A large-load six-degree-of-freedom parallel robot and a control method thereof

Abstract

The present application relates to the technical field of large-load six-degree-of-freedom robots, and specifically provides a large-load six-degree-of-freedom parallel robot, wherein the large-load six-degree-of-freedom parallel robot comprises an upper platform, a lower platform, an outer-ring leg group connected between the upper platform and the lower platform, and an inner-ring leg group connected between the upper platform and the lower platform; the outer-ring leg group is connected between the outer ring of the upper platform and the lower platform and is used for controlling the pose of the upper platform, and the inner-ring leg group is connected between the inner ring of the upper platform and the lower platform and is used for controlling the stress of the upper platform. Meanwhile, the present application also provides a control method of the large-load six-degree-of-freedom parallel robot, wherein the outer-ring leg group is controlled in position by a PID controller, and the inner-ring leg group is controlled in force by a interval two-type fuzzy neural network intelligent controller. The present application can greatly increase the bearing capacity of the upper platform of the six-degree-of-freedom robot under the premise of using electric cylinder legs.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of large-load six-degree-of-freedom robots, in particular to a large-load six-degree-of-freedom parallel robot and a control method thereof. BACKGROUND

[0002] Parallel robots have high stiffness, no cumulative motion error, high precision, good dynamic response, large carrying capacity, low motion inertia and many other advantages, and have been widely used in optical precision pointing, micro-positioning and motion simulation of large equipment such as aircraft.

[0003] A parallel robot generally includes an upper platform, a lower platform and a support leg connecting the upper and lower platforms, and the support leg is usually composed of an electric cylinder or a hydraulic cylinder. When the support leg driving mode is an electric cylinder, the servo control precision of the upper platform is high, but the output force is limited, and the large equipment such as aircraft and large optical instruments cannot be pointed and positioned. When the support leg driving mode is a hydraulic cylinder, the upper platform can bear heavy load, but the hydraulic system is large in size, easy to leak, difficult to maintain and has high requirements for the use site. The above shortcomings greatly limit the application of the parallel robot.

[0004] At present, with the continuous development of optical instruments, new requirements are put forward for the carrying capacity, dynamic response and control precision of the parallel robot. However, as known from the above, the output force of the electric cylinder is limited and cannot bear heavy load. The hydraulic system also has a large number of limitations that are difficult to overcome, which greatly limits the application of the parallel robot. In order to solve the above problems, the following patents appear in the prior art:

[0005] 1. Patent No. 201810138365.9, Patent Name "Large-load six-degree-of-freedom compliant parallel robot system", which adopts six small electric cylinders to drive the movement of the moving platform to realize the six degrees of freedom of three translations and three rotations, and also includes a core component of a compliant spherical hinge, which can combine the advantages of the traditional six-degree-of-freedom platform and the advantages of the compliant mechanism to improve the positioning accuracy and carrying capacity of the mechanism.

[0006] 2. Patent No. 201210530430.5, Patent Name "Double-parallel heavy-load static balance motion simulation table", which is an invention patent. The upper platform of the mechanism is controlled by six structurally identical electric cylinder support legs, and the gravity of the load is carried by three hydraulic or pneumatic cylinder support legs. The simulation table in this invention has strong carrying capacity and fast dynamic response speed.

[0007] However, the prior art uses electric cylinder legs to improve the control accuracy of the upper platform, but only six legs are used to control the movement and position of the upper platform, and the load bearing capacity is limited; in the comparative document 2, three small legs are added to bear the gravity of the load, but at the same time, a weighing platform and other components are also added, greatly increasing the complexity of the robot structure; at the same time, the control method of each leg is not disclosed in the comparative document.

[0008] In summary, how to design a parallel robot that can increase the load bearing capacity of the upper platform of a six-degree-of-freedom robot under the premise of using electric cylinder legs is a problem that needs to be solved at present. SUMMARY

[0009] To solve the above problems, a large-load six-degree-of-freedom parallel robot and its control method are provided, which can use electric cylinder legs to ensure control accuracy and increase the load bearing capacity of the upper platform through the cooperation of outer legs and inner legs.

[0010] A large-load six-degree-of-freedom parallel robot comprises an upper platform, a lower platform, and a leg assembly connected between the upper platform and the lower platform, the leg assembly comprising an outer ring leg assembly and an inner ring leg assembly located inside the outer ring leg assembly; the outer ring leg assembly is connected between the outer ring of the upper platform and the lower platform and is used to control the pose of the upper platform, and the inner ring leg assembly is connected between the inner ring of the upper platform and the lower platform and is used to control the stress of the upper platform; the outer ring leg assembly and the inner ring leg assembly each comprise six outer legs of the same length and six inner legs of the same length, the upper and lower ends of the outer legs and the inner legs are each movably connected to the upper platform and the lower platform through a hinge, and the outer legs and the inner legs further comprise an electric cylinder and a force sensor located between the hinge and the electric cylinder at the upper platform, and the force sensor provides a force feedback signal of a force control loop.

