Control system for autonomous construction machinery

Abstract

This invention provides a control device for autonomous construction machinery equipped with multiple pumps and multiple hydraulic actuators, thereby improving the controllability of the autonomous construction machinery and enabling the desired autonomous operation. [Solution] The control device (control controller 40) of the automated construction machine (hydraulic excavator 1) stores the predicted angular displacement value and rotation angle information (attitude information) as history, and uses the history of the predicted angular displacement value, the history of the rotation angle information, and the current rotation angle information to calculate (correct) the predicted angular displacement value from the present time to a predetermined time after the time has elapsed.

Description

[Technical Field]

[0001] This invention relates to a control device for an autonomous construction machine. [Background technology]

[0002] Construction machinery such as hydraulic excavators are used not only in civil engineering but also in a variety of other tasks. In recent years, however, there has been progress in the development of autonomous construction machinery that performs tasks semi-automatically or automatically. These autonomous construction machines can reduce the amount of work that operators must perform, thus reducing the burden on operators and is expected to lead to an improvement in working conditions.

[0003] In these types of autonomous construction machines, work is performed by controlling the machine's body and work equipment to the desired position. However, due to the hydraulic system, there is a delay in the operation, which becomes a problem during feedback control. This delay refers to the time lag between inputting an action command to the hydraulic system to position the machine body and work equipment in the desired position during autonomous driving, and the construction machine actually performing the action command as intended.

[0004] In contrast, there is a method for predicting and controlling the future posture of construction machinery after wasted time (for example, Patent Document 1). In Patent Document 1, the gain is adjusted based on blade load data, and this is used to determine the predicted height of the blade tip at a future point in time. Based on the determined predicted height at the future point in time and the current target height, the blade's movement is controlled so that the deviation between the predicted height and the target height at the present time becomes small. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] International Publication No. 2015 / 083469 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] However, in hydraulic systems of construction machinery equipped with multiple pumps and hydraulic actuators, such as hydraulic excavators, the degree of deviation from the target command value varies due to factors other than load, such as the upper limit of pump flow rate and flow division. Therefore, in such hydraulic systems, the gain cannot be uniquely determined by the load. As a result, the predicted attitude may deviate from the actual attitude. Furthermore, Patent Document 1 does not correct this deviation in attitude. For the reasons stated above, in the prior art of Patent Document 1, the responsiveness of the operation may deteriorate in autonomous construction machinery equipped with multiple pumps and hydraulic actuators.

[0007] The present invention is based on the above-mentioned matters and aims to provide a control device for an automated construction machine equipped with multiple pumps and multiple hydraulic actuators that can improve the controllability of the automated construction machine and realize the desired automated operation. [Means for solving the problem]

[0008] To achieve the above objective, the control device for an automated construction machine of the present invention generates angular velocity command values ​​for each joint based on rotation angle information of each joint of a multi-jointed work device provided on the construction machine and the target posture of the work device, and operates a plurality of actuators that drive the work device based on the angular velocity command values ​​of each joint, wherein the control device calculates a predicted angular displacement value of the amount of rotation that each joint is expected to rotate from the present time to a predetermined time, based on past values ​​of the angular velocity command values ​​of each joint, and compares the predicted angular displacement value from the predetermined time before and Based on the rotation angle information from a predetermined time prior, the predicted angular movement amount from the present to the predetermined time is corrected; based on the predicted angular movement amount from the present to the predetermined time and the rotation angle information at the present, the future posture of the work device at the predetermined time from the present is calculated; based on the future posture of the work device and the target posture of the work device, the angular velocity command values ​​for each joint to be output at the present are generated; and automatic operation control is performed to operate the plurality of actuators based on the angular velocity command values ​​for each joint to be output at the present. [Effects of the Invention]

[0009] According to the present invention, based on past values ​​of the angular velocity command values ​​of each joint of the work device, a predicted angular displacement value is calculated for each joint that is expected to rotate from the present time to a predetermined time later. By correcting the predicted angular displacement value to match the actual angular displacement in the past, it becomes possible to predict the angular displacement more accurately. This improves the controllability of automated construction machinery equipped with multiple pumps and multiple hydraulic actuators, enabling the realization of desired automated operations.

[0010] Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0011] [Figure 1] A perspective view showing the configuration of a hydraulic excavator in an embodiment. [Figure 2A]Side view of a hydraulic excavator in an embodiment. [Figure 2B] Top view of a hydraulic excavator in an embodiment. [Figure 3] Configuration diagram of the control system for a hydraulic excavator in an embodiment. [Figure 4] Detailed configuration diagram of the solenoid valve unit in the embodiment. [Figure 5] Detailed configuration diagram of the solenoid valve unit in the embodiment. [Figure 6] Hardware configuration diagram of controller 40 in the embodiment. [Figure 7] Functional block diagram of the controller 40 in the embodiment. [Figure 8] Functional block diagram of the target attitude calculation unit 40e in the embodiment. [Figure 9] Calculation flow diagram of the target attitude number calculation unit 40e-1 in the embodiment. [Figure 10] An explanatory diagram of the target attitude selected by the target attitude selection unit 40e-2 in the embodiment. [Figure 11] Functional block diagram of the angular displacement prediction unit 40b in the embodiment. [Figure 12] A diagram illustrating the calculation process of the dead time calculation unit 40b-1 in the embodiment. [Figure 13] An explanatory diagram illustrating the definitions of symbols used when calculating correction parameters in the embodiment. [Figure 14] A diagram illustrating the calculation process of the correction parameter calculation unit 40b-2 in the embodiment. [Figure 15] A diagram illustrating the calculations performed by the angular velocity command integration unit 40b-22 in the embodiment. [Figure 16] A diagram illustrating the calculation process of the future state prediction unit 40b-3 in the embodiment. [Figure 17] A diagram illustrating the calculation contents of the operation control unit 40a in the embodiment. [Figure 18] A diagram illustrating the calculation contents of the angular velocity command calculation unit 40a-1 in the embodiment. [Figure 19] A diagram illustrating the calculation process of the proportional valve output calculation unit 40c in the embodiment. [Figure 20] A diagram illustrating the time difference between the angular velocity command value and the actual angular velocity in the embodiment. [Figure 21] Graph of angular velocity command value and boom 8 angle in the embodiment (from time t-2twα to time t). [Figure 22] Graph of angular velocity command value and boom 8 angle in the embodiment (from time t-twα to time t+twα). [Figure 23] A graph of the angular velocity command value and the angle of boom 8 in the embodiment (from time t to time t+twα), and the difference in angular velocity command value with and without correction. [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described below with reference to the drawings. In each figure, parts having the same function are denoted by the same reference numerals, and their repeated descriptions may be omitted. In the following description of this paper, a hydraulic excavator equipped with a bucket 10 as a work tool (attachment) at the tip of the work device is used as an example, but the invention can also be applied to hydraulic excavators equipped with work tools other than buckets, and to construction machinery other than hydraulic excavators.

[0013] Furthermore, in the following explanations in this paper, when there are multiple identical components, an alphabet letter may be added to the end of the code (number), but the alphabet letter may be omitted and the multiple components may be referred to collectively. For example, when there are two travel hydraulic motors 3a and 3b (right travel hydraulic motor 3a, left travel hydraulic motor 3b), they may be referred to collectively as travel hydraulic motor 3.

