EP3953771A1 - Unité de travail hydrostatique et procédé pour sa commande - Google Patents

Unité de travail hydrostatique et procédé pour sa commande

Info

Publication number
EP3953771A1
EP3953771A1 EP20719358.2A EP20719358A EP3953771A1 EP 3953771 A1 EP3953771 A1 EP 3953771A1 EP 20719358 A EP20719358 A EP 20719358A EP 3953771 A1 EP3953771 A1 EP 3953771A1
Authority
EP
European Patent Office
Prior art keywords
model
actuators
adaptation
hydrostatic
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP20719358.2A
Other languages
German (de)
English (en)
Inventor
Frank Bender
Jochen Mayer
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Robert Bosch GmbH
Original Assignee
Robert Bosch GmbH
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Robert Bosch GmbH filed Critical Robert Bosch GmbH
Publication of EP3953771A1 publication Critical patent/EP3953771A1/fr
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/048—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators using a predictor

Definitions

  • the invention relates to a hydrostatic implement according to the preamble of
  • Claim 1 and a method for controlling it according to Claim 8.
  • a generic working device has a large number of rotational and / or translational degrees of freedom.
  • the working device consists of a chain of hydrostatic actuators.
  • the working device must work collision-free with the working machine and its surroundings during its range of motion. Furthermore, the tool arranged at the tip of the working device should be precisely and conveniently controllable in all three coordinates. The control should be possible even for inexperienced operators - such as a driver of the mobile work machine - with high dynamics and positioning accuracy.
  • the aging effects of the hydraulic components can lead to a deterioration in behavior over the service life of the equipment.
  • Deviations between the real system and the model of the system shown in the feedforward control can lead to a deterioration in the positioning accuracy.
  • the oscillation of the implement can also be stimulated, since the deviations between the target and actual behavior must be compensated by the controller in a closed circuit.
  • a prerequisite for such an assistance function is generally the acquisition of the
  • Joint angle and generally the kinematics configuration, in particular via angle sensors, cylinder stroke sensors or an inertial sensor system.
  • the invention is based on the object of creating a process-reliable controllable hydrostatic work device, as well as a method for process-reliable control of the work device.
  • the first object is achieved by a hydrostatic working device with the features of patent claim 1, the second by a method with the features of patent claim 8.
  • a hydrostatic work device for a mobile work machine in particular for an excavator, crawler excavator, mobile excavator, mini excavator, backhoe loader or a concrete pump or forest machine, has hydrostatic actuators, in particular kinematically coupled.
  • the actuators can be rotary or translatory, for example rotary motors or lifting cylinders.
  • a tool or a tool holder is preferably provided at an end section of the implement.
  • the work device has an operating device via which a movement request to the actuators can be detected.
  • the operating device is preferably signal-connected to a control device of the implement.
  • Movement request is an input variable in the control device of the Working device stored, at least kinematic model of the actuators.
  • the actuators are from the control device, in particular by means of a pressure medium source and a
  • Valve arrangement controllable at least as a function of an output variable of the control device, in particular of the model.
  • a detection device is provided, by means of which a kinematic state of the actuators that can be assigned or assigned to the movement requirement or output variable can be detected.
  • the working device in particular its control device, has an adaptation or
  • the control device is capable of learning according to the invention.
  • an adaptation or learning algorithm is implemented by means of the adaptation device.
  • the model has at least one state space model of the working device, in particular, one static and one dynamic each, in combination with a recursive least squares algorithm (RLS algorithm) that has an adaptive effect on the state space model or models.
  • RLS algorithm recursive least squares algorithm
  • the model has characteristic curves and / or characteristic diagrams of the implement with fixed support points, the values of which can be learned and / or adapted via an RLS algorithm.
  • the learning or adaptation algorithm is formed by at least one neural network and a deep learning method. Weights can be learned and adjusted using backpropagation. For this purpose, acquisition data from the acquisition device can preferably be stored in the control device.
  • a fourth variant of the learning or adaptation algorithm is designed as reinforcement learning (RL). This means that transfer behavior can be learned to a large extent without using the model.
  • the adaptation device can be designed in such a way that the adaptation can be carried out periodically or batch-wise, in particular after completion of a work task, or continuously.
  • the learning or adaptation algorithm is preferably stored in the adaptation device for execution.
  • control device is generic. It is preferably generic in its structure. This has the advantage that it can be adapted to configurations of the working device and / or the mobile working machine via the adapting device.
  • one or more parameters of the model can be adapted via the adaptation device, in particular as a function of the movement requirement and the detected and assigned state.
  • the adaptation device is designed in such a way that it uses it as a function of the movement requirement and the detected, assignable or