[0011] Preferably, the hinge is a Hooke joint; the center point connecting line of the Hooke joint at the outer leg and the center point connecting line of the Hooke joint at the inner leg are respectively an outer hexagon and an inner hexagon of a half-symmetrical hexagonal structure, and the edges of the outer hexagon and the inner hexagon are arranged in parallel.

[0012] A control method for a large-load six-degree-of-freedom parallel robot, which controls the above-mentioned six-degree-of-freedom parallel robot, the outer legs are controlled by a PID controller for position control to control the pose of the upper platform; the inner legs are controlled by a interval type-2 fuzzy neural network intelligent controller for force control.

[0013] Preferably, the desired length of the outer leg is determined by the inverse kinematics of the desired pose of the upper platform; and the desired force control value of the inner leg is determined by the inverse dynamics of the pose of the upper platform.

[0014] Preferably, the control process of the outer leg is as follows:

[0015] S1: obtain the desired length Li of the outer leg by kinematic inverse solution of the desired pose of the upper platform, wherein i represents different outer legs, i = 1, 2, …, 6;

[0016] S2: PID control is performed on the corresponding outer leg;

[0017] S3: judge the actual length of the outer leg, and feed back the length information, i.e. the position information, to Li and compare it with Li;

[0018] S4: when the actual length of the outer leg is consistent with the desired length, the position control is completed; when the actual length of the outer leg is not consistent with the desired length, the next round of control is performed through iteration until the actual length of the outer leg is equal to the desired length of the outer leg.

[0019] Preferably, the actual length of the outer leg in step S3 is judged and fed back by the motor and the encoder, specifically:

[0020] S31: drive the electric cylinder to move by the motor;

[0021] S32: detect the number of revolutions of the motor by the encoder to judge the actual length of the outer leg;

[0022] S33: feed back the actual length information, i.e. the position information, of the outer leg 8 to Li by the encoder.

[0023] Preferably, the next round of control in step S4 is performed through iteration, specifically:

[0024] S41: feed back the actual length information, i.e. the position information, of the outer leg to Li and obtain the difference from Li;

[0025] S42: introduce the difference between the actual length and the desired length of the outer leg into the PID control process;

[0026] S43: obtain and feed back a new actual length by the motor and the encoder;

[0027] When there is a difference between the actual length and the desired length of the outer leg, repeat steps S41-S43 repeatedly;

[0028] S4n: when the actual length of the outer leg is equal to the desired length of the outer leg, the position control of the upper platform is completed.

[0029] Preferably, the control of the inner leg is synchronized with the control of the outer leg, and the control process of the inner leg is as follows:

[0030] S1: determining the force control expected value Fi of the inner support leg from the dynamic inverse solution of the pose of the upper platform, wherein i represents different inner support legs, i=7, 8, …, 12;

[0031] S2: interval type-2 fuzzy neural network control is performed on the corresponding inner support leg;

[0032] S3: judging the force control actual value of the inner support leg, and feeding back the force control actual value to Fi and comparing it with Fi;

[0033] S4: when the force control actual value is consistent with the force control expected value, the force control is completed; when the force control actual value is not consistent with the force control expected value, the next round of control is performed through iteration until the force control actual value of the inner support leg is equal to the force control expected value of the inner support leg.

[0034] Preferably, in step S3, the force control actual value of the inner support leg is judged, and specifically:

[0035] S31: moving the electric cylinder through the motor drive;

[0036] S32: measuring the force control actual value of the inner support leg through the force sensor.

[0037] Preferably, in step S4, the next round of control is performed through iteration, and specifically:

[0038] S41: feeding back the force control actual value information of the inner support leg to Fi and obtaining the difference value with Fi;

[0039] S42: introducing the difference value between the force control actual value and the force control expected value of the inner support leg into the interval type-2 fuzzy neural network control process;

[0040] S43: obtaining a new force control actual value through the motor and the force sensor;

[0041] When there is a difference between the force control actual value and the force control expected value of the inner support leg, steps S41-S43 are repeatedly repeated;

[0042] S4n: when the force control actual value of the inner support leg is equal to the force control expected value of the inner support leg, the force control of the inner support leg is completed.

