Control device and control method

Abstract

The control device is provided with: an operation command unit for controlling each cylinder (44-46) for driving the excavator to be controlled; an ideal speed calculation unit that calculates an ideal speed to be generated in the next time step on the basis of the target position and the current position of the bucket distal end (43a); a reference speed calculation unit that calculates, on the basis of the quasi-static characteristics of each cylinder, a reference speed that is a speed closest to an ideal speed in the range of actuator forces that can be generated when PID control is performed on each cylinder, and a reference generated force that is a generated force when PID control is performed at the reference speed; and an operation amount calculation unit that calculates an operation amount by referring to the generated force and the reference speed on the basis of the quasi-static characteristics of each cylinder.

Description

Technical Field

[0001] The present invention relates to a control device and a control method for controlling the position of a control object driven by a hydraulic actuator or a force acting on the control object. Background Art

[0002] A control method has been developed to automatically move a bucket, working device, etc. in a machine such as an excavator driven by a hydraulic actuator to a target position. The response characteristics of a hydraulic actuator including a hydraulic actuator have strong nonlinearity, which varies greatly based on the opening and closing of a relief valve. Therefore, it is difficult to achieve automatic positioning control using a simple control law.

[0003] Furthermore, when force control is performed on a machine driven by a hydraulic actuator, specifically, when admittance control is performed on a force acting on a control object in contact with the external environment, it is difficult to achieve this with a simple control law due to the strong nonlinearity of the hydraulic actuator.

[0004] In Patent Document 1, based on prior information on the dynamic characteristics of the controlled object, the prediction equation for predicting the position and speed of the controlled object in the next step and the equation representing the sliding mode control law are solved as a set of simultaneous equations, and the control algorithm is simplified by calculating the operating quantity, that is, the valve opening instruction for the hydraulic actuator.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Publication No. 2021-121717

[0008] In the control method of Patent Document 1, the dynamic characteristics of the controlled object are used when calculating the valve opening instruction for the hydraulic actuator. However, when the prior information of the dynamic characteristics of the controlled object is incorrect or there is a time lag in the response of the hydraulic actuator, the accuracy of the prediction equation may not be guaranteed, and the desired control characteristics may not be obtained. Summary of the invention

[0009] An object of the present invention is to provide a control device and a control method that can achieve high control performance in position control or force control even when the accuracy of prior information related to the dynamic characteristics of the controlled object is low and when there is a time lag in the response of the hydraulic actuator.

[0010] The control device involved in the first aspect of the present invention includes: an action command unit that controls a hydraulic actuator that drives a machine that is a control object; an ideal speed calculation unit that calculates the ideal speed that the machine should generate in the next step based on the target position and current position of the machine; a reference speed calculation unit that calculates a reference speed and a reference generated force based on the quasi-static characteristics of the hydraulic actuator, the reference speed being a speed that is closest to the ideal speed among the target speeds that can calculate the actuator force that can be generated when the hydraulic actuator implements PID control or PD control, and the reference generated force being the actuator force that should be generated when PID control or PD control is implemented using the reference speed as the target speed; and an operation amount calculation unit that calculates the operation amount of the hydraulic actuator based on the quasi-static characteristics of the hydraulic actuator, the reference generated force and the reference speed.

[0011] In addition, the control method involved in the second aspect of the present invention includes: an ideal speed calculation step, calculating the ideal speed that should be generated by the machine in the next step according to the target position and current position of the machine driven by the hydraulic actuator as the control object; a reference speed calculation step, calculating the reference speed and reference generated force according to the quasi-static characteristics of the hydraulic actuator, the reference speed being the speed closest to the ideal speed among the target speeds that can calculate the actuator force that can be generated when the hydraulic actuator implements PID control or PD control, and the reference generated force being the actuator force that should be generated when PID control or PD control is implemented using the reference speed as the target speed; and an operation amount calculation step, calculating the operation amount of the hydraulic actuator according to the quasi-static characteristics of the hydraulic actuator, the reference generated force and the reference speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a schematic diagram showing the structure of a hydraulic actuator according to the first embodiment of the present invention.

[0013] Figure 2 This is a side view schematically showing the excavator according to the first embodiment.

[0014] Figure 3 This is a block diagram showing the configuration of the control system according to the first embodiment.

[0015] Figure 4 This is a conceptual diagram showing a method of controlling a plurality of hydraulic actuators.

[0016] Figure 5 This is a flowchart showing the flow of position control according to the first embodiment.

[0017] Figure 61 is a diagram showing the posture of the hydraulic excavator and the trajectory of the target position according to the numerical example.

[0018] Fig. 7A It is a conceptual diagram showing the flow of processing in the control law, and is a diagram of the state of the conventional control law.

[0019] Figure 7B It is a conceptual diagram showing the flow of processing in the control law, and is a diagram of the state of the control law according to the first embodiment of the present invention.

[0020] Figure 8 It is a diagram showing simulation results based on a conventional control law and a control law according to the first embodiment of the present invention.

[0021] Fig.9A 1 is a diagram showing simulation results for comparing the influence of modeling errors, and is a diagram of the control law according to the first embodiment of the present invention without regenerative loop compensation.

[0022] Fig. 9B 1 is a diagram showing simulation results for comparing the influence of modeling errors, and is a diagram of the control law according to the first embodiment of the present invention with regenerative loop compensation.

[0023] Fig. 9C It is a diagram showing simulation results for comparing the influence of modeling errors, and is a diagram of the state of conventional control laws.

[0024] Fig.10 This is a block diagram showing the structure of a control system according to the second embodiment.

[0025] Fig.11 This is a functional block diagram showing the configuration of a control system according to the second embodiment.

[0026] Fig.12 This is a flowchart showing the flow of force control according to the second embodiment.

[0027] Fig.13 This is a conceptual diagram showing the structure of a hydraulic testing machine according to the numerical example.

[0028] Fig.14A It is a diagram showing simulation results related to a numerical example in the case of step input, and is a diagram for a case where the hardness of the contact environment is 50HS.

[0029] Fig. 14B It is a diagram showing simulation results related to a numerical example in the case of a step-type input, and is a diagram for a case where the hardness of the contact environment is 65HS.

[0030] Fig. 14CIt is a diagram showing simulation results related to a numerical example in the case of a step-type input, and is a diagram for a case where the hardness of the contact environment is 70HS.

[0031] Fig.15A It is a diagram showing simulation results related to a numerical example in the case of a sine wave input, and is a diagram for a case where the hardness of the contact environment is 50HS.

[0032] Fig. 15B It is a diagram showing simulation results related to a numerical example in the case of a sine wave input, and is a diagram for a case where the hardness of the contact environment is 65HS.

[0033] Fig. 15C It is a diagram showing simulation results related to a numerical example in the case of a sine wave input, and is a diagram for a case where the hardness of the contact environment is 70HS. DETAILED DESCRIPTION

[0034] <Implementation Method 1>

[0035] Hereinafter, among the control devices and control methods according to the embodiments of the present invention, a control device and a control method concerning position control will be described with reference to the drawings.

