DE102007060856A1 - Lane determining method, involves determining stopping point of movable objects e.g. lorry, with sensor arrangement, and determining lanes from stopping points of movable objects with statistic process - Google Patents

Abstract

The method involves determining a stopping point of movable objects e.g. lorry, with a sensor arrangement (11). Lanes are determined from the stopping points of the movable objects with a statistic process, where the statistic process is considered as model assumptions. A normal operation of a loading yard under a loading crane is monitored in a measuring interval, where the measuring interval is defined by a preset necessary number of monitoring cases. The lanes are determined in a sensor coordinate system of the sensor arrangement. An independent claim is also included for a sensor arrangement for determining lanes.

Description

A Sensor arrangement, which is mounted for example on a crane, serves to measure (or estimate) the position and position movable objects, such as the crane itself or a cargo, z. B. a container. Other uses include measurement the position and position of a vehicle or a moving component of the crane itself.

at For example, the crane can be a loading crane. Loading cranes are used at freight transhipment sites, storage areas, used in assembly halls and shipyards as well as in track construction. at A loading crane for motor vehicles is opposite the ground inclined to the loading crane, so that water can drain off. Farther are on the ground under the loading crane tracks for trucks marked.

A The design of a loading crane is a gantry crane. This spans a loading and working area like a portal. In general, run its sidewalls with wheels on two parallel Rails. On the crane bridge, the horizontal part of the Portal crane, a trolley moves with a hoist. alternative Also, a rail crane can be mounted on the crane bridge be.

Farther come as loading crane also a bridge crane, a half portal crane, a gantry crane and a portal crane in consideration.

portable Parts of a crane are z. B. the trolley or the spreader - a Hoist, with which containers can be taken.

in the Context of a loading crane serve the measurements of the sensor array as a basis to signal truck drivers where to stop to have. Furthermore, due to such measurements, the crane itself to be controlled.

The Sensor arrangement may for example consist of one or more of the the following elements: a 3D laser scanner, a tiltable 2D laser scanner or a video camera. The Elements The sensor arrangement are usually such in the structure attached to the crane that - in the case of a gantry crane - several Traces for trucks or railroad tracks overlined become.

• – Das Sensor-Koordinatensystem eines oder mehrerer Laserscanner, die in der Sensoranordnung verbaut sind,
• – das Kamera-Koordinatensystem einer oder mehrerer Kameras, sofern diese im Rahmen der Sensoranordnung verbaut sind,
• – das Kran-Koordinatensystem des Krans, bezüglich dem z. B. Laufkatze und Spreader lokalisiert werden,
• – das Boden-Koordinatensystem, bezüglich dessen ggf. Spuren für Lastwagen oder Gleise für Eisenbahnwaggons gegeben sind, welche beladen und entladen werden.

In order for the sensor assembly to be used in the manner described, it is necessary to calibrate it. This means that the following coordinate systems must be related to each other:

• The sensor coordinate system of one or more laser scanners installed in the sensor arrangement,
• The camera coordinate system of one or more cameras, provided that they are installed in the frame of the sensor arrangement,
• - The crane coordinate system of the crane, with respect to the z. Trolley and spreader are located,
• - The ground coordinate system, with respect to which, if necessary, tracks for trucks or railroad railroad tracks are given, which are loaded and unloaded.

The Calibration is solved in the prior art, for example by that a specially prepared and prepared for this purpose Calibration body placed on the ground in the area of the crane and with respect to the crane or crane coordinate system manually survey a surveyor. additionally Can lanes manually with respect to the crane or on to measure the calibration body. The sensor arrangement then captures the calibration body, from which Coordinate transformations between the sensor coordinate system and the other coordinate systems to let. The disadvantage here is that the current operation of the crane for calibration to be interrupted for a long time got to.

in the Context of the automation of the loading crane system is a specific one Calibration of the sensor arrangement for the respective work processes desirable. Often found in the area the loading crane structuring features, which only optically can be detected. An example of this is Lane markings painted on the ground indicating the location of the lanes Show. When integrating an optical sensor (such as a camera) in the sensor array, a coordinate transformation between the camera coordinate system and the other coordinate systems become.

outgoing from this calibration remains the location of the lanes, on which vehicles drive in the loading crane system. An automatic Detection of lanes is described in the applications described in large cranes due to their large dimensions and non-standardized visual features. So can For example, the lane markings will be different. Therefore It is necessary in the prior art, more, in their optical and geometric features accurately defined calibration into the system in order to carry out the calibration on the basis of this. Also a manual measurement of the lane markings means here additional effort.