[0043] The present application has the following beneficial effects:

[0044] 1、The present application is provided with six outer support legs and six inner support legs, wherein the pose of the upper platform can be controlled through the six outer support legs, the stress of the upper platform can be controlled through the inner support legs, and the load capacity of the upper platform can be greatly improved through the cooperation of the inner support legs and the outer support legs.

[0045] 2. The parallel arrangement of the outer and inner hexagons formed by the Hooke hinges corresponding to the outer and inner legs can improve the support stability of the outer and inner legs, further ensure the control performance of the controller, and improve the environmental adaptability of the parallel robot.

[0046] 3. In this invention, the outer outriggers are position-controlled by a PID controller, which can ensure the control accuracy of the upper platform's posture; the inner outriggers are controlled by an interval type II fuzzy neural network intelligent controller, which can realize the control of the upper platform's posture and prevent excessive internal stress between the outer outriggers. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of a six-degree-of-freedom parallel robot.

[0048] Figure 2 This is the control flowchart of a six-degree-of-freedom parallel robot.

[0049] Figure 3 This is a schematic diagram of a six-degree-of-freedom parallel robot.

[0050] Figure 4 This is a schematic diagram of the Hooke hinges at the outer and inner legs.

[0051] The attached diagram is labeled as follows: 1. Upper platform; 2. Force sensor; 3. Electric cylinder; 4. Lower platform; 5. Hooke hinge; 6. Outer hexagon; 7. Inner hexagon; 8. Outer leg; 9. Inner leg. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the following description is provided in conjunction with the appendix. Figures 1-4 The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and do not constitute a limitation thereof.

[0053] A high-load, six-DOF parallel robot, such as Figure 1 , 3 As shown, it includes an upper platform 1 and a lower platform 4. The inner diameter of the upper platform 1 is smaller than the inner diameter of the lower platform 4. A support leg assembly is provided between the upper platform 1 and the lower platform 4. The support leg assembly includes an outer ring support leg assembly and an inner ring support leg assembly. The outer ring support leg assembly is connected between the outer rings of the upper platform 1 and the lower platform 4 and is used to control the position and posture of the upper platform 1. The inner ring support leg assembly is connected between the inner rings of the upper platform 1 and the lower platform 4 and is used to control the force on the upper platform 1. The inner ring support leg assembly is located inside the outer ring support leg assembly and is set near the middle of the upper platform 1 and the lower platform 4. The outer ring support leg assembly is set near the edge of the upper platform 1 and the lower platform 4.

[0054] like Figure 1 and Figure 3As shown, the outer leg group and the inner leg group respectively include six outer legs 8 with the same length and six inner legs 9 with the same length, the length of the inner leg 9 is less than the length of the outer leg 8; the upper and lower ends of the outer leg 8 and the inner leg 9 are movably connected with the upper platform 1 and the lower platform 4 through the hinge respectively, the hinge in the embodiment is a hooke joint 5; as shown Figure 4 As shown, the center point connection line of the hooke joint 5 at the outer leg 8 and the center point connection line of the hooke joint 5 at the inner leg 9 are respectively an outer hexagon 6 and an inner hexagon 7 with a half-symmetrical hexagonal structure, the edges of the outer hexagon 6 and the inner hexagon 7 are arranged in parallel, which can improve the stability of the outer leg 8 and the inner leg 9.

[0055] The outer leg 8 and the inner leg 9 further include an electric cylinder 3 and a force sensor 2 between the hooke joint 5 at the upper platform 1 and the electric cylinder 3, the force sensor 2 is used to provide a force feedback signal of the force control loop; the electric cylinder 3 includes a motor, and an encoder is installed at the motor to provide feedback for position control.

[0056] A control method of a large-load six-degree-of-freedom parallel robot, the parallel robot is controlled as follows: the outer leg 8 is controlled by a PID controller to control the pose of the upper platform 1, the PID controller can ensure the control accuracy of the pose of the upper platform 1; the expected length of the outer leg 8 is determined by the inverse kinematics of the expected pose of the upper platform 1; the inner leg 9 is controlled by an interval type-2 fuzzy neural network intelligent controller, the control of the interval type-2 fuzzy neural network intelligent controller can prevent excessive internal stress between the 12 legs; the force control expected value of the inner leg 9 is determined by the inverse dynamics of the pose of the upper platform 1, specifically, the force expected value of the six inner legs is determined by the inverse dynamics of the pose of the upper platform according to the expected pose of the upper platform, and the force feedback signal is provided by the force sensor.