[0036] <Control Algorithm>

[0037] First, the control law involved in this embodiment, that is, the control algorithm, is described. This control algorithm takes a machine driven by a hydraulic actuator as a control object and determines the operation amount u of the hydraulic actuator when the control object is moved to a target position. Specifically, Figure 1 The oil pressure actuator 1 shown is considered as a hydraulic actuator.

[0038] like Figure 1 As shown, the hydraulic actuator 1 includes a hydraulic pump 11a for supplying hydraulic pressure, a pump relief valve 11b, a drain valve 11c, a pump check valve 11d, a main control valve 12 for controlling hydraulic pressure, a rod-side relief valve 13a, a rod-side check valve 13b, a head-side relief valve 14a, a head-side check valve 14b, and a cylinder 15 that operates based on hydraulic pressure control. In addition, the hydraulic actuator 1 includes a regeneration circuit 16, and the regeneration circuit 16 includes a check valve 16a and a flow control valve 16b.

[0039] <Mathematical Preparation>

[0040] In this embodiment, the closed unit sphere B and the function represented by the following equations (1), (2), (3), and (4) are used.

[0041]

[0042] In addition, a function having a set value as an argument is handled as in the following equation (5).

[0043]

[0044] Here, X represents a real closed interval.

[0045] <Control Law>

[0046] The hydraulic actuator is represented by the following equations (6a), (6b), and (6c).

[0047]

[0048] f∈Γ(v,u) (6c)

[0049] Here, p represents the position of the controlled object, v represents the velocity of the controlled object, and M represents the mass of the controlled object.

[0050] The actuator force f and the external force g act on the control object of the above equation. The set-valued function Γ is a quasi-static model of the hydraulic actuator, which is defined as a set-valued function from the current speed v and the valve opening instruction, that is, the operating amount u, to the actuator force f. In equation (19) of the well-known document 1 (R. Kikuuwe, et al., "A nonsmooth quasi-static modeling approach for hydraulic actuators," J. Dyn. Sys., Meas., Control, vol. 143, no. 12, p. 121002, 2021) and equation (24) of the well-known document 2 (Y Yamamoto, et al., "Asliding-mode set-point position controller for hydraulic excavators," IEEE Access, vol. 9, pp. 153735-153749, 2021), it is expressed in equation (24) from Figure 1 The quasi-static model of the hydraulic actuator 1 driven in the state where the regeneration circuit 16 is removed from the hydraulic circuit of the hydraulic actuator 1, that is, when the regeneration circuit 16 is always closed, is the specific form of the set-valued function Γ.

[0051] The operand u∈B as the second argument of the function Γ is represented by Figure 1 The four variables of the opening degree of the main control valve 12 are shown. The operation amount u and the opening degree u*∈[0,1](*∈{ph,pr,th,tr}) of the main control valve 12 have the relationship of the following equation (7).

[0052] u ph =u tr =max(0,u),u pt =u th = -min(0,u) (7)

[0053] According to the above relational expression (7), the manipulated variable u as a positive control input becomes a command to extend the cylinder, and the manipulated variable u as a negative control input becomes a command to retract the cylinder.

[0054] In this embodiment, the position control law for the hydraulic actuator represented by the following equations (8a), (8b), (8c), and (8d) is used.

[0055]

[0056] Here, K, L, D are PID gains, H is the time constant, p d is the target position, p is the current position, v r is the reference speed, u is the instruction to the actuator, f^ is the reference force generated by the actuator, and a is the error integral value of the PID controller. In addition, g is the estimated value of the external force. If the estimated value of the external force cannot be obtained, g = 0 can also be set. If the estimated value can be obtained, setting the value to g will obtain better control performance.

[0057] In addition, the function Θ is an inverse function of the second argument of the set-valued function Γ, and is a set-valued function having a relationship with the set-valued function Γ as shown in the following formula (9).

[0058]

[0059] Formula (8a) is based on the speed p · and reference speed v r The PID control law is constructed. In addition, equation (8b) is a sliding mode control law with the following equation (10) as the switching plane.

[0060]

[0061] In this embodiment, the PID control law of formula (8a) is combined with the sliding mode control law of formula (8b). Hereinafter, this method is referred to as differential algebraic relaxation. Since the right side of the sliding mode control law of formula (8b) is a set value, the value cannot be uniquely determined and is not suitable for single implementation. However, by combining it with the PID control law of formula (8a), such a problem can be avoided, and the reference generated force f^ can be calculated (operated (the same applies below)). In addition, in each formula, symbols such as dots attached to symbols are attached after the symbols in this article, but what they represent is the same.

[0062] <Discrete-time algorithm>

[0063] In order to realize the above control law, a control algorithm in discrete time is considered. By using the backward Euler method to discretize the control law of equation (8) (equations (8a) to (8d)), the following equation (11) (equations (11a) to (11f)) can be obtained.

[0064]

[0065] a k =2a k-1 -a k-2 Ten T 2 (v r，k -v k ) (11d)

[0066] v k =(p k -p k-1 ) / T (11e)

[0067] v d，k =(p d，k -p d，k-1 ) / T (11f)

[0068] Here, T represents the sampling interval and k∈Z represents the index of discrete time.

[0069] Using equations (11a) and (11d), we can eliminate v from equation (11b). r，k and a k After that, we get the following formula (12) (formula (12a) to formula (12c)).

[0070]

[0071] Here, A = ΔLT 2 +KT+D (as shown in Formula 12, =Δ is equivalent to the symbol formed by placing Δ on = in the up and down directions).

[0072] Here, v s，k can be interpreted as the ideal speed that should be achieved in the next step by the sliding mode control law. In addition, v f，k It can be interpreted as a provisional target speed for making the generated force of the hydraulic actuator zero when the PID control law is implemented on the hydraulic actuator.

[0073] The well-known document 3 (R. Kikuuwe, Y. Yamamoto, and B. Brogliato, “Implicit implementation of nonsmooth controllers to nonsmooth actuators,” IEEE Trans. Autom. Control, 10.1109 / TAC.2022.3163124, 2022) shows that the following formula (13) holds.

[0074]

[0075] By using formula (13), formula (12a) can be rewritten as the following formula (14).

[0076]

[0077] Thus, we can get the k , v k , p d，k , v d，k}Get the control input u k The algorithm of the following formula (15) (formula (15a) to formula (15f)).

[0078] v s，k :=(p d，k +Hv d，k -p k ) / H (15a)

[0079] v f，k : =vk-((K / T+L)(a k-1 -a k-2 )+La k-1 -g k ) / A (15b)

[0080]

[0081] a k :=2a k-1 -a k-2 +T 2 (v r，k -v k ) (15e)

[0082]

[0083] Here, the function Γ s And function Θ s It is a single-valued function that has a relationship with function Γ and function Θ as shown in the following equations (16) and (17).

