RU2841578C2 - Excavating tool tracking system - Google Patents

Abstract

FIELD: machine building.

SUBSTANCE: invention relates to a system and a method for tracking a working equipment tool, preferably an excavating tool comprising wear elements, such as an excavator bucket. System comprises one or more sensors mounted on the working equipment and facing towards the implement, and a processor configured to receive data relating to the tool from one or more sensors; generating a three-dimensional representation of at least a portion of the implement using the obtained data; comparison of generated three-dimensional representation with previously generated three-dimensional representation; identifying one or more events such as wear and loss of at least part of the implement by comparing the generated three-dimensional representation with a previously generated three-dimensional representation, wherein processor is configured to generate three-dimensional representation by aligning a plurality of three-dimensional models by combining three-dimensional models in a common coordinate system, wherein generation of a three-dimensional representation of at least part of the implement using data received from one or more sensors, comprises using a processor configured to collect a plurality of vector scans during a certain period of time in order to generate a three-dimensional representation using motion estimation data, wherein the movement evaluation data are derived from data received from one or more sensors as the tool moves in the field of view of one or more sensors.

EFFECT: control of tool wear of working equipment.

57 cl, 6 dwg

Description

Field of technology to which the present invention relates

[0001] The present invention relates to a system for monitoring wear elements and a method for monitoring wear elements. In particular, the present invention relates, inter alia, to a system for monitoring wear elements and a method for monitoring wear and/or the presence (or absence) of one or more earthmoving tools, such as a bucket tooth, a bucket guard segment, or an excavator bucket adapter.

Prior art of the present invention

[0002] References to prior art contained in this document should not be construed as an admission that such art is prior art.

[0003] Many types of operations involve the use of wear elements, usually in the form of expendable, removable components designed to wear out to protect some equipment. One notable application of wear elements is in the mining industry, where buckets and other excavator components are equipped with wear elements that are installed in high-wear areas, such as the cutting edge of the excavator bucket, to protect the bucket itself. Such wear elements often include excavator bucket teeth assemblies and bucket shrouds.

[0004] Excavator bucket teeth as an assembly, mounted on the cutting edge of an excavator bucket, and other elements of the same kind, usually include a removable cutting tooth, an adapter body and an adapter head, which is attached to the cutting edge of the bucket or other element of the same kind by welding or other means. Removable intertooth segments are usually located between the teeth of the excavator bucket as an assembly for the purpose of protecting the edge of the bucket. A socket-shaped recess is usually provided at the rear end of the teeth for receiving a front part of the adapter head suitable for insertion into it, and a locking system is usually used to secure the tooth on the adapter for subsequent removal.

[0005] During operation, such wear elements are typically subject to significant wear and prolonged exposure to various forces. As the various components wear, the locking system may become loose, increasing the risk of losing the excavation element or the entire adapter and tooth assembly. This results in long downtimes required to replace the lost wear elements; and in cases where the components are not repairable, this may result in damage and/or long downtimes during subsequent processing operations, especially if the disconnection remains undetected. For example, if a wear element becomes disconnected from an excavator bucket, the wear element may damage other equipment on the mine site when it is inadvertently caught, such as by a rock crusher. In addition, excavation operations with disconnected or heavily worn wear elements are inherently less efficient.

[0006] In an attempt to prevent accidental detachment of wear elements, scheduled maintenance work is carried out on work sites according to a certain schedule. Other technologies are also used for continuous monitoring and reporting of loss of wear elements. However, these technologies are usually complex and are not suitable for all conditions that arise, for example, on a mine site.

[0007] One such system for monitoring wear and/or loss of wear elements is described in PCT Publication No. WO 2018/009955 A1, the contents of which are incorporated herein by reference in their entirety. While it provides some effectiveness, particularly in controlled environmental conditions, it has been found that over long periods of use, the ability to detect loss of earthmoving tools as they wear out is lost. In addition, the described system is significantly limited by the resolution of the sensor device, and in any case, it is aimed at tracking teeth and is therefore not capable of effectively monitoring other wear elements, such as bucket guards.

The purpose of the present invention

[0008] The object of the present invention is to provide a wear element tracking system and a method for using the same that would eliminate or mitigate one or more of the disadvantages or problems described above, or that would represent at least an acceptable alternative.

[0009] Other preferred objects of the present invention will become apparent from the following description.

Brief summary of the present invention

[0010] According to one embodiment, although not necessarily the only or broadest, the present invention relates to a system for tracking a work equipment tool, the system including:

one or more sensors mounted on the work equipment and facing the implement; and

a processor configured to:

obtaining data related to the tool from one or more sensors;

generating a three-dimensional representation of at least a portion of the tool using the obtained data;

comparing the generated three-dimensional representation with a previously generated three-dimensional representation;

identifying one or more events such as wear and loss of at least a portion of the tool by comparing the generated three-dimensional representation with a previously generated three-dimensional representation; and

issuing an indication of wear or loss in the event of identification of wear or loss of at least part of the tool.

[0011] In a preferred embodiment, the tool is characterized by the presence of wear parts. These wear parts are preferably designed so that they can be replaced. In a preferred embodiment, the tool is an earthmoving tool.

[0012] One or more sensors may include at least one sensor capable of receiving data representing a three-dimensional shape of the surface of the earthmoving implement. One or more sensors may include a time-of-flight sensor. One or more sensors may include a rangefinder sensor. One or more sensors may include a laser rangefinder sensor. One or more sensors may include a light detection, identification, and ranging (LIDAR) sensor. The LIDAR sensor may be a three-dimensional LIDAR sensor. One or more sensors may include a multi-layer, time-of-flight, scanning laser rangefinder sensor.

[0013] One or more sensors may include stereoscopic vision sensors. The stereoscopic vision sensors may provide data in various spectra, including the visible spectrum and/or the thermal infrared spectrum. The one or more sensors may include non-time-of-flight ranging systems. The non-time-of-flight ranging systems may include structured three-dimensional illumination-based ranging systems. The one or more sensors may include radar. The one or more sensors may include ultrasonic sensors. The one or more sensors may include sensors configured to detect a radiation pattern. The radiation pattern may be created and/or modified by an earthmoving implement. The one or more sensors may use magnetic resonance imaging (MRI). The one or more sensors may include acoustic sensors. The one or more sensors may include a two-dimensional sensor, wherein a three-dimensional representation is derived from the two-dimensional sensor data. A 3D representation can be inferred from 2D sensor data based on light analysis and/or machine learning.

[0014] One or more sensors may include a single sensor configured to measure and output data representing a three-dimensional shape of a surface of at least a portion of an earthmoving implement. Alternatively, one or more sensors may include a plurality of sensors. The plurality of sensors may include optical photodetectors, such as cameras. The plurality of sensors may include at least a pair of two-dimensional scanning rangefinders oriented relative to each other at different angles to perform measurements. The two-dimensional scanning rangefinders may include laser rangefinders. The two-dimensional scanning rangefinders may be oriented relative to each other at an angle of about 90° to perform measurements. Generating a three-dimensional representation of at least a portion of an earthmoving implement using data obtained from the one or more sensors may include configuring a processor to collect a plurality of vector scans over a certain period of time to generate the three-dimensional representation. The processor may be configured to collect a plurality of vector scans over a period of time to generate a three-dimensional representation using motion estimation data.

[0015] The processor may be configured to combine data from sensors that are read in different ways and have different fidelity and/or noise characteristics to generate a three-dimensional representation. The combination of sensor data may involve using a combinatorial algorithm. For example, data from a LIDAR sensor and a radar may be combined using a combinatorial algorithm, such as a Kalman filter.

