A modular gas-electric hybrid bionic robot and its bionic mechanical arm

Abstract

The invention discloses a modular gas-electric hybrid bionic robot and a bionic mechanical arm thereof, comprising a mechanical arm body, wherein the mechanical arm body comprises a shoulder module, an upper arm structure and a forearm structure which are connected in sequence; the shoulder module comprises a shoulder joint motor 1, a right-angle connecting frame, a shoulder joint motor 2, a connecting piece and a shoulder joint motor 3; the upper arm structure comprises an upper arm support, a pneumatic tendon 1, a pneumatic tendon 2, an elbow flange 1, an elbow flange 2 and an elbow posture sensor; the forearm structure comprises an elbow joint motor, an elbow flange 3, a pneumatic tendon 3, a wrist flange, a wrist posture sensor and a supporting rod; in the forearm structure of the invention, the pronation / supination of the wrist, the abduction / internal rotation, the extension / extension of the shoulder and the rotation of the upper arm / forearm are all driven by a motor as a driving source; the elbow bending action and the wrist bending action are driven by a plurality of pneumatic tendons, which is more in line with the human body structure and realizes the coordinated work and flexible movement of muscles.

Description

Technical Field

[0001] The invention belongs to the technical field of robots, and in particular is a modular gas-electric hybrid bionic robot and a bionic mechanical arm thereof. Background Art

[0002] With the rapid development of science and technology, the application of robotics in life and industry is becoming more and more extensive. Especially in tasks with high precision, high flexibility and high efficiency, the advantages of robots are fully utilized. Bionic robotic arms are new intelligent robotic arms designed based on the structure and function of human arms. With the in-depth study of human biomechanical characteristics, bionic robotic arms have become a new hot spot in robotics research. At present, bionic robotic arm technology is in its infancy, and related theories and key technologies are still being improved.

[0003] Most of the current bionic robotic arms are only partially bionic, and the overall bionic degree is not high. In terms of the driving source, most of the existing bionic robotic arms use motors as actuators, which is different from the working principle of human muscles. The motors are large in size and have uneven output torque, making it difficult to achieve muscle coordination and flexible movement. In addition, the motor as a driving source also limits the bionic robotic arm's goal of achieving lightweight. These factors restrict the improvement of the movement performance of the bionic robotic arm. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a modular gas-electric hybrid bionic robot and a bionic mechanical arm thereof in view of the deficiencies of the above-mentioned prior art. The pronation / supination of the wrist in the forearm structure of the modular gas-electric hybrid bionic robot and the bionic mechanical arm thereof, the abduction / internal rotation, extension / extension of part of the structure in the shoulder module and the entire upper arm structure and forearm structure, and the rotation of the upper arm structure and the forearm structure are driven by various joint motors as driving sources. However, the elbow bending movement and the wrist bending movement are realized by driving multiple pneumatic tendons. The pneumatic tendons are used as the driving source, that is, pneumatic artificial muscles (capable of contracting like human muscles) are used, which are more in line with the human body structure and can realize the coordinated work and flexible movement of muscles. The overall structure is lighter in weight than the existing ones that all use motors as driving sources.

[0005] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0006] A modular gas-electric hybrid bionic mechanical arm, comprising a mechanical arm body, wherein the mechanical arm body comprises a shoulder module, an upper arm structure and a forearm structure connected in sequence;

[0007] The shoulder module includes a shoulder joint motor 1, a right-angle connecting frame, a shoulder joint motor 2, a connecting piece and a shoulder joint motor 3, the shell of the shoulder joint motor 1 is used to be connected to the fuselage fixing frame of the bionic robot, the output end of the shoulder joint motor 1 is connected to one end of the right-angle connecting frame, the other end of the right-angle connecting frame is connected to the shell of the shoulder joint motor 2, the output end of the shoulder joint motor 2 is connected to the connecting piece, the connecting piece is connected to the shell of the shoulder joint motor 3, and the output end of the shoulder joint motor 3 is connected to the upper arm structure;

[0008] The upper arm structure includes an upper arm support, pneumatic tendon 1, pneumatic tendon 2, elbow flange 1, elbow flange 2 and an elbow posture sensor, the output end of the shoulder joint motor 3 is connected to the top of the upper arm support through a connecting hanging plate, the bottom end of the upper arm support is rotatably connected to elbow flange 1, elbow flange 1 is connected to elbow flange 2, the side of elbow flange 2 is connected with an elbow posture sensor, one end of the pneumatic tendon 1 is rotatably connected to the top of one side of the upper arm support, and the other end is rotatably connected to one side of the elbow flange 1, one end of the pneumatic tendon 2 is rotatably connected to the top of the other side of the upper arm support, and the other end is connected to a rope, a pulley is connected to the bottom of the other side of the upper arm support, and the rope is connected to the other side of the elbow flange 1 after being wound around the pulley;

[0009] The forearm structure includes an elbow joint motor, an elbow flange three, a pneumatic tendon three, a wrist flange, a wrist posture sensor and a support rod. The elbow flange two is connected to the outer shell of the elbow joint motor, the output end of the elbow joint motor is connected to the elbow flange three, the elbow flange three is rotatably connected to one end of the pneumatic tendon three, the other end of the pneumatic tendon three is rotatably connected to the wrist flange, the wrist flange is connected to the wrist posture sensor, one end of the support rod is connected to the elbow flange three, and the other end of the support rod is connected to the wrist flange through a universal joint.