[0084]

[0085] Θ s (v, f)∈Θ(v, sat Γ(v，β) (f)) (17)

[0086] As described above, the set-valued function Γ can be obtained from the quasi-static characteristics of the hydraulic actuator (the relationship between the velocity v at steady speed, the generated force f, and the operation amount u as the valve opening instruction). In addition, the set-valued function θ can be obtained from the set-valued function Γ by using equation (9). Using these relationships, equations (16) and (17), the single-valued function Γ can be obtained s And the single-valued function Θ s In the well-known document 2, formula (34) and Section III.C respectively indicate that Figure 1 The single-valued function Γ of the hydraulic actuator 1 driven in the state where the regeneration circuit 16 is removed from the hydraulic circuit of the hydraulic actuator 1 shown, that is, in the state where the regeneration circuit 16 is always closed s And the single-valued function Θ s specific form.

[0087] The force f that the hydraulic actuator 1 can generate at time k is k and produces a force f k The target speed v r，k The pair of real numbers that satisfies the following two equations (18a) and (18b) {f k ^, v r，k}.

[0088]

[0089] There are infinite pairs like the above. On the other hand, the pairs {f k ^, v r，k} is an infinite pair of {f k ^, v r，k} r，k Closest to v s，kThat is, equations (15b), (15c), and (15d) are the following steps (reference speed calculation step): Calculate the reference speed v according to the quasi-static characteristics of the hydraulic actuator r，k and the reference force f k ^, reference speed v r，k The target speed v is used to calculate the actuator force that can be generated when the hydraulic actuator 1 is subjected to PID control. f，k The closest to the ideal speed v s，k The speed of the reference force f k ^ is to set the reference speed v r，k As the target speed v f，k The actuator force that should be generated when PID control is implemented.

[0090] In addition, when L=0 in the algorithm of formula (15), a k In this case, the use of e k =(a k -a k-1 ) / T defines a new variable e k The algorithm of equation (15) is modified into the following equations (19a) to (19f).

[0091] v s，k :=(p d，k +Hv d，k -p k ) / H (19a)

[0092] v f，k :=v k -(Ke k-1 -g k ) / A (19b)

[0093]

[0094] e k :=e k-1 +T(v r，k -v k ) (19e)

[0095]

[0096] Here, A=ΔKT+D (=Δ is equivalent to the symbol obtained by placing Δ above = in the vertical direction).

[0097] The algorithm expressed by the above equation (19) is obtained by replacing a with e in equation (8). ·And set L = 0, discretize using the backward Euler method, and it can be derived through the same derivation process as the above equation (15). Through equation (19), the PID control of equation (15) can be replaced by PD control with integral gain L = 0.

[0098] In the control law based on the quasi-static model as in the present embodiment, i.e., the conventional control law involved in the well-known document 2, a double-implicit implementation scheme is adopted to deal with the set-valuedness of the sliding mode control law. In the double-implicit implementation scheme, the control law and the dynamic characteristic model of the controlled object are discretized by the backward Euler method and the control algorithm is constructed by combining them. In the control method based on the differential algebraic relaxation involved in the present embodiment, since the dynamic characteristic model is not required in the implementation, the modeling error can be suppressed more than the conventional control method. In addition, by adjusting the gain of the PID controller coupled by the differential algebraic relaxation, the sensitivity to the modeling error and the dead time can be adjusted.

[0099] As shown in the algorithm of equation (19), in the position control according to the present embodiment, according to the target position p of the hydraulic actuator, d，k and the current position p k , in more detail, according to the target position p d，k 、Current location p k and the target velocity v derived from these positions d，k 、Current speed v k , calculate the ideal velocity v that the hydraulic actuator should produce at the next time step s，k (Ideal speed calculation step, equation (19a)) Furthermore, based on the quasi-static model Γ of the hydraulic actuator, the reference speed v is calculated. r，k and the reference force f k ^, reference speed v r，k The target speed v is used to calculate the actuator force f that can be generated when the hydraulic actuator is subjected to PID control or PD control. f，k The closest to the ideal speed v s，k The speed of the reference force f k ^ is based on the reference speed v r，k The actuator force to be generated when implementing PID control or PD control as the target speed (refer to the speed calculation step, equations (19b) to (19e)). In addition, based on the quasi-static model Γ of the hydraulic actuator, the reference generated force f is calculated. k ^ and reference speed v r，k , calculate the operating amount u of the hydraulic actuator k(Operation amount calculation steps, formula (19f)).

[0100] <Compensation for the influence of the regeneration circuit>

[0101] like Figure 1 As shown, the hydraulic actuator 1 includes a regeneration circuit 16. The regeneration circuit 16 uses the potential energy of the connecting rod (link) to send the working fluid (working oil) from the rod side chamber to the head side chamber. If the cross-sectional area of ​​the rod side chamber is A r , let the cross-sectional area of ​​the head side chamber be A h , and A - =(A r +A h ) / 2, the pressure difference between the rod side chamber and the head side chamber generated by the external force g can be approximated as g / A - Since the flow rate of the flow control valve 16b is proportional to the square root of the pressure difference between the two chambers, the flow rate q of the working oil passing through the regeneration circuit is reg The estimated value of is obtained by the following formula (20).

[0102]

[0103] Here, U reg Indicates the opening degree of the flow control valve 16b of the regeneration circuit 16. Coefficient C reg By C reg a reg √2 / p, p is the mass density of the working oil, a reg is the maximum opening area of ​​the flow control valve 16b, C reg is the outflow coefficient.

[0104] The flow rate q based on the regeneration network 16 reg The increment of cylinder velocity v reg is found to be v reg =q reg / A h In addition, the variable v reg The check valve 16a of the regeneration circuit 16 has a non-negative value. reg By expanding the control law of equation (8) to compensate for the influence of the regeneration circuit 16, the control laws shown in the following equations (21a), (21b), and (21c) can be obtained.

[0105]

[0106] By importing the above variable V reg When the hydraulic oil is sent to the head side chamber through the regeneration circuit 16, the switching plane becomes p d -p+H(p d· -v r -V reg ) = 0. Thus, since the switching plane is reached at a speed lower than vreg compared to the case without compensation, control can be performed to suppress an excessive speed increase.

[0107] Used to implement the variable V reg The control law algorithm is obtained by applying the same process as the derivation of the algorithm of equation (15) to the control law of equation (21) for discretization. The derived algorithm becomes an algorithm in which equation (15a) of the algorithm expressed by equation (15) is replaced by the following equation (22).

[0108] v s，k :=(p d，k +H(v d，k -v reg，k )-p k ) / H (22)

[0109] Variable V reg，k The value of can be calculated for each time step related to the calculation using equation (20). At this time, the gravity acting on the cylinder 15 derived from the joint angle and the mass of the link can be used as the external force g.

[0110] <Control system configuration>

[0111] In the present embodiment, as an example of a specific configuration of a control system, a case where the position control of a bucket of an excavator driven by a plurality of hydraulic actuators is performed will be described.

[0112] The position control device according to the present embodiment, that is, the control unit 50 is mounted on Figure 2 The excavator 30 shown in the figure is used to control the movement of the excavator 30.

