DE102018200232A1 - Method for operating a work machine and work machine - Google Patents

Abstract

The present invention relates to a method for operating a working machine (300) with at least one working tool (318) and a correspondingly equipped working machine, wherein the following steps are provided:
- detecting (200) at least one movement trajectory of the at least one work tool (318) and / or the work machine (300) in an ongoing work process with an experienced operator;
- analyzing (205) the detected at least one motion trajectory based on at least one predeterminable motion feature;
- learning (235) the at least one predeterminable motion feature for the experienced operator;
- comparing (245) the at least one learned motion feature with a corresponding motion feature of a less experienced operator;
Assisting (240) the less experienced operator based on the results of the comparison (245).

Description

The invention relates to a method for operating a work machine and a work machine, according to the preambles of the respective independent claims. The present invention also relates to a computer program, a machine-readable data carrier for storing the computer program and an electronic control unit, by means of which the method according to the invention can be carried out.

State of the art

Out DE10 2005 019 820 A1 discloses an operating method for work machines having at least one work tool, wherein at least one operating characteristic of the work machine indicating an application of the working tool of the work machine is detected. Based on the detected operating characteristic, the operation of the work machine is adapted to a new application of the work tool.

The operational data of the work machine thus obtained may be the hydraulic pressure, the hydraulic flow, the working tool position, or any other information necessary for determining an application being executed, the level of training or experience of the operator used tool are suitable. This data can be obtained when the work machine and the respective work tool are executing the respective application. The level of experience of the operator may also be determined based on measured data from fluid pressure sensors, fluid flow sensors, and tool position sensors.

The level of experience of the operator may also be identified by generating an experience level signature, comparing the experience level signature thus generated with other known experience level signatures to determine if appropriate a match. The work machine can then be switched to a suitable operating mode based on both the identified application and the determined experience level of the respective operator. As a result, the efficiency when using here affected tools can be maximized.

Disclosure of the invention

The invention is based on the idea to learn the control behavior of experienced operators and to use the knowledge thus learned to assist inexperienced operators. This is based on the finding that an optimal simultaneous movement of a preferably mobile or movable machine in multi-axis movements of tools, especially in unstructured environments, only long-standing experienced or experienced operators succeed without any necessary corrective interventions. The inexperienced operator is more likely to sequentially control the movement of the work machine and tools, and must reposition them repeatedly, or oversteer or understeer the tools during the work process.

In addition, an inexperienced or inexperienced operator is usually overwhelmed at the same time to pay attention to the static and dynamic stability of the machine, to avoid collisions with obstacles and to keep the machine at an optimal operating point or to move along optimal Bewegungsstrajektorien.

• - Erfassen wenigstens einer Bewegungstrajektorie des mindestens einen Arbeitswerkzeugs und/oder der Arbeitsmaschine in einem laufenden Arbeitsprozess mit einem erfahrenen Bediener;
• - Analysieren der erfassten wenigstens einen Bewegungstrajektorie auf der Grundlage wenigstens eines vorgebbaren Bewegungsmerkmals;
• - Lernen des wenigstens einen vorgebbaren Bewegungsmerkmals für den erfahrenen Bediener;
• - Vergleichen des wenigstens einen erlernten Bewegungsmerkmals mit einem entsprechenden Bewegungsmerkmal eines weniger erfahrenen Bedieners;
• - Assistieren des weniger erfahrenen Bedieners auf der Grundlage der Ergebnisse des Vergleichs.

The proposed method for operating a work machine with at least one work tool provides in particular the following steps:

• Detecting at least one movement trajectory of the at least one working tool and / or the working machine in a running working process with an experienced operator;
• - analyzing the detected at least one movement trajectory on the basis of at least one predetermined movement feature;
• - learning the at least one predetermined movement feature for the experienced operator;
• Comparing the at least one learned motion feature with a corresponding motion feature of a less experienced operator;
• Assist the less experienced operator based on the results of the comparison.