[0010] As a further improved technical solution of the present invention, in the shoulder module, the top of the shell of the shoulder joint motor three is connected to a shoulder posture sensor through a first IMU fixed support, and the shoulder posture sensor adopts an IMU 9-axis posture sensor.

[0011] As a further improved technical solution of the present invention, in the upper arm structure, the elbow posture sensor adopts an IMU 9-axis posture sensor, and the side of the elbow flange plate 2 is connected to the elbow posture sensor through a second IMU fixed support; the upper arm support and the elbow flange plate 1 are both connected to a double earring support, the double earring at one end of the pneumatic tendon 1 is rotatably connected to the double earring support at the top of one side of the upper arm support, and the double earring at the other end is rotatably connected to the double earring support on the elbow flange plate 1, the double earring at one end of the pneumatic tendon 2 is rotatably connected to the double earring support at the top of the other side of the upper arm support, the bottom of the other side of the upper arm support is connected to a pulley through a pulley support, and the rope is connected to the rope rack on the other side of the elbow flange plate 1 after being wrapped around the pulley.

[0012] As a further improved technical solution of the present invention, there are two pneumatic tendons 1, which are arranged side by side, and there is one pneumatic tendon 2.

[0013] As a further improved technical solution of the present invention, in the forearm structure, the elbow flange three is rotatably connected to one end of the pneumatic tendon three through a joint bearing, and the other end of the pneumatic tendon three is rotatably connected to the wrist flange through a joint bearing, the wrist flange is connected to a wrist sensor fixing plate, and the wrist sensor fixing plate is connected to a wrist posture sensor.

[0014] As a further improved technical solution of the present invention, there are four pneumatic muscles three, which are located around the support rod, and one end of the support rod is connected to the elbow flange three through a coupling.

[0015] As a further improved technical solution of the present invention, it also includes a control cabinet, on which a motion control system is arranged, the motion control system includes a controller, a current module, a power supply and multiple proportional pressure valves, the power supply is respectively connected to the controller, the current module, the pressure valve and the shoulder joint motor 1, the shoulder joint motor 2 and the shoulder joint motor 3 in the robotic arm body, the controller is respectively connected to the current module and the shoulder joint motor 1, the shoulder joint motor 2 and the shoulder joint motor 3 in the robotic arm body, the current module is connected to all the proportional pressure valves, and the pneumatic tendon 1, the pneumatic tendon 2 and the pneumatic tendon 3 in the robotic arm body are respectively connected to the proportional pressure valves.

[0016] In order to achieve the above technical purpose, another technical solution adopted by the present invention is:

[0017] A modular gas-electric hybrid bionic robot comprises a modular gas-electric hybrid bionic mechanical arm, a manipulator, a head and a body fixing frame, wherein the end of the forearm structure of the modular gas-electric hybrid bionic mechanical arm is connected to the manipulator, the head is connected to the body fixing frame, the body fixing frame is connected to the shell of a shoulder joint motor in the modular gas-electric hybrid bionic mechanical arm, a deep learning camera is arranged on the head, a laser radar is arranged on the body fixing frame, and both the deep learning camera and the laser radar are connected to a power supply.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] The design structure of the robot arm of the present invention is an SRS configuration, which means that the central axes of the three degrees of freedom of the shoulder and the three degrees of freedom of the wrist intersect at one point, wherein the central axes of the three degrees of freedom of the shoulder intersect at one point, which means that the axes of the motion axes of the shoulder joint motor 1, the shoulder joint motor 2 and the shoulder joint motor 3 intersect at one point, and the central axes of the three degrees of freedom of the wrist intersect at one point, which means that the central axes of the motion axes of the three degrees of freedom provided by the elbow joint motor and the four pneumatic tendons 3 of the wrist intersect at one point. The robot arm adopts a series redundant design (series means one degree of freedom connected to one degree of freedom, and redundancy means seven degrees of freedom), and has seven degrees of freedom, namely: the shoulder joint motor 1, the shoulder joint motor 2, and the shoulder joint motor 3 respectively realize the three degrees of freedom movement of the shoulder, the elbow joint motor realizes the rotation of the wrist, and realizes one degree of freedom movement, the two pneumatic tendons 1 and the one pneumatic tendon 2 of the elbow drive the elbow to rotate, and realize one degree of freedom movement, and the four pneumatic tendons 3 of the wrist can drive the wrist to perform double-degree-of-freedom movement, and realize two degrees of freedom movement. Therefore, it can provide greater operational freedom and adapt to complex operating environments. Especially in a small space, the SRS robot arm can achieve more flexible operation.

[0020] High redundancy: Due to the existence of seven degrees of freedom, the SRS robot arm has redundancy, which can realize motion planning of different paths and optimize the efficiency of task completion. At the same time, it can avoid the failure of a single joint causing task failure, thus improving the reliability of the system.

[0021] Strong ability to handle complex tasks: By adjusting the movement of the joints, the robot arm can complete multiple tasks at the same time, such as obstacle avoidance and precise end positioning, which is suitable for high-demand industrial environments and complex operating tasks.

[0022] Wide operating range: Due to its tandem structure, the end effector of the SRS robot arm can cover a larger working range and can perform tasks ranging from small-scale fine movements to large-scale handling tasks.

[0023] Good repeatability: The joints of the robot arm can achieve high-precision movements through precise control and coordination, which is suitable for application scenarios with high precision requirements, such as industrial processing and assembly.