[0057] As shown Figure 2 The control process of the outer leg 8 is as follows:

[0058] S1: the expected length Li of the outer leg 8 is obtained by the inverse kinematics of the expected pose of the upper platform 1, where i represents different outer legs 8, i = 1, 2, …, 6;

[0059] S2: PID control is performed on the corresponding outer leg 8;

[0060] S3: the actual length of the outer leg 8 is judged, and the length information, i.e. the position information, is fed back to Li and compared with Li; the length information of the outer leg 8 corresponds to the position information of the upper platform 1;

[0061] The actual length of the outer leg 8 is judged and fed back by the motor and the encoder, specifically:

[0062] S31: driving the motor to drive the cylinder 3 to move;

[0063] S32: detecting the number of rotations of the motor by the encoder to determine the actual length of the outer leg 8;

[0064] S33: feeding back the actual length information of the outer leg 8, i.e. the position information, to Li.

[0065] S4: when the actual length of the outer leg 8 is consistent with the expected length, the position control of the upper platform 1 is completed; when the actual length of the outer leg 8 is not consistent with the expected length, the next round of control is performed in an iterative manner until the actual length of the outer leg 8 is equal to the expected length of the outer leg 8;

[0066] wherein the next round of control is performed in an iterative manner, specifically:

[0067] S41: feeding back the actual length information of the outer leg 8, i.e. the position information, to Li and obtaining the difference from Li;

[0068] S42: introducing the difference between the actual length of the outer leg 8 and the expected length into the PID control process;

[0069] S43: obtaining and feeding back a new actual length by the motor and the encoder;

[0070] when there is a difference between the actual length of the outer leg 8 and the expected length, repeatedly repeating steps S41-S43;

[0071] S4n: when the actual length of the outer leg 8 is equal to the expected length of the outer leg 8, the position control of the upper platform 1 is completed.

[0072] According to the above method, the six outer legs 8 are controlled one by one.

[0073] The control of the inner leg 9 is synchronized with the control of the outer leg 8, and the control process of the inner leg 9 is as follows:

[0074] S1: determining the force control expected value Fi of the inner leg 9 from the dynamic inverse solution of the pose of the upper platform 1, wherein i represents different inner legs 9, i = 7, 8,..., 12;

[0075] S2: performing interval type-2 fuzzy neural network control on the corresponding inner leg 9;

[0076] S3: determining the force control actual value of the inner leg 9, and feeding back the force control actual value to Fi and comparing it with Fi;

[0077] wherein the force control actual value of the inner leg 9 is determined, specifically:

[0078] S31: moving the electric cylinder 3 by the motor;

[0079] S32: measuring the force control actual value of the inner leg 9 by the force sensor 2.

[0080] S4: when the force control actual value of the inner leg 9 is consistent with the force control expected value of the inner leg 9, the control of the force of the inner leg 9 is completed; when the force control actual value of the inner leg 9 is not consistent with the force control expected value of the inner leg 9, the next round of control is performed by iteration until the force control actual value of the inner leg 9 is equal to the force control expected value of the inner leg 9.

[0081] wherein the next round of control is performed by iteration, specifically:

[0082] S41: feeding back the force control actual value information of the inner leg 9 to Fi and obtaining the difference with Fi;

[0083] S42: introducing the difference between the force control actual value and the force control expected value into the interval type fuzzy neural network control process;

[0084] S43: obtaining a new force control actual value by the motor and the force sensor 2;

[0085] when the force control actual value of the inner leg 9 and the force control expected value exist a difference, repeatedly repeating steps S41-S43;

[0086] S4n: when the force control actual value of the inner leg 9 = the force control expected value of the inner leg 9, the force control of the inner leg 9 is completed.

[0087] According to the above method, the six inner legs 9 are controlled one by one.

[0088] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

[0089] The specific embodiments of the present application do not constitute a limitation on the scope of protection of the present application. Any various other corresponding changes and modifications made according to the technical concept of the present application should be included in the protection scope of the claims of the present application.