According to the proposed method, there is an analysis of preferably spatial (3D) movement trajectories of at least one tool and / or the movement of the present work machine in a running work process. In this case, empirically predeterminable movement characteristics, eg speeds, accelerations, positions, etc., can be learned, controlled by experienced operators, specific to the work process and tool and / or machine specific. In other words, the movement characteristics can be learned for various work processes, tools and / or machines or the respective movement characteristics can be recorded and analyzed, if an experienced operator has the Machine operated. The features thus learned can be compared with corresponding, already analyzed features of a current and less experienced operator and appropriate hints are given to the less experienced operator to optimize the spatial motion trajectories of tools and mobile work machines in the work process. The output of the notes can be displayed by means of a visual control assistant, eg on a display device (display).

Alternatively or additionally, individual, learned features of an experienced operator can be taken over by the inexperienced operator, and thus individual movements can be precontrolled and / or limited and / or completely taken over by the control system.

The advantage of the proposed method lies in a more accurate, faster, safer and more fuel-efficient driving and guiding of tools and mobile machines. In addition, an optimal control of a work machine for a particular work process can be trained or trained to a less experienced or less experienced operator.

For detecting a spatial movement trajectory known per se, already used in construction machines direct measurement and positioning systems or position systems can be used. Thus, known measuring and position systems for construction machines, in which by means of inclination and / or angle sensors and / or gyro and / or acceleration sensors, e.g. on an excavator arm and a blade whose position relative to an excavator cabin and / or can be determined absolutely to the environment or to the terrain, are used. Alternatively or additionally, GNSS systems ("Global Navigation Satellite Systems") come into consideration, which allow an absolute position determination of a tool and / or the machine, relative to the environment or to the surrounding terrain.

Also, measuring devices known per se which provide the route from e.g. Determine an excavator arranged adjusting cylinders of a tool by means of inductive, magnetic or ohmic measuring principles are used. Thus it can be closed on the known geometric data of excavator arm and blade on the position of a cutting edge of the bucket teeth, relative to the working machine.

Alternatively or additionally, known LIDAR ("Light Detection And Ranging") measuring systems can be considered, which provide an optical measuring method for locating and measuring the distance of objects in the room. The LIDAR method uses ultraviolet, infrared, or visible light rays. As a measurement method, e.g. a "time of flight" measurement in which one or more light pulses are emitted and reflected on an existing object, wherein the time is detected until the reception of the reflected signal.

Alternatively or additionally, for detecting the spatial movement trajectory or for learning the control behavior of an experienced operator, known methods for viewing the gaze direction of persons can be used, e.g. Method for detecting the direction of sight of persons by means of a camera. For this purpose, a so-called personalization, i. Face recognition and learning of the line of sight of the person take place, whereby typical facial features are detected and recognized. The person to be trained has to fix a symbol / point on a display device with their eyes, which is then moved over the display surface of the display device several times. Ultimately, this learning process must be done once for each person on the respective machine. During operation of the machine, the camera recognizes the specific person and, due to the personalized learning data, can associate the viewing direction with millimeter precision to a position on the display device. The camera (e.g., an infrared camera) and the display are positioned defined to each other.

Alternatively or additionally, methods and measuring systems known per se in the field of stereoscopy can also be used, in which two or more images of the same working scene are taken from different camera positions. From the location of a particular scene point in the at least two images can be determined its spatial position with knowledge of the intrinsic and extrinsic calibration parameters of the camera.

Alternatively or additionally, spatial (3D) road surface profiles can be created by means of which a spatial surface profile map of the ground in the field of view of the camera can be created directly and very accurately. Thus, in off-road applications, "driving tunnels" or "suitable paths" can be recommended to the driver or predicted the trajectory at the current steering angle position and displayed on a display device by means of the known "augmented reality" method. by means of a lane black in the picture. This also allows real-time operation, e.g. warn of bad passable areas. In order to obtain a horizontal image of the 3D road surface, also the proper movement of the working machine must be considered.