[0024] The designed structure takes into account the installation position of the sensor. Power system: The shoulder movement is realized by three shoulder joint motors. Specifically, the axes of the three motion axes intersect at one point, which is convenient for establishing the DH model of the robot and the inverse kinematic solution of the robot.

[0025] This mechanical arm is designed to imitate the human body's mechanical arm. The human body's shoulder has three degrees of freedom. The shoulder of the mechanical arm of the present invention is realized by three shoulder joint motors to realize tasks such as rotation of three degrees of freedom. The axes of the motion axes of shoulder joint motor 1, shoulder joint motor 2 and shoulder joint motor 3 intersect at one point. The elbow adopts three pneumatic tendons to realize one degree of freedom movement. The designed structure adopts pneumatic artificial muscles (which can contract like human muscles) to realize the extraction of heavy objects. And because the pneumatic artificial muscles have impact resistance, they will not cause damage to the mechanical structure itself when subjected to external interference and impact. The elbow motor realizes the rotation of one degree of freedom of the wrist. The wrist adopts a two-degree-of-freedom motion platform composed of four pneumatic tendons three (pneumatic artificial muscles) (that is, two pneumatic tendons provide one degree of freedom, and four pneumatic tendons provide two degrees of freedom), which can realize the requirements of two-degree-of-freedom movement, and has the advantages of impact resistance and large load-bearing capacity. The elbow can lift 20kg, the wrist can lift 10kg, and the robotic arm itself weighs 15.08kg and is 1.26 times longer than an adult's arm.

[0026] The modularized gas-electric hybrid bionic robot and the bionic mechanical arm of the present invention can drive the pronation / supination of the wrist in the forearm structure through the elbow joint motor, can drive the abduction / internal rotation of the partial structure in the shoulder module, the forearm structure and the upper arm structure through the shoulder joint motor 1, drive the extension / extension of the partial structure in the shoulder module, the forearm structure and the upper arm structure through the shoulder joint motor 2, and drive the rotation (pronation / supination) of the forearm structure and the upper arm structure through the shoulder joint motor 3. However, the elbow bending action and the wrist bending action are driven by seven pneumatic tendons. The pneumatic tendons are used as the driving source, that is, pneumatic artificial muscles (which can contract like human muscles) are used, which are more in line with the human body structure, can realize the coordinated work and flexible movement of muscles, and the overall structure is lighter than the existing ones that all use motors as the driving source, and the bionic mechanical arm has higher movement performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a schematic diagram of the structure of a modular gas-electric hybrid bionic robotic arm of the present invention;

[0028] Figure 2 This is a schematic diagram of the structure of a modular gas-electric hybrid bionic robotic arm of the present invention;

[0029] Figure 3 This is a schematic diagram of the structure of a modular gas-electric hybrid bionic robotic arm of the present invention;

[0030] Figure 4 This is a schematic diagram of the structure of a modular gas-electric hybrid bionic robotic arm of the present invention;

[0031] Figure 5 It is a structural schematic diagram of the rotation of the forearm structure of the modular gas-electric hybrid bionic mechanical arm of the present invention;

[0032] Figure 6 This is a schematic diagram of the structure of a modular gas-electric hybrid bionic robot of the present invention;

[0033] Figure 7 This is a schematic diagram of the control cabinet structure of the present invention;

[0034] Figure 8 It is a schematic diagram of the information monitoring and processing process of the present invention;

[0035] In the figure:

[0036] A. Bionic robotic arm; B. Head; C. Body fixing frame;

[0037] A1, shoulder joint motor 1; A2, right angle connecting frame; A3, shoulder joint motor 2; A4, shoulder joint motor 3; A5, first IMU fixed support; A6, shoulder posture sensor; A7, double earrings; A8, pneumatic tendon 1; A9, upper arm support; A10, pneumatic tendon 2; A11, elbow flange 1; A12, elbow flange 2; A13, elbow posture sensor; A14, elbow joint motor machine; A15, elbow flange three; A16, joint bearing; A17, pneumatic tendon three; A18, support rod; A19, wrist flange; A20, wrist sensor fixing plate; A21, wrist posture sensor; A22, rigid coupling; A23, universal joint; A24, connector; A25, double earring support; A26, connecting hanging plate; A27, rope; A28, pulley; A29, pulley support;

[0038] 1. Controller; 2. Current module; 3. 24V DC power supply; 4. 48V DC power supply; 5. Socket; 6. Circuit breaker; 7. Terminal block; 8. Fixed base plate; 9. Proportional pressure valve. DETAILED DESCRIPTION

[0039] The specific embodiments of the present invention are further described below according to the accompanying drawings:

[0040] A modular gas-electric hybrid bionic mechanical arm comprises a mechanical arm body, wherein the mechanical arm body comprises a shoulder module, an upper arm structure and a forearm structure which are connected in sequence.