Claims

1. A large load six degree of freedom parallel robot comprising an upper platform (1), a lower platform (4) and a leg assembly connected between the upper platform (1) and the lower platform (4), characterized in that, The leg assembly comprises an outer ring leg group and an inner ring leg group located inside the outer ring leg group; the outer ring leg group is connected between the outer ring of the upper platform (1) and the lower platform (4) and is used for controlling the pose of the upper platform (1), and the inner ring leg group is connected between the inner ring of the upper platform (1) and the lower platform (4) and is used for controlling the force of the upper platform (1); the outer ring leg group and the inner ring leg group each comprise six outer legs (8) with the same length and six inner legs (9) with the same length, the upper and lower ends of the outer legs (8) and the inner legs (9) are movably connected to the upper platform (1) and the lower platform (4) through hinge members respectively, and the outer legs (8) and the inner legs (9) further comprise electric cylinders (3) and force sensors (2) located between the hinge members and the electric cylinders (3) at the upper platform (1), and the force feedback signal of the force control loop is provided through the force sensors (2); The outer legs (8) are controlled in position by a PID controller to control the pose of the upper platform (1); and the inner legs (9) are controlled in force by a type-2 fuzzy neural network intelligent controller; The expected length of the outer legs (8) is determined by the inverse kinematics of the expected pose of the upper platform (1); and the expected force control value of the inner legs (9) is determined by the inverse dynamics of the pose of the upper platform (1); The control process of the outer legs (8) is as follows: S1: the expected length Li of the outer legs (8) is obtained from the inverse kinematics of the expected pose of the upper platform (1), wherein i represents different outer legs (8), i=1, 2, …, 6; S2: the PID control is performed on the corresponding outer leg (8); S3: the actual length of the outer leg (8) is judged, and the length information, i.e. the position information, is fed back to Li and compared with Li; S4: when the actual length of the outer leg (8) is consistent with the expected length, the position control is completed; when the actual length of the outer leg (8) is not consistent with the expected length, the next round of control is performed in an iterative manner until the actual length of the outer leg (8) is equal to the expected length of the outer leg (8); The control of the inner legs (9) is synchronous with the control of the outer legs (8), and the control process of the inner legs (9) is as follows: S1: the expected force control value Fi of the inner legs (9) is determined from the inverse dynamics of the pose of the upper platform (1), wherein i represents different inner legs (9), i=7, 8, …, 12; S2: the type-2 fuzzy neural network control is performed on the corresponding inner leg (9); S3: the actual force control value of the inner leg (9) is judged, and the actual force control value is fed back to Fi and compared with Fi; S4: when the actual force control value of the inner leg (9) is consistent with the expected force control value, the force control is completed; when the actual force control value of the inner leg (9) is not consistent with the expected force control value, the next round of control is performed in an iterative manner until the actual force control value is equal to the expected force control value.

2. The large payload six degree of freedom parallel robot of claim 1, wherein, The hinge piece is a hooke hinge (5); the center point connecting lines of the hooke hinge (5) at the outer leg (8) and the center point connecting lines of the hooke hinge (5) at the inner leg (9) are respectively outer hexagons (6) and inner hexagons (7) of half-symmetrical hexagonal structures, and the edges of the outer hexagons (6) and the inner hexagons (7) are arranged in parallel.

3. The large payload six degree of freedom parallel robot of claim 1, wherein, In step S3, the actual length of the outer leg (8) is judged by the motor and the encoder, and the specific process is as follows: S31: the motor drives the electric cylinder (3) to move; S32: the number of revolutions of the motor is detected by the encoder to judge the actual length of the outer leg (8); S33: the actual length information, i.e. the position information, of the outer leg (8) is fed back to Li.

4. The large payload six degree of freedom parallel robot of claim 3, wherein, In step S4, the next round of control is performed in an iterative manner, and the specific process is as follows: S41: the actual length information, i.e. the position information, of the outer leg (8) is fed back to Li and the difference from Li is obtained; S42: the difference between the actual length and the expected length of the outer leg (8) is introduced into the PID control process; S43: a new actual length is obtained and fed back by the motor and the encoder; When there is a difference between the actual length and the expected length of the outer leg (8), steps S41-S43 are repeatedly repeated; S4n: when the actual length of the outer leg (8) is equal to the expected length of the outer leg (8), the position control of the upper platform (1) is completed.

5. The large payload six degrees of freedom parallel robot according to claim 1 or 4, characterized in that, In step S3, the actual force control value of the inner leg (9) is judged, and the specific process is as follows: S31: the motor drives the electric cylinder (3) to move; S32: the actual force control value of the inner leg (9) is measured by the force sensor (2).

6. The large payload six degree of freedom parallel robot of claim 5, wherein, In step S4, the next round of control is performed in an iterative manner, and the specific process is as follows: S41: the actual force control value information of the inner leg (9) is fed back to Fi and the difference from Fi is obtained; S42: the difference between the actual force control value and the expected force control value is introduced into the interval type fuzzy neural network control process; S43: a new actual force control value is obtained by the motor and the force sensor (2); When there is a difference between the actual force control value and the expected force control value of the inner leg (9), steps S41-S43 are repeatedly repeated; S4n: when the actual force control value of the inner leg (9) is equal to the expected force control value of the inner leg (9), the force control of the inner leg (9) is completed.

Images