With the proposed method, work processes can also be optimized in order to increase the safety of persons and to achieve a wear-reduced and resource-saving use of machines and tools as well as an effective collision avoidance.

The invention can be used in particular in construction machines or agricultural machines equipped with at least one tool.

The computer program according to the invention is set up to carry out each step of the method, in particular if it runs on a computing device or a control device. It allows the implementation of the method according to the invention in an electronic control unit, without having to make structural changes to this. For this purpose, the machine-readable data carrier is provided on which the computer program according to the invention is stored. By loading the computer program according to the invention onto an electronic control unit, the electronic control unit according to the invention is obtained, which is set up to control a working machine affected here by means of the method according to the invention.

Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

It is understood that the features mentioned above and those yet to be explained below can be used not only in the particular combination indicated, but also in other combinations or in isolation, without departing from the scope of the present invention.

list of figures

• 1 shows an embodiment of the method according to the invention with reference to a flowchart.
• 2 shows a set up for the application of erfindungsgebmäßen method, schematically illustrated work machine.

Description of exemplary embodiments

The in 1 shown method for operating a mobile work machine, such as a shovel, is based initially on the assumption that an experienced operator of the machine, the static and dynamics, eg inertia, the reaction times, present attacks a tool and the performance or the performance limits, the tipping tendencies , the vibration behavior, etc. of the machine knows. When controlling the machine, he takes this knowledge into account with foresight or feedforward. In the method, the control behavior of the experienced operator is recorded depending on the working sequence, together with the sensory movement trajectories or the respective motion kinematics of the machine and of the respective tool or tools 200 and analyzed according to empirically definable criteria 205 , These criteria 210 serve as an indicator for the level of experience or the level of experience of the respective operator.

• - Das Umfeld wird mittels Umfeld-Sensoren wie z.B. Kamera und/oder LIDAR (= „Light Detection And Ranging“) und/oder Radar usw. erfasst, wodurch insbesondere auch die Bodenoberflächenstruktur und darauf befindliche, im Vorfeld bereits klassifizierte Objekte erkannt werden können. Aus dem erfassten Umfeld wird mittels maschinellen Sehens und Objektklassifikation erkannt, dass es sich z.B. um eine Baugrube handelt, die ausgehoben wird. Bei der erkannten Baugrube wird zudem erkannt, dass diese systematisch bzw. schrittweise größer bzw. tiefer wird und und dass eine Arbeitsmaschine bzw. ein entsprechendes Fahrzeug vorhanden ist, welches zunehmend mit Abraum beladen wird;
• - Die Fahrzeugposition bzw. -ausrichtung wird mittels GNSS, RTK (Echtzeitkinematik), Beschleunigungssensoren sowie eines Gyrosensors bestimmt;
• - Die Bewegungen der Werkzeuge werden direkt z.B. mittels Weg- und/oder Winkel- und/oder Beschleunigungs- usw. Sensoren und/oder indirekt mittels Kamera und/oder LIDAR usw. gemessen.

From the movement kinematics of the working machine and the tools thus detected, the proposed method can conclude individual movement processes of the (entire) working process 215 , In this case, it is also possible to use, for example, global position data of the machine recorded by means of GPS in an environment which is also sensory. Thus, certain work processes or work sequences can be detected, for example, excavation of soil until its unloading on a transport vehicle. This recognition is based in particular on the following, detected by means of suitable sensors 220 and data evaluated by a computer 225 for generating motion trajectories for the machine and / or the tools:

• - The environment is detected by means of environment sensors such as camera and / or LIDAR (= "Light Detection And Ranging") and / or radar, etc., which in particular the ground surface structure and located thereon, previously classified objects can be detected. By means of machine vision and object classification, it is recognized from the recorded environment that it is, for example, an excavation that is excavated. In the recognized excavation is also recognized that this systematically or gradually larger or deeper and and that a work machine or a corresponding vehicle is present, which is increasingly loaded with overburden;
• The vehicle position or orientation is determined by means of GNSS, RTK (real-time kinematics), acceleration sensors and a gyro sensor;
• The movements of the tools are measured directly, for example by means of displacement and / or angle and / or acceleration, etc. sensors and / or indirectly by means of camera and / or LIDAR etc.