[0041] like Figure 1-4 As shown, the shoulder module includes a shoulder joint motor A1, a right-angle connecting frame A2, a shoulder joint motor A3, a connecting piece A24 and a shoulder joint motor A4, the shell of the shoulder joint motor A1 is used to be fixedly connected to the fuselage fixing frame C of the bionic robot, the output end of the shoulder joint motor A1 is fixedly connected to one end of the right-angle connecting frame A2, the shoulder joint motor A1 is used to drive the shoulder joint motor A1 to rotate, the other end of the right-angle connecting frame A2 is fixedly connected to the shell of the shoulder joint motor A3, the output end of the shoulder joint motor A3 is fixedly connected to the connecting piece A24, the connecting piece A24 is fixedly connected to the shell of the shoulder joint motor A4, the output end of the shoulder joint motor A4 is connected to the upper arm structure, and the shoulder joint motor A4 is used to drive the upper arm structure to rotate; the top of the shell of the shoulder joint motor A4 is connected to a shoulder posture sensor A6 through a first IMU fixed support A5, and the shoulder posture sensor A6 adopts an IMU 9-axis posture sensor.

[0042] like Figure 1-4 As shown, the upper arm structure includes a connecting plate A26, a double earring support A25, a double earring A7, an upper arm support A9, a pneumatic tendon A8, a pneumatic tendon A10, an elbow flange A11, an elbow flange A12 and an elbow posture sensor A13, the output end of the shoulder joint motor A4 is connected to the top of the upper arm support A9 through the connecting plate A26, the bottom end of the upper arm support A9 is rotatably connected to the elbow flange A11, and the bottom of the elbow flange A11 is fixedly connected to the elbow flange A12. Fixed connection, the side of the elbow flange A12 is fixedly connected with an elbow posture sensor A13, one end of the pneumatic tendon A8 is rotationally connected to the top of one side of the upper arm support A9, and the other end is rotationally connected to one side of the elbow flange A11, one end of the pneumatic tendon A10 is rotationally connected to the top of the other side of the upper arm support A9, and the other end is connected to a rope, a pulley is connected to the bottom of the other side of the upper arm support A9, and the rope A27 is wound around the pulley and connected to the other side of the elbow flange A11.

[0043] like Figure 1-4As shown, in the upper arm structure, the side of the elbow flange A12 is connected to the elbow posture sensor A13 through the second IMU fixed support, and the elbow posture sensor A13 adopts the IMU 9-axis posture sensor; the upper arm support A9 and the elbow flange A11 are both connected to the double earring support A25, the double earring A7 connected to one end of the pneumatic tendon A8 is rotatably connected to the double earring support A25 at the top of one side of the upper arm support A9, and the double earring A7 connected to the other end is rotatably connected to the double earring support A25 on the elbow flange A11, the double earring A7 at one end of the pneumatic tendon A10 is rotatably connected to the double earring support A25 at the top of the other side of the upper arm support A9, and the bottom of the other side of the upper arm support A9 is connected to a pulley A28 through a pulley support A29, and the rope A27 is wound around the pulley A28 and connected to the rope rack on the other side of the elbow flange A11.

[0044] like Figure 1-4 As shown, in the upper arm structure, there are two pneumatic tendons A8, specifically the pneumatic tendons DMSP-20-200N-RM-CM, and the two pneumatic tendons A8 are arranged side by side. There is one pneumatic tendon A10, specifically the pneumatic tendon DMSP-20-150N-RM-CM.

[0045] like Figure 1-4 As shown, the forearm structure includes an elbow joint motor A14, an elbow flange plate A15, a pneumatic tendon plate A17, a wrist flange plate A19, a wrist posture sensor A21 and a support rod A18. The elbow flange plate A12 is connected to the housing of the elbow joint motor A14, and the output end of the elbow joint motor A14 is connected to the elbow flange plate A15. The elbow joint motor A14 is used to drive the elbow flange plate A15 to rotate, thereby driving the pneumatic tendon plate A17 and the wrist flange plate A19. A19, wrist posture sensor A21 and support rod A18 rotate, elbow flange three A15 is rotationally connected to one end of pneumatic tendon three A17, the other end of pneumatic tendon three A17 is rotationally connected to wrist flange A19, wrist flange A19 is connected with wrist posture sensor A21, one end of support rod A18 is connected to elbow flange three A15 through rigid coupling A22, the other end of support rod A18 is rotationally connected to wrist flange A19 through universal joint A23.

[0046] like Figure 1-4As shown, in the forearm structure, the elbow flange A15 is rotatably connected to one end of the pneumatic muscle A17 through a joint bearing A16, and the other end of the pneumatic muscle A17 is rotatably connected to the wrist flange A19 through a joint bearing A16. The wrist flange A19 is connected to a wrist sensor fixing plate A20, and the wrist sensor fixing plate A20 is connected to a wrist posture sensor A21. The wrist posture sensor A21 adopts an IMU 9-axis posture sensor.

[0047] like Figure 1-4 As shown, in the forearm structure, there are four pneumatic muscles A17, which are located around the outer side of the support rod A18. Specifically, the pneumatic muscles are DMSP-10-100N-RM-CM.