The Determination of lanes serves the purpose of automating the Loader crane system to be able to predict where trucks be located. The disadvantage here is that the current Operation of the loading crane system for manual surveying the lanes must be interrupted.

It thus sets itself the task, a method and a sensor arrangement to identify lanes indicating a fault reduce the ongoing operation of the loading crane.

These The object is achieved by the method and the sensor arrangement according to the independent claims solved. Preferred developments emerge from the dependent Claims.

at the method for determining lanes with a sensor arrangement be with the sensor arrangement points of residence of mobile Determined objects. The lanes are subsequently determined from the residence points.

The Sensor arrangement for determining lanes has a data memory on, in the residence points of moving objects storable are, and further a computing unit, which for determining the lanes is set up from the stored residence points.

the Method and the sensor arrangement is based on the knowledge that the moving objects (such as vehicles, in particular lorries, moving parts of the loading crane, in particular trolleys or spreaders, or freight items, in particular containers) essentially Stop there, where the lanes on the floor of the loading yard are drawn. From the location of moving objects can be closed so on the location of the lanes.

It become thus those structures, which the movement behavior to control the moving objects, from the movement behavior of inferred from moving objects.

This offers a number of advantages. As a 3D laser scanner, which only evaluates distance measurement data, lane markings, which on located at the bottom of the loading yard, can not detect It is now possible, the laser scanner yet to determine the lanes by this the residence points of the trucks determined. So the lanes are directly in the sensor coordinate system the sensor arrangement determined. This eliminates manual Ver measurements the lanes and a determination of a coordinate transformation between a ground coordinate system and the sensor coordinate system. Rather, the coordinate transformations between the used Coordinate systems without manual measurement automatically determined. This is more accurate and reduces the effort of installation and connection of the sensor arrangement.

With the manual measurement of the lanes eliminates the effort for creating, interim storage, placing and putting away the calibration body. This saves costs. Furthermore, the ongoing operation of the loading crane system is not disturbed, since the evaluation of the driving behavior of the truck and the determination lanes in the background during ongoing operation respectively.

One Another advantage is that waives about a video camera can be to the only optically detectable lane markings determine. Finally, provide the method and the sensor arrangement also the advantage of subsequent changes in the plant, when the lanes are redefined, automatically to recognize. The sensor assembly is capable of becoming recalibrate automatically in this case.

in the Below, embodiments of the invention are based on explained in more detail by figures. Show it:

1 Crane with a sensor arrangement and a cargo under the crane,

2 Distance measurement data of the sensor arrangement,

3 a crane coordinate system, a sensor coordinate system and a ground coordinate system,

4 a flowchart of a calibration of the sensor arrangement.

1 shows a crane 10 , At the crane 10 is a sensor arrangement 11 attached, which in the in 1 case shown consists of two elements. Also shown is a cargo 12 For example, a container on a truck, which through the sensor assembly 11 is detected. Also in 1 you can see wheels 14 with which the crane 10 can be moved on rails. A floor 15 under the crane 10 is inclined, so that water can drain. On the ground 15 are lane markers 13 attached, which mark tracks for vehicles.

2 shows measured data of the sensor arrangement 11 , in this case distance measurement data of a laser scanner. In the measurement data can be ground measurement data 21 of parts of the soil 15 U.N the crane 10 as well as crane measurement data 22 of parts of the crane 10 identify. This allows a geometric measurement of the crane and its working space.

The measurement data can be generated, for example, by pivoting a 2D laser scanner over the parts of the crane 10 and the parts of the soil 15 be won.