It can be seen from the vehicle and tool movements in a presently assumed working scenario that it is a picking up and unloading of earth or overburden onto a vehicle This means that the movement patterns of the excavator, excavator arm and bucket indicate picking, lifting, panning, discharging and unloading. From the detected movements, a dynamic map is created 230 in which the positions and movements of the machine and the tools are depicted as spatial movement trajectories. It is also possible to map data relating to the global 3D position as well as the movement of the working edge of the tool and / or the TCP (Tool Center Point) in the global 3D environment in the dynamic map.

The features or movement patterns required for the recognition of the working sequences or even parts thereof can be determined in advance (ie "offline") by means of a model calculation 235 be learned. At runtime of the system (ie, "online"), the model thus learned is fed with the above-mentioned actual data, so that by means of the model, the associated operating sequence or parts thereof are output as a result to the less experienced operator 240.

• - Kriterium „Trajektorie“: Anhand der Art und Weise der Steuerung kann erkannt werden, ob der Bediener in der Lage ist, beliebige 3D-Trajektorien von Fahrzeug und/oder Werkzeugen zu bedienen. So bedient z.B. ein unerfahrener Bediener die Joysticks eines Baggers nur auf der NNord-Süd bzw. West-Ost-Achse, d.h. sozusagen nur eindimensional, wohingegen ein erfahrener Bediener die möglichen Stellbereiche, z.B. durch Nord-West, Nord-Ost, usw., d.h. mehrere Bewegungen gleichzeitig gesteuert in mehreren Raumrichtungen überlagert. Ist eine sich wiederholende gesamte Arbeitssequenz, z.B. vom Aufnehmen bis zum Abladen bekannt, dann kann eine theoretische Kraftstoffverbrauchs- und/oder Zeit- und/oder Weg- optimale Trajektorie berechnet werden. Die so berechnete Trajektorie kann mit einer tatsächlich vorliegenden Trajektorie verglichen werden, um daraus die Differenz von Kraftstoffverbrauch, Zeitaufwand, Weg, usw. zu ermitteln.