[0048] The modular gas-electric hybrid bionic mechanical arm A of this embodiment also includes: Figure 7 The control cabinet shown in the figure is provided with a motion control system, which includes a controller 1 (Beckhoff CX5130), a current module 2 (current module EL4024), a power supply (including a 24V DC power supply 3 and a 48V DC power supply 4), a socket 5, an air switch 6, a terminal block 7, a fixed base plate 8 and 7 proportional pressure valves 9. The controller 1 (Beckhoff CX5130), the current module 2 (current module EL4024), the power supply (including a 24V DC power supply 3 and a 48V DC power supply 4), the socket 5, the air switch 6, the terminal block 7 and the 7 proportional pressure valves are all arranged on the fixed base plate 8. The circuit breaker 6 is connected to the power supply, and the power supply is respectively connected to the controller 1, the current module 2, the pressure valve 9, and the shoulder joint motor A1, the shoulder joint motor A3, and the shoulder joint motor A4 in the robot body, so as to supply power to all the electrical equipment in the modular gas-electric hybrid bionic robot A. The controller 1 is respectively connected to the current module 2 and the shoulder joint motor A1, the shoulder joint motor A3, and the shoulder joint motor A4 in the modular gas-electric hybrid bionic robot A. The current module 2 is connected to all the proportional pressure valves 9, and the two pneumatic tendons A8, the one pneumatic tendon A10, and the four pneumatic tendons A17 in the modular gas-electric hybrid bionic robot A are respectively connected to the proportional pressure valve 9. The power supply in this embodiment includes a 24V DC power supply 3 and a 48V DC power supply 4. The two types of power supplies supply power to all the equipment. The terminal block is responsible for dividing the power supply into multiple paths. The controller 1 is connected to the current module 2 and each joint motor, the current module is connected to all the proportional pressure valves 9, and the proportional pressure valves 9 are connected to the pneumatic tendons.

[0049] This embodiment also provides a modular gas-electric hybrid bionic robot, such as Figure 6As shown, it specifically includes two modular gas-electric hybrid bionic manipulators A, a manipulator (not shown in the figure), a head B, a fuselage fixing frame C and a host computer (TwinCAT3 host computer). The forearm structure end of the modular gas-electric hybrid bionic manipulator A is connected to the manipulator (i.e., the end effector), the head B is connected to the fuselage fixing frame C, the fuselage fixing frame C is connected to the shell of the shoulder joint motor A1 in the modular gas-electric hybrid bionic manipulator A, a deep learning camera is provided on the head B, a laser radar is provided on the fuselage fixing frame C, and the deep learning camera and the laser radar are both connected to the power supply. The deep learning camera and the laser radar are both connected to the host computer, and the controller 1 in the two modular gas-electric hybrid bionic manipulators A is also connected to the host computer. The relevant data is read by the deep learning camera and sent to the host computer, the host computer processes the data, and makes instructions through the controller 1 to realize the completion of the corresponding goal of the manipulator.

[0050] The pneumatic muscle of this embodiment is a pressure-controlled actuator, and its length shortens as the pressure increases.

[0051] The modular gas-electric hybrid bionic robot of this embodiment can be provided with two modular gas-electric hybrid bionic mechanical arms A, and the two modular gas-electric hybrid bionic mechanical arms A are respectively connected to the two sides of the fuselage fixing frame C, and can respectively simulate the upper limbs of the experimenter.

[0052] When performing elbow flexion movements, Figure 5 As shown, the TwinCAT3 host computer sends an angle signal through the controller 1, and controls the pressure change and flow rate change of the proportional pressure valve 9 output gas through the current module EL4024. The proportional pressure valve 9 can also transmit the real-time pressure value to the controller 1. At this time, the two pneumatic tendons DMSP-20-200N-RM-CM (pneumatic tendon A8) contract at the same time, and the pneumatic tendon DMSP-20-150N-RM-CM (pneumatic tendon A10) extends. The two form an antagonistic effect to achieve elbow bending. The proportional pressure valve 9 provides air pressure through the air source and is used in conjunction with the air source device, such as an air pump, compressed air, etc. The air source device provides pressure to the pneumatic tendon through the proportional pressure valve 9.

[0053] When the posture sensor of this embodiment detects that it is close to the target angle, the target value of the system is set to the required flow rate and pressure by adjusting the proportional coefficient, integral coefficient and differential coefficient of the PID control algorithm in the controller 1. The PID control algorithm intelligently adjusts the opening of the pressure valve according to the feedback signal of the posture sensor to achieve stable control, so that the actual flow rate and pressure of the system gradually approach the target value, and the controller 1 automatically issues a stop command. When the extension command is executed, the pneumatic tendon A8 stretches (the air pressure value decreases), and the pneumatic tendon A10 contracts (the air pressure value increases). The PID control algorithm is adjusted through the feedback signal. When the posture sensor detects that the set angle is reached, the controller 1 automatically issues a stop command.

[0054] The shoulder joint motor 1 A1, shoulder joint motor 2 A3 and shoulder joint motor 3 A4 in the shoulder module of the modular gas-electric hybrid bionic robotic arm A of this embodiment can respectively realize the abduction / internal rotation, extension / extension and rotation (pronation / supination) of some structures in the shoulder module, forearm structure and upper arm structure, etc., all of which are transmitted by the controller 1 through CAN communication, and the shoulder posture sensor A6 reads the angular velocity and position of the joint.

[0055] The pneumatic muscles in the forearm structure are similar. The diagonal pneumatic muscles A17 can form an antagonistic effect, and the adjacent pneumatic muscles A17 form a synergistic effect. The two degrees of freedom are integrated together to form a coupling phenomenon, which reduces the size while realizing the rotation of the wrist flange A19 in various directions, thereby realizing the rotation of the manipulator in various directions.

[0056] Motor part control: The elbow joint motor A14 adopts CAN communication control, which can control the speed, angle, acceleration, relative angle, absolute angle, etc., and drive the pronation / supination of the wrist in the forearm structure. In addition, the abduction / internal rotation of the partial structure, forearm structure and upper arm structure in the shoulder module is driven by the shoulder joint motor 1 A1 as the driving source, the extension / extension of the partial structure, forearm structure and upper arm structure in the shoulder module is driven by the shoulder joint motor 2 A3 as the driving source, and the rotation (pronation / supination) of the forearm structure and upper arm structure is driven by the shoulder joint motor 3 A4 as the driving source.