[0113] The excavator 30 includes a lower traveling body 31, an upper slewing body 32, and a working device 40. The lower traveling body 31 is a part that allows the excavator 30 to travel, and is, for example, a crawler. The upper slewing body 32 is rotatably mounted on the lower traveling body 31 via a slewing motor 47. The upper slewing body 32 is provided with a cab for an operator to operate the excavator 30, the working device 40, and the like.

[0114] The working device 40 includes a boom 41 rotatably mounted on the upper swing body 32 , an arm 42 rotatably mounted on the boom 41 , and a bucket 43 rotatably mounted on the arm 42 for performing excavation or the like.

[0115] In addition, the working device 40 includes a boom cylinder 44 connected to the upper rotating body 32 and the boom 41 and moving the boom 41, an arm cylinder 45 connected to the boom 41 and moving the arm 42, a bucket cylinder 46 connected to the arm 42 and the bucket 43 and moving the bucket 43, and a swing motor 47 connected to the lower traveling body 31 and the upper rotating body 32 and moving the upper rotating body 32. The boom cylinder 44, the arm cylinder 45, the bucket cylinder 46, and the swing motor 47 (hereinafter also referred to as the actuators 44 to 47) are hydraulic actuators driven by hydraulic pressure.

[0116] The control unit 50 is mounted on the excavator 30. Figure 3 As shown in the block diagram of , it includes a control unit 51, a storage unit 52, a display unit 53, and an input unit 54. The control unit 50 is connected to the position sensor 48, the speed sensor 49, the actuators 44 to 47, etc., and performs motion control of each part of the excavator 30.

[0117] The control unit 51 is a computer device composed of a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc., and controls the operation of the excavator 30. The control unit 51 operates the CPU by reading various operation programs and data stored in the ROM of the control unit 51, the storage unit 52, etc. into the RAM, thereby realizing Figure 3 The functions of the control unit 51 are shown. Thus, the control unit 51 operates as an ideal speed calculation unit 511 (ideal speed calculation unit), a reference speed calculation unit 512 (reference calculation calculation unit), an operation amount calculation unit 513 (operation amount calculation unit), an external force estimation unit 514, and an action instruction unit 515.

[0118] The ideal speed calculation unit 511, the reference speed calculation unit 512 and the operation amount calculation unit 513 calculate the operation amount u of each actuator 44 to 47 used to move the working device 40 based on the outputs of the position sensor 48 and the speed sensor 49, that is, the measured values ​​and the target position of the bucket 43, more specifically, the target position of the far end of the bucket 43 (hereinafter referred to as the bucket far end 43a), which is a specified part of the excavator 30.

[0119] The external force estimating unit 514 estimates the magnitude of the external force acting on each actuator 44 to 47. The method of estimating the external force is not particularly limited, and for example, the external force may be estimated based on the output of a sensor (not shown) provided in each actuator 44 to 47.

[0120] The motion command unit 515 controls the actuators 44 to 47 based on the operation amount u calculated by the operation amount calculation unit 513 to operate the upper swing body 32 , the boom 41 , the arm 42 , and the bucket 43 .

[0121] The storage unit 52 is a nonvolatile memory such as a hard disk or a flash memory, and stores various setting parameters, a control algorithm for calculating the operation amount u, and the like.

[0122] The display unit 53 is a display device such as a liquid crystal panel or organic EL (Electroluminescence), and displays various setting parameters, detection values ​​of the position sensor 48 and the speed sensor 49, etc. The display unit 53 according to this embodiment is a liquid crystal panel provided in the cab of the excavator 30 .

[0123] The input unit 54 is an input device for inputting various setting parameters for operating the excavator 30 , such as a target position of the bucket tip 43 a. The input unit 54 is, for example, a touch screen disposed on the display unit 53 .

[0124] The position sensor 48 is a sensor for detecting the position of the working device 40 of the excavator 30 as the control object, and the position sensor 48 involved in this embodiment includes a swing angle sensor 484 for detecting the angle of the upper swing body 32, a boom angle sensor 481 for detecting the angle of the boom, an arm angle sensor 482 for detecting the angle of the arm, and a bucket angle sensor 483 for detecting the angle of the bucket. The control unit 51 calculates the lengths of the boom cylinder 44, the arm cylinder 45, and the bucket cylinder 46 (hereinafter also referred to as each cylinder 44 to 46) and the angle of the swing motor 47 based on the angles of each part of the working device 40 detected by the boom angle sensor 481, the arm angle sensor 482, the bucket angle sensor 483, and the swing angle sensor 484, and performs position control of the bucket distal end 43a based on the calculated lengths of each cylinder 44 to 46 and the angle of the swing motor 47.

[0125] The speed sensor 49 is a sensor for detecting the speed of each part of the working device 40. The speed sensor 49 involved in this embodiment includes a rotation angular velocity sensor 494 for detecting the angular velocity of the upper rotating body 32, a boom speed sensor 491 for detecting the extension and retraction speed of the boom cylinder, a boom speed sensor 492 for detecting the extension and retraction speed of the boom cylinder, and a bucket speed sensor 493 for detecting the extension and retraction speed of the bucket cylinder.

[0126] Next, the position control process of the bucket distal end 43a performed by the control unit 50 according to the present embodiment will be described. Figure 4As shown in the block diagram, the control algorithm for calculating the operation amount u of each cylinder 44 to 46 as a hydraulic actuator is used as an example for explanation. In addition, in the following embodiment, only the cylinders 44 to 46 are used, but it is not limited to this. It is also possible to configure a controller that adds the rotation angle of the upper rotating body 32, and according to the target position p d The position of the bucket distal end 43a is controlled by using the coordinates of Figure 3 As shown, the rotation angle sensor 484 as the position sensor 48 and the rotation angular velocity sensor 494 as the velocity sensor 49 are used to actuate the rotation motor 47 as the hydraulic actuator and combine with the actions of the boom 41, the arm 42, and the bucket 43 to control the position of the bucket distal end 43a.

[0127] In the control algorithm of the present invention, the control unit 51 calculates the operation amount u of each cylinder 44 to 46 based on the given target position of the control object, the desired behavior, the current position, the current speed, etc. The desired behavior is, for example, the time constant of the convergence movement of the control object to the target position.

[0128] <Control Flow>

[0129] Reference Figure 5 The flow chart of FIG. 1 will specifically describe the process of position control of the bucket distal end 43a of the excavator 30.

[0130] The operator of the excavator 30 inputs and sets the target position p of the bucket distal end 43a through the input unit 54. d (Step S11). Target position p d The input may be a coordinate input by a touch screen, i.e., input unit 54, or the coordinates of the bucket end 43a manually moved by a joystick may be used as the target position p. d Alternatively, the target position p may be set by reading the coordinates on the design surface pre-stored in the storage unit 52. d .

[0131] In addition, the operator inputs and sets parameters such as the time constant H indicating a desired behavior. These parameters may be set by reading values ​​stored in advance in the storage unit 52 .

[0132] The control unit 51 controls the input target position p d , calculate the target position p by inverse kinematics d The target length of each cylinder 44 to 46 on the upper surface is obtained (step S12). Figure 4 As shown, the position of the bucket tip 43a is controlled by controlling and extending each cylinder 44 to 46 so that the length of each cylinder 44 to 46 reaches a target length.