• - Kriterium „Zielführung“: Hierbei kann bestimmt werden, wieviele Iterationen, Zwischenhalte, Umwege, usw. erforderlich sind, um das Werkzeug in die Zielposition zu steuern. Jede erkannte Iteration bzw. Korrektur kann dabei einen Zähler erhöhen. So steuert ein erfahrener Bediener auf Anhieb mit einer bezüglich der Form der Trajektorie und Geschwindigkeit des Werkzeugs kontinuierlichen und gleichförmigen Trajektorie das Werkzeug zum Werkstück. Einen unerfahrener Bediener erkennt man an der ungleichmäßigen Form der Trajektorie sowie an Arbeitsvorgängen wie das Zwischenhalten, Umkehrungen, Bewegungswiederholungen, Umkreisungen in Ziel-Nähe usw.
• - Kriterium „Arbeitsvorgang“: Hierbei kann erkannt werden, wie effizient das Werkzeug eingesetzt wird. So kann z.B. bestimmt werden, ob die Baggerschaufel mit einem Aufnahmevorgang voll beladen wird oder nicht bzw. der Bediener den Aufnahmevorgang wiederholt usw.
• - Kriterium „Präzision“: Liegen die Zieldaten für das Werkstück vor, z.B. die Maße der Baugrube, so kann erkannt werden, wie genau der Bediener gearbeitet hat. Die Vorgabe kann anhand von 3D-Lokalisierungsdaten (GNSS) des Werkstücks erfolgen. Die Messung erfolgt entweder am Werkstück selbst, z.B. mittels Kamera und/oder LIDAR, oder durch die Summe der Bewegungsräume des Werkzeugs, z.B. der Position der Werkzeugschneidkante der Baggerschaufel. Die Zieldaten können über ein BIM („Building Information Modeling“ = Baustellenplanung) oder vor Ort in Geräte vorgegeben werden. Die Bewertung kann auch eine Ebenheit einer Baugrube bewerten oder auch ein Unterschreiten und/oder Überschreiten von Zieldaten. So muss eine zu tiefe Baugrube wieder aufgefüllt und verfestigt werden oder es wird mehr Baumaterial benötigt.
• - Kriterium „Stabilität“: Hierbei kann bestimmt werden, wie nahe der Bediener die Maschine an die Stabilitätsgrenzen führt bzw. diese überschreitet. Diese Grenzen können eine zulässige Arbeitslast des Werkzeugs und/oder zulässige Neigung des Fahrzeugs auch auf unebenen Untergründen usw. sein. Gleichfalls können abrupte Bewegungen, Stopps, Beschleunigungen usw. zu einer Instabilität führen.
• - Kriterium „Sicherheit“: Hierbei kann mittels maschinellem Sehen erkannter Objekte (z.B. Personen, Fahrzeuge usw.) erkannt werden, wie oft der Bediener das eigene Fahrzeug und/oder Werkzeug auf Kollisionskurs und/oder kritischen Bremswegen von Fahrzeug- und/oder Werkzeugbewegungen gebracht hat oder es gar zu Ausweichbewegungen der Objekte gekommen ist. Kollisionen können ebenfalls damit erkannt werden.
• - Kriterium „Dynamik bzw. Geschwindigkeit“: Neben der Geschwindigkeit der Bewegungen von Fahrzeug und Werkzeugen wird hierbei auch das Beschleunigungsverhalten mit betrachtet. So ist es für den genauen und schnellen Arbeitsprozess vorteilhaft, die Beschleunigung vor der Zielerreichung zu reduzieren, um z.B. ein Überschwingen der Werkzeuge bzgl. der Zielposition zu vermeiden. Zudem würde ein ungebremstes Bewegen der Werkzeuge in die Anschläge zu höherem Verschleiß und (hydraulischem) Energieverbrauch führen. D.h. für die Bewertung des Bedieners spielt nicht nur die Geschwindigkeit entlang einer optimalen Bewegungstrajektorie ein Rolle, sondern auch, ob es zu Überschwingen und/oder Anschlägen kommt. D.h. das Optimum der Geschwindigkeit ist die maximale Geschwindigkeit bzw. Geschwindigkeitsverlauf über der optimalen Trajektorie bzw. Zeit von Start zum Ziel, die nicht zu einem Überschwingen oder Fahren in die Anschläge führt. Ein erfahrener Bediener kennt die Verzögerungen und die Trägheit seines Fahrzeugs und/oder Werkzeuge. Aufgrund seiner Erfahrungen mit unterschiedlichen Lasten kann er zukünftige Bewegungen abschätzen und entsprechend vorsteuern (vorausschauende bzw. prädiktive Steuerung). So startet der erfahrene Bediener bei einer Schwenkbewegung des Baggers rechtzeitig den Auslauf, um ein Überschwingen des Baggerarms bzw. der Last zu vermeiden, oder er nimmt rechtzeitig die Vortriebsleistung weg, um ein Fahren des Baggerarms in die Anschlagspositionen zu vermeiden.