[0016] One or more sensors may be mounted on the work equipment so that the earthmoving implement is in their line of sight. One or more sensors may be located on the implement itself, where a portion of the implement of interest is in their line of sight. During operation of the work equipment, the line of sight may be continuous. Alternatively, during operation of the work equipment, the line of sight may be intermittent. One or more sensors may be mounted on movable elements of the work equipment. One or more sensors may be mounted on a movable boom of the work equipment. The work equipment may be an excavator. One or more sensors may be mounted on a stick of the excavator. One or more sensors may be mounted on a boom of the excavator. One or more sensors may be mounted on a body or a cabin of the excavator. One or more sensors may be mounted on a bucket of the excavator.

[0017] The processor may be configured to generate a three-dimensional representation of at least a portion of the earthmoving implement by combining data obtained from one or more sensors with a motion estimate. The motion estimate may be derived from the sensor data. The motion estimate may be derived from the sensor data as the earthmoving implement moves in the field of view.

[0018] The processor may be configured with an additional capability of pre-processing the received data to generate a three-dimensional representation. The processor may be configured with an additional capability of pre-processing the received data within a specified range. The pre-processing may include range selection. The pre-processing may include multiple scans using interlaced scanning. Multiple scans using interlaced scanning may provide a wider effective field of view.

[0019] The pre-processing may include determining that the earth-moving implement is substantially within the field of view of one or more sensors. This evaluation may include identifying whether the sensor data indicates that the earth-moving implement is identified as present or absent at selected points. The evaluation may include determining a ratio of points at which the earth-moving implement is suspected to be present to points at which the earth-moving implement is suspected to be absent. The evaluation may include comparing the ratio to a predetermined threshold. The predetermined threshold may be based on known geometry of the earth-moving implement. Alternatively, the evaluation may be based on a state machine. The state machine may comprise one or more of the following states: wear features are not visible; wear features are partially visible; wear features are completely visible; wear features are partially outside the field of view of the sensors; and wear features are completely outside the field of view of the sensor. The determination of the condition may be based on heuristic rules that identify conditions of the spatial distribution of 3D points corresponding to each condition. The assessment may also be supplemented by a rejection mechanism that discards data indicating that wear features may still be embedded in the drift wall or may be covered by material that is of no interest. The rejection mechanism may check for empty data within the known dimensions of the tool. The rejection mechanism may check the approximate shape (e.g., flat, spherical, or elliptical) of the tool by examining the results of principal component analysis of the 3D points.

[0020] The processor may be configured to generate a three-dimensional representation by combining a plurality of sensor data sets obtained at different times into a single three-dimensional model. The combining may include combining sensor data over a period of time starting from the time when the earthmoving implement was determined to be substantially within the field of view of one or more sensors. The combining may include voxelizing the sensor data. Points from separate data sets related to a single voxel may be merged into a single point. Points from separate data sets related to a single voxel may be merged into a single point using a statistical model. Points from separate data sets related to a single voxel may be merged into a single point representing a median.

[0021] The processor may be configured to generate a three-dimensional representation by combining multiple sets of sensor data obtained at different times into a single two-dimensional model, such as a range map. The combining may include combining sensor data over a period of time, starting from the time when the earthmoving implement was determined to be substantially within the field of view of one or more sensors. The combining may include projecting the three-dimensional data onto a flat, cylindrical, spherical, or other continuous surface to obtain a two-dimensional gridded representation. Points from individual sets of data that relate to a single pixel may be merged using a statistical model. The statistical model may include an cumulative average and/or a Kalman filter. The Kalman filter may be one-dimensional.

[0022] The processor may be configured to generate a three-dimensional representation by aligning a plurality of three-dimensional models. The alignment may involve aligning different three-dimensional models in a common coordinate system. The alignment may involve using a selected model as a reference model and aligning the remaining three-dimensional models to this selected model. The alignment of the generated models may involve using an ICP (Iterative Closest Point Algorithm) or NDT (Normal Distribution Transformation) process. The alignment process may be characterized by constraints, such as, for example, assumed degrees of freedom. The alignment may involve determining a fractional linear transformation matrix. The fractional linear transformation matrix may be based on the agreement of key points, geometric features, or axes between the reference model and the intermediate models. The determination of the axes for the transformation may be based on principal component analysis (PCA).

[0023] The processor may be configured to further convert the generated three-dimensional representation into two-dimensional range data. The two-dimensional range data may be an image. The generated three-dimensional representation may be converted into a two-dimensional image by selecting a plane of the three-dimensional representation and indicating range data perpendicular to this plane with different image characteristics. Different image characteristics may include different colors or brightness. Range data perpendicular to the selected plane may be indicated using a color gradient mapped to a rectangular coordinate. The two-dimensional image may be filtered using, for example, an opening/closing or widening/narrowing filter. A plurality of two-dimensional images may be combined over a certain period of time to reduce noise.

[0024] The processor may be configured with an additional capability of comparing the generated three-dimensional representation with a previously generated three-dimensional representation by comparing two-dimensional images that include range data. The two-dimensional images may be compared by subtracting the images. The variable time bases may include a first time base that is shorter than a second time base. The second time base may be at least twice as long as the first time base. The second time base may be an order of magnitude longer than the first time base. The first time base may be characterized by a duration of less than an hour. The second time base may exceed 12 hours.

[0025] The processor may be configured to further identify one or both of such events as wear and loss of at least a portion of the earthmoving implement by analyzing the result of the comparison of the two-dimensional images. A significant difference in this comparison with the first time base may indicate loss of at least a portion of the earthmoving implement. A less significant difference in this comparison with the second time base may indicate wear of at least a portion of the earthmoving implement.

[0026] The analysis may include creating a difference image. The difference image may be divided into separate zones. These zones may correspond to zones of interest on the earth-moving implement. The zones of interest may include the expected location of wear features on the earth-moving implement. The wear features may include one or more features such as teeth, adapters, shrouds, and liners. The difference image may be divided into separate zones based on a given geometric model of the earth-moving implement. The difference image may be divided into separate zones using feature boundary detection analysis. Feature boundary detection analysis may be used to identify features of substantially vertical lines. The difference image may be divided into separate vertical zones.

[0027] The analysis may include measuring changes in the difference image in each zone. Measuring changes in the difference image in each zone may include determining a number of pixels. Determining a number of pixels may include counting adjacent pixels. The number of adjacent pixels may indicate areas of wear and/or loss in that zone. Adjacent pixels may be counted row by row. The number of pixels counted in a row may be compared to a threshold to indicate whether wear and/or loss may have occurred. The threshold may be predetermined. The threshold may be adaptive, such as based on a comparison of a value over time or obtained by machine learning. Machine learning may be based on feedback from an operator. The analysis may include using a convolution process. The convolution process may include using a convolution filter. The convolution filter may provide the location and magnitude of changes where the differences are not due to loss but, for example, only due to a change in depth. Noise reduction can also be accomplished using an image mask derived from an image created at an earlier time base and superimposed on the current image. This mask prevents analysis of parts of the image that are considered irrelevant.

[0028] The processor may be configured to provide an indication of the identified wear or loss to the operator of the work equipment. The data provided may comprise an alarm signal. The alarm signal may be one or more of an acoustic alarm signal, a visual alarm signal, and a tactile alarm signal. The alarm signal may be transmitted to the operator of the work equipment. The alarm signal may also or instead be transmitted remotely. In a preferred embodiment, the alarm signal is transmitted remotely to a periphery outside the equipment. The alarm signal may include a first alarm signal indicating wear and a second alarm signal different from the first alarm signal and indicating loss. The indication of wear or loss may be used by control systems of the work equipment to adapt its operation.