  • assigned state parameters of the model are recursively estimable.
  • the adaptation device has one for the adaptation
  • the model has a static component, in particular a static input non-linearity, a value lookup table or a characteristic field, and a dynamic component, in particular a linear state space model. Both parts are summarized in a Hammerstein model, for example.
  • the working device has a control device by means of which a deviation of the state from the movement request can be regulated.
  • a deviation of the state from the movement request can be regulated.
  • it causes an increasing improvement of the learned / adapted model / parameters, so that the deviation between the movement requirement and the actually achieved (assigned) state becomes increasingly smaller.
  • the adaptation device is after the adaptation
  • control device can also be deactivated in a further development and, in particular, can be activated if necessary.
  • a defined start-up cycle of the working device in particular standardized, is stored for execution in the control device.
  • the model in particular its parameters, can thus be adapted particularly quickly and in a manner that is easy to reproduce and analyze.
  • the work device has a hydrostatic pump and a
  • Valve arrangement for supplying pressure medium to the actuators.
  • a mobile work machine has a work device that is designed according to at least one of the aforementioned aspects of the invention.
  • the applicant reserves the right to make a patent claim or a patent application on such a machine.
  • a method for controlling a hydrostatic work device which is designed according to at least one aspect of the preceding description, has the steps of "detecting the movement request”, “determining the output variable of the control device as a function of the model”, “controlling the actuators as a function of the output variable”, “Detection of the actuator status that can be assigned or assigned to the motion request or the output variable”.
  • a step is “adapting the model or its inversion as a function of the movement requirement and / or the output variable, as well as as a function of the detected, assignable or assigned state”.
  • Another advantage of the method is that it can be applied to work equipment. It can be transferred to work equipment from any supplier with the advantages already mentioned.
  • the step of adapting in particular by means of the learning or adapting device, can take place in the course of commissioning, a revision or in a working operation of the work device. It can take place periodically or batchwise or continuously, in particular it can be stored in the adaptation device for periodic, batchwise or continuous execution.
  • the adapt step is carried out in particular by adapting parameters of the model.
  • the method has a step “correcting a deviation of the state reached as a function of the movement request from the assigned movement request”. This is preferably done via the aforementioned
  • This step / the control device thus brings about an increasing improvement in the learned / adapted parameters and the discrepancy between the movement requirement and the assigned state becomes increasingly smaller.
  • the step of regulating, in particular the control device can be largely or completely dispensed with. This enables faster movements of the working device and high positioning accuracy.
  • Figure 1 is a logic circuit diagram of a working device according to the invention.
  • FIG. 2 the working device according to FIG. 1 in a working operation, following the adaptation operation
  • FIG. 3 shows a logic circuit diagram of a working device according to the invention in
  • FIG. 4 shows the implement according to FIG. 3 in a working operation, following the
  • a hydrostatic working device 1 has a hydrostatic-mechanical unit 2, consisting of hydrostatic equipment 4 and mechanical
  • kinematic elements 6 a control device 8 (ECU), and a
  • Operator interface 10 such as a joystick (Human Machine Interface, HMI).
  • the hydrostatic equipment 4 has components for supplying pressure medium, such as, for example, a hydrostatic pump, valve arrangement or a valve block, which are shown in the form of blocks according to FIG. 1.
  • a hydrostatic pump for example, a hydrostatic pump, valve arrangement or a valve block, which are shown in the form of blocks according to FIG. 1.
  • Elements 6 have the hydrostatic actuators, such as rotary motors or lifting cylinders. These are also shown in accordance with FIG. 1 in the form of blocks. Furthermore, the hydrostatic-mechanical unit 2 has a detection device 12 by means of which at least positions and speeds and possibly accelerations of the mechanical-kinematic elements 6 can be detected. The detection by means of the detection device 12 can (in addition also) relate to the components of the hydrostatic equipment 4.
  • the operator interface 10 is connected to an interpretation device 14 for interpreting an operator request with regard to the movement of the work device 1.
  • the interpretation device 14 is connected to a model of inverse kinematics 16 of the mechanical-kinematic elements 6. Kinematics parameters, symbolized by the comparatively thick arrow, enter the kinematics 16.
  • a learning or adaptation device 18 with a corresponding learning or adaptation algorithm is also stored for execution.
  • the control device 8 also has a gray box pre-control 20, a control device 22 and a signal processing device 27.
  • An output of the inverse kinematics 16 is connected to an input of the gray box pilot control 20.
  • An output of the adaptation device 18 is connected to an input of the gray box pilot control 20.
  • the output of the inverse kinematics 16 is connected to an input of the control device 22 by means of an operator 26.
  • An output of the control device 22 is connected to the output of the gray box precontrol by means of an operator 24.