Based on one or more of the following criteria, the abilities of a respective operator can be determined:

• - criterion "trajectory": Based on the type of control can be detected whether the operator is able to operate any 3D trajectories of vehicle and / or tools. For example, an inexperienced operator operates the joysticks of an excavator only on the north-south or west-east axis, ie, so to speak only one-dimensionally, whereas an experienced operator the possible parking areas, for example by north-west, north-east, etc., ie several movements simultaneously controlled superimposed in several spatial directions. If a repetitive overall operating sequence, eg from picking up to unloading, is known, then a theoretical fuel consumption and / or time and / or path-optimal trajectory can be calculated. The trajectory calculated in this way can be compared with an actual trajectory in order to determine the difference between fuel consumption, time required, travel, etc.

• - "route guidance" criterion: Here it can be determined how many iterations, intermediate stops, detours, etc. are required to steer the tool to the destination position. Each detected iteration or correction can increase a counter. Thus, an experienced operator immediately controls the tool to the workpiece with a trajectory that is continuous and uniform with respect to the shape of the trajectory and speed of the tool. An inexperienced operator can be recognized by the non-uniform shape of the trajectory as well as by operations such as stopping, reversals, repetition of motion, orbits around the target, etc.
• - Criterion "work process": It can be seen how efficiently the tool is used. For example, it can be determined whether or not the bucket is fully loaded with a picking operation, or the operator repeats picking, etc.
• - Criterion "Precision": If the target data for the workpiece is available, eg the dimensions of the excavation, it can be recognized how exactly the operator worked. The default can be based on 3D localization data (GNSS) of the workpiece. The measurement takes place either on the workpiece itself, for example by means of a camera and / or LIDAR, or by the sum of the movement spaces of the tool, for example the position of the tool cutting edge of the excavator bucket. The target data can be specified via a BIM ("Building Information Modeling") or on site in devices. The evaluation can also evaluate a flatness of a pit or a falling below and / or exceeding of target data. For example, a pit that is too deep must be refilled and consolidated, or more building material is required.
• - "Stability" criterion: This can be used to determine how close the operator approaches or exceeds the stability limits. These limits may be an allowable workload of the tool and / or allowable inclination of the vehicle even on uneven ground, etc. Likewise, abrupt movements, stops, accelerations, etc. can lead to instability.
• - "Safety" criterion: It can be detected by machine vision of recognized objects (eg persons, vehicles, etc.) how often the operator has brought his own vehicle and / or tool on collision course and / or critical braking distances of vehicle and / or tool movements has or even evasive movements of the objects has come. Collisions can also be detected.
• - Criterion "dynamics or speed": In addition to the speed of the movements of the vehicle and tools, the acceleration behavior is also considered. So it is advantageous for the accurate and fast work process to reduce the acceleration before reaching the target, for example, to avoid overshooting the tools with respect to the target position. In addition, an unrestrained movement of the tools would be in the attacks lead to higher wear and (hydraulic) energy consumption. That is, for the evaluation of the operator not only the speed along an optimal movement trajectory plays a role, but also whether it comes to overshoot and / or attacks. That is, the optimum of the speed is the maximum speed or speed course over the optimal trajectory or time from start to finish, which does not lead to an overshoot or drive in the stops. An experienced operator knows the delays and inertia of his vehicle and / or tools. Due to his experience with different loads, he can estimate future movements and control them accordingly (predictive or predictive control). Thus, the experienced operator starts the spout in time with a pivoting movement of the excavator in order to avoid overshooting of the excavator arm or the load, or he takes off the propulsion power in good time, to avoid driving the excavator arm in the stop positions.

By comparison 245 the from the mentioned model calculation 235 resulting data and in the created dynamic map 230 actual data can be determined on the basis of the aforementioned criteria, the general experience of a present operator in the control of a particular machine. Not all imaginable work processes have to be recorded or measured, as usually only one, eg the above-mentioned exemplary work scenario, is sufficient.