[0029] The system may further include a vehicle identification system. The vehicle identification system may include one or more sensors that identify a vehicle. The vehicle identification system may use a processor to perform an operation to identify the vehicle. The vehicle identification system provides the ability to identify the appropriate vehicle when a fact of loss of a part of the earthmoving implement, such as a wear element, is determined. The vehicle identification system facilitates the identification of the appropriate vehicle to which the detached wear element or other element of the same kind can be delivered during a delivery operation performed by the working equipment.

[0030] The processor may be configured with an additional capability of recording and/or transmitting coordinates of a global navigation satellite system (GNSS) upon identification of the loss of at least a portion of the implement. The GNSS system may include a GPS (global positioning system).

[0031] In a preferred embodiment, the processor is located on the work equipment. However, the processor may be located remotely. The processor may include one or more network servers. The work equipment may include a processor for local data processing, and also communicate with one or more network processors for remote data processing.

[0032] According to another embodiment of the present invention, there is provided a method for monitoring one or more wear elements of a work equipment tool, the method comprising:

obtaining data related to the tool from one or more sensors installed on the work equipment;

generating a three-dimensional representation of at least a portion of the tool using data obtained from one or more sensors;

comparing the generated 3D representation with a previously generated 3D representation;

identifying one or more events such as wear and loss of at least a portion of the tool based on the result of performing the comparison step; and

issuing an indication of wear or loss of at least a portion of the implement after such wear or loss is identified in the stage of identifying one or more such events as wear and loss of at least a portion of the earthmoving implement.

[0033] In a preferred embodiment, one or more wear elements are replaceable. In a preferred embodiment, the implement is an earthmoving implement.

[0034] The step of receiving data related to the earthmoving implement from one or more sensors may include receiving three-dimensional data related to at least a portion of the earthmoving implement. The step of receiving data may include receiving data from a single sensor. Alternatively, the step of receiving data may include receiving information from a plurality of sensors. The proposed method may further include performing a step of converting data from the plurality of sensors into three-dimensional data.

[0035] The step of generating a three-dimensional representation of at least a portion of the earthmoving implement may include combining data obtained from one or more sensors with a motion estimate. The proposed method may further include deriving a motion estimate from data from a sensor, preferably from data obtained as the earthmoving implement moves in the field of view of the sensor.

[0036] The proposed method may further include pre-processing the data obtained from the sensors before performing the three-dimensional representation generation step. The proposed method may further include a step of pre-processing the data obtained from the sensors by identifying the data from the sensors within a given range, preferably by a range selection method. The pre-processing step may include multiple scans using interlaced scanning. Multiple scans using interlaced scanning may provide a wider effective field of view of the data from the sensors.

[0037] The proposed method may further include a step of determining that the earthmoving implement is substantially within the field of view of one or more sensors. The step of evaluating may include identifying whether the data from the sensors indicates that the earthmoving implement is identified as present or absent at selected points. The step of evaluating may include determining a ratio of points at which the earthmoving implement is suspected to be located to points at which the earthmoving implement is suspected to be absent. The step of evaluating may include comparing this ratio to a predetermined threshold value. The predetermined threshold value may be based on a known geometry of the earthmoving implement.

[0038] The proposed method may further include a step of combining a plurality of sets of sensor data obtained at different times into a single three-dimensional model. The step of combining may include combining sensor data over a certain period of time, starting from the moment when it was determined that the earthmoving implement is located substantially within the field of view of one or more sensors. The step of combining may include voxelizing the sensor data. The proposed method may further include a step of merging individual sets of data related to a single voxel into a single point. Points from individual sets of data related to a single voxel may be merged into a single point using a statistical model. Points from individual sets of data related to a single voxel may be merged into a single point representing a median.

[0039] The proposed method may further include a step of aligning a plurality of three-dimensional models. The step of aligning may include aligning different three-dimensional models in a common coordinate system. The step of aligning may include using a selected model as a reference model and aligning the remaining three-dimensional models to this selected model. The step of aligning the generated models may include using an ICP (Iterative Nearest Point Algorithm) or NDT (Normal Distribution Transformation) process. The ICP process may be characterized by constraints, such as, for example, assumed degrees of freedom. The step of aligning may include determining a fractional linear transformation matrix. The fractional linear transformation matrix may be based on establishing a correspondence between key points, geometric features, or axes of the reference model and intermediate models.

[0040] The proposed method may further include converting the generated three-dimensional representation into two-dimensional range data. The step of converting into two-dimensional range data may include creating an image. The step of converting may include selecting a plane of the three-dimensional representation and indicating range data perpendicular to this plane with different image characteristics. Different image characteristics may include different colors. Range data perpendicular to the selected plane may be indicated using a color gradient mapped to a rectangular coordinate. The proposed method may further include filtering the two-dimensional image. The filtering step may include using one or more filters such as an opening/closing filter or an expanding/contracting filter. The filtering may include reducing noise by combining a plurality of two-dimensional images over a certain period of time.

[0041] The step of comparing the generated three-dimensional representation with the previously generated three-dimensional representation may include comparing two-dimensional images that include range data. The step of comparing may include comparing the two-dimensional images by variable time bases. The step of comparing may include subtracting one image from another. The variable time bases may include a first time base that is shorter than the second time base. The proposed method may further include a step of analyzing the result of comparing the two-dimensional images.

[0042] The proposed method may further include a step of creating a difference image. The proposed method may further include a step of dividing the difference image into separate zones. These zones may correspond to zones of interest of the earthmoving implement. The zones of interest may include the expected location of wear features of the earthmoving implement. The wear features may include one or more such features as teeth, adapters, protective segments and liners. The difference image may be divided into separate zones based on a given geometric model of the earthmoving implement. The step of dividing the difference image into separate zones may include using an analysis by the method of detecting the edges of the image elements. The analysis by the method of detecting the edges of the image elements may be used to identify features of substantially vertical lines. The difference image may be divided into separate vertical zones.

[0043] The comparison step may further include a step of measuring changes in the difference image in each selected zone. Measuring changes in the difference image in each zone may include determining a number of pixels. The step of determining the number of pixels may include counting adjacent pixels. The number of adjacent pixels may indicate areas of wear and/or loss in that zone. The step of determining the number of pixels may include counting pixels in rows. The number of pixels counted in a row may be compared with a threshold value to indicate whether wear and/or loss may have occurred. The threshold value may be predetermined. The threshold value may be adaptive, for example based on a comparison of a value over time or obtained by machine learning. Machine learning may rely on feedback from an operator.

[0044] The step of issuing an indication of wear or loss may include issuing an indication to the operator of the working equipment. The step of issuing an indication may include issuing an alarm signal. The alarm signal may be one or more of such signals as an acoustic alarm signal, a visual alarm signal, and a tactile alarm signal. The alarm signal may be transmitted to the operator of the working equipment. The step of issuing an indication may include transmitting an indication of loss and/or wear and/or an alarm signal remotely. The alarm signal may include a first alarm signal indicating wear and a second alarm signal different from the first alarm signal and indicating loss. The proposed method may further include a step of using the indication of wear or loss in the control systems of the working equipment to adapt its operation.

[0045] The proposed method may further include a step of identifying the associated vehicle. The proposed method may further include identifying the associated vehicle in the event of a loss event being determined. The proposed method may further include determining the associated vehicle to which the detached wear element or other element of the same kind may be delivered during the delivery operation performed by the work equipment.