The crane measurement data 22 are here as rectangles in 3D heraussegmentierte sidewalls of the crane 10 , At the ground measurement data 21 are correspondingly out-segmented ground points. Of these, only a subset may need to be used to achieve sufficient accuracy; This saves computing time and storage requirements.

3 shows the crane again 10 his wheels 14 , the sensor arrangement 11 as well as the ground 15 and the lane markers 13 , Additionally marked are a ground coordinate system 16 , a crane coordinate system 17 and a sensor coordinate system 18 the sensor arrangement 11 ,

4 shows a flow chart for calibration of the sensor array 11 , In a first step 1 detects the sensor arrangement 11 Measurement data of their environment as in 2 shown.

These are distance measurement data. In a second step 2 are in the measured data in 2 shown soil measurement data 21 as well as the crane measurement data 22 identified. This can be done computer-aided using plan drawings, which with respect to the installation position of the sensor array 11 or their elements and other elements of the crane 10 are accurate enough. The measured data are given here as 3D measuring points, which are all initially in the sensor coordinate system 18 the sensor arrangement 11 available.

In a third step 3 become the crane measurement data 22 used to get off the ground metrics 21 a ground coordinate system 16 to investigate. In a fourth step 4 becomes a coordinate transformation between a sensor coordinate system 18 the sensor array and the ground coordinate system 16 calculated, based on which the sensor arrangement 11 is calibrated.

The method for calibrating the sensor arrangement 11 uses symmetries, such as surface symmetry or translation symmetry in three-dimensional space. Such symmetries are shown by the crane 10 on - especially in the design as a gantry crane. The symmetries are calculated from the crane measurement data 22 and possibly the soil measurement data 21 extracted. This is based on the knowledge that one through the crane measurement data 22 represented surface of 3D measuring points has a normal vector, as the y-direction vector of the crane coordinate system 17 can be used. Furthermore, one can take advantage of the fact that one through the ground measurement data 21 represented surface of 3D measuring points has a normal vector, as the z-direction vector of the ground coordinate system 16 can be chosen. Surprisingly, the x-direction vector of the ground coordinate system 16 identical to the x direction vector of the crane coordinate system 17 to get voted. This x-direction vector is both parallel to the ground 15 as well as parallel to one by the crane measurement data 22 represented area. This becomes clear with the example of the gantry crane. Since this runs on rails, they run so well parallel to the ground as well as parallel to the inner walls of the gantry crane. The x-direction vector of both the ground coordinate system 16 as well as the crane coordinate system 17 can thus be selected parallel to the rails. Thus, the x-direction vector is both the normal vector of the ground measurement data 21 represented surface of 3D measurement points as well as the normal vector of the crane measurement data 22 represented surface of 3D measuring points vertically. On the basis of this, these coordinate systems can be successively developed.

In the following, an algorithm suitable for this purpose is given by way of example. Any other algorithms are possible. In the alternative, the algorithm calculates the crane coordinate system 17 , which is for the derivation of the soil coordinate system 16 is used.

In the scenario of a gantry crane in a harbor are called crane metrics 22 Identified 3D measurement points that belong to the sidewall (the so-called "sill bar"), on the one hand 3D measurement points that belong to the seaward sidewall and, on the other hand, 3D measurement points that belong to the landside sidewall. For gantry cranes on the mainland, any other characterization of the two sides of the gantry crane can be chosen.

A major axis transformation (described in: Jonathon Shlens: "A Tutorial on Principal Component Analysis" , available on the internet on 02.07.2007 under http://www.snl.salk.edu/~shlens/pub/notes/pca.pdf on the 3D measuring points of the seaward side wall of the crane 10 as well as on the 3D measuring points of the side wall of the crane 10 allows in each case the formation of a normal vector in the center of gravity of the respective 3D measuring points. From these two normal vectors, the y-direction vector of the Crane coordinate system 17 educated. In this case, it is chosen as a convention that the y direction vector faces the sea side.

Also by major axis transformation will be on the 3D measurement points in the soil measurement data 21 a normal vector is formed and as a z-direction vector of the ground coordinate system 16 selected.