[0133] As an ideal speed calculation step, the ideal speed calculation unit 511 of the control unit 51 calculates the ideal speed v to be generated at the next time step. s，k (Step S13). More specifically, the control unit 51 obtains angle data of each part of the working device 40 from the boom angle sensor 481, the arm angle sensor 482, and the bucket angle sensor 483 (hereinafter also referred to as each angle sensor 481 to 483). The control unit 51 calculates the current position p as the length of each cylinder 44 to 46 based on the obtained angle data. k The control unit 51 calculates the current position p k (m), the set target position p d，k (m), target speed v d，k (m / s), time constant H (s) and equation (15a) or equation (22), calculate the ideal speed v of each cylinder 44 to 46 s，k The control period is not particularly limited, and is, for example, 10 ms (milliseconds).

[0134] Next, as a reference speed calculation step, the reference speed calculation unit 512 of the control unit 51 calculates the reference speed vk of the next time step based on the current speed vk (m / sec), the set PID gains K, L, D, the control period T (sec), and equations (15b) to (15e) or equations (19b) to (19e). r，k and the reference force f k More specifically, the reference speed calculation unit 512 obtains the current speed v of each cylinder 44 to 46 from the boom speed sensor 491, the arm speed sensor 492, and the bucket speed sensor 493. k Furthermore, the reference speed calculation unit 512 calculates the reference speed v based on the quasi-static model Γ of the hydraulic actuator. r，k and the reference force f k ^, reference speed v r，k The target speed v is used to calculate the actuator force f that can be generated when the hydraulic actuator is subjected to PID control or PD control. f，k The closest to the ideal speed v s，k The speed of the reference force f k ^ is based on the reference speed v r，k The actuator force that should be generated when PID control or PD control is performed as the target speed.

[0135] Next, as an operation amount calculation step, the operation amount calculation unit 513 of the control unit 51 calculates the operation amount u of each cylinder 44 to 46. k(Step S15) Specifically, the operation amount calculation unit 513 calculates the operation amount u based on the reference speed vr,k and the reference generated force fk^ calculated in step S14 and the quasi-static model Γ of the hydraulic actuator. k (Formula (15f), (17)).

[0136] The motion instruction unit 515 of the control unit 51 sends the calculated operation amount u k The output is given to each cylinder 44 to 46 to operate the boom 41, the arm 42, and the bucket 43, thereby moving the bucket distal end 43a (step S16).

[0137] When the next time step is reached, the control unit 51 re-acquires the current position p of each cylinder 44 to 46 from the position sensor 48. k . At the current position p obtained k With the target position p d，k When the difference between φ and φ is equal to or smaller than a predetermined threshold value set in advance (“Yes” in step S17 ), the control unit 51 ends the position control.

[0138] In addition, at the current position p obtained k With the target position p d，k When the difference between the operation amount u and the operation amount u is greater than a predetermined threshold value (“No” in step S17), the control unit 51 returns to the ideal speed calculation step in step S13. Then, the control unit 51 calculates the operation amount u at this time step according to the control algorithm. k By repeating this, the control unit 51 performs position control so that the position of the bucket distal end 43a becomes the target position p d .

[0139] As described in detail above, in the control device and control method involved in the position control of the present embodiment, since the reference speed and the actuator force are calculated by combining the PID control law or the PD control law that makes the speed of the controlled object follow the reference speed with the sliding mode control law, and the operation amount of the hydraulic actuator is calculated based on the quasi-static characteristics of the hydraulic actuator, it is possible to perform control without relying on the dynamic characteristic model of the controlled object. Therefore, even when the accuracy of the prior information on the dynamic characteristics of the controlled object is low and when there is a time lag in the response of the hydraulic actuator, high control performance can be obtained.

[0140] <Numerical example>

[0141] The following describes a simulation of the position control of a hydraulic actuator using the control law involved in the present embodiment. In this example, a real-time simulator of a 13-ton hydraulic excavator was used for the simulation. In addition, in this example, the control law involved in the present embodiment is applied to the actuator of each axis by using kinematics to control the position of the far end of the boom. In addition, the sampling interval of the real-time simulator is 0.1ms, and the sampling interval of the controller is 10ms. The real-time simulator and the controller are connected via UDP / IP (User Datagram Protocol / Internet Protocol), and can send and receive joint angle information, control inputs, etc.

[0142] The force generated by the cylinder in the real-time simulator is calculated based on the quasi-static model of the hydraulic actuator described in the above-mentioned well-known document 1. In addition, the force generated by the boom cylinder is calculated based on the quasi-static model including the regeneration circuit described in the well-known document 1. In this example, in order to simulate the dynamic characteristics of the hydraulic actuator and the dead time in the hydraulic actuator, a filter shown by the following formula (23) is set between the actuator model and the controller.

[0143]

[0144] Here, u f is the filtered control input. In addition, the dead time Td, cutoff frequency ω0 and attenuation ratio ζ are set to T d =300msec, ω0=94.2rad / sec, ζ=1.

[0145] Figure 6 The posture and target position q of the excavator 30 are shown as follows: d The target position q d and the target position q d The time change rate of is as shown in the following equation (24), and the magnitude of the time change rate of the target position qd is constant.

[0146]

[0147] In addition, as a parameter of this example, both the boom 41 and the arm 42 of the excavator 30 are K=3×10 5 N / m, B=3×10 5 N·s / m, H=1.0s. In addition, the parameters of the actuator model in the control law are set to the same values ​​as the parameters of the controlled object.

[0148] In order to compare the control method involved in this embodiment with the conventional control method, a control simulation was performed under the same conditions. Fig. 7A The conventional sliding mode control law based on double implicit implementation (the control law involved in the well-known document 2) shown in FIG. 1 utilizes the control law B. Figure 7B The PID control law of this embodiment is shown in FIG. Since the control law A contains a dead time compensator, its look-ahead time T d ^Set to 300ms (=T d The simulation was performed for 100 ms, 150 ms, and 0 ms, respectively. In addition, the time constant H representing the slope of the switching plane is 1.0 s.

[0149] The control law B is expressed by the following equation (25) (equation (25a), equation (25b)).

[0150]

[0151] Here, the proportional gain K p , differential gain K d And integral gain K i are set to 3×10 8 N / m, 3×10 8 N·s / m and 0. These gain values ​​were adjusted by trial and error through repeated simulations.

[0152] Figure 8 is a graph showing the simulation results. Figure 8 As shown, the control method involved in this embodiment (the suggested method in the figure) in which the regenerative circuit compensation is set to be effective is in the entire area between the target position and the boom end position ( Figure 8 The tip position of the shovel. The same is true in the following figures) with less error, which is the best result. The control method (control law B) involved in this embodiment without regenerative circuit compensation increases the error due to the influence of the regenerative circuit during the vertical descent and horizontal retraction when the regenerative circuit is opened.