[0046] The proposed method may further include a step of transmitting data related to the tool, obtained from one or more sensors installed on the working equipment, to a server. The server may perform the steps of generating, comparing, identifying and/or issuing an indication. The proposed method may further include receiving data from the server indicating wear or loss of at least a portion of the tool.

[0047] According to another embodiment of the present invention, a system for monitoring wear elements of a tool, preferably an earthmoving tool of working equipment, is provided, the system including:

one or more sensors mounted on the work equipment that is equipped with a tool located in the reading field of view; and

a processor configured to implement a method for tracking one or more wear elements of a work equipment tool according to the description provided above.

[0048] According to another embodiment of the present invention, there is provided an excavation equipment, such as an excavator, comprising:

earthmoving implement;

one or more wear elements located on an earthmoving implement;

one or more sensors facing one or more wear elements of the earthmoving implement;

a processor communicating with one or more sensors, wherein the processor is configured to:

receiving data from one or more sensors;

generating a three-dimensional representation of at least a portion of the tool using data obtained from one or more sensors;

comparing the generated three-dimensional representation with a previously generated three-dimensional representation;

identifying one or more such events as wear and loss of one or more wear elements by comparing the generated three-dimensional representation with a previously generated three-dimensional representation; and

issuing an indication of wear or loss of one or more wear elements upon identification of such wear or loss.

[0049] The excavator may be a variety of earthmoving and earthmoving equipment, including, for example, crawler excavators, wheeled single-bucket loaders, hydraulic single-bucket excavators, electric-powered rope excavators, rope-scraper excavator buckets, backhoe diggers, underground loaders, bucket wheel excavators and other similar machines.

[0050] Other features and advantages of the present invention will become apparent from the following detailed description.

Brief description of the figures

[0051] Preferred embodiments of the present invention will be described in more detail below, by way of example only, with reference to the accompanying drawings, where:

Fig. 1 illustrates a system for monitoring wear elements of an earthmoving implement;

Fig. 2 shows an example of data from an earthmoving implement sensor;

Fig. 3 shows an example of a schematic representation of an earthmoving tool generated from sensor data;

Fig. 4 shows an example of a comparison of a three-dimensional representation of an earthmoving tool with a previously generated three-dimensional representation;

Fig. 5 shows a thermal image of the earthmoving tool shown in Fig. 4; and

Fig. 6 shows a schematic diagram illustrating one example of the implementation of a wear element tracking system.

Detailed disclosure of the present invention

[0052] Fig. 1 illustrates a system 10 for tracking an earthmoving tool 20 of a working equipment in the form of an excavator 30. It should be understood that the present invention can be applied to other types of vehicles or working equipment. The illustrated excavator 30 is a crawler-type excavator 30. However, it should be understood that the excavator 30 can be another type of machine equipped with an earthmoving tool 20, including, for example, wheeled single-bucket loaders, hydraulic single-bucket excavators, electric-powered rope excavators, rope-scraper buckets, backhoe diggers, underground loaders, bucket wheel excavators and other similar machines. Although the illustrated implement is represented by an earthmoving implement 20, it should also be understood that the present invention can be applied to implements of other types, in particular to implements with replaceable wear parts, such as construction implements, production implements, processing tools or other implements of a similar nature.

[0053] The excavator 30 illustrated in Fig. 1 is provided with a movable boom 40 including a boom 42 and a handle 44. On the movable boom 40, in particular on the handle 44 of the movable boom, one or more sensors 50 can be mounted, in the field of view 52 of which at least a part of the earthmoving implement 20 is located. Depending on the angles between the shafts of the movable boom, the earthmoving implement 20 may not always be in the field of view 52 of the sensor 50, and in the preferred embodiment, the sensors are mounted in a certain position and directed towards the earthmoving implement in such a way that the earthmoving implement 20 moves into its field of view 52 during normal working operations, for example, during a dumping operation.

[0054] The sensor 50 communicates with the processor 60, which is preferably located on the excavator 30, and even more preferably in the excavator cabin 70. However, the processor 60 may also be located remotely so that data from the sensor 50 is transmitted outside the vehicle to a remote terminal. The processor 60 may also be located on the excavator 30 so that processed information, such as analysis results or alarms, are transmitted to the remote terminal for remote monitoring and evaluation.

[0055] The sensor 50 is preferably configured to collect data representing a three-dimensional model of the current state of the earthmoving implement 20, such as, for example, a point cloud, a probability cloud, a surface model, or another model of the same kind. In one of the preferred embodiments, the sensor 50 is a multi-layer, time-of-flight, scanning laser rangefinder sensor (such as, for example, the SICK LD MRS-8000 sensor). However, it should be noted that alternative sensors that may also be used include, but are not limited to, stereoscopic vision systems (operating in both the visible spectrum and any other spectral region, such as thermal imagers), structured three-dimensional illumination-based ranging systems (not time-of-flight), radar, ultrasonic sensors, and devices that can infer structure based on passive or indirect methods, such as detection of the radiation pattern created or modified by the earthmoving tool 20 (e.g., by MRI or passive acoustic analysis).

[0056] In one preferred embodiment, the sensor 30 is a single three-dimensional sensor, but it should also be understood that one or more non-three-dimensional sensors may be used, such as two or more two-dimensional scanners or three-dimensional sensors with a limited field of view that create a complete three-dimensional model in a short period of time. Examples of such configurations include, but are not limited to, a monocular structure consisting of motion-aware sensor systems, including event camera solutions, or a pair of two-dimensional scanning laser rangefinders oriented at a certain angle relative to each other (preferably about 90°) such that one sensor obtains a motion estimate by tracking an earthmoving implement, while the other sensor collects time-varying vector scans of the earthmoving implement, which are then combined into a three-dimensional model using the motion estimate data.

[0057] The entire area of interest of the earthmoving implement may not be captured in a single scan or frame of the sensor 50, but the sensor data may be combined with motion estimates derived from the sensor data as the earthmoving implement 20 moves within the field of view 52 of the sensor 50 to generate a three-dimensional model.

[0058] Depending on the location of the sensor 50, there will likely be an expected spread of ordered distance data that can be considered acceptable, relative to data collected closer or further from the sensor 50 than allows for their reliable rejection. This data may be caused by dirt, dust, the wall of the excavation, or other elements that are not relevant parts of the earthmoving tool 20. Therefore, the processor 60 can pre-process the data received from the sensor using a range selection method, which can significantly reduce the amount of data requiring complex processing.

[0059] Other pre-processing steps may also be provided, depending on the sensor. For example, for the SICK LDMRS-8000 sensor, 50 multiple scans may be provided using interlaced scanning in order to present the data with a wider vertical field of view for analysis. Similarly, a noise reduction operation based on point clustering may be provided for this sensor. Other pre-processing steps may also be used, depending on the type, and in some cases even the brand, of the sensor 50 used.

[0060] If the relevant parts of the earthmoving implement 20 to be tracked by the system 10 are not always in the field of view, then it is necessary to determine when to start and stop collecting data from the sensor 50 for analysis by the processor 60. The moment when the collection of relevant data begins may be referred to as a "trigger" point or event. The trigger point may be identified by analyzing each frame or scanned image of the sensor data and determining the ratio of points where the earthmoving implement 20 is expected to be present to points where the earthmoving implement 20 is expected to be absent. This ratio may be compared to a predetermined threshold value, preferably taking into account the known geometry of the earthmoving implement 20 or a state machine.