  • An output of the operator 24 is connected to the hydrostatic equipment 4 for controlling it, as well as to the adaptation device 18.
  • the hydrostatic equipment 4 is connected to the mechanical-kinematic elements 6 by pressure medium.
  • the detection device 12 is connected to a signal processing device 27. Its output is connected to the output of the inverse kinematics 16 by means of the operator 26, an output of the operator 26 being connected to an input of the control device 22.
  • the output of the signal processing device 27 is also connected to the inverse kinematics 16.
  • kinematics parameters of the mechanical-kinematic elements 6 or the hydrostatic actuators of the working device 1 are included in the inverse kinematics 16.
  • the joystick 10 is deflected by the operator.
  • the interpretation device 14 interprets this as a driver's request and delivers a signal to the inverse kinematics 16.
  • the inverse kinematics 16 outputs an output signal to the gray box precontrol 20. In a first run, this generates a signal to the hydrostatic equipment 4, in particular to the pumps and valves, so that the actuators of the mechanical-kinematic elements 6 are supplied with pressure medium and moved.
  • the detection device 12 detects this
  • Movement / positions / changes in position which is / are assigned to the original movement request and forwards them to the signal processing device 27.
  • the detected positions, speeds and possibly accelerations are sent from the signal processing device 27 to the operator 26 and thus to the
  • Control device 22 passed on. For the final adjustment of the deviation of the recorded position, speed, acceleration from the original and
  • the output signal of the control device 22 is sent to the operator 24 and again to the hydrostatic equipment 4 until the deviation is below a specified threshold.
  • Adaptation device 18 and the inverse kinematics 16. In the learning or
  • the adaptation device 18 intervenes in the gray box precontrol 20 and thus changes its output signal to the operator 24.
  • FIG. 2 shows the same hydrostatic working device 1 after the end of the learning or adaptation operation.
  • the adaptation device 18 is accordingly deactivated and no longer has access to the gray box pilot control 20 (arrow crossed through).
  • the control device 22 is deactivated (dashed illustration), or no longer has to intervene, since the gray box precontrol 20, in particular a model stored therein, is optimized by the previous intervention of the adaptation device 18 in such a way that none or at least none is sufficiently large Deviation of the
  • FIGS. 3 and 4 show a second exemplary embodiment of a hydrostatic working device 101, in each case in the learning or adaptation mode (FIG. 3) and in normal working mode (FIG. 4). The features of the control device 8 and 8 are shown in more detail
  • Adaptation device 18 As a model of the working device 101, they have a Hammerstein model with a static input non-linearity 28 (characteristic field), and thus connected in series, a linear, dynamic state space model 30. In addition, the
  • Control device 8, or adaptation device 18 has an executable method of least squares 32.
  • the operator interface 10 is connected to the inputs of the devices 4, 28 and 32.
  • the movement request y des is sent to this in the form of the deflection of the joystick 10. This leads to a changed pressure medium supply on the hydrostatic equipment 4 and to a movement, speed and acceleration on the mechanical-kinematic elements 6 (hydrostatic actuators) that are determined by the detection device 12 is captured.
  • the actual, the motion request y of the associated state y act of the actuators (6) is returned to the device 32.
  • the fact stored for executing the method of least squares compares the current state y act with the original motion request y of and fits in the map 28 and in the dynamic
  • State space model 30 parameters, for example parameters A, B, C.
  • the output of the adapter 18 is then a predicted movement y pred .
  • the operation of the device 32 continues until a termination condition is reached, which is formulated as a function of the reached, assigned state y act and the predicted movement y pred .
  • Figure 4 shows the hydrostatic working device 101 canceled, that is, after completion of the learning or adaptation operation according to Figure 3. Then, as shown in FIG 4 the feedback of the reached, the associated state y act on the adjusting means 18 is interrupted. The movement request y of the operator interface 10 is then in turn entered into the models 28, 30 and only a control signal Drv Dmd is sent to the hydrostatic equipment 4, from which the corresponding pressure medium supply to the mechanical
  • Movement requirement y of and that of movement requirement y of the assigned, reached state y act is so low due to the learning or adaptation phase carried out and the adapted parameters A, B, C that there is a need for regulation and others Adjustment can be dispensed with. There is therefore only a pure control mode with the advantages mentioned above.
  • a hydrostatic work device with a learning and adaptation device via which a particularly kinematic-hydraulic model stored in a control device for controlling the work device can be adapted or optimized so that a discrepancy between a detected state of the work device and a movement requirement assigned to the state is small , is negligible, or zero.
  • a method for controlling the implement with one step is also disclosed
  • Adapting or optimizing the model by means of the adapting device in such a way that a deviation of the detected state of the work device from the movement requirement assigned to the state is small, negligible or zero.
  • a deviation of the detected state of the work device from the movement requirement assigned to the state is small, negligible or zero.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Operation Control Of Excavators (AREA)
  • Numerical Control (AREA)