[0153] Control law A results in an error that always occurs during horizontal retraction. In addition, at the look-ahead time T d When ^=0 (no dead time compensation), the blade tip position always vibrates. Control law B also results in vibration in the entire range.

[0154] <Verification of the impact of modeling errors>

[0155] A simulation was performed to verify the effect of the parameter error of the hydraulic actuator model on the control performance. Based on the values ​​of the overflow pressure and outflow coefficient on the simulator side, an error of {-20%, 0, +20%} was randomly added to the parameters on the control law side. The number of trials was set to 100. In addition, for comparison, the control method involved in this embodiment and the control law A based on the quasi-static model of the hydraulic actuator were simulated in the same manner.

[0156] Fig.9A , Fig. 9B , Fig. 9C is a diagram showing the simulation results. Fig. 9C Compared with the control law B without regenerative loop compensation Fig.9A , with regenerative loop compensation using control law B Fig. 9B For comparison, Fig.9A , Fig. 9B The trajectory of the simulation results (the trajectory with parameter errors) is concentrated near the trajectory in the ideal case without parameter errors. It can be seen that the control performance of the control method involved in this embodiment changes little relative to the parameter error. This result is considered to be because the control method involved in this embodiment does not rely on the dynamic characteristics model of the controlled object and has low model dependence. In addition, since control law A performs dead time compensation based on a quasi-static model and a dynamic characteristics model of the controlled object, it is considered to more significantly show the influence of parameter errors. From Fig.9A and Fig. 9B It can be seen from the comparison that even in the presence of parameter errors, by compensating for the influence of the regeneration circuit, the shovel tip track can be brought closer to the target position q d track.

[0157] As described above, according to the control device and control method involved in the position control of the present embodiment, since the position control is performed by the sliding mode control algebraically coupled with the PID control or the PD control, it becomes a structure that does not depend on the dynamic characteristic model of the control object, and therefore, it is difficult to be affected by parameter errors, dead time, etc. Therefore, even a hydraulic actuator with a long dead time can be properly controlled. In addition, since it is based on the quasi-static model of the hydraulic actuator, it can handle the strong nonlinearity of the hydraulic actuator.

[0158] Furthermore, the control device and control method according to the present embodiment are extended to compensate for the influence of the regenerative circuit, so they can also be applied to a hydraulic actuator including a regenerative circuit.

[0159] In the above embodiment, the speed sensor 49 is used to measure the extension and contraction speed of each cylinder 44 to 46 , but the present invention is not limited thereto. For example, the control unit 51 may calculate the speed of each cylinder 44 to 46 based on the angle data measured by each angle sensor 481 to 483 .

[0160] <Implementation Method 2>

[0161] In the above-mentioned first embodiment, the case where the position control is performed on the control object driven by the hydraulic actuator is described, but the force control of the control object can also be performed using the same control law as in the first embodiment. In this embodiment, a control device and a control method for the case where the force control is performed on the control object driven by the hydraulic actuator are described.

[0162] Specifically, the force control device and the force control method are described by taking as an example the case where the admittance control is a control object driven by the same hydraulic actuator 1 as in Embodiment 1. The control system according to this embodiment has a feedback loop for performing force control. To achieve this, the control unit 51' is different from Embodiment 1 in that it has a reference position operation unit 516 (reference position calculation unit). The other structures are the same as in Embodiment 1, so the same symbols are attached and the description thereof is omitted.

[0163] like Fig.10 As shown in the block diagram of FIG. 1 , the control system according to the present embodiment converts the force (hereinafter also referred to as the reaction force) f exerted by the environment on the hydraulic actuator 1 to be controlled e The measured value of and the target value of the force applied by the hydraulic actuator 1 to the environment, that is, the target applied force f d As input, it controls the action of the hydraulic actuator 1. Fig.10 As shown, the admittance control of the present embodiment is a force control based on position control. More specifically, in the admittance control involved in the present embodiment, based on the internal position controller that performs the position control involved in embodiment 1, the position p of the rod of the hydraulic actuator 1 as the control object is controlled to follow the position q of the virtual object having the target dynamic characteristics of the control object. The target dynamic characteristics of the virtual object are dynamic characteristics defined by the target inertia and target viscosity of the control object, and the virtual object involved in the present embodiment is represented as a mass damping system that models the dynamic characteristics of the hydraulic actuator 1. In addition, it is assumed that a target applied force fd and a reaction force fe act on the virtual object, and the reaction force fe is a force measured as a force from the environment that the hydraulic actuator 1 is subjected to due to the contact between the control object and the highly rigid external environment.

[0164] Hereinafter, a control law according to the present embodiment, that is, a control algorithm for controlling a machine driven by a hydraulic actuator and determining an operation amount u of the hydraulic actuator when force control is performed when the controlled object contacts a high-rigidity environment, will be described.

[0165] <Mathematical Preparation>

[0166] In this embodiment, a function called Normal Cone represented by the following equation (26) is used.

[0167]

[0168] <Control Law>

[0169] The hydraulic actuator to be controlled is expressed by the following equation (27) (equation (27a), equation (27b), equation (27c)).

[0170]

[0171] f∈Γ(v,u) (27c)

[0172] Here, p represents the position of the controlled object, v represents the velocity of the controlled object, and M represents the mass of the controlled object.

[0173] The control object in the above equation is acted on by the actuator's generated force f, external force (disturbance) g, and the reaction force f from the environment. e . The set-valued function Γ is the same quasi-static model of the hydraulic actuator as in implementation mode 1, and is defined as a set-valued function from the current speed v and the valve opening instruction, i.e., the operating amount u, to the force f generated by the actuator. The control input u∈B determines the opening of the flow control valve. When u>0, the flow control valve opens and an oil flow in the direction of extending the hydraulic actuator is generated inside the hydraulic actuator. When u<0, the flow control valve opens and an oil flow in the direction of contracting the hydraulic actuator is generated inside the hydraulic actuator. In addition, the premise that M is known is not set. p can be obtained by the position sensor.

[0174] In this embodiment, the admittance control law represented by each equation of the following equation (28) is used.

[0175]

[0176] Here, g^ is the estimated value of interference, q, B v and M v is the position, viscosity and mass of the virtual object rod. Equation (28a) expresses the force f based on the target d and the measured reaction force f from the environment eThe third term on the right side of equation (28a) is N[-vm,vm](q · ) Based on the definition of the normal cone of equation (26), the velocity q of the virtual object is · Restricted to the interval [-v m , v m ] effect. Equations (28b) to (28d) are position controllers that make the position p of the controlled object follow the position q of the virtual object, and represent the same control law as the position controller involved in Implementation Example 1. In addition, the position controller has gains K, B and time constant H of PID as adjustable parameters.

[0177] <Discrete-time algorithm>

[0178] For {f d，k , f e，k , p k , g k ^} Calculate the control input u from the force controller involved in this embodiment to the hydraulic actuator k The algorithm can be obtained by following the same discretization process as in Implementation 1, which is expressed as follows

[0179] Various forms of formula 29.