[0061] For example, in the case where an excavator bucket is being analyzed, measurements across the entire earthmoving implement 20 from one side of the bucket to the other side of it can be used to simply divide the data from the sensor into areas where a large amount of data is expected to be obtained (for example, where wear elements in the form of teeth are located) and into areas where a smaller amount of data is expected to be obtained (for example, where wear elements in the form of guard segments are located). The ratio of these points can be determined by performing a relatively simple division operation or by using a more complex algorithm, such as, for example, a fuzzy logic "AND" operation using an algebraic product. This value is then compared with a threshold value that compares the number of teeth with the number of guard segments, as well as with the expected field of view of the sensor in the frame where the earthmoving implement 20 can be seen, in order to obtain a ratio for comparison.

[0062] On the other hand, the state machine may contain the following states: wear features are not visible; wear features are partially visible; wear features are completely visible; wear features are partially outside the field of view of the sensors; and/or wear features are completely outside the field of view of the sensor. The determination of the state may be based on heuristic rules, according to which conditions of the spatial distribution of three-dimensional points corresponding to each state are identified. The evaluation may also be supplemented by a rejection mechanism that discards data indicating that wear features may still be obscured, for example, by their entry into the wall of the excavation or by covering with material that is identified as not of any interest. This rejection mechanism may check for the presence of empty data within the known dimensions of the tool. The rejection mechanism may also check the approximate shape (e.g., flat, spherical or elliptical) of the tool by examining the results of a principal component analysis of the three-dimensional points.

[0063] Fig. 2 shows an example of data 100 from a sensor 50 for one frame of an image of a portion of an earthmoving tool 20 of a bucket (not shown) obtained by a scanning laser range finder immediately after reaching a trigger point. The data 100 includes clearly identifiable wear features in the form of teeth 110 and protective segments 120. Although not clearly expressed in Fig. 2, each point contains range information relative to the sensor 50, thereby providing a sufficient amount of data to generate a three-dimensional representation of the earthmoving tool 20.

[0064] To improve reliability, a buffer (preferably a ring buffer) of sensor data may be maintained prior to trigger point detection, and subsequent analysis may be performed on scanned images in this buffer unless a particular scanned image is rejected due to lack of integrity (e.g., insufficient data points, failure to track key features used to create a 3D model, etc.).

[0065] Although it is not essential, multiple sets of data from the sensor 50 are preferably combined over relatively short periods of time to create a more effective three-dimensional representation of the earthmoving implement 20. For most locations of the sensor 50, sensing methods and applications of this technology, it can be assumed that the earthmoving implement 20 will not be visible all the time and that the data from the sensor will vary in its quality due to, for example, noise in the signals, temporary interference (such as, for example, dust or excavated material) and current weather conditions (such as fog or rain). Accordingly, it is advantageous to combine multiple sets of data from the sensor 50 over a given period of time to provide a better representation of the earthmoving implement 20 than using a single set of data. In one preferred embodiment, sensor data is used regarding a single moldboard movement. This is determined by the buffer sizes and the trigger event, as well as the ability to "lock" the movement on the earthmoving implement 20 when processing data during the motion tracking. If no new data is received for a specified period of time, then a data processing event can be triggered.

[0066] A reading method with a corresponding three-dimensional representation broken down into volumetric elements (voxels) can be used to combine multiple sets of data from a sensor into a single model. In one preferred embodiment, in which a scanning laser rangefinder is used, voxelization based on spherical frames can be used, which encodes a description with a ray tracing effect. Multiple points from different scanned images that are included in the same voxel are reduced to a single point using an appropriate statistical model (such as, for example, a median model). The resolution of the model can be determined based on the required accuracy of reproduction of the results of measuring wear or loss and the capabilities of the sensor 50 during its use. Alternatively, the data can be immediately combined into a two-dimensional image with a coordinate grid with a similar statistical fusion of data from multiple scanned images. A statistical model such as an integral average can be used. The performance can also be improved using other statistical models, such as the one-dimensional Kalman filter.

[0067] Fig. 3 shows an example of a three-dimensional representation 200 of an earthmoving implement 20 created by combining multiple sets of three-dimensional sensor data measured with respect to a single moldboard movement using voxelization of spherical coordinates. The representation 200 includes clearly identifiable wear features in the form of teeth 210 and protective segments 220. Once the three-dimensional representation 200 is generated, it can be compared with a previously generated three-dimensional representation. The current three-dimensional representation 200 is also preferably saved so that it can be used as a previously generated three-dimensional representation when performing such comparisons in the future.

[0068] Over time, during the operation of the earthmoving implement 20, a plurality of three-dimensional representations of it are collected. Depending on the installation configuration of the sensor 50, these representations may be collected in a common coordinate system due to a certain location of the sensor 50 and the earthmoving implement 20 relative to each other. Otherwise, they may be in different spatial coordinate systems or cannot be easily aligned for comparison. In such cases, the collected three-dimensional representations 200 of the models of the earthmoving implement 20 are preferably transformed so that they are aligned in a common coordinate system, allowing the processor 60 to perform an accurate comparison.

[0069] A variety of approaches may be used to align the plurality of three-dimensional representations 200 so that they are aligned in a common coordinate system. A preferred approach involves using a reference three-dimensional representation 200 and aligning all other three-dimensional representations 200 to this reference representation. This reference three-dimensional representation may be simply the first three-dimensional representation generated during or after startup, or any other representation generated at any stage of the process, as long as it is used properly.

[0070] The alignment is preferably performed using an ICP (Iterative Closest Point) process. The ICP process is preferably limited in the degrees of freedom assumed. For example, the bucket of a hydraulic front shovel excavator can only move in two dimensions and rotate about one axis relative to the sensor 50 mounted on the handle 44. Another example of an alignment algorithm is the calculation of a fractional linear transformation matrix based on the matching of key points between the reference representation and intermediate representations in a two-dimensional image space, supplemented by a stage of appropriate color normalization for range alignment or rotation about an unlimited axis.

[0071] Another example of a suitable alignment algorithm is the NDT (Normal Distribution Transformation) process. Another example of a suitable alignment algorithm involves alignment in two steps: first by applying a coarse alignment mechanism, such as a rotation determined by alignment according to a given coordinate system along the main axis of the point cloud determined by the principal component analysis (PCA) method, together with translation based on centroids, boundaries, statistical or other geometric features; and then by subsequently applying a fine alignment mechanism, such as applying the ICP or NDT process.

[0072] After alignment (if necessary), the processor 60 compares the newly generated three-dimensional representation with at least one previously generated three-dimensional representation. This comparison serves primarily to detect changes in the earthmoving implement 20 over a given period of time. A significant change in the three-dimensional representation 200 of the earthmoving implement 20 over a relatively short period of time, such as between moldboard movements, indicates an event of loss or breakage of the earthmoving implement 20. A less significant change over a longer period of time indicates abrasive wear of the earthmoving implement 20.

[0073] In one preferred embodiment, the three-dimensional representations 200 are transformed into a two-dimensional range map using image processing techniques such as, for example, superimposing a Cartesian coordinate system on the three-dimensional model and projecting the range measurements obtained from the model onto a pair of axes of the Cartesian coordinate system and coloring each pixel according to the range value along a third, perpendicular axis. Additional filtering operations, such as opening/closing or widening/narrowing, may be applied to the image to fill in gaps caused by noise or to otherwise generally improve the quality of the image. Combining images over appropriate periods of time may also be used to further reduce the impact of transient noise, such as shifting the average of images in a window of only a few minutes or removing pixels that are only observed in a small portion of the images.

[0074] The processor 60 then compares the images over variable time bases. For example, when an image subtraction operation is performed, differences in the images are highlighted. These comparisons can be performed to detect wear and loss events during time periods allocated to the object of interest. For example, when the current image is compared with the last previous image, significant changes in the condition of the earthmoving implement 20 are highlighted that indicate an event of loss or breakage of the earthmoving implement 20. When an image created in a small moving average window, for example, with only a few moldboard events, is compared with a similar image obtained the day before or even earlier, wear of the earthmoving implement 20 should be evident, both in the depth (color) of the difference image and in the change in the boundary of the features of the earthmoving implement 20 in the difference image.