[0014] <Configuration of a hydraulic excavator> The configuration of the hydraulic excavator in this embodiment will be explained using Figures 1 and 2A and 2B.

[0015] Figure 1 is a perspective view showing the configuration of a hydraulic excavator 1 according to an embodiment. The hydraulic excavator 1 comprises an articulated front working device 1A, a vehicle body 1B, and a controller (not shown in Figure 1).

[0016] The front working device 1A includes a boom cylinder 5, an arm cylinder 6, a bucket cylinder 7, a boom 8, an arm 9, a bucket 10, a bucket link 13, a boom angle sensor 30, an arm angle sensor 31, and a bucket angle sensor 32. The boom 8, arm 9, and bucket 10 are multiple driven members that rotate vertically, and the front working device 1A is constructed by connecting these.

[0017] The vehicle body 1B has a lower traveling body 11 and an upper rotating body 12. The lower traveling body 11 is driven by a pair of left and right hydraulic motors 3a (Figure 3, etc.) and 3b. The upper rotating body 12 is mounted on top of the lower traveling body 11 and is configured to rotate.

[0018] The base of the boom 8 is rotatably supported at the front of the upper slewing body 12 via a boom pin. An arm 9 is rotatably connected to the tip of the boom 8 via an arm pin. A bucket 10 is rotatably connected to the tip of the arm 9 via a bucket pin P (Figure 2A). The boom 8 is driven by a boom cylinder 5, the arm 9 is driven by an arm cylinder 6, and the bucket 10 is driven by a bucket cylinder 7.

[0019] Figure 2A is a side view of the hydraulic excavator 1. As shown in Figure 2A, the rotation angle of the boom 8 is α, the rotation angle of the arm 9 is β, and the rotation angle of the bucket 10 is γ, when the X-axis is defined parallel to the longitudinal direction of the lower traveling body 11 and the Z-axis is defined perpendicular to the X-axis. The inclination angle of the upper rotating body 12 (vehicle body 1B) with respect to a reference plane (e.g., the horizontal plane) is φ.

[0020] Figure 2B is a top view of the hydraulic excavator 1. As shown in Figure 2B, the relative angle between the upper rotating body 12 and the lower traveling body 11 is θ.

[0021] Returning to the explanation of Figure 1, the boom angle sensor 30 is attached to the boom pin so as to be able to measure the rotation angle α of the boom 8. The arm angle sensor 31 is attached to the arm pin so as to be able to measure the rotation angle β of the arm 9. The bucket angle sensor 32 is attached to the bucket link 13 so as to be able to measure the rotation angle γ of the bucket 10. Note that angle sensors 30, 31, and 32 can each be replaced with angle sensors relative to a reference plane (e.g., a horizontal plane).

[0022] The upper rotating body 12 includes a driver's cab 120, a hydraulic pump 2, a rotating hydraulic motor 4, an engine 18, a vehicle body tilt angle sensor 33, a rotating angle sensor 34, and a tank 200. The driver's cab 120 is equipped with a right operating lever 22a, a left operating lever 22b, a right travel lever 23a, a left travel lever 23b, and an engine speed setting device 480. Hereinafter, in this specification, the right operating lever 22a and the left operating lever 22b may be referred to as operating device 22, and the right travel lever 23a and the left travel lever 23b may be referred to as operating device 23.

[0023] The slewing hydraulic motor 4 rotates the upper slewing body 12. That is, the upper slewing body 12 is included in the driven members driven by the slewing hydraulic motor 4. Hereinafter, in this specification, the travel hydraulic motors 3a and 3b, the slewing hydraulic motor 4, the boom cylinder 5, the arm cylinder 6, and the bucket cylinder 7 may be collectively referred to as "actuators".

[0024] The vehicle body tilt angle sensor 33 is attached to any position on the upper rotating body 12 and detects the tilt angle φ of the upper rotating body 12 (vehicle body 1B) with respect to a reference plane (e.g., the horizontal plane). The rotation angle sensor 34 is provided on the rotation center axis of the upper rotating body 12 and measures the relative angle θ between the upper rotating body 12 and the lower traveling body 11.

[0025] Engine 18 is the prime mover (power source) and drives the hydraulic pump 2 and the pilot pump described later. The engine speed setting device 480 is a device for setting the rotational speed of engine 18. The engine speed setting device 480 is a dial-type variable resistor, and the voltage output from the engine speed setting device 480 changes depending on the operation of the dial.

[0026] The above describes the configuration of the hydraulic excavator 1 in the embodiment.

[0027] <Configuration of the hydraulic excavator control system> Next, the configuration of the control system for the hydraulic excavator in the embodiment will be explained using Figure 3.

[0028] Figure 3 is a configuration diagram showing the configuration of the control system of the hydraulic excavator 1. As shown in Figure 3, the control system of the hydraulic excavator 1 includes hydraulic pumps 2a and 2b, flow control valves 15a to 15f, load detection devices 16a to 16l, engine 18, regulators 2aa and 2ba, operating right lever 22a and operating left lever 22b (operating device 22), travel right lever 23a and travel left lever 23b (operating device 23), gate lock lever (not shown), lock valve 39, control controller (also simply called controller) 40 as a control device, operating devices 45a and 45b, operating devices 46a and 46b, operating devices 47a and 47b, pilot pump 48, pump line 143, pilot lines 144a to 149b, hydraulic drive unit 150a to 155b, solenoid valve unit 160, tank 200, engine controller 470, and engine speed detection device 490.

[0029] The operating devices 22 and 23, as well as the gate lock lever, are located in the operator's cab 120 and are operated by the operator. The operating devices 22 and 23 are electric lever type and generate electrical signals corresponding to the amount and direction of operation by the operator. These generated electrical signals are input to the controller 40 via the operating devices 45a to 47b. The controller 40 outputs electrical signals to the solenoid valve unit 160 to drive the solenoid proportional valves corresponding to the operations input to the operating devices 22 and 23.

[0030] Operating devices 45a to 47b are located inside the driver's cab 120. Operating device 47a is connected to the right travel lever 23a and outputs a signal to the controller 40 for operating the right travel hydraulic motor 3a. Operating device 47b is connected to the left travel lever 23b and outputs a signal to the controller 40 for operating the left travel hydraulic motor 3b. Operating devices 45a and 46a are connected to a common right operating lever 22a and output signals to the controller 40 for operating the boom cylinder 5 and bucket cylinder 7. Operating devices 45b and 46b are connected to a common left operating lever 22b and output signals to the controller 40 for operating the arm cylinder 6 and slewing hydraulic motor 4.

[0031] The engine 18 drives the hydraulic pumps 2a and 2b and the pilot pump 48. The hydraulic pumps 2a and 2b are variable displacement pumps whose capacity is controlled by regulators 2aa and 2ba, respectively. The pilot pump 48 is a fixed displacement pump. The engine controller 470 controls the rotational speed of the engine 18 in accordance with the control signal from the controller 40. The engine rotational speed detection device 490 is a rotation sensor for detecting the rotational speed of the engine 18.

[0032] The hydraulic pump 2 and pilot pump 48 draw in and discharge hydraulic fluid from the tank 200. In this embodiment, control signals output from the controller 40 are input to regulators 2aa and 2ba. The detailed configuration of regulators 2aa and 2ba is omitted, but the discharge flow rates of hydraulic pumps 2a and 2b are controlled according to the control signals.