Subsequently, the x-direction vector of both the ground coordinate system results 16 as well as the crane coordinate system 17 from the cross product of the y-direction vector of the crane coordinate system 17 and the z-direction vector of the ground coordinate system 16 ,

In the next step, the z direction vector of the crane coordinate system becomes the cross product of the x direction vector and the y direction vector of the crane coordinate system 17 calculated. The result is the y-direction vector of the ground coordinate system 16 from the cross product of the z-direction vector and the x-direction vector of the ground coordinate system 16 ,

As the origin of the crane coordinate system 17 First, the middle between the foot points of the crane measurement data 22 selected normal vectors.

Subsequently, the 3D measurement points in the crane measurement data 22 in the crane coordinate system, which belong to the seaward side wall and the side wall on the landside 17 transformed. In the course of this, their expansion in the x-direction is also determined as minimum and maximum. Then the crane coordinate system becomes 17 shifted in its x-direction so that its origin lies in the middle between the determined minimum and maximum. This is the crane coordinate system 17 completely determined.

As the origin of the ground coordinate system 16 First, the focus of the 3D measurement points in the soil measurement data 21 certainly. Subsequently, the z-axis of the crane coordinate system 17 with the plane passing through the 3D measurement points in the ground measurement data 21 is formed, cut. The intersection as the origin of the ground coordinate system 16 selected.

Now, lanes can be under the crane 10 manually between tracks on which the wheels 14 of the crane 10 run, be measured. The location of the center between the tracks relative to the sensor array 11 results from the previously determined coordinate systems.

Based on the determined coordinate systems in each case a coordinate transformation to the sensor coordinate system 18 the sensor arrangement 11 calculated. With this, the sensor arrangement 11 calibrated.

Determination of lanes:

To determine the lanes is observed over a sufficiently long period of time, where moving objects (such as vehicles, especially trucks, moving parts of the loading crane, especially trolleys or spreaders, or freight, especially containers) are located on the loading yard in the crane system. The measurement interval may alternatively be defined by a predetermined required number of observation cases. In the measuring interval, normal operation is observed in the loading yard under the loading crane. From the measurement data of the sensor arrangement obtained here 11 It is determined in which areas in the loading yard the probabilities of residence of the moving objects pile up. By means of statistical methods, the location of the lanes can be estimated with sufficient accuracy from the residence points or residence probabilities. The statistical method can take model assumptions into account, for example the model assumption that the lanes on the ground 15 are drawn in parallel or the model assuming that the lanes on the ground 15 are each drawn the same width.

hereby it becomes possible to dispense with a video camera and the determination of the lanes instead with a 3D laser scanner perform. For this purpose, the 3D laser scanner takes during the defined measuring interval three-dimensional scans from the loading yard in the loading crane system in its normal operation. Out of the scans subsequently residence points of moving objects, in particular Truck, determined. These residence points over the statistics of all scans across accumulation points on which correspond to the lanes. The parameters indicating the location of the lanes describe, thus become from the observed residence points inferred from the moving objects.

With the manual measurement of the lanes is thus considerable installation effort in the connection of the sensor array 11 saved. The sensor arrangement 11 can adapt continuously to new conditions in the loading crane system.

Another advantage is that the lanes are located directly in the sensor coordinate system ( 18 ) be determined.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited non-patent literature

• - Jonathon Shlens: "A Tutorial on Principal Component Analysis" [0038]
• - http: //www.snl.salk.edu/~shlens/pub/notes/pca.pdf [0038]

Claims (5)

Method for detecting lanes with a sensor arrangement ( 11 ), - in which with the sensor arrangement ( 11 ) Points of residence of moving objects are determined, - in which the lanes are determined from the residence points. The method of claim 1, wherein the lanes determined from the residence points using a statistical procedure become The method of claim 2, wherein the model assumptions be taken into account. The method of claim 1, wherein the movable Objects vehicles, especially trucks, moving parts a loading crane, in particular trolleys or spreaders, or freight items, especially containers are. Sensor arrangement ( 11 ) for determining lanes, - with a data memory, in the residence points of moving objects can be stored, - with a computing unit, arranged to determine the lanes from the stored residence points.