[0075] Fig. 4 is a visualization of an example 300 of a comparison of a three-dimensional representation of an earthmoving implement with a previously generated three-dimensional representation in a two-dimensional image format. The representation difference image 300 includes clearly identifiable wear features in the form of teeth 310 and protective segments 320. In this example, a relatively long period of time is provided between the compared representations, where both wear in the form of relatively small dark areas 330 along the perimeter of the wear features and tooth tip loss in the form of a relatively larger dark block 340 of one of the teeth 310 can be seen. Depth wear is also visible. Fig. 5 is a thermal photographic image of the earthmoving implement 20 shown in Fig. 4 illustrating tooth tip loss 340 highlighted in the difference image 300.

[0076] After the comparison, the processor 60 may identify wear and/or loss of the earthmoving implement 20. The process for such identification may be selected so that it is suitable for the earthmoving implement 20 being monitored and the type of determination required. In one preferred embodiment, in which the determination is to identify wear and/or loss of at least a portion of the illustrated earthmoving implement 20, a process of image convolution with a kernel scaled and weighted in an appropriate manner may be applied to the difference image. Alternatively, a pixel row counting algorithm may be applied to the difference image.

[0077] The application of a convolution filter to the difference image is performed using a square kernel with a linearly increasing weight in the direction from the edge to the center. The kernel size is chosen so that it is approximately the same as the size of the object that can reasonably be expected to be lost, or slightly larger when converted to pixels through scaling and resolution of the difference image. An analysis of the value obtained from the convolution operation is used to identify the loss or amount of wear and to determine the location of the wear or loss zone. This value is compared with a predetermined threshold value or an adaptive threshold value. One example of a predetermined threshold value may be based on a portion of the maximum possible result in the case of a severe loss. It may be manually adjusted to change the sensitivity of the detection. One example of an adaptive threshold value may be based on a comparison of a value over time and the identification of changes in the value that would indicate statistical outliers, or on a machine learning technique in which the threshold value is determined through feedback to the operator regarding the accuracy of the detection.

[0078] The image used for comparison is preferably divided into vertical zones corresponding to the expected location of the teeth and protective segments, based on a given geometric model of the earthmoving tool 20, which is usually determined by knowledge of its geometry. Such zoning is preferably performed automatically, for example using an algorithm for detecting the boundaries of image elements and monitoring the features of substantially vertical lines.

[0079] For each vertical zone, starting from the edge of the image adjacent to the tops of the teeth and iterating row by row toward the base of the teeth, each adjacent row of pixel values in the difference image is preferably counted that indicate its absence in the most recent model when compared to the earlier model. The number of missing rows for each pixel is compared with any known threshold value, such as, for example, a predetermined threshold value or an adaptive threshold value.

[0080] One example of a predetermined threshold value may be based on the difference between the length of a fully worn element of the earthmoving implement 20 when it is mounted on the working member and the length of the working member or fastener without the element of the earthmoving implement 20, converted into pixels by scaling and resolving the image of the difference. Examples of adaptive threshold values may be based on comparing a value over time and identifying changes in this value that may indicate statistical outliers, or on a learning technique in which the threshold value is determined through feedback to the operator regarding the accuracy of the detection.

[0081] After detecting wear or loss, the processor 60 provides an indication of such wear or loss. This output may be in a variety of forms, but in a preferred embodiment, the output comprises an alarm signal for sending a notification of the identified wear or loss event. In a preferred embodiment, the notification is sent to at least the operator of the work equipment 30.

[0082] In addition to simply notifying about wear or loss, the processor 60 is preferably configured to identify and provide an indication of any other useful characteristics detected, such as identifying the occurrence of excessive wear (e.g., faster than expected wear of at least a portion of the earthmoving implement 20). This can be implemented by comparing pixels or groups of pixels, including moving and overlapping windows of pixels, in the difference images constructed from variable time bases with some predetermined baseline indicators indicating an acceptable wear rate. The acceptable wear rate can be provided either for the entire earthmoving implement 20, or for specific parts of the earthmoving implement 20, such as, for example, one or more specific wear elements.

[0083] Other useful notifications may be generated as a result of multivariate and spatial correlation with data received from other systems. For example, wear rate may be correlated with one or more types of extracted material, time of day, vehicle driver, etc. Such notifications may be useful for interpretive analysis, for example, by providing the system with auxiliary input data for automatic detection of changes in the properties of the extracted material.

[0084] The output data may include an alarm signal. For example, the alarm signal may be generated upon detection of an event of loss of the earthmoving implement 20.

A user interface may be provided. Such a user interface may provide an alarm signal to the operator of the working equipment 30. Output data may also be transmitted to other systems of the working equipment, including control systems of the working equipment 30. Examples of how alarm signals may be transmitted to the operator include, but are not limited to, the transmission of one or more signals such as an acoustic alarm signal, a visual alarm signal, and a tactile feedback alarm signal. The alarm signals preferably distinguish between wear and loss events. The output data preferably also informs the operator which part of the earthmoving implement 20 has been identified as lost or worn. Such information is also preferably available through an application programming interface and/or a digital output for use by other systems, such as a control system, or during remote operations.

[0085] A vehicle identification system may be provided, preferably a system based on radio frequency identification (RFID), or a global positioning system (GPS), or a vehicle monitoring system (FMS), which may use a combination of methods, such as an FMS for trucks. With such a vehicle identification system, an associated vehicle may be identified, such as, for example, a transport vehicle, into which the material extracted by the working equipment 30 is loaded. When an event of loss of an element of the earthmoving implement 20 occurs, the vehicle identification system may be used to identify the vehicle that most likely contains the lost part of the earthmoving implement 20.

[0086] The processor 60 may also be configured to provide a tracking history. Such a tracking history allows the operator to view difference image information, three-dimensional models, data or images from the sensors 50 themselves (if applicable, depending on the sensing method) and/or images from adjacent sensors (such as a thermal imager) to help the operator identify the current state of the earthmoving implement 20 and/or changes in state over previous periods. Such a tracking history may be used to analyze the loss event, whereby manual processing of data from previous periods may be used to compensate for any delays in detection by the system. For example, a lower false alarm rate may be achieved by widening the averaging window and comparison periods at the expense of a possibly longer delay between the actual occurrence and the system-identified loss event of an element of the earthmoving implement 20.

[0087] The processor may also be configured to transmit data. The data is preferably transmitted to a remote terminal. Such data may include one or more three-dimensional representations of the earthmoving implement 20, difference image information, and alarms. Such data is preferably sent from the working equipment 30 to a remote server or cloud environment for additional processing, analysis, monitoring, and/or reporting. The remote server or cloud environment may complement the local processor of the working equipment or even perform certain data processing instead of the local processor on the working equipment. Examples of some indicators that can be extracted from such information include an estimate of the wear rate and service life of the earthmoving implement 20. Such information can be supplemented with information from other sources, such as, for example, specific digging energy and economic constraints or indicators and prices for parts of the earthmoving implement to obtain recommendations regarding replacement periods for the earthmoving implements 20, or it can be connected as input data to other systems, such as, for example, automatic ordering systems for the earthmoving implements 20. The processor can be configured to receive input data from the operator of the work equipment indicating the occurrence of a wear or loss event. Upon receipt of such input data, an analysis can be performed and/or a corresponding notification can be transmitted remotely. This notification can be used to alert about the need for maintenance of the work equipment. Such an "on-demand" approach can mean that longer intervals between maintenance inspections can be provided for the work equipment implement.