[0033] The pressurized oil discharged from the hydraulic pump 2 is supplied to the boom cylinder 5 via flow control valve 15a, to the arm cylinder 6 via flow control valve 15b, to the bucket cylinder 7 via flow control valve 15c, to the slewing hydraulic motor 4 via flow control valve 15d, to the right travel hydraulic motor 3a via flow control valve 15e, and to the left travel hydraulic motor 3b via flow control valve 15f. The supplied pressurized oil causes the boom cylinder 5, arm cylinder 6, and bucket cylinder 7 to extend and retract, causing the boom 8, arm 9, and bucket 10 to rotate, respectively, and changing the position and orientation of the bucket 10. In addition, the supplied pressurized oil causes the slewing hydraulic motor 4 to rotate, causing the upper slewing body 12 to slewing relative to the lower travel body 11. Finally, the supplied pressurized oil causes the right travel hydraulic motor 3a and the left travel hydraulic motor 3b to rotate, causing the lower travel body 11 to travel.

[0034] To enable the detection of cylinder pressure, the boom cylinder 5 is equipped with load detection devices 16a and 16b, the arm cylinder 6 is equipped with load detection devices 16c and 16d, and the bucket cylinder 7 is equipped with load detection devices 16e and 16f. In this embodiment, load detection devices 16a to 16f are pressure sensors that detect the pressure on the bottom side and the rod side of the boom cylinder 5, arm cylinder 6, and bucket cylinder 7, respectively, and output them as electrical signals to the controller 40. In addition, the right travel hydraulic motor 3a is equipped with load detection devices 16g and 16h, the left travel hydraulic motor 3b is equipped with load detection devices 16i and 16j, and the slewing hydraulic motor 4 is equipped with load detection devices 16k and 16l. In this embodiment, load detection devices 16g to 16l are pressure sensors that detect the pressure in the respective input and output lines of the right travel hydraulic motor 3a, the left travel hydraulic motor 3b, and the slewing hydraulic motor 4, respectively, and output them as electrical signals to the controller 40. Note that, due to space limitations, the connection lines from the load detection devices 16a to 16l to the controller 40 are not shown in Figure 3.

[0035] The pump line 143 is the discharge pipeline of the pilot pump 48, and after passing through the lock valve 39, it is connected to each electromagnetic proportional valve in the electromagnetic valve unit 160. In this embodiment, the lock valve 39 is an electromagnetic switching valve, and its electromagnetic drive unit is electrically connected to the lock lever position detector of the gate lock lever. The gate lock lever has a center of rotation, and the operator can manually rotate the gate lock lever. The gate lock lever can be configured to rotate, for example, from an angle that contacts the upper limit stopper to an angle that contacts the lower limit stopper. The lock lever position detector is a switch sensor, and it can be configured so that a switch is pressed by the gate lock lever when the gate lock lever hits the lower limit stopper. As a result, the lock lever position detector outputs a signal to the lock valve 39 according to the position of the gate lock lever. If the gate lock lever is in the locked position, the lock valve 39 closes and the pump line 143 is shut off, and if it is in the unlocked position, the lock valve 39 opens and the pump line 143 is opened. In other words, when the pump line 143 is blocked, the operation by the control devices 22 and 23 is disabled, and operations such as driving, slewing, and excavating are prohibited.

[0036] The above describes the configuration of the control system for the hydraulic excavator 1 in this embodiment.

[0037] <Configuration of the solenoid valve unit in the control system of a hydraulic excavator> Next, the detailed configuration of the solenoid valve unit 160 shown in Figure 3 will be explained using Figures 4 and 5.

[0038] Figures 4 and 5 show the detailed configuration of the solenoid valve unit 160. The primary port side of the solenoid valve unit 160 is connected to the pilot pump 48 via the pump line 143. The solenoid valve unit 160 has solenoid proportional valves 54a to 59b that reduce the pilot pressure from the pilot pump 48 and output it to the pilot lines 144a to 149b. The solenoid proportional valves 54a to 59b are used as control signals to drive the flow control valves 15a to 15f in response to the input electrical signals.

[0039] As shown in Figure 4, solenoid proportional valves 54a and 54b are connected to the hydraulic drive units 150a and 150b of the flow control valve 15a via pilot lines 144a and 144b. Solenoid proportional valves 55a and 55b are connected to the hydraulic drive units 151a and 151b of the flow control valve 15b via pilot lines 145a and 145b. Solenoid proportional valves 56a and 56b are connected to the hydraulic drive units 152a and 152b of the flow control valve 15c via pilot lines 146a and 146b.

[0040] As shown in Figure 5, the solenoid proportional valves 57a and 57b are connected to the hydraulic drive units 153a and 153b of the flow control valve 15d via pilot lines 147a and 147b. The solenoid proportional valves 58a and 58b are connected to the hydraulic drive units 154a and 154b of the flow control valve 15e via pilot lines 148a and 148b. The solenoid proportional valves 59a and 59b are connected to the hydraulic drive units 155a and 155b of the flow control valve 15f via pilot lines 149a and 149b.

[0041] The solenoid proportional valves 54a to 59b have their opening degree at the minimum (zero) when de-energized, and their opening degree increases as the current, which is the control signal from the controller 40, increases. In this way, the opening degree of each solenoid proportional valve 54a to 59b corresponds to the control signal from the controller 40.

[0042] By outputting a control signal from the controller 40 to the solenoid valve unit 160 to drive the solenoid proportional valves 54a to 59b, pilot pressure can be generated even when there is no operator input from the corresponding operating devices 22 and 23, thereby forcibly generating the operation of each actuator.

[0043] The above describes the detailed configuration of the solenoid valve unit 160 shown in Figure 3.

[0044] <Controller Hardware Configuration> Next, the hardware configuration of the hydraulic excavator controller in the embodiment will be explained using Figure 6.

[0045] Figure 6 is a hardware configuration diagram of the controller 40. As shown in Figure 6, the controller 40 is a computer device and comprises an input unit 91, a central processing unit (CPU) 92 which is a processor, read-only memory (ROM) 93, random access memory (RAM) 94, and an output unit 95.

[0046] The control system shown in Figure 6 includes operating devices 22 and 23 connected to a controller 40 (control device), an attitude detection device 50, an engine speed setting device 480, load detection devices 16a to 16l, electromagnetic proportional valves 54 to 59, and an engine controller 470. The attitude detection device 50 consists of the boom angle sensor 30, arm angle sensor 31, bucket angle sensor 32, vehicle body tilt angle sensor 33, and slewing angle sensor 34. These angle sensors 30, 31, 32, 33, and 34 function as attitude sensors that detect the rotation angle (hereinafter also simply referred to as angle) of each joint of the front work device 1A and the upper slewing body 12 as attitude information.

[0047] The input unit 91 receives signals from the attitude detection device 50, signals indicating the manipulated amount from the operating devices 22 and 23, signals from the engine speed setting device 480, and signals from the load detection devices 16a to 16l, and converts them so that the CPU 92 can perform calculations. The ROM 93 is a recording medium that stores a control program for executing the control contents described later, and various information necessary for executing control calculations. The CPU 92 performs predetermined calculation processing on the signals received from the input unit 91, ROM 93, and RAM 94 according to the control program stored in the ROM 93. The output unit 95 creates an output signal according to the calculation results of the CPU 92, and outputs this signal to the engine controller 470 and electromagnetic proportional valves 54a to 59b to control the driving of the engine 18 and each actuator.