[0057] Objective: Use controller 1 to control the four joint motors of the robot arm through CAN communication so that the joints of the robot arm move at predetermined angles.

[0058] Specific control methods:

[0059] (1) Hardware connection and configuration:

[0060] Connect the 4 joint motors to the servo drive and ensure that the power supply and CAN communication lines are connected correctly.

[0061] Open the configuration software of controller 1 (e.g. TwinCAT).

[0062] (2) Controller 1 configuration:

[0063] In the configuration software, select the appropriate CAN communication interface module and assign the node address of each module.

[0064] (3) Encoder configuration:

[0065] If the joint motor uses an encoder to feedback position information, configure the encoder parameters so that controller 1 (Beckhoff controller) can accurately read the motor position.

[0066] (4) Write motion control program:

[0067] Use Beckhoff programming environment (such as TwinCAT PLC programming environment) to write motion control logic.

[0068] Define the target angle and motion speed as input parameters for control.

[0069] (5) CAN communication protocol definition:

[0070] Define the CAN communication protocol, for example, define a CAN frame as a data structure for transmitting joint target angles and velocities.

[0071] (6) Generate CAN message:

[0072] In the motion control program, the corresponding CAN message is generated and the data of the joint target angle and speed are filled into the CAN message.

[0073] Send this CAN message to the corresponding CAN communication interface module.

[0074] (7) Driver parsing message:

[0075] The servo drivers of the joint motors receive CAN messages via the CAN bus.

[0076] The driver parses the CAN message and extracts the target angle and speed information.

[0077] (8) Control motor movement:

[0078] The servo driver controls the joint motor to start moving based on the analyzed target angle and speed information.

[0079] (9) Feedback and adjustment:

[0080] The actual angle of the motor can be obtained in real time through feedback devices such as encoders.

[0081] The error between the target angle and the actual angle is calculated in controller 1, and then PID control is performed to generate corresponding control commands for adjustment.

[0082] (10) Exception handling and security:

[0083] Write code to handle abnormal situations, such as motor overload, communication loss, etc.

[0084] Set up safety control logic to ensure emergency stop of motor movement when necessary to ensure safe operation.

[0085] (11) Testing and debugging:

[0086] Load the program in controller 1 and control the joint motor through CAN communication to perform motion control test.

[0087] Gradually adjust the PID parameters and motion control logic to ensure that the robot arm moves at the expected angle, and make necessary optimizations and adjustments.

[0088] (12) System Integration:

[0089] Integrate the written program into the entire robotic arm control system to achieve coordinated movement with other joints and sensors.

[0090] The working principle of the bionic mechanical arm A in the modular gas-electric hybrid bionic robot of this embodiment is:

[0091] The motion parameters of the bionic robot arm A in this embodiment are provided by the electromyographic signal collector. Figure 8 As shown, the electromyographic signal is collected by the MYO electromyographic signal collector through the real-time movement of the experimenter's upper limb arm, and transmitted to the host computer through the data line. The host computer processes the signal collected by MYO to obtain the spatial position information and velocity acceleration and other motion information of the corresponding bionic manipulator A in the modular gas-electric hybrid bionic robot. The motion trajectory of the obtained position and motion information is modeled by the path planning algorithm RRT, and the obtained results are fitted to obtain the path planning function. The path planning function is fed back to the controller 1 on the control cabinet of the bionic manipulator A. The controller 1 controls each joint motor and each pneumatic tendon in the bionic manipulator A to start moving according to the path planning function. Figure 8 The information monitoring system in the robot consists of a deep learning camera on the robot's head B, multiple IMU 9-axis position sensors fixed on the bionic robot arm A, and an exciting radar installed on the fuselage fixing frame C. Its structure is as follows Figure 6As shown, the deep learning camera collects information and then obtains the obstacle contour (target contour) on the path passed by the bionic manipulator A during the movement. The deep learning camera and the laser radar installed on the fuselage fixing frame C can work together to send signals to the controller 1. The controller 1 integrates and processes the information obtained by the deep learning camera and the laser radar installed on the fuselage fixing frame C to obtain the obstacle contour, the distance between the manipulator body and the obstacle, and other information to determine whether the bionic manipulator A will encounter an obstacle during the movement. The multiple IMU 9-axis posture sensors fixed on the bionic manipulator A not only obtain the motion posture, speed, acceleration, and time information of the bionic manipulator A in Cartesian space in real time, but also obtain the angle, angular velocity, and angular acceleration information of the bionic manipulator A in the joint space, so that the manipulator moves at a stable speed and acceleration, and runs or grabs the target object with the correct posture. The controller 1 also includes an iterative learning control module and a path control module. The iterative learning control module is the core part of the entire control system. The iterative learning control module has been learned through a large amount of manually collected data (database) before. The information of the joint position and Cartesian space posture of the bionic robot arm A obtained by the IMU 9-axis posture sensor is input into the iterative learning control module in the controller 1 to obtain the air pressure required for the pneumatic tendon (before the iterative learning process, the air pressure and the state of the end trajectory of the pneumatic tendon are obtained and input into the iterative loop control network (iterative learning control module). When the network training is completed, only the position needs to be input, and the controller 1 will determine the air pressure that needs to be input for control). After the speed, acceleration and time information of the bionic robot arm A in the joint space are input into the iterative learning control module in the controller 1, the speed of air pressure change can be obtained. The distance information obtained by the deep learning camera and lidar is sent to the controller 1. The position information can be obtained through the iterative learning control module in the controller 1. The target air pressure value, the speed of air pressure change and the position information are fed back to the path control module. The path control module analyzes the obtained data and changes the path trajectory function in real time (changes the size of the pneumatic tendon input air pressure and the speed of the input air pressure to change the position and speed of the terminal trajectory. After iterating to the next position, the distance to the obstacle is judged, and the pressure and speed to be input are re-determined). This cycle ensures that the bionic robot arm A completes the task in the correct posture without hitting any obstacles.