The method described is particularly suitable for use in repetitive work sequences and sequences of motion. The number of movement possibilities, e.g. of an excavator with respect to certain operations, is determined by the dimensions and the kinematics of the moving parts of the tool (excavator arm and shovel). The described model for evaluating the driver recognizes at runtime, which work process is concerned. Then, based on the above criteria, it is determined whether there are better patterns of movement from already learned or trained models of one or more experienced operators. These better patterns may then be displayed on a display device to the inexperienced operator 250, namely the virtual and / or predictive movement of the machine and / or the tools and / or the associated timely (predictive) control actions. The entire work process or workflow can be assisted only with respect to individual components, e.g. timely deceleration, e.g. to avoid overshoot. If the operator intends to execute the work process several times in succession or to repeat a previous work process or work step which he or the system has recorded by means of the global dynamic 3D map, then the latter can reach the automatic start mode when the previous starting point has been reached shut down / execute, whereby the system automatically corrects the corresponding parts of the working sequence with learned control commands of the experienced operator. Given enough learned motion patterns of experienced operators, the model may issue a motion and / or control recommendation for each possible trajectory or intervene in an automatic mode to improve.

The method also allows attention recognition. In particular, in repetitive, monotonous operations can be determined with the evaluation method, whether the operator against his best work sequence is currently (increasingly) deteriorated. Corresponding warnings and / or improving interventions in the system are conceivable. Information to a BIM system, etc. can be made to evaluate the operator specific work performance. Based on this, the selection of operators can be made to specific machines and work processes.

With a correlation search, e.g. By means of the known method of "data mining" or the "metadata analysis", a connection with the wear or the failure of machines and tools and / or their parts or parts thereof can be produced from the operator-specific movement patterns. On the one hand, it can then be recognized which (frequent) movement patterns and associated control processes contribute the largest share to the aging or damage of vehicle or machine parts, and which operators in which work process due to a specific movement pattern or parts thereof have the least wear or tear. Caused damage. These motion patterns could be used as reference patterns for an experienced operator.

If the machine has a person recognition or personalization, e.g. via a face recognition, the last patterns / models can be selected and further learned with previously learned movement patterns or user behavior. Otherwise, the personalization must otherwise take place via the respective human-machine interface (HMI).

Furthermore, with the knowledge of the machine-dependent control capabilities, the operator can again optimize the site planning (BIM) become. It can be optimized who, when, on which machine, which work processes, at which typical working hours for the respective work sequences, slower or faster than a predetermined planning.

In addition, in addition to the previously mentioned sensor information, also force and / or torque sensors can be used and mapped in the operator and / or machine and / or tool and / or process-specific force and / or torque curves.

An optimized according to the described method travel route or Fahrtrajektorie can be transferred from one machine or vehicle to another machine or vehicle. The environment, i. e.g. the inclination of the roadway, to be considered as further information in the dynamic map. If the environment changes, a tool change may be necessary. When which tool an experienced operator uses, can also be learned and notified or displayed to a novice operator with the same or similar pattern as a recommendation.

The method described can be realized in the form of a control program for an electronic control unit for controlling a working machine involved here or in the form of one or more corresponding electronic control units (ECUs).

2 schematically shows a set up for carrying out the method described above working machine 300 using a block diagram. The machine 300 In this embodiment, for carrying out the method described, the following functional components, namely a control unit, are included 305 , a sensor 310 with several sensors, eg an angle sensor 312 (Representative of possibly further sensors), as well as an actuator or motor 315 with several drive components, eg (again only representative) of an internal combustion engine 317 and a tool 318 ,

The sensors 310 includes tilt and angle sensors in the present embodiment, as well as inductive or magnetic sensors arranged on respective positioning cylinders of the tool, by means of which the position of the tool relative to a machine operator's cab and absolutely relative to the environment is determined. The position or orientation of the machine is determined in the present case by means of an acceleration sensor and a gyro sensor. In addition, an absolute position determination of the tool and the machine, relative to the surrounding terrain, by means of a GNSS system mentioned above, a said LIDAR system and / or an optical camera system take place.