[0048] Although the controller 40 in Figure 6 is equipped with semiconductor memory ROM 93 and RAM 94 as storage devices, other types of storage devices can be used as replacements; for example, magnetic storage devices such as hard disk drives may also be included.

[0049] The above describes the hardware configuration of the hydraulic excavator controller in the embodiment.

[0050] <Functional blocks inside the controller> Next, the internal functional blocks of the controller 40 in this embodiment will be explained using Figure 7.

[0051] Figure 7 is a functional block diagram of the controller 40. Note that only functional blocks related to automatic operation are shown here, so functional blocks that calculate the output to the electromagnetic proportional valves 54a to 59b from the signals from the operating devices 22 and 23, functional blocks that calculate the output to the engine controller 470 from the signal from the engine speed setting device 480, and functional blocks that switch the output between automatic and manual operation are omitted.

[0052] The control controller 40 includes an motion control unit 40a, an angular movement prediction unit 40b, a proportional valve output calculation unit 40c, a numerical value holding unit 40d, and a target attitude calculation unit 40e.

[0053] The motion control unit 40a calculates the angular velocity command value for each driven member from the attitude information (angle) α(t), β(t), γ(t), θ(t), φ(t), which are signals transmitted from the attitude detection device 50 and converted by the input unit 91, the future attitude and correction parameters transmitted from the angular movement prediction unit 40b (described later), and the target attitude information transmitted from the target attitude calculation unit 40e (described later). The motion control unit 40a then transmits the angular velocity command value information to the proportional valve output calculation unit 40c and the numerical value holding unit 40d. The correction parameter is a dimensionless numerical value that indicates how much the actual angular movement was compared to the previously calculated predicted value of the angular movement. Details will be described later.

[0054] The angular displacement prediction unit 40b calculates the future attitude and correction parameters from the attitude information (angle) α(t), β(t), γ(t), θ(t), φ(t), which are signals transmitted from the attitude detection device 50 and converted by the input unit 91, and from the past values ​​of the angular velocity command value information transmitted from the numerical value holding unit 40d. The angular displacement prediction unit 40b then transmits the future attitude and correction parameters to the motion control unit 40a.

[0055] The proportional valve output calculation unit 40c receives information on angular velocity command values ​​transmitted from the operation control unit 40a and calculates operation commands to be output to the electromagnetic proportional valves 54 to 59. The proportional valve output calculation unit 40c then outputs the operation commands to the electromagnetic proportional valves 54 to 59 via the output unit 95.

[0056] The numerical value holding unit 40d receives information on the angular velocity command value transmitted from the operation control unit 40a and holds it for one step of the controller's control cycle. The numerical value holding unit 40d then transmits the past values ​​of the held angular velocity command value information to the angular movement amount prediction unit 40b.

[0057] The target attitude calculation unit 40e calculates the target attitude from the attitude information (angles) α(t), β(t), γ(t), θ(t), φ(t), which are signals transmitted from the attitude detection device 50 and converted by the input unit 91. The target attitude calculation unit 40e then transmits the target attitude information to the motion control unit 40a.

[0058] The above describes the internal functional blocks of the controller 40 in this embodiment.

[0059] <Target posture calculation section> Next, the function of the target attitude calculation unit 40e in the embodiment will be explained using Figure 8. Figure 8 is a functional block diagram of the target attitude calculation unit 40e in the embodiment. Here, Δt is the calculation period of the controller 40. Also, the target attitude number N(t) and target attitude P tgt (t) will be explained later using a different diagram.

[0060] The target attitude calculation unit 40e includes a target attitude number calculation unit 40e-1, a target attitude selection unit 40e-2, a numerical value holding unit 40e-3, and a numerical value holding unit 40e-4.

[0061] The target attitude number calculation unit 40e-1 receives attitude information α(t), β(t), γ(t), θ(t), φ(t), the target attitude number N(t-Δt) from the previous step transmitted from the numerical value holding unit 40e-3, and the target attitude P from the previous step transmitted from the numerical value holding unit 40e-4. tgt The (t-Δt) signal is received, and the target attitude number N(t) is calculated. The target attitude number calculation unit 40e-1 then transmits the target attitude number N(t) to the target attitude selection unit 40e-2 and the numerical value holding unit 40e-3.

[0062] The target attitude selection unit 40e-2 receives the target attitude number N(t) transmitted from the target attitude number calculation unit 40e-1, and determines the target attitude P corresponding to the received target attitude number N(t). tgt (t) is selected. Then, the target attitude selection unit 40e-2 selects the target attitude P as the output of the entire target attitude calculation unit 40e. tgt (t) is output, along with the target attitude P tgt (t) is also transmitted to the numerical value holding unit 40e-4.

[0063] The numerical value holding unit 40e-3 receives the target attitude number N(t) transmitted from the target attitude number calculation unit 40e-1 and holds it for one step of the controller's control cycle. Then, the numerical value holding unit 40e-3 transmits the target attitude number N(t-Δt) from the previous step, which it has held, to the target attitude number calculation unit 40e-1.

[0064] The numerical value holding unit 40e-4 receives the target attitude P transmitted from the target attitude selection unit 40e-2. tgt (t) is received and held for one step of the controller's control cycle. The numerical holding unit 40e-4 then holds the target attitude P from the previous step that was held. tgt (t-Δt) is transmitted to the target attitude number calculation unit 40e-1.

[0065] The above describes the functions of the target attitude calculation unit 40e in the embodiment.

[0066] Next, the calculation flow of the target attitude number calculation unit 40e-1 in the embodiment will be described with reference to FIG. 9. FIG. 9 shows a flowchart of the target attitude number calculation unit 40e-1 in the embodiment, which is repeatedly executed, for example, while the controller 40 is operating. Steps S101 to S107 shown in FIG. 9 constitute one control cycle.

[0067] In S101, the calculation of the target attitude number calculation unit 40e-1 is started.

[0068] In S102, the current attitude P(t) is obtained. The current attitude P(t) is a vector expressed as (α(t), β(t), γ(t), θ(t)) using the attitude information α(t), β(t), γ(t), and θ(t).

[0069] In S103, the target attitude P one step before tgt (t-Δt) and the current attitude P(t), it is determined whether the Euclidean norm (hereinafter referred to as "norm") of the difference is equal to or less than the threshold value C th . The target attitude P one step before tgt (t-Δt) and the current attitude P(t) is equal to or less than the threshold value C th , that is, when it is determined that the current attitude P(t) has approached the target attitude P one step before tgt (t-Δt), the determination in S103 is Yes, and the process proceeds to S104. The target attitude P one step before tgt (t-Δt) and the current attitude P(t) is larger than the threshold value C th , that is, when it is determined that the current attitude P(t) has not approached the target attitude P one step before tgt (t-Δt), the determination in S103 is No, and the process proceeds to the process of S105.

[0070] In S104, the target attitude number N(t) is set to a value obtained by adding 1 to the target attitude number N(t-Δt) one step before.

[0071] In S105, the target attitude number N(t) is set to the target attitude number N(t-Δt) from the previous step.

[0072] In S106, it is determined whether the target attitude number N(t) is greater than 10. If the target attitude number N(t) is greater than 10, S106 is determined to be Yes, and the process proceeds to S107. If the target attitude number N(t) is 10 or less, S106 is determined to be No, and the process in this step ends.