[0092] The above-mentioned path planning algorithm RRT is used to model the motion trajectory of the obtained position and motion information as follows:

[0093] The implementation process of RRT:

[0094] 1. Initialization phase:

[0095] Set up the basic elements for the algorithm:

[0096] Start point and target point: Define the start point and target point in the joint space of the bionic robot A. The start point is the current joint angle configuration of the bionic robot A, and the target point is the target position that the end effector needs to reach.

[0097] Workspace and obstacle information: defines the workspace of the bionic robot A and contains models or definitions of obstacles. Obstacles will be used for collision detection to prevent the bionic robot A from colliding with them.

[0098] Initialize the search tree: The starting point is used as the root node of the search tree, and the tree structure is initialized to record the motion path of the bionic robot arm A.

[0099] 2. Sampling stage:

[0100] The algorithm randomly generates a new sampling point from the joint space or Cartesian space of the bionic robot A. This sampling point can represent the state of the bionic robot A under a certain joint angle configuration or the position of the end effector.

[0101] 3. Find the nearest node:

[0102] After getting a new random sampling point, RRT finds the node closest to the sampling point from the existing search tree. The nearest neighbor search is calculated using Euclidean distance or other metrics.

[0103] 4. Path extension:

[0104] Expanding from the nearest node to the sampling point, the algorithm generates a new feasible node from the nearest node to the sampling point under the motion constraints of the bionic robot A. The motion of the robot A follows its physical motion model and joint limits.

[0105] The extended path is then collision checked to ensure that the path segment does not collide with obstacles in the environment. If the extended path is safe, the new feasible node is added to the tree.

[0106] 5. Path optimization:

[0107] The path is continuously optimized to make the generated path as close to the global optimum as possible. Whenever a new node is added to the tree, the algorithm searches for other existing nodes in its local area and reconnects these nodes to optimize the path length or path cost.

[0108] The algorithm checks all nodes around the new feasible node to see if it can reconnect them to make the path from the new feasible node shorter or better. If it finds a better connection than the new feasible node, it reconnects the nodes to the tree.

[0109] 6. Iteratively build the tree:

[0110] RRT will repeat the steps of random sampling, path expansion and path optimization, and continue to expand to the target area. With the increase of sampling points, the tree will gradually fill the feasible workspace of bionic robot A and explore more and more feasible paths.

[0111] 7. Convergence and path extraction:

[0112] When a node in the tree successfully reaches the target point, RRT will trace back from the target point to the starting point and extract the path. This path has been optimized many times and is close to the optimal solution. In addition, the algorithm does not stop immediately, but continues to optimize, further shortening the path or reducing the motion cost until the predetermined termination conditions (such as the number of iterations, time limit, etc.) are met.

[0113] 8. Post-processing:

[0114] RRT can generate a feasible and optimal path, but in practical applications, the path often needs to be further smoothed. In order to improve the motion efficiency of robot A, the path is optimized by using curve interpolation and removing redundant nodes to make it more in line with the actual execution requirements of bionic robot A.

[0115] The protection scope of the present invention includes but is not limited to the above embodiments. The protection scope of the present invention shall be based on the claims. Any replacement, deformation, and improvement of the technology that can be easily thought of by technicians in this field shall fall within the protection scope of the present invention.