Abstract

L'invention concerne une unité de travail hydrostatique pour une machine de travail mobile, présentant des actionneurs hydrostatiques et un dispositif d'utilisation via lequel une requête de déplacement au niveau des actionneurs peut être détectée, qui est une grandeur d'entrée d'un modèle cinématique enregistré dans un dispositif de commande des actionneurs, les actionneurs pouvant être pilotés en fonction d'une grandeur de sortie du dispositif de commande, un dispositif de détection via lequel un état cinématique des actionneurs, pouvant être associé ou associé à la requête de déplacement ou la grandeur de sortie, étant prévu. L'invention concerne également un procédé pour la commande de l'unité de travail.
EP20719358.2A 2019-04-12 2020-04-06 Unité de travail hydrostatique et procédé pour sa commande Pending EP3953771A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102019205297.5A DE102019205297A1 (de) 2019-04-12 2019-04-12 Hydrostatisches Arbeitsgerät und Verfahren zu dessen Steuerung
PCT/EP2020/059728 WO2020207953A1 (fr) 2019-04-12 2020-04-06 Unité de travail hydrostatique et procédé pour sa commande

Publications (1)

Publication Number Publication Date
EP3953771A1 true EP3953771A1 (fr) 2022-02-16

Family

ID=70289749

Family Applications (1)

Application Number Title Priority Date Filing Date
EP20719358.2A Pending EP3953771A1 (fr) 2019-04-12 2020-04-06 Unité de travail hydrostatique et procédé pour sa commande

Country Status (5)

Country Link
EP (1) EP3953771A1 (fr)
KR (1) KR20210151812A (fr)
CN (1) CN113646709A (fr)
DE (1) DE102019205297A1 (fr)
WO (1) WO2020207953A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102021204544A1 (de) 2021-05-05 2022-11-10 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren zum Betreiben eines hydraulischen Zylinders einer Arbeitsmaschine
DE102021205386A1 (de) 2021-05-27 2022-12-01 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren zum Betreiben eines hydraulischen Zylinders einer Arbeitsmaschine