[0073] In S107, the target attitude number N(t) is set to 1.

[0074] The above is the calculation flow of the target attitude number calculation unit 40e-1 in the embodiment.

[0075] Next, the calculation details of the target attitude selection unit 40e-2 in the embodiment will be explained using Figure 10. Figure 10 shows the target attitude selected by the target attitude selection unit 40e-2 in the embodiment.

[0076] The ROM 93 in the controller 40 contains information on the 1st to 10th target attitudes (P tgt1 From P tgt10 ) is stored. Target postures 1 through 10 are a chronological arrangement of target postures when the hydraulic excavator 1 operates automatically. The target posture selection unit 40e-2 selects the N(t) target posture from the target posture number calculation unit 40e-1 to the current target posture P tgt Set to (t). For example, if target attitude number N(t)=3, then the third target attitude P tgt3 =( α tgt3 , β tgt3 gamma tgt3 θ tgt3 ) Target posture P tgt Set to (t).

[0077] The above describes the calculation contents of the target attitude selection unit 40e-2 in the embodiment.

[0078] If the calculations performed by the target posture calculation unit 40e above enable the subsequent calculations to "automatically move toward the target posture," then the automatic operation can be repeated such as first automatically moving toward the first target posture, then automatically moving toward the next target posture (the second target posture) as the current posture approaches the first target posture, and so on until it approaches the tenth target posture. In other words, the automatic operation of "approaching the first to the tenth target postures in order" can be repeated.

[0079] <Angular displacement prediction unit> Next, the functional blocks of the angular movement prediction unit 40b in the embodiment will be described using Figure 11. Figure 11 is a functional block diagram of the angular movement prediction unit 40b in the embodiment.

[0080] The angular movement prediction unit 40b comprises a dead time calculation unit 40b-1, a correction parameter calculation unit 40b-2, and a future state prediction unit 40b-3. Dead time is the time difference that occurs from the time the controller 40 outputs current to each electromagnetic proportional valve until the opening degree of each electromagnetic proportional valve actually opens to a value equivalent to the current value, each actuator operates, and each driven member reaches a desired angular velocity. In other words, it is the time it takes to operate each actuator based on the angular velocity command value of each driven member until the angular velocity of each joint of the front work device 1A reaches the angular velocity command value.

[0081] The dead time calculation unit 40b-1 receives information on the angular velocity command value and calculates the dead time. The dead time calculation unit 40b-1 then transmits the dead time information to the correction parameter calculation unit 40b-2 and the future state prediction unit 40b-3.

[0082] The correction parameter calculation unit 40b-2 receives the angular velocity command value, attitude information α(t), β(t), γ(t), θ(t), φ(t), and dead time information transmitted from the dead time calculation unit 40b-1, and calculates the correction parameters. The correction parameter calculation unit 40b-2 then transmits the correction parameter information to the future state prediction unit 40b-3. The correction parameter calculation unit 40b-2 also transmits the correction parameter information as the overall output of the angular displacement prediction unit 40b.

[0083] The future state prediction unit 40b-3 receives attitude information α(t), β(t), γ(t), θ(t), φ(t), angular velocity command value, dead time information transmitted from dead time calculation unit 40b-1, and correction parameter information transmitted from correction parameter calculation unit 40b-2, and calculates the future attitude. Then, the future state prediction unit 40b-3 transmits the future attitude as the output of the entire angular displacement prediction unit 40b.

[0084] The above describes the functional blocks of the angular movement prediction unit 40b in the embodiment.

[0085] <Calculation details of the waste time calculation unit 40b-1> Next, the calculation details of the dead time calculation unit 40b-1 in the embodiment will be explained using Figure 12. Figure 12 shows the calculation details of the dead time calculation unit 40b-1 in the embodiment.

[0086] As shown in Figure 12, the dead time calculation unit 40b-1 has a table that converts the angular velocity command value of each driven member into the dead time of each driven member, and calculates the dead time of each driven member from the angular velocity command value of each driven member. Specifically, the angular velocity command value of the boom 8 is input into the table and the dead time t of the boom 8 is calculated. wα The angular velocity command value of arm 9 is entered into the table, and the dead time t of arm 9 is calculated. wβ The angular velocity command value of bucket 10 is entered into the table, and the dead time t of bucket 10 is calculated. wγ The angular velocity command value of the upper rotating body 12 relative to the lower traveling body 11 is entered into the table, and the dead time t of the upper rotating body 12 relative to the lower traveling body 11 is calculated. wθThese are calculated individually. Then, the dead time calculation unit 40b-1 outputs each of the dead times thus obtained.

[0087] The above describes the calculations performed by the dead time calculation unit 40b-1 in this embodiment.

[0088] <Definition of symbols> Before describing the next functional block, the definitions of the symbols used in the following description will be explained using Figure 13. Figure 13 shows the definitions of the symbols used when calculating the correction parameters in the embodiment. Since the same calculation is performed for the boom 8, arm 9, bucket 10, and the relative angles of the upper slewing body 12 and the lower traveling body 11, only the symbols related to the boom 8 will be explained here. In the calculation of the relative angles of the arm 9, bucket 10, upper slewing body 12 and the lower traveling body 11, the α part in the following symbols is replaced with β, γ, and θ respectively. Also, the parts of electromagnetic proportional valve 54a and electromagnetic proportional valve 54b are replaced with electromagnetic proportional valve 55a and electromagnetic proportional valve 55b, electromagnetic proportional valve 56a and electromagnetic proportional valve 56b, and electromagnetic proportional valve 57a and electromagnetic proportional valve 57b, respectively.

[0089] As shown in Figure 13, time tt wα The actual angular displacement of angle α(t) from time t to time t is Δα(tt wα , t), from time t to time t+t wα The predicted angular displacement of angle α(t) up to Δα e (t, t+t) wα ), from time t to time t+t wα The predicted angular displacement of angle α(t) up to Δα' (correction to avoid division by zero) is used. e (t, t+t) wα ), the value of the correction parameter k α (t), from time t to time t+t wα The corrected angular displacement prediction value of angle α(t) up to Δα ec (t, t+t) wα ), the boom angular velocity command value to be output at time t is ω αc (t) The provisional value of the boom angular velocity command value to be output at time t is ω' αc(t), predicted at time t, time t+t wα The predicted value of the posture (future posture) in α ec (t, t+t) wα )

[0090] The above is a definition of the symbols used in the following explanation.

[0091] <Calculation details of the correction parameter calculation unit 40b-2> Next, the calculation details of the correction parameter calculation unit 40b-2 in the embodiment will be explained using Figures 14 and 15. Figure 14 shows the calculation details of the correction parameter calculation unit 40b-2 in the embodiment. Figure 15 shows the mathematical formula representing the calculation performed by the angular velocity command integration unit 40b-22 shown in Figure 14.

[0092] The correction parameter calculation unit 40b-2 includes an actual angular displacement calculation unit 40b-21 and an angular velocity command integration unit 40b-22.

[0093] The actual angular displacement calculation unit 40b-21 calculates the attitude information (angle) α(t) and the dead time t. wα It receives information, calculates the actual angular displacement, and transmits it.