Claims

1. A modular gas-electric hybrid bionic robotic arm, characterized in that: The robot comprises a mechanical arm body, wherein the mechanical arm body comprises a shoulder module, an upper arm structure and a forearm structure connected in sequence; The shoulder module comprises a shoulder joint motor 1 (A1), a right-angle connecting frame (A2), a shoulder joint motor 2 (A3) and a shoulder joint motor 3 (A4); the shell of the shoulder joint motor 1 (A1) is used to be connected to the fuselage fixing frame (C) of the bionic robot; the output end of the shoulder joint motor 1 (A1) is connected to one end of the right-angle connecting frame (A2); the other end of the right-angle connecting frame (A2) is connected to the shell of the shoulder joint motor 2 (A3); the output end of the shoulder joint motor 2 (A3) is connected to a connecting piece (A24); the connecting piece (A24) is connected to the shell of the shoulder joint motor 3 (A4); and the output end of the shoulder joint motor 3 (A4) is connected to the upper arm structure; The upper arm structure comprises an upper arm support (A9), a pneumatic muscle 1 (A8), a pneumatic muscle 2 (A10), an elbow flange 1 (A11), an elbow flange 2 (A12) and an elbow posture sensor (A13); the output end of the shoulder joint motor 3 (A4) is connected to the top of the upper arm support (A9); the bottom end of the upper arm support (A9) is rotatably connected to the elbow flange 1 (A11); the elbow flange 1 (A11) is connected to the elbow flange 2 (A12); the side of the elbow flange 2 (A12) is connected to the elbow posture sensor (A13). A posture sensor (A13), wherein one end of the pneumatic tendon 1 (A8) is rotationally connected to the top of one side of the upper arm support (A9), and the other end is rotationally connected to one side of the elbow flange 1 (A11); one end of the pneumatic tendon 2 (A10) is rotationally connected to the top of the other side of the upper arm support (A9), and the other end is connected to a rope (A27); a pulley (A28) is connected to the bottom of the other side of the upper arm support (A9); the rope (A27) is wound around the pulley (A28) and connected to the other side of the elbow flange 1 (A11); The forearm structure comprises an elbow joint motor (A14), an elbow flange plate three (A15), a pneumatic tendon plate three (A17), a wrist flange plate (A19), a wrist posture sensor (A21) and a support rod (A18); the elbow flange plate two (A12) is connected to the outer shell of the elbow joint motor (A14); the output end of the elbow joint motor (A14) is connected to the elbow flange plate three (A15); the elbow flange plate three (A15) is rotationally connected to one end of the pneumatic tendon plate three (A17); the other end of the pneumatic tendon plate three (A17) is rotationally connected to the wrist flange plate (A19); the wrist flange plate (A19) is connected to the wrist posture sensor (A21); one end of the support rod (A18) is connected to the elbow flange plate three (A15); and the other end of the support rod (A18) is connected to the wrist flange plate (A19) via a universal joint (A23).

2. The modular gas-electric hybrid bionic robotic arm according to claim 1, characterized in that: In the shoulder module, the top of the shell of the shoulder joint motor three (A4) is connected to a shoulder posture sensor (A6) through a first IMU fixed support (A5), and the shoulder posture sensor (A6) adopts an IMU 9-axis posture sensor.

3. The modular gas-electric hybrid bionic robotic arm according to claim 1, characterized in that: In the upper arm structure, the elbow posture sensor (A13) adopts an IMU 9-axis posture sensor, and the side of the elbow flange plate 2 (A12) is connected to the elbow posture sensor (A13) through a second IMU fixed support; the upper arm support (A9) and the elbow flange plate 1 (A11) are both connected to a double earring support (A25), the double earring (A7) at one end of the pneumatic tendon 1 (A8) is rotatably connected to the double earring support (A25) at the top of one side of the upper arm support (A9), and the double earring (A7) at the other end is connected to the elbow flange plate The double earring (A25) on the first (A11) is rotatably connected, the double earring (A7) at one end of the pneumatic tendon second (A10) is rotatably connected to the double earring support (A25) at the top of the other side of the upper arm support (A9), the bottom of the other side of the upper arm support (A9) is connected to a pulley (A28) through a pulley support (A29), and the rope (A27) is wound around the pulley (A28) and connected to the rope rack on the other side of the elbow flange (A11).

4. The modular gas-electric hybrid bionic robotic arm according to claim 1, characterized in that: There are two pneumatic tendons 1 (A8), which are arranged side by side, and there is one pneumatic tendon 2 (A10).

5. The modular gas-electric hybrid bionic robotic arm according to claim 1, characterized in that: In the forearm structure, the elbow flange plate three (A15) is rotatably connected to one end of the pneumatic tendon three (A17) via a joint bearing (A16), and the other end of the pneumatic tendon three (A17) is rotatably connected to the wrist flange plate (A19) via a joint bearing (A16), and the wrist flange plate (A19) is connected to a wrist sensor fixing plate (A20), and the wrist sensor fixing plate (A20) is connected to a wrist posture sensor (A21).

6. The modular gas-electric hybrid bionic robotic arm according to claim 5, characterized in that: There are four pneumatic muscles three (A17) and they are located around the support rod (A18). One end of the support rod (A18) is connected to the elbow flange three (A15) through a coupling.

7. The modular gas-electric hybrid bionic robotic arm according to claim 1, characterized in that: The invention also includes a control cabinet, on which a motion control system is arranged. The motion control system includes a controller (1), a current module (2), a power supply and a plurality of proportional pressure valves (9). The power supply is respectively connected to the controller (1), the current module (2), the pressure valves and the shoulder joint motor 1 (A1), the shoulder joint motor 2 (A3) and the shoulder joint motor 3 (A4) in the robot body. The controller (1) is respectively connected to the current module (2) and the shoulder joint motor 1 (A1), the shoulder joint motor 2 (A3) and the shoulder joint motor 3 (A4) in the robot body. The current module (2) is connected to all the proportional pressure valves (9). The pneumatic tendon 1 (A8), the pneumatic tendon 2 (A10) and the pneumatic tendon 3 (A17) in the robot body are respectively connected to the proportional pressure valves (9).

8. A modular gas-electric hybrid bionic robot, characterized in that: The modular gas-electric hybrid bionic robot arm comprises a robot, a head (B) and a body bracket (C) as described in any one of claims 1 to 7, wherein the end of the forearm structure of the modular gas-electric hybrid bionic robot arm is connected to the robot, the head (B) is connected to the body bracket (C), the body bracket (C) is connected to the outer shell of the shoulder joint motor 1 (A1) in the modular gas-electric hybrid bionic robot arm, a deep learning camera is arranged on the head (B), a laser radar is arranged on the body bracket (C), and the deep learning camera and the laser radar are both connected to a power supply.

Images