[0088] In addition to determining wear or loss, the processor may be configured to determine the time when a wear element should be replaced by identifying positive rather than negative changes in the difference image. Such information may be used to determine the service life of wear parts and/or replacement patterns. Analysis of the difference image may also be used to recognize the shape of the wear part. This may be used to identify the wear part in use for the purpose of determining operator preferences and/or actions. A suitability analysis may be performed in which wear and/or loss characteristics of identified wear elements may be determined. After the suitability analysis is performed, recommendations may be issued for replacement of specific wear elements.

[0089] Fig. 6 is a schematic diagram illustrating one example of an implementation of a wear feature tracking system that includes sensors 500, a processor 600, and an output 700. The processor 600 is configured to: receive data related to an earthmoving implement 20 from one or more sensors at 610; generate a three-dimensional representation of at least a portion of the implement using the received data at 620; compare the generated three-dimensional representation with a previously generated three-dimensional representation 630 at 640; identify one or more events such as wear and loss of at least a portion of the earthmoving implement by comparing the generated three-dimensional representation with the previously generated three-dimensional representation at 650; and provide 700 an indication of wear or loss at 660 if wear or loss of at least a portion of the earthmoving implement is identified.

[0090] The present invention provides an advantage in that it provides a tracking system 10 and a corresponding method for identifying lost and worn wear elements of an earthmoving implement 20. This can improve productivity, since excavation work with damaged earthmoving implements 20, such as with worn or disconnected wear elements, is less efficient by default. Moreover, identifying a loss event in an earthmoving implement 20 allows for a quick replacement of the lost element, which prevents other potential problems from occurring on the work site, such as damage to associated equipment.

[0091] The tracking system 10 also provides for a preventive maintenance mode, whereby wear parts of the earthmoving implement 20 can be tracked and replaced when they reach a predetermined wear condition to avoid unplanned downtime.

[0092] In a preferred embodiment, a three-dimensional representation of the earthmoving implement 20 can be created and fixed in an appropriate coordinate system. This provides the advantage that the algorithm can be independent of the reading method used to create the representation. By combining multiple models over short periods of time, noise in the three-dimensional representation can be reduced. Two three-dimensional representations generated at different points in time by a relatively simple computationally subtraction operation (such as, for example, subtraction of a range image) can highlight differences in the condition of the earthmoving implement 20 in the time period between the collection of the respective data sets. Repeating this operation over a variable time scale can be used to detect different scales of wear and to obtain different degrees of sensitivity to major changes (such as, for example, a loss event).

[0093] It should be borne in mind that the proposed system and method can be applied to any earthmoving implement 20, for example, for tracking wear parts installed on buckets, a backhoe ditcher, a front shovel excavator, a wheeled single-bucket loader, a rotary bucket excavator and drilling rigs.

[0094] Depending on the embodiment, minimal, if any, prior art knowledge of the geometry of the earth-moving implement 20 is required. The output data may be presented in a relatively simple format (such as, for example, a range difference pattern) that can be queried using standard image processing techniques to obtain a large amount of information about the condition of the earth-moving implement 20, compared to, for example, relatively simple linear measurements of tooth length. The output data may also be easily combined with other data sources to significantly increase the utility of the measurement to obtain more in-depth knowledge of the earth-moving implement 20. The system and method provide a high degree of reliability due to a low, and typically easily controlled, false alarm rate. The output data may also be presented in a format suitable for operator alerting and external monitoring and/or compact display of information.

[0095] Adjectives used in this document such as "first" and "second," "left" and "right," "upper" and "lower," and other adjectives of a similar nature may be used solely to distinguish one element or action from another element or action, without requiring or implying that in practice such relationship or order necessarily be as stated. Where the context permits, reference to an integer, component, or step (and the like) shall not be construed as being limited to only one integer, component, or step, but may include one or more integers, components, or steps, and the like.

[0096] The above description of various embodiments of the present invention is provided for those skilled in the art. It is not intended to be exhaustive or to limit the claimed invention to a single embodiment. As noted above, many alternative embodiments of the present invention and the possibility of making various changes thereto will be apparent to those skilled in the art. Accordingly, although some specific alternative embodiments of the claimed invention have been described, other embodiments will be apparent to those skilled in the art or can be devised with relative ease. It is intended that the present invention cover all alternative embodiments, modifications and variations made to the claimed invention that are presented herein, as well as other embodiments thereof that fall within the spirit and scope of the invention described above.

[0097] As used herein, the terms “comprises,” “comprising,” “includes,” “including,” or other terms of similar nature are intended to mean a non-exclusive inclusion, meaning that a method, system, or device comprising the listed elements does not exclusively include those elements, but may also include other elements not listed.

Claims (72)

1. A system for tracking the implement of working equipment, wherein this system includes: one or more sensors mounted on the work equipment and facing the implement; and a processor configured to: obtaining data related to the tool from one or more sensors; generating a three-dimensional representation of at least a portion of the tool using the obtained data; comparing the generated three-dimensional representation with a previously generated three-dimensional representation; identifying one or more events such as wear and loss of at least a portion of a tool by comparing the generated three-dimensional representation with a previously generated three-dimensional representation, wherein the processor is configured to generate a three-dimensional representation by aligning a plurality of three-dimensional models using a method of combining three-dimensional models in a common coordinate system, wherein generating a three-dimensional representation of at least a portion of an implement using data obtained from one or more sensors comprises using a processor configured to collect a plurality of vector scans over a period of time to generate the three-dimensional representation using motion estimation data, wherein the motion estimation data is derived from data obtained from the one or more sensors as the implement moves within the field of view of the one or more sensors. 2. The tracking system of claim 1, wherein the implement is an earthmoving implement with replaceable wear parts. 3. The tracking system according to claim 1 or 2, wherein one or more sensors include at least one sensor configured to obtain data displaying the three-dimensional shape of the surface of the tool. 4. The tracking system of claim 3, wherein one or more sensors include a multi-layer, time-of-flight, scanning laser rangefinder sensor. 5. The tracking system of claim 3, wherein the one or more sensors include a two-dimensional sensor that provides sufficient two-dimensional data to infer a three-dimensional shape of the surface of the implement. 6. A tracking system according to any of the preceding paragraphs 1-5, in which the processor is configured to combine data from sensors read in different ways and having different fidelity and/or noise characteristics in order to generate a three-dimensional representation. 7. The tracking system according to claim 6, wherein the processor is configured to combine data using a combinatorial algorithm. 8. A tracking system according to any of the preceding paragraphs 1-7, in which one or more sensors are located on the working equipment such that the earthmoving implement is in their line of sight. 9. The tracking system according to item 8, in which one or more sensors installed on the working equipment are located on a movable bar of the working equipment. 10. A tracking system according to any one of the preceding paragraphs 1-9, in which the processor is configured with the additional capability of pre-processing data received from one or more sensors before generating the three-dimensional representation. 11. The tracking system of claim 10, wherein the pre-processing includes range selection. 12. A tracking system according to claim 10 or 11, wherein the pre-processing provides for multiple scans using interlaced scanning. 13. The tracking system of claim 1, wherein the alignment involves using an ICP (Iterative Closest Point) process or an NDT (Normal Distribution Transformation) process. 14. A tracking system according to any one of the preceding paragraphs 1-13, wherein the alignment comprises determining a fractional linear transformation matrix. 15. A tracking system according to any one of the preceding paragraphs 1-14, in which the processor is configured with the additional capability of converting the generated three-dimensional representation into two-dimensional range data. 16. The tracking system of claim 15, wherein the processor is configured with the additional capability of comparing the generated three-dimensional representation with a previously generated three-dimensional representation by comparing two-dimensional range data. 17. The tracking system according to claim 15 or 16, in which the processor is configured with the additional capability of identifying one or more events such as wear and loss of at least a portion of the weapon by analyzing the result of comparing two-dimensional images that include two-dimensional range data. 18. The tracking system of claim 17, wherein the analysis comprises creating a difference image divided into separate zones that correspond to the zones of interest on the weapon. 19. The tracking system according to claim 18, in which the difference image is divided into separate zones based on a given geometric model of the weapon and/or analysis using the method of detecting the boundaries of image elements. 20. A tracking system according to any one of the preceding paragraphs 17-19, wherein the analysis comprises measuring changes in the difference image in each zone by determining the number of pixels and/or applying an image convolution process. 21. A tracking system according to any of the preceding paragraphs 17-20, wherein the analysis includes noise suppression using an image mask, which prevents analysis of a portion of the image that is deemed irrelevant. 22. A tracking system according to any of the preceding paragraphs 1-21, in which the processor is configured to provide an indication of wear or loss of a part of the tool. 23. The tracking system according to any one of the preceding paragraphs 1-22, further comprising a vehicle identification system comprising one or more sensors that provide identification of the associated vehicle upon establishing the fact of loss of at least a part of the tool. 24. A tracking system according to any one of the preceding paragraphs 1-23, in which the processor is configured with the additional capability of recording and/or transmitting global navigation satellite system (GNSS) coordinates upon identifying the loss of at least a portion of the weapon. 25. A method for tracking one or more wear elements of a work equipment tool, wherein this method provides for: obtaining data related to the tool from one or more sensors installed on the work equipment; generating a three-dimensional representation of at least a portion of the tool using data obtained from one or more sensors, wherein the generation of the three-dimensional representation is carried out by aligning a plurality of three-dimensional models using a method of combining three-dimensional models in a common coordinate system; comparing the generated three-dimensional representation with a previously generated three-dimensional representation; and identifying one or more events such as wear and loss of at least a portion of the tool based on the result of performing the comparison step, wherein generating a three-dimensional representation of at least a portion of an implement using data obtained from one or more sensors comprises using a processor configured to collect a plurality of vector scans over a period of time to generate the three-dimensional representation using motion estimation data, wherein the motion estimation data is derived from data obtained from the one or more sensors as the implement moves within the field of view of the one or more sensors. 26. The method according to claim 25, wherein the tool is an earthmoving tool containing wear elements, and the step of identifying one or more such events as wear and loss of at least a portion of the earthmoving tool involves identifying one or more such events as wear and loss of at least wear elements of the tool. 27. The method according to claim 25 or 26, wherein the step of obtaining data relating to the tool from one or more sensors comprises obtaining three-dimensional data relating to at least a portion of the tool. 28. The method according to any of the preceding paragraphs 25-27, further comprising pre-processing the data from the sensor before the start of the three-dimensional representation generation stage. 29. The method of claim 28, wherein the pre-processing comprises multiple scans using interlaced scanning. 30. The method according to any one of the preceding paragraphs 25-29, further comprising the step of determining that the earthmoving implement is located substantially within the field of view of one or more sensors. 31. The method of claim 30, wherein the step of assessing that the weapon is within the field of view of one or more sensors includes identifying whether the data from the sensor indicates that at selected points the weapon is identified as present or absent. 32. The method according to claim 30 or 31, wherein the step of assessing that the weapon is in the field of view of one or more sensors involves determining the ratio of points at which the weapon is assumed to be located to points at which the weapon is assumed to be absent. 33. The method according to any of the preceding paragraphs 30-32, in which the step of assessing that the weapon is in the field of view of one or more sensors includes comparing said ratio with a given threshold value based on known geometry of the weapon. 34. The method according to any one of the preceding paragraphs 25-33, further comprising the step of combining multiple sets of sensor data obtained at different times into a single three-dimensional model. 35. The method of claim 34, wherein the step of combining multiple sets of sensor data comprises combining sensor data over a period of time starting from the moment when it has been determined that the implement is substantially within the field of view of one or more sensors. 36. The method according to claim 34 or 35, wherein the combining step comprises voxelizing the data from the sensor. 37. The method according to claim 25, wherein the leveling step involves using an ICP (Iterative Nearest Point) or NDT (Normal Distribution Transformation) process. 38. The method according to any one of the preceding paragraphs 25-37, in which the leveling step comprises determining a fractional-linear transformation matrix. 39. The method according to claim 38, wherein the fractional linear transformation matrix is based on matching key points between the reference model and intermediate models. 40. The method according to any one of the preceding paragraphs 25-39, further comprising converting the three-dimensional representation into two-dimensional range data. 41. The method according to claim 40, wherein the step of converting into two-dimensional range data comprises creating an image by selecting a three-dimensional representation plane and indicating range data perpendicular to said plane with different image characteristics. 42. The method of claim 41, further comprising filtering the two-dimensional image by applying one or more opening/closing and expanding/contracting filters and reducing noise by combining a plurality of two-dimensional images over a certain period of time. 43. The method according to claim 41 or 42, wherein the step of comparing the generated three-dimensional representation with the previously generated three-dimensional representation comprises comparing two-dimensional images that include range data. 44. The method according to any of the preceding paragraphs 41-43, in which the comparison step comprises comparing two-dimensional images over variable time bases. 45. The method of claim 44, wherein the variable time base includes a first time base that is shorter than the second time base. 46. The method according to any one of the preceding paragraphs 41-45, further comprising the steps of creating a difference image and dividing the difference image into separate zones that correspond to zones of the tool of interest. 47. The method of claim 46, wherein the areas of interest include the expected location of tool wear elements. 48. The method according to claim 47, wherein the tool is an earthmoving tool and the wear elements are one or more such elements as teeth, adapters, protective segments and inserts. 49. The method according to any of the preceding paragraphs 46-48, in which the difference image is divided into separate zones based on a given geometric model of the weapon and/or the results of an analysis using the method of detecting the boundaries of image elements. 50. The method according to claim 49, wherein the comparison step further comprises a step of measuring changes in the difference image in each selected zone. 51. The method according to claim 50, wherein measuring changes in the difference image in each zone comprises performing a step of determining the number of pixels by counting adjacent pixels and/or applying a convolution process. 52. The method of claim 51, wherein the step of counting adjacent pixels comprises counting pixels in rows, and the number of pixels counted in a row is compared with a threshold value to indicate whether wear and/or loss has occurred. 53. The method according to any one of the preceding paragraphs 25-52, further comprising the step of providing an indication of wear or loss of at least a portion of the implement after performing the step of identifying one or more such events as wear and loss of at least a portion of the implement, during which such wear or loss was identified. 54. The method according to claim 53, wherein the step of issuing an indication of wear or loss comprises issuing an alarm signal to the operator of the working equipment and/or transmitting one or more of the following types of information: an indication of loss, wear and an alarm signal to a remote terminal. 55. The method according to paragraph 53 or 54, further comprising the step of using wear or loss indication in the control systems of the working equipment to adapt its operation. 56. The method according to any of the preceding paragraphs 25-55, further comprising the step of identifying the associated vehicle when a loss event is indicated. 57. A system for monitoring wear elements for a work equipment tool, the system comprising: one or more sensors mounted on the work equipment that is equipped with a tool located in the reading field of view; and a processor configured to implement the method according to any of the preceding paragraphs 25-56.