[0094] The actual angle movement calculation unit 40b-21 includes a dead time holding unit 40b-211. The dead time holding unit 40b-211 stores the received attitude information (angle) for a dead time t wα The data is retained as history for that period. Then, as shown in the formula in (1) of the figure, the current posture information α(t) value is used to obtain the posture information α(tt) from before the dead time, which is retained as the dead time history. wα By subtracting the value of ), we can obtain the actual angular displacement Δα(tt) from before the dead time to the present. wα Calculate t).

[0095] In the angular velocity command integration unit 40b-22, the angular velocity command value ω αc (t-Δt) and dead time t wα Based on the information received, the predicted angular displacement of boom 8 from the present until the end of the dead time (with zero division avoidance correction) Δα' is calculated.e (t, t+t) wα Calculate and output the result.

[0096] The angular velocity command integration unit 40b-22 includes a parts integration calculation unit 40b-221, a dead time holding unit 40b-222, and an absolute minimum value guarantee unit 40b-223.

[0097] The integral by parts unit 40b-221 calculates the angular velocity command value ω from before the dead time to the present, as shown in Figure 15 (2). αc By integrating (t), we can obtain the predicted angular displacement Δα from the present time to the time elapsed. e (t, t+t) wα The system calculates the time and transmits that information to the dead time holding unit 40b-222.

[0098] The dead time holding unit 40b-222 receives the angular displacement prediction value Δα from the partial integration calculation unit 40b-221. e (t, t+t) wα ) wasted time wα The history is retained, and the predicted angular displacement Δα is calculated from before the dead time until the present. e (tt wα The information is then sent to the absolute minimum value guarantee unit 40b-223.

[0099] The absolute minimum value guarantee unit 40b-223, in order to prevent division by zero in subsequent calculations, uses the predicted angular displacement value Δα transmitted from the dead time holding unit 40b-222, as shown in (3) of Figure 15. e (tt wα The absolute value of t) is the threshold C α If the following conditions are met, the sign remains the same, but the absolute value is the threshold C. α The output is corrected to achieve the following: This is the predicted angular displacement value (corrected to avoid division by zero) Δα'. e (tt wα , t)

[0100] Based on the above calculations, the correction parameter calculation unit 40b-2 calculates the actual angular movement amount Δα(tt) from before the dead time to the present from the actual angular movement amount calculation unit 40b-21. wαt) is the angular displacement amount predicted by the angular velocity command integral unit 40b-22 to be the amount of angular displacement that the boom 8 is expected to move from before the dead time to the present (division by zero avoidance correction) Δα'. e (tt wα The following values ​​were obtained: , t) and t). The correction parameter calculation unit 40b-2 calculates the actual angular displacement Δα(tt) mentioned above. wα , t) is the predicted value of angular displacement (correction to avoid division by zero) Δα' e (t, t+t) wα By dividing by ), the correction parameter k α Calculate (t).

[0101] The above describes the calculation contents of the correction parameter calculation unit 40b-2 in the embodiment.

[0102] <Calculation details of the future state prediction unit 40b-3> Next, the calculations of the future state prediction unit 40b-3 in the embodiment will be explained using Figure 16. Figure 16 shows the calculations of the future state prediction unit 40b-3 in the embodiment.

[0103] The future state prediction unit 40b-3 includes an angular velocity command integration unit 40b-31. The angular velocity command integration unit 40b-31 contains the angular velocity command value ω αc (t-Δt) and dead time t wα The following is input. Then, as shown in (4) in the figure, the angular velocity command value ω from before the dead time to the present. αc Integrating (t) gives the predicted angular displacement Δα from the present to the time after the dead period. e (t, t+t) wα The integral method in the angular velocity command integral unit 40b-31 is the same as that of the integral by parts unit 40b-221 in this embodiment.

[0104] In the future state prediction unit 40b-3, the predicted angular displacement value Δα from the present to the time after the delay is calculated. e (t, t+t) wα ) with correction parameter k α By multiplying by (t), the corrected angular displacement prediction value Δα is obtained. ec (t, t+t)wα ) is obtained. Then, the corrected predicted angular movement amount Δα ec (t, t+t wα ) is added to the current posture information α(t), thereby obtaining the predicted posture value at time t+t wα as of time t (this is referred to as the future posture at time t+t wα as of time t) α ec (t, t+t wα ) is calculated.

[0105] The above is the calculation content of the future state prediction unit 40b-3 in the embodiment.

[0106] <Calculation content of motion control unit 40a> Next, the calculation content of the motion control unit 40a in the embodiment will be described with reference to Fig. 17. Fig. 17 shows the calculation content of the motion control unit 40a in the embodiment.

[0107] The motion control unit 40a includes an angular velocity command calculation unit 40a-1. The angular velocity command calculation unit 40a-1 calculates, based on the future posture α at time t+t wα as of time t ec (t, t+t wα ) and the target posture α of the boom 8 tgt (t), the provisional value ω' of the angular velocity command value of the boom 8 to be output at time t αc (t) is calculated.

[0108] The provisional value ω' of the angular velocity command value output from the angular velocity command calculation unit 40a-1 αc (t) is divided by the correction parameter k α (t), so as to obtain the angular velocity command value ω that is finally output from the motion control unit 40a at time t αc (t).

[0109] The above is the calculation content of the motion control unit 40a in the embodiment.

[0110] <Calculation content of angular velocity command calculation unit 40a-1> Next, the calculation contents of the angular velocity command calculation unit 40a-1 in the embodiment will be explained using Figure 18. Figure 18 shows the calculation contents of the angular velocity command calculation unit 40a-1 in the embodiment.

[0111] The angular velocity command calculation unit 40a-1 calculates the target attitude α of the boom 8. tgt (t) to time t+t at time t wα Future posture α ec (t, t+t) wα Subtract the gain G from the value. p By multiplying by this, the provisional value ω' of the angular velocity command value of boom 8 output at time t is obtained. αc This performs what is known as P control (proportional control), which calculates (t).

[0112] The above describes the calculation contents of the angular velocity command calculation unit 40a-1 in the embodiment.

[0113] <Calculation details of proportional valve output calculation unit 40c> Next, the calculation details of the proportional valve output calculation unit 40c in the embodiment will be explained using Figure 19. Figure 19 shows the calculation details of the proportional valve output calculation unit 40c in the embodiment.

[0114] As shown in Figure 19, the proportional valve output calculation unit 40c calculates the angular velocity command value ω αc (t) has a table that converts the output current to the solenoid proportional valve 54a and solenoid proportional valve 54b, and the angular velocity command value ω αc The output currents to the solenoid proportional valves 54a and 54b are calculated as operation commands from the value of (t). This results in the angular velocity command value ω αc Automatic operation control is performed to automatically operate multiple actuators by controlling the operation of electromagnetic proportional valves 54a and 54b according to (t).

[0115] The above describes the calculation contents of the proportional valve output calculation unit 40c in the embodiment.

[0116] <Effects of the Embodiment> As described above, according to this embodiment, the amount of angular movement is predicted from the angular velocity command value (predicted angular movement value), and by comparing it with the actual amount of angular movement, a correction is applied to the predicted angular movement value for the next calculation. Therefore, even if the actual angular velocity fluctuates due to various factors, the amount of angular movement can be predicted with greater accuracy. Furthermore, by applying a correction to the angular velocity command value that has been calculated once, it becomes easier to achieve the desired angular velocity. For example, if the actual amount of angular movement is half of the predicted value, the angular velocity command value will be doubled due to the correction, so the final angular velocity can be brought closer to the desired value. This will be explained using Figures 20 to 23. Here again, for the same reasons as above, the explanation will be limited to the operation of the boom 8.