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3561667B2 (ja) * 1999-11-18 2004-09-02 新キャタピラー三菱株式会社 油圧ポンプの制御装置
US7496414B2 (en) * 2006-09-13 2009-02-24 Rockwell Automation Technologies, Inc. Dynamic controller utilizing a hybrid model
US8244438B2 (en) * 2008-01-31 2012-08-14 Caterpillar Inc. Tool control system
CA2883185A1 (fr) * 2012-08-27 2014-03-06 Ekso Bionics, Inc. Systeme d'actionneur hydraulique
US20140263607A1 (en) * 2013-03-15 2014-09-18 Clark Equipment Company Scannable codes to display machine information
WO2017129200A1 (fr) * 2016-01-28 2017-08-03 MAX-PLANCK-Gesellschaft zur Förderung der Wissenschaften e.V. Système d'optimisation et de commande de mouvement continu dans le monde réel
US9945096B2 (en) * 2016-02-10 2018-04-17 Deere & Company Force-based work vehicle blade pitch control
FR3052187B1 (fr) * 2016-06-06 2018-06-15 Peugeot Citroen Automobiles Sa Procede de compensation de l'usure cinematique d'une commande d'un actionneur de turbocompresseur a geometrie variable
FR3054635B1 (fr) * 2016-07-29 2019-05-17 Clesse Industries Detendeur pour la detente d’un gaz liquefie a compensation de pression integree
US11144683B2 (en) * 2016-12-06 2021-10-12 General Electric Company Real-time adaptation of system high fidelity model in feature space
US10526766B2 (en) * 2017-07-31 2020-01-07 Deere & Company Work machines and methods and systems to control and determine a position of an associated implement

Also Published As

Publication number Publication date
CN113646709A (zh) 2021-11-12
WO2020207953A1 (fr) 2020-10-15
DE102019205297A1 (de) 2020-10-15
KR20210151812A (ko) 2021-12-14

Similar Documents

Publication Publication Date Title
DE69618385T2 (de) Baggersteuersystem mit Arbeitsbereichbegrenzung für Baumaschinen
DE112008002587T5 (de) Hydraulikmanagement für begrenzte Geräte von Baumaschinen
AT514115A1 (de) Elektrohydraulischer Steuerkreis
WO2021122718A1 (fr) Procédé de commande d'un vérin hydraulique d'un engin de travail
DE102018214272B4 (de) Robotersystem
WO2020207953A1 (fr) Unité de travail hydrostatique et procédé pour sa commande
DE102017213650A1 (de) Verfahren zum Regeln eines hydraulischen Systems, Regeleinheit für ein hydraulisches System und hydraulisches System
DE10256441A1 (de) System und Verfahren zur Steuerung eines hydraulischen Strömungsmittels
DE102019113765A1 (de) Verfahren zur Steuerung eines hydraulischen Systems einer mobilen Arbeitsmaschine und mobile Arbeitsmaschine
EP3997527B1 (fr) Planification de trajectoires en temps réel destinée à des pompes de pistons axiaux en construction à plateau orientable avec prise en compte systématique de limitations de système
EP3839256A1 (fr) Procédé de fonctionnement d'une pompe à débit variable à vitesse variable
WO2018210448A2 (fr) Procédé de fonctionnement d'une presse à poudre à régulation de position et presse à poudre pour mettre en œuvre le procédé
DE202006021039U1 (de) Einrichtung zum Regeln einer Spritzgießmaschine
DE102016214708A1 (de) Stetigventileinheit, hydraulische Achse und Verfahren zum Betreiben einer hydraulischen Achse
DE102018211738A1 (de) Echtzeitfähige Ansteuerstrategie für hydraulische Systeme unter systematischer Berücksichtigung von Stell(raten)- und Zustandsgrößenbeschränkungen
WO2022248177A1 (fr) Procédé de commande d'un vérin hydraulique d'un engin de travail
EP3592991B1 (fr) Procédé pour la commande d'un actionneur hydraulique, dispositif de commande et commande d'actionneur
EP3118085B1 (fr) Systeme de direction pour une machine de travail
EP3292444B1 (fr) Procédé de fonctionnement d'un mécanisme de commande électrique ou fluidique et commande de déplacement d'un mécanisme de commande
EP3988801A1 (fr) Procédé de fonctionnement d'un entraînement hydraulique
WO2019048246A1 (fr) Évaluation automatique d'un comportement de machine
WO2022084173A1 (fr) Procédé de fonctionnement d'un entraînement hydraulique
DE102013211484A1 (de) Verfahren zur Schwingungsdämpfung bei einer mobilen Arbeitsmaschine
WO2023104484A1 (fr) Procédé d'entraînement d'un système d'apprentissage automatique associé à un vérin hydraulique
EP2836879B1 (fr) Dispositif de régulation pour une unité de cylindre hydraulique avec commande séparée de chaque soupape

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20211112

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20230102