[0180]

[0181] q k :=q k-1 +v q，k T (29b)

[0182] v k :=(p k -p k-1 ) / T (29c)

[0183] v s，k :=(q k -p k +Hv q，k ) / H (29d)

[0184]

[0185] e k :=e k-1 +(v r，k -v k )T (29h)

[0186]

[0187] Here, A=ΔKT+B, k is the index of discrete time, and T represents the sampling interval (regarding the sign of =Δ, as described above).

[0188] As shown in the above-mentioned formula (29), the algorithm of force control involved in this embodiment is an algorithm obtained by adding formulas (29a) to (29b) for constituting a force controller to formulas (29c) to (29i) which are equivalent to the algorithm of position control of formula (19) involved in embodiment 1. In this way, in the control device and control method involved in force control of this embodiment, force control is performed using an internal position controller based on the position control method involved in embodiment 1. Therefore, even when the accuracy of the prior information on the dynamic characteristics of the control object is low and when there is a time lag in the response of the hydraulic actuator, high control performance can be obtained.

[0189] <Control system configuration>

[0190] like Fig.11 As shown in the block diagram of , the control unit 51 ′ of the control unit 50 ′ according to the present embodiment is different from the control unit 51 according to the first embodiment in that it includes a reference position calculation unit 516 (reference position calculation unit).

[0191] The reference position calculation unit 516 calculates the reference position using a virtual object having mechanical dynamic characteristics of a hydraulic actuator as a control object. Specifically, as shown in the above algorithm, the target applied force set as the target value of the applied force of the hydraulic actuator on the environment and the measured value of the reaction force from the environment acting on the hydraulic actuator measured by the force sensor 60 of the hydraulic actuator are input into the model of the virtual object. In addition, the reference position of the control object, more specifically, the rod of the hydraulic actuator in the virtual object in the next time step is calculated. The force sensor 60 for measuring the force acting on the hydraulic actuator can be built into the hydraulic actuator or can be arranged outside the hydraulic actuator.

[0192] In addition, the ideal speed calculation unit 511 involved in this embodiment calculates the ideal speed using the reference position calculated by the reference position calculation unit 516. More specifically, the ideal speed calculation unit 511 calculates the ideal speed using the reference position as the target position in the algorithm involved in the first embodiment. Fig.10 As shown, the reference position q, which is the output of the reference position calculation unit 516, is input to the control law according to the first embodiment operating as an internal position controller and control is executed, thereby performing overall force control of the controlled object.

[0193] <Control Flow>

[0194] Fig.12FIG. 1 is a flow chart showing the process of force control according to the present embodiment. Fig.12 As shown, in the force control according to the present embodiment, the target applied force f of the hydraulic actuator is first set. d (Step S31). Next, a force f is applied according to the set target. d and the force f acting on the hydraulic actuator e The reference position q is calculated based on the measured value (step S32). Parameters such as the time constant H given at the start of control are given in advance in the same manner as in the first embodiment.

[0195] More specifically, as a reference position calculation step, the reference position calculation unit 516 of the control unit 51' calculates the reference position that the rod of the hydraulic actuator, which is the control target, should be at in the next time step. The algorithm for calculating the reference position is shown in the above equations (29a) to (29b).

[0196] The processing of steps S33 to S36 is similar to the processing of steps S13 to S16 according to the first embodiment when the reference position calculated in step S31 is input as the target position ( Figure 5 )same.

[0197] In steps S33 to S36, the internal position controller calculates the operation amount u and controls the action of the hydraulic actuator as the control object. The control unit 51' repeatedly performs the processing of steps S31 to S37 until the control processing is terminated based on the end instruction of the operator or the end of the specified action time ("No" in step S37). In addition, the control system terminates the control processing when the specified end condition, such as the end instruction of the operator, the end of the specified action time, etc., is met ("Yes" in step S37). Based on the above processing, the control unit 51' involved in this embodiment can perform force control of the control object.

[0198] As described above, in the control device and control method for performing force control according to the present embodiment, since the reference speed and the actuator force are calculated by combining the PID control law or the PD control law for making the speed of the controlled object follow the reference speed with the sliding mode control law, and the force control of the controlled object is performed by using the control law according to the first embodiment for calculating the operation amount of the hydraulic actuator based on the quasi-static characteristics of the hydraulic actuator, control can be performed without relying on the dynamic characteristic model of the controlled object. Therefore, even when the accuracy of the prior information on the dynamic characteristics of the controlled object is low and when there is a time lag in the response of the hydraulic actuator, high control performance can be obtained.

[0199] <Numerical example>

[0200] Next, a force control simulation of a hydraulic actuator using the control law according to the present embodiment will be described. Fig.13 The hydraulic tester shown in the figure performs the admittance control according to the present embodiment. The hydraulic tester includes an electromagnetic proportional flow control valve (the top level of the integrated valve), a relief valve, and a check valve. The flow rate of the working oil supplied from the pump unit is constant, which is 4.17×10 -4 m 3 / s, the maximum output of the hydraulic cylinder is 1.56×10 -3 N. In addition, the sampling interval is T = 0.01s.

[0201] The control unit obtains the rod position p and the reaction force f from the environment from the linear encoder and load sensor of the hydraulic cylinder. e The control input u is converted into the input voltage of the flow control valve based on the D / A conversion board.

[0202] like Fig.13 As shown in the figure, a rubber plate fixed to a rigid body wall is used as the environment in contact with the hydraulic cylinder. In addition, a metal plate is placed in front of the rubber plate in such a way that only the load button portion of the load sensor contacts the environment. In this example, a contact experiment with three types of rubber plates with different hardnesses was conducted for comparison. The hardness of the rubber plates is Shore A hardness of about 50 HS (Shore A50), Shore A hardness of about 65 HS (Shore A65), and Shore A hardness of about 70 HS (Shore A70). The rubber plates with smaller numbers are softer materials.

[0203] The parameters involved in the control algorithm of the admittance control in this example are K = 2.5 × 10 3 N / m, B=3.0×10 2 N·s / m, H=0.5s. The parameters of the virtual object are B v =7.5×10 3 N·s / m,M v = 5 kg. These parameters are determined by trial and error so that the generated force f does not become unstable.

[0204] Fig.14A , Fig. 14B , Fig. 14C represents the force f applied relative to the step-like target d The result of contact force control. Fig.14A , Fig. 14B , Fig. 14C As shown, for all hardness environments, the measured reaction force f can follow the target applied force f d .

[0205] Fig.15A , Fig. 15B , Fig. 15C represents the force f applied to the target in the form of a sine wave d The result of contact force control. Fig.15A , Fig. 15B , Fig. 15C As shown, for all hardness environments, the measured reaction force f can follow the target applied force f d .

[0206] As described above, according to the force control method according to the present embodiment, it is possible to make the force applied to the environment by the hydraulic actuator in contact with the high-rigidity environment appropriately follow the target applied force.

[0207] The present invention is well suited for position control and force control of a machine operated by a hydraulic actuator, and is particularly well suited for automatic positioning control and admittance control of an engineering machine operated by a hydraulic actuator.