[0117] <Definition of wasted time and the problems caused by wasted time> First, we will explain the definition of wasted time and the problems that arise from it using Figure 20.

[0118] Figure 20 shows the angular velocity command value ω at time t, which is set by the controller 40. αc After calculating (t), it takes time for the actual angular velocity to reach that value, wα This indicates that it takes 2 seconds. This time difference occurs because it takes time from the moment the controller 40 outputs current to the electromagnetic proportional valves 54a and 54b at time t until the opening degree of each electromagnetic proportional valve actually opens to the amount of current, pressurized oil physically flows into the bottom or rod side of the boom cylinder 5, causing the boom cylinder 5 to extend or contract, and the boom 8 to reach the desired angular velocity. This time difference is dead time.

[0119] This dead time must be compensated for when calculating the angular velocity command value. For example, at time t, the target attitude α of boom 8 tgt Even if there is a difference between (t) and the current posture α(t), time t+t wα If the difference becomes 0, then the angular velocity command value ω at time t αc (t) must be 0. Therefore, time t+t wα The orientation at time t needs to be predicted as accurately as possible.

[0120] The above is the definition of wasted time and the problems caused by wasted time. This embodiment compensates for this wasted time, that is, the time t + t wα This concerns "posture prediction."

[0121] <Correction parameter k> α (t) Meaning of the calculation at the time of calculation > Next, the correction parameter k in the embodiment α The calculation of (t) will be explained using Figure 21. Figure 21 shows the time t-2t in the embodiment. wα Angular velocity command value ω from time t αc The graphs for (t) and the angle of boom 8 are shown.

[0122] In this embodiment, the controller 40, at time t, "dead time t from time t" wα The system calculates the "predicted value of the angular displacement until the specified time has elapsed." At that time, the angular velocity command value ω shown in the graph above is used. αc (t) at time tt wα By integrating from time t, we can predict the angular displacement Δα. e (t, t+t) wα ) is being calculated. This is because, due to the effect of dead time, "from time t to time t+t wα The actual angular velocity up to the time tt wα This is because it becomes the angular velocity command value from time t. Therefore, by integrating the latter, the predicted angular displacement value Δα is obtained. e (t, t+t) wα ) can be calculated. And this value can be used to find t wα By keeping it as a history of seconds, "time tt wα Predicted angular displacement value Δα e (tt wα The value of t can be calculated (dead time holding unit 40b-222).

[0123] Meanwhile, the controller 40 will send a message at time t, "time tt wα The actual angular displacement Δα(tt) from time t to time t wαThe calculation also includes ", t)" (actual angular movement amount calculation unit 40b-21). Specifically, the attitude α(t) at a certain time is calculated as t wα Keep it as a history of seconds and time tt wα Posture α(tt wα (dead time holding unit 40b-211) is used to determine the actual angular displacement Δα(tt) by subtracting this from the current posture α(t). wα We are looking for t).

[0124] Then, the controller 40 controls the actual angular movement amount Δα(tt wα , t) at time tt wα Predicted angular displacement value Δα e (tt wα By dividing by t, the correction parameter k α (t) is calculated (correction parameter calculation unit 40b-2). In other words, "time tt wα The time predicted by tt wα Predicted angular displacement Δα from time t to time t e (tt wα "Time tt" for ", t)" wα The actual angular displacement Δα(tt) from time t to time t wα The magnification of t) is the correction parameter k α (t)

[0125] The above describes the correction parameter k in the embodiment. α (t) This is the calculation performed during the calculation.

[0126] <Correction parameter k> α Correction of the predicted angular displacement value by (t) > Next, the correction of the predicted angular displacement value using the correction parameter in the embodiment will be explained using Figure 22. Figure 22 shows the time tt in the embodiment. wα From time t+t wα Angular velocity command value ω up to αc This is a graph of (t) and a graph of the angle of boom 8.

[0127] In this embodiment, the controller 40 calculates the "predicted value of the angular movement amount until dead time has elapsed from time t" at time t. At that time, similar to Figure 21, the angular velocity command value ω shown in the graph above. αc (t) at time tt wα By integrating from time t, we can predict the angular displacement Δα. e (t, t+t) wα The integral of angular velocity command is calculated (angular velocity command integral section 40b-31).

[0128] Next, the controller 40 calculates the predicted angular displacement value Δα. e (t, t+t) wα ) Correction parameter k α Multiply by (t). This gives the corrected angular displacement prediction value Δα. ec (t, t+t) wα (Future state prediction unit 40b-3) is seeking ).

[0129] Corrected angular displacement prediction value Δα ec (t, t+t) wα ) reflects the ratio of the actual angular movement to the predicted angular movement up to that point. Therefore, for example, if some load is applied to boom 8 and the actual angular velocity changes to the angular velocity command value ω αc Even if it is slower with respect to (t), by reflecting that multiplier, time t becomes time t+t wα It is possible to predict posture more accurately.

[0130] This results in the angular velocity command value ω αc (t) can be calculated more appropriately. This point will be explained using Figure 23. Figure 23 shows the angular velocity command value and the angle of the boom 8 from time t to time t+t in the embodiment. wα The graph shows the difference in angular velocity command values ​​with and without correction.

[0131] For example, the target attitude α shown in Figure 23 tgt (t), current posture α(t), predicted angular displacement Δα e (t, t+t) wα ), corrected angular displacement prediction value Δα ec(t, t+t) wα Assume that the current posture α(t) and the predicted angular displacement Δα are as follows. e (t, t+t) wα The value obtained by adding ) is the target posture α tgt (t) becomes greater than the predicted angular displacement Δα, which is corrected for the current posture α(t). ec (t, t+t) wα The value obtained by adding ) is the target posture α tgt (t) is equal to (t). In this case, if the angular velocity command calculation unit 40a-1 is calculated based on the former, the angular velocity command value ω is obtained as shown in equation (5) in Figure 23. αc When (t) is a negative value, and the angular velocity command calculation unit 40a-1 performs the calculation based on the latter, the angular velocity command value ω is obtained as shown in equation (6) in Figure 23. αc (t) becomes 0. Here, the one being corrected is time t+t wα Setting the angular velocity command value to 0 at time t is more appropriate because it allows for a more accurate prediction of the attitude.

[0132] As described above, the angular velocity command value ω of boom 8 αc (t) can be calculated more accurately. Furthermore, by performing similar calculations on other actuators, it becomes possible to calculate the angular velocity command value for each actuator more accurately, even when operating multiple actuators. As a result, the controllability of construction machinery is improved, and the desired automated operation can be achieved.

[0133] Furthermore, in this embodiment, the desired angular velocity is the provisional value ω' of the angular velocity command value. αc (t) However, the ratio of the angular velocity command to the actual angular velocity is also corrected by the correction parameter k α This is represented by (t). Therefore, for example, if the actual angular velocity is only half of the desired angular velocity command, the actual angular velocity can be made to match the desired angular velocity command by correcting the angular velocity command to twice its value (effect shown in Figure 17).

