道科学研究院 | Volume measurement method and device based on three-dimensional laser scanning |
| CN104180771A (en) * | 2014-09-09 | 2014-12-03 | 南京金创有色金属科技发展有限公司 | High-speed and high-precision tank volume measurement method and device based on three-dimensional laser scanning |
| CN108007406A (en) * | 2017-11-22 | 2018-05-08 | 湖南三德科技股份有限公司 | Compartment localization method and system |
| CN109902857A (en) * | 2019-01-22 | 2019-06-18 | 江苏徐工工程机械研究院有限公司 | Automatic planning method and system for loading point of transport vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112347664A (en) | 2021-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107766405A (en) | Automotive vehicle road model defines system | |
| US10928508B2 (en) | Camera and radar fusion | |
| CN115236628B (en) | Method for detecting residual cargoes in carriage based on laser radar | |
| CN110837814B (en) | Vehicle navigation method, device and computer readable storage medium | |
| CN110243380B (en) | Map matching method based on multi-sensor data and angle feature recognition | |
| CN112499298B (en) | Transport robot for loading | |
| CN108828518B (en) | Method for positioning stacking car in wagon carriage | |
| CN110815202B (en) | Obstacle detection method and device | |
| JP2009527751A (en) | Object detection method using swivelable sensor device | |
| CN111587215B (en) | Method and system for generating landing solutions for containers on landing surfaces | |
| CN112070838A (en) | Object identification and positioning method and device based on two-dimensional-three-dimensional fusion characteristics | |
| CN112744217B (en) | Collision detection method, travel path recommendation device, and storage medium | |
| CN112085843B (en) | Tunnel class target feature real-time extraction and measurement method and device | |
| CN114325760B (en) | Highway tunnel quality inspection obstacle avoidance method and system based on multi-line laser radar | |
| CN117590362B (en) | Multi-laser radar external parameter calibration method, device and equipment | |
| CN112347664B (en) | Modeling method and device for irregular loading space | |
| CN114638974A (en) | Target object identification method, system, medium and electronic terminal | |
| JP6947699B2 (en) | Information processing device and point cloud position correction method, and program | |
| CN115856923A (en) | Measuring method, device, equipment and storage medium for unloading of mine truck | |
| CN115971004A (en) | Intelligent putty spraying method and system for carriage | |
| CN115273066A (en) | Shelf identification method and device, electronic equipment and machine-readable storage medium | |
| CN113138594A (en) | Automatic driving method and device | |
| CN117152672B (en) | Cabin driving area dividing method and system based on point cloud polygon fitting | |
| CN114519711B (en) | Method, system, medium and electronic terminal for measuring steel coil in reservoir area | |
| CN118225074B (en) | Self-adaptive map updating method and device for bulk cargo ship cabin cleaning robot |
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 |