[0208] The control device involved in the first aspect of the present invention includes: an action command unit that controls a hydraulic actuator that drives a machine that is a control object; an ideal speed calculation unit that calculates the ideal speed that the machine should generate in the next step based on the target position and current position of the machine; a reference speed calculation unit that calculates a reference speed and a reference generated force based on the quasi-static characteristics of the hydraulic actuator, the reference speed being a speed that is closest to the ideal speed among the target speeds that can calculate the actuator force that can be generated when the hydraulic actuator implements PID control or PD control, and the reference generated force being the actuator force that should be generated when PID control or PD control is implemented using the reference speed as the target speed; and an operation amount calculation unit that calculates the operation amount of the hydraulic actuator based on the quasi-static characteristics of the hydraulic actuator, the reference generated force and the reference speed.

[0209] Furthermore, the hydraulic actuator may include a regeneration circuit, and the ideal speed may be calculated based on the target position of the machine, the current position, an opening of a flow control valve of the regeneration circuit, and an estimated value of a flow rate of the working fluid passing through the regeneration circuit.

[0210] In addition, the machine may be driven by a plurality of the hydraulic actuators, the operation amount calculation unit may calculate the operation amount for each of the hydraulic actuators, and the action instruction unit may move a specified part of the machine to a target position by controlling each of the hydraulic actuators according to the calculated operation amount.

[0211] Furthermore, the control device may further include an external force estimating unit for estimating the external force applied to the machine; wherein the reference speed calculating unit may calculate the reference speed and the reference generated force based on the external force estimated by the external force estimating unit.

[0212] Furthermore, the hydraulic actuator may be a hydraulic actuator.

[0213] In addition, the control device may also include: a reference position calculation unit, which inputs the reaction force received by the hydraulic actuator from the environment contacted by the machine and the target value of the force applied by the hydraulic actuator to the environment, i.e., the target applied force, into a virtual object having the target dynamic characteristics of the machine, and calculates the reference position of the machine in the virtual object at the next moment step; wherein the ideal speed calculation unit can calculate the ideal speed by using the reference position as the target position.

[0214] In addition, the control method involved in the second aspect of the present invention includes: an ideal speed calculation step, calculating the ideal speed that should be generated by the machine in the next step according to the target position and current position of the machine driven by the hydraulic actuator as the control object; a reference speed calculation step, calculating the reference speed and reference generated force according to the quasi-static characteristics of the hydraulic actuator, the reference speed being the speed closest to the ideal speed among the target speeds that can calculate the actuator force that can be generated when the hydraulic actuator implements PID control or PD control, and the reference generated force being the actuator force that should be generated when PID control or PD control is implemented using the reference speed as the target speed; and an operation amount calculation step, calculating the operation amount of the hydraulic actuator according to the quasi-static characteristics of the hydraulic actuator, the reference generated force and the reference speed.

[0215] In addition, the control method may also include: a reference position calculation step, inputting the reaction force received by the hydraulic actuator from the environment to which the machine contacts and the target value of the force applied by the hydraulic actuator to the environment, i.e., the target applied force, into a virtual object having the target dynamic characteristics of the machine, and calculating the reference position in the virtual object where the machine should be at the next moment; wherein, in the ideal speed calculation step, the reference position can be used as the target position to calculate the ideal speed.

[0216] According to the control device and control method of the present invention, since the reference speed and the actuator force are calculated by combining the PID control law or the PD control law that makes the speed of the controlled object follow the reference speed with the sliding mode control law, and the operation amount of the hydraulic actuator is calculated based on the quasi-static characteristics of the hydraulic actuator, it is possible to perform control without relying on the dynamic characteristic model of the controlled object. Therefore, even when the accuracy of the prior information on the dynamic characteristics of the controlled object is low and when there is a time lag in the response of the hydraulic actuator, high control performance can be obtained.

Claims

1. A control device, characterized in that include: An action command unit controls a hydraulic actuator that drives a machine that is a control object; an ideal speed calculation unit, which calculates an ideal speed that the machine should produce at the next time step according to the target position and the current position of the machine; a reference speed calculation unit that calculates a reference speed and a reference generated force based on the quasi-static characteristics of the hydraulic actuator, wherein the reference speed is a speed closest to the ideal speed among target speeds from which an actuator force that can be generated when the hydraulic actuator is subjected to PID control or PD control, and the reference generated force is an actuator force that should be generated when the PID control or PD control is performed using the reference speed as the target speed; as well as, The operation amount calculation unit calculates the operation amount of the hydraulic actuator based on the quasi-static characteristics of the hydraulic actuator, the reference generated force, and the reference speed.

2. The control device according to claim 1, characterized in that: The hydraulic actuator includes a regenerative circuit, The ideal speed is calculated based on the target position of the machine, the current position, an opening of a flow control valve of the regeneration circuit, and an estimated value of a flow rate of a working fluid passing through the regeneration circuit.

3. The control device according to claim 1, characterized in that: said machine being driven by a plurality of said hydraulic actuators, The operation amount calculation unit calculates the operation amount for each of the hydraulic actuators. The motion command unit controls each of the hydraulic actuators according to the calculated operation amount to move a predetermined portion of the machine to a target position.

4. The control device according to claim 1, characterized in that Also includes: an external force estimating unit for estimating an external force applied to the machine; wherein: The reference speed calculation unit calculates the reference speed and the reference generated force based on the external force estimated by the external force estimation unit.

5. The control device according to any one of claims 1 to 4, characterized in that: The hydraulic actuator is an oil pressure actuator.

6. The control device according to claim 1, characterized in that Also includes: The reference position calculation unit inputs the reaction force received by the hydraulic actuator from the environment in contact with the machine and the target value of the force applied by the hydraulic actuator to the environment, i.e., the target applied force, into a virtual object having the target dynamic characteristics of the machine, and calculates the reference position of the machine in the virtual object at the next time step; wherein, The ideal speed calculation unit calculates the ideal speed using the reference position as the target position.

7. A control method, characterized in that include: an ideal speed calculation step of calculating an ideal speed that should be generated by the machine at the next time step based on the target position and the current position of the machine driven by the hydraulic actuator as a control object; a reference speed calculation step of calculating a reference speed and a reference generated force based on the quasi-static characteristics of the hydraulic actuator, wherein the reference speed is a speed closest to the ideal speed among target speeds from which the actuator force that can be generated when the hydraulic actuator is subjected to PID control or PD control can be calculated, and the reference generated force is an actuator force that should be generated when the PID control or PD control is performed using the reference speed as the target speed; as well as, The operation amount calculation step calculates the operation amount of the hydraulic actuator based on the quasi-static characteristics of the hydraulic actuator, the reference generated force, and the reference speed.

8. The control method according to claim 7, characterized in that Also includes: A reference position calculation step is to input the reaction force of the hydraulic actuator from the environment in contact with the machine and the target value of the force applied by the hydraulic actuator to the environment, i.e., the target applied force, into a virtual object having the target dynamic characteristics of the machine, and calculate the reference position of the machine in the virtual object at the next moment; wherein, In the ideal speed calculation step, the ideal speed is calculated using the reference position as the target position.

Images