[0134] As described above, the control device (control controller 40) of the automated construction machine (hydraulic excavator 1) of this embodiment generates angular velocity command values ​​for each joint (each driven member) based on rotation angle information (posture information) of each joint of the multi-jointed work device (front work device 1A having a boom 8, arm 9, and bucket 10) provided on the construction machine and the target posture of the work device (operation control unit 40a), and operates a plurality of actuators that drive the work device based on the angular velocity command values ​​of each joint. The control device for the machine, the control device (control controller 40), calculates a predicted angular displacement value (angle displacement prediction unit 40b) that is predicted to occur when each joint rotates from the present time to a predetermined time (dead time) later, based on the past values ​​of the angular velocity command values ​​of each joint (integration), and corrects the predicted angular displacement value from the present time to a predetermined time (dead time) later (corrected angular displacement prediction value Δα) based on the predicted angular displacement value from a predetermined time (dead time) earlier and the rotation angle information from a predetermined time (dead time) earlier. ec (t, t+t) wα The correction parameter calculation unit 40b-2 and the future state prediction unit 40b-3 calculate the future posture of the work device at the predetermined time from the present time to the predetermined time (dead time) later (angle movement prediction unit 40b) by adding together the predicted angular movement amount from the present time to the predetermined time (dead time) later and the rotation angle information at the present time (angle movement amount prediction unit 40b), generate the angular velocity command value of each joint to be output at the present time based on the future posture of the work device and the target posture of the work device (operation control unit 40a), and perform automatic operation control to operate the plurality of actuators (automatically) based on the angular velocity command value of each joint to be output at the present time.

[0135] The control device (control controller 40) adjusts the correction parameter (k) based on the predicted angular movement amount from a predetermined time (dead time) before (from before the predetermined time to the present), the rotation angle information from a predetermined time (dead time), and the rotation angle information at the present time. α (t)) is calculated (Correction parameter calculation unit 40b-2: Figure 14), and the correction parameter (k αBased on (t), the predicted angular displacement value from the present time to after the predetermined time (dead time) is corrected by integrating the correction parameter with the predicted angular displacement value from the present time to after the predetermined time (dead time) (corrected angular displacement value Δα ec (t, t+t) wα ))(Future state prediction unit 40b-3: Figure 16).

[0136] The control device (control controller 40) calculates the difference between the rotation angle information from a predetermined time (dead time) before and the rotation angle information at the current time (actual angle movement amount Δα(tt) wα , t)) is the predicted angular displacement value (predicted angular displacement value (zero division avoidance correction) Δα' from the predetermined time (dead time) before (from the predetermined time to the present time) e (tt wα By dividing by t), the correction parameter (k α (t)) is calculated (correction parameter calculation unit 40b-2: Figure 14).

[0137] The control device (control controller 40) generates a provisional value of the angular velocity command value of each joint to be output at the present time based on the future posture of the work device and the target posture of the work device, and the provisional value of the angular velocity command value and the correction parameter (k α Based on (t), the system generates the angular velocity command values ​​for each joint to be output at the present time (operation control unit 40a: Figure 17).

[0138] In other words, the control device (control controller 40) of the automated construction machine (hydraulic excavator 1) of this embodiment stores the predicted angular displacement and rotation angle information (attitude information) as history, and uses at least the history of the predicted angular displacement, the history of the rotation angle information, and the current rotation angle information to calculate (correct) the predicted angular displacement from the present time to a predetermined time after the time has elapsed.

[0139] According to this embodiment, based on past values ​​of the angular velocity command values ​​of each joint of the work device, a predicted angular displacement value is calculated for each joint that is expected to rotate from the present time to a predetermined time later. By correcting the predicted angular displacement value to match the actual angular displacement in the past, it becomes possible to predict the angular displacement more accurately. This improves the controllability of an automated construction machine equipped with multiple pumps and multiple hydraulic actuators, enabling the realization of desired automated operation.

[0140] <Variations of the Embodiment> It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail to make the present invention easier to understand, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. For example, the integral calculation in the partial integration calculation unit 40b-221 in this embodiment is performed on the angular velocity command value ω αc It is also possible to calculate it using angular acceleration, angular jerk, etc., in addition to (t).

[0141] Furthermore, some or all of the functions of the controller in the above-described embodiment may be implemented in hardware, for example, by designing them as integrated circuits. Alternatively, they may be implemented in software by having a processor interpret and execute programs that implement each function. Information such as programs, tables, and files that implement each function can be stored in storage devices such as hard disks, SSDs (Solid State Drives), or recording media such as IC cards, SD cards, or DVDs, in addition to the storage device within the controller. [Explanation of Symbols]

[0142] 1. Hydraulic excavator (autonomous construction machine) 1A…Front work device 1B... Vehicle body 8... Boom 9... Arm 10…bucket 11…Lower running body 12…Upper rotating body 16a~16l...Load detection device 18…Engine (power source) 22, 23...Operating device 30... Boom angle sensor 31... Arm angle sensor 32…Bucket angle sensor 33... Vehicle tilt angle sensor 34... Swivel angle sensor 40…Control controller (controller) 50…Posture detection device 54, 55, 56, 57, 58, 59... Solenoid proportional valves 470…Engine Controller 480...Engine speed setting device

Claims

1. A control device for an automated construction machine that generates angular velocity command values ​​for each joint based on rotation angle information of each joint of a multi-jointed work device provided on the construction machine and the target posture of the work device, and operates a plurality of actuators that drive the work device based on the angular velocity command values ​​of each joint, The control device is Based on the past values ​​of the angular velocity command values ​​for each of the aforementioned joints, a predicted angular displacement value is calculated for each of the aforementioned joints that is expected to rotate from the present time to a predetermined time later. Based on the predicted angular displacement amount from the predetermined time prior and the rotation angle information from the predetermined time prior, the predicted angular displacement amount from the present time to the predetermined time prior is corrected. Based on the predicted angular movement amount from the present moment to the predetermined time, and the rotation angle information at the present moment, the future orientation of the work device at the predetermined time is calculated. Based on the future orientation of the work device and the target orientation of the work device, the angular velocity command values ​​of each joint to be output at the present time are generated. A control device for an automated construction machine, characterized by performing automatic motion control to operate the plurality of actuators based on the angular velocity command values ​​of each joint output at the present time.

2. A control device for an automated construction machine according to claim 1, The control device for an automated construction machine is characterized by calculating correction parameters based on the predicted angular displacement amount from a predetermined time prior, the rotation angle information from a predetermined time prior, and the rotation angle information at the present time, and correcting the predicted angular displacement amount from the present time to the predetermined time prior based on the correction parameters.

3. A control device for an automated construction machine according to claim 2, The control device for an automated construction machine is characterized by calculating the correction parameter by dividing the difference between the rotation angle information from a predetermined time ago and the rotation angle information at the present time by the predicted value of the angle movement amount from a predetermined time ago.

4. A control device for an automated construction machine according to claim 1, The control device for an automated construction machine is characterized in that the predetermined time is the time it takes to operate the plurality of actuators based on the angular velocity command value until the angular velocity of each joint of the work device reaches the angular velocity command value.

5. A control device for an automated construction machine according to claim 2, The control device for an automated construction machine is characterized by generating a provisional value for the angular velocity command value of each joint to be output at the present time based on the future posture of the work device and the target posture of the work device, and generating the angular velocity command value of each joint to be output at the present time based on the provisional value for the angular velocity command value and the correction parameter.

Images