The control unit 305 includes a trajectory calculation module 320 which is based on the sensor technology 310 collected data calculates the movement trajectories of the machine and the tool. The trajectories thus calculated serve as input data of a processor unit 325 ,

The processor unit 325 includes a model calculation module 330 , by means of which, on the basis of said input data, the features or movement patterns required for particular working sequences or even parts thereof can be learned by means of a model calculation and subsequently recognized. By means of a comparator module 335 From this, in the described manner, the level of experience of a given operator is determined. The resulting level of experience is in the present embodiment by means of an output unit 340 on the one hand to a display device (display) to the respective operator output. In addition, these data become a machine or tool control 345 fed by means of the actuators 315 is operated to operate the machine or the tool according to the present level of experience. The actuators in the present embodiment comprises an internal combustion engine drive for moving the machine and a hydraulic drive for moving the tool, for operating a braking system of the machine and for operating a (servo) steering system of the machine.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102005019820 A1 [0002]

Claims (14)

Method for operating a work machine (300) with at least one work tool (318), characterized by the following steps: - detecting (200) at least one movement trajectory of the at least one work tool (318) and / or the work machine (300) in a running work process with a experienced server; - analyzing (205) the detected at least one motion trajectory based on at least one predeterminable motion feature; - learning (235) the at least one predeterminable motion feature for the experienced operator; - comparing (245) the at least one learned motion feature with a corresponding motion feature of a less experienced operator; Assisting (240) the less experienced operator based on the results of the comparison (245). Method according to Claim 1 , characterized in that empirically predeterminable movement characteristics determined by the experienced operator are learned in a process-specific and / or tool-specific and / or machine-specific manner. Method according to Claim 1 or 2 , characterized in that the results of the comparison are output to the less experienced operator for optimizing the spatial movement of tools and / or the work machine in the work process (250). Method according to Claim 3 characterized in that an output (250) of indications to the less experienced operator is by means of a visual control assistant. Method according to Claim 4 , characterized in that the instructions are issued to the less experienced operator on a display device (250). Method according to Claim 1 or 2 , characterized in that the results of the comparison (245) are output to the less experienced operator by pre-controlling and / or by limiting individual movements. Method according to Claim 1 or 2 , characterized in that the results of the comparison (245) are completely taken over by a control system of the work machine (300). Method according to one of the preceding claims, characterized in that the at least one movement trajectory of the at least one working tool (318) and / or the working machine (300) is detected by means of a sensor system (310) Method according to Claim 8 , characterized in that the sensor system (310) comprises one or more of the following sensors: - at least one tilt and / or angle sensor arranged on the at least one working tool (318); - acceleration sensor; - gyrosensor; - A GNSS system for absolute positioning of the at least one working tool (318) and / or the working machine (300); - An inductive, magnetic or ohmic measuring device for determining the path of the at least one working tool (318) arranged adjusting cylinders; - A LIDAR measuring system for the optical measurement of the spatial distance of the at least one working tool (318) and / or the working machine (300); - A camera sensor for viewing direction detection of an operator of the work machine (300); - A stereoscopic camera sensor, in which the spatial position of the at least one working tool (318) and / or the working machine (300) can be determined from at least two images; - A camera sensor for creating a spatial road surface profile. Computer program, which is set up, each step of a method according to one of Claims 1 to 9 perform. Machine-readable data carrier on which a computer program according to Claim 10 is stored. An electronic control unit (305) adapted to operate a work machine (300) by a method according to any one of Claims 1 to 9 to control. Working machine (300) with at least one working tool (318) and with a sensor system (310) with at least one sensor for detecting at least one movement trajectory of the at least one working tool (318) and / or the working machine (300) in a running working process, characterized by a Control unit (305), which comprises a trajectory calculation module (320) and a processor unit (325), which determines by means of a model calculation data relating to the experience level of a present operator. Working machine after Claim 13 , characterized in that the data determined with respect to the level of experience of the present operator of a machine or Tool control (345) are supplied, by means of an actuator (315) of the working machine (300) is controlled according to a present